In vitro investigations on uptake and

Transcription

1 In vitro investigations on uptake and toxicity of cyanobacterial toxins Dissertation Zur Erlangung des akademischen Grades des Doktors der Naturwissenschaften an der Universität Konstanz Fachbereich Biologie vorgelegt von Andreas Fischer Tag der mündlichen Prüfung: Referent: Prof. Dr. Daniel Dietrich Referent: Prof. Dr. Karl-Otto Rothhaupt Referent: Prof. Dr. Christof Hauck

2 2

3 Publications & Presentations Peer reviewed articles Fischer, A., Hoeger, S. J., Feurstein, D., Stemmer, K., Knobeloch, D., Nussler, A., and Dietrich, D. (2010). The role of organic anion transporting polypeptides (OATPs/SLCOs) in the toxicity of different microcystin congeners in vitro: A comparison of primary human hepatocytes and OATP-transfected HEK293 cells. Toxicol Appl Pharmacol, in press. Fischer, A. and Dietrich, D. (in preparation). Inhibitory capacity of Adda on protein phosphatase 1 and 2A. Fischer, A., Feurstein, D., Knobeloch, D., Nussler, A., and Dietrich, D. (in preparation). In vitro toxicity of cylindrospermopsin on primary human hepatocytes and OATP-expressing HEK293 cells. Fischer, A., Hoeger, S. J., Fastner, J., Robertson, A., and Dietrich, D. (in preparation). Detection of microcystins and β-n-methylamino-l-alanine in bluegreen algae supplements. Feurstein, D., Holst, K., Fischer, A., and Dietrich, D. R. (2009). Oatp-associated uptake and toxicity of microcystins in primary murine whole brain cells. Toxicol Appl Pharmacol 234, Book section / Conference proceedings Dietrich, D. R., Fischer, A., Michel, C., and Hoeger, S. J. (2008). Toxin mixture in cyanobacterial blooms - a critical comparison of reality with current procedures employed in human health risk assessment. In Proceedings of the Interagency, International Symposium on Cyanobacterial Harmful Algal Blooms. (K. H. Hudnell, Ed.), pp Advances in Experimental Medicine & Biology. Abstracts & Poster presentations Fischer, A., Hoeger, S. J., Feurstein, D. J., Ernst, B., and Dietrich, D. R. (2007). Importance of organic anion transporting polypeptides (OATPs) for the toxicity of single microcystin congeners in vitro. 7th International Conference on Toxic Cyanobacteria, Rio das Pedras, Rio de Janeiro State, Brazil. Feurstein, D., Fischer, A., and Dietrich, D. (2007). In vitro toxicity of microcystins in primary murine whole brain and neuronal cultures. 7th International Conference on Toxic Cyanobacteria, Rio das Pedras, Rio de Janeiro State, Brazil. Feurstein, D., Fischer, A., and Dietrich, D. (2008). Microcystin congener-specific in vitro neurotoxicity. Toxicology Letters 180,

4 Contents 1 General Introduction Cyanobacteria Systematics Ecology, Morphology and Physiology Toxic Cyanobacterial Blooms and Monitoring Cyanobacterial Toxins Oligopeptides Microcystins Nodularins Alkaloids Cylindrospermopsins Saxitoxins Anatoxins Other Cyanobacterial Toxins β-n-methylamino-l-alanine Lipopolysaccharides Objectives Inhibitory capacity of Adda on protein phosphatase 1 and 2A Abstract Introduction Material & Methods Results Discussion The role of organic anion transporting polypeptides (OATPs/SLCOs) in the toxicity of different microcystin congeners in vitro: A comparison of primary human hepatocytes and OATP-transfected HEK293 cells Abstract Introduction Material & Methods Results Discussion Supplemental information

5 5 In vitro toxicity of cylindrospermopsin in primary human hepatocytes and OATPexpressing HEK293 cells Abstract Introduction Material & Methods Results Discussion Detection of microcystins and β-n-methylamino-l-alanine in blue-green algae supplements Abstract Introduction Material & Methods Results Discussion General Discussion Summary Zusammenfassung Abbreviations References Apendix

6 Chapter I General Introduction 1 General Introduction 1.1 Cyanobacteria Systematics The phylum Cyanobacteria belongs to the superkingdom Eubacteria and is systematically classified into the non-filamentous orders Chroococcales and Pleurocapsales and the filamentous orders Oscillatoriales, Nostocales and Stigonematales (Castenholz and Waterbury, 1989; van den Hoek et al., 1993). A newer classification according to the National Center for Biotechnology Information (NCBI, Taxonomy Browser: Cyanobacteria, retrieved on March 17 th, 2009) includes two additional orders: Gloeobacterales and Prochlorales Ecology, Morphology and Physiology The gram-negative cynaobacteria constitute the most diverse and widespread of the phototrophic prokaryotes (Skulberg et al., 1993; Codd, 1995). They represent a considerable proportion of the marine phytoplankton and play a crucial role in photosynthetic primary production and nitrogen fixation (van den Hoek et al., 1993; Paerl, 2000). Cyanobacteria occur worldwide in nearly any given habitat. They can be found in wetlands and arid deserts, in hot springs and glaciers, even in arctic ponds and ice. However, the majority inhabits salt-, brakish- and in particular freshwater. In addition, they often represent the pioneer organisms that colonize bare areas of rock and soil (van den Hoek et al., 1993; Mur et al., 1999; Hitzfeld et al., 2000; Oliver and Ganf, 2000; Oren, 2000; Pentecost and Whitton, 2000; Vincent, 2000; Ward and Castenholz, 2000; Wynn-Williams, 2000). Their basic algal-like morphology is as diverse as their habitats: It comprises unicellular, pseudoparenchymatic, colony forming and filamentous forms with branched and unbranched trichomes (Skulberg et al., 1993; Mur et al., 1999; Whitton and Potts, 2000; Graham et al., 2009). 6

7 Chapter I General Introduction Cyanobacteria are considered not only one of the oldest life forms on earth, as fossils were dated back to 3.3 to 3.5 billion years ago (Schopf and Packer, 1987), but also the very first oxygenic photosynthesizers. This ability led to the formation of an oxygenic atmosphere, hence, paving the way for obligat aerobic prokaryotes and especially for eukaryotes (van den Hoek et al., 1993; Graham et al., 2009). Unlike eukaryotic plants, cyanobacterial thylakoids are freely located in the cytoplasm arranged concentrically and equidistantly near the cell periphery and are typically not stacked (van den Hoek et al., 1993; Mur et al., 1999; Graham et al., 2009). The accessory pigments are imbedded in phycobilisomes on the surface of the thylakoids (van den Hoek et al., 1993; Mur et al., 1999; Graham et al., 2009). In addition to chlorophyll a and carotenoids they take advantage from the use of further accessory pigments, especially phycoerythrin, phycocyanin and allophycocyanin to perform oxygenic photosynthesis. These phycobiliproteins absorp light in the range of 400 and 700 nm, thus, include the green light ( nm) that is inaccessable for green algae. Hence, they are able to live in the shadow of other phytoplankton and/or under limited light conditions (van den Hoek et al., 1993; Mur et al., 1999). Moreover, cyanobacteria are capable of altering the constituency of the phycobiliproteins and the size of the light harvesting antennae in dependence to light quality and other environmental influences. This chromatic adaptation enables them to absorp light even more efficiently (Mur et al., 1999; Oliver and Ganf, 2000; Graham et al., 2009). In order to avoid light inhibition or competitive nutrient limitation many cyanobacterial species are capable of regulating buoyancy by generation and degeneration of gas vesicles, cytoplasmatic cylindrical inclusions whose protein walls are only permeable to gases, or by accumulation of assimilation products that cause an increase in density and / or collapse of those gas vesicles as a result of invreased turgor pressure (van den Hoek et al., 1993; Oliver and Ganf, 2000; Graham et al., 2009). In stratified, non-circulating water bodies buoyancy regulation allows planktonic species to vertically adjust their position and move within the water column. This in turn enables them to exploit less competitive and nutrient rich niches like the metalimnion, where nutrients often accumulate due to the sudden change in density (Mur et al., 1999; Oliver and Ganf). A 7

8 Chapter I General Introduction further prerequisite for the formation of a metalimnetic population is a euphotic metalimnion (Lampert and Sommer, 1999; Oliver and Ganf, 2000). Moreover, these nutrients, as well as metabolites may be very effectively stored as e.g. cyanophycean starch, lipid globules, cyanophycin granules, polyphosphate bodies, carboxysomes, etc., enabling cyanobacteria to outlast temporary nutritional poverty (van den Hoek et al., 1993; Oliver and Ganf, 2000; Graham et al., 2009). Under adverse conditions, e.g. at the end of the growing season, vegetative cells of filamentous cyanobacteria with the exception of species in the order Oscillatoriales may be differentiated into akinetes, thickwalled resting cells containing storage granules. These cells are able to outlast years of dormancy (van den Hoek et al., 1993; Mur et al., 1999; Graham et al., 2009). In fact, Aphanizomenon flos-aquae and species of the genus Anabaena may stay viable in anoxic sediments for up to 18 and 64 years, respectively (van den Hoek et al., 1993). Most akinete forming species are also capable of producing heterocysts, cells specialized in nitrogen fixation that posess likewise thick walls and a hyaline protoplast. Heterocysts completely lack the oxygen-generating photosystem II and deplete diffusing oxygen by enhanced respiration in order to protect the contained nitrogenase, the highly oxygen-susceptible enzyme system that reduces atmospheric dinitrogen to ammonium ion (van den Hoek et al., 1993; Mur et al., 1999; Oliver and Ganf, 2000; Graham et al., 2009) Toxic Cyanobacterial Blooms and Monitoring As a result of their high diversity, adaptability and specialization cyanobacteria may gain competitive advantage over other photoautotrophic organisms. Thus, under favored conditions cyanobacteria often become the dominant phytoplankton of surface waters and form blooms, mass developments of one or several cyanobacterial species capable of buoyancy regulation. Blooms occur especially in eutrophic waterbodies and are an increasingly observed phenomenon due to anthropogenic nutrient input like sewage water, phosphate containing detergents and fertilizers (Bartram et al., 1999; Mur et al., 1999; Oliver and Ganf, 2000). However, mass developments may also occur in mesoand oligotrophic waterbodies: e.g. Planktothrix rubescens forms metalimnetic blooms in moderately nutrient-rich lakes (Lampert and Sommer, 1999; Mur et 8

9 Chapter I General Introduction al., 1999; Chorus, 2001). In addition, although poorly understood climatic changes, especially global warming, are considered to favour occurence, persistence and distribution of cyanobacterial blooms (Paul, 2008). Besides aesthetic nuisance caused by surface scum, discolouration of the water, bad taste and odour blooms may pose a serious threat for human and animal health as they may contain highly potent toxins (Bartram et al., 1999; Kuiper-Goodman et al., 1999). Amongst a total of over 150 cyanobacterial genera including approximately 2,000 species about 40 toxigenic species are known (Skulberg et al., 1993). Although this represents a rather small proportion it was estimated that approximately 25-75% of bloom isolates are capable of producing toxins (Lawton and Codd, 1991; Chorus, 2001). This demonstrates the widespread distribution of toxigenic species and hence the importance to monitor surface waters where cyanobacterial blooms regularly occur, especially those used as a drinking water source or for recreational purposes. However, monitoring is hampered by the highly variable nature of blooms: A bloom may be dominated by a single species or be contemporaneously composed of several species of which some or the majority may be non-toxic. Even within a certain species there may be a variety of toxic and non-toxic strains at the same time (Vezie et al., 1998; Sivonen and Jones, 1999; Pereira et al.; Moreno et al., 2004; Molica et al., 2005). Thus, there is a high risk to overlook toxigenic species, especially when monitoring is based on species identification only. Naturally phytoplankton communities are never static in their composition. Instead they are usually characterized by a seasonal succession, i.e. the species composition is changing during the course of seasons (Lampert and Sommer, 1999). This applies for cyanobacteria as well: in a single waterbody blooms of different cyanobacterial species have been reported to occur. Usually those blooms are temporally seperated or in succession, but have occasionally been found to be overlapping, too (Carmichael et al., 2000; Pereira et al., 2000; Hoeger et al., 2004). Accordingly, seasonal changes in toxin production occur as well. However, size and density of a bloom do not necessarily correlate with the amount of toxin produced, thus, are not a reliable indicator for the presence 9

10 Chapter I General Introduction of toxins (Watanabe et al., 1992; Vezie et al., 1998; Sivonen and Jones, 1999; Hoeger et al., 2004; Dietrich et al., 2008). Besides seasonal changes in the abundance of cyanobacterial populations and toxins their spacial distribution may vary as it succumbs different influences like surface and underwater topography, stratification, currents and wind (Watanabe et al., 1992; Lampert and Sommer, 1999; Oliver and Ganf, 2000; Falconer, 2005a). Consequently, both temporal and spacial distribution of cyanobacterial populations and hence the toxins potentially produced have to be taken into account for the determination of monitoring sites and frequency. 1.2 Cyanobacterial Toxins Cyanobacteria produce a multitude of secondary metabolites, whose physiological functions and ecological regulations remained predominantly unknown to date. However, some of them revealed to be toxic towards aquatic and terrestrial organisms, especially mammals (Sivonen and Jones, 1999; Dow and Swoboda, 2000; Kaebernick and Neilan, 2001; Welker and von Dohren, 2006). Several different toxins and toxin congeners may be simultaneously produced by a single species or strain (Harada et al., 1991a; Sivonen et al., 1992; Park et al., 1993; Sivonen and Jones, 1999; Harada et al., 2001; Welker and von Dohren, 2006). In addition, cyanotoxin production appears to be variable and facultative, i.e. its pattern may alter qualitatively and quantitatively in a certain strain or species in response to several environmental and physiological factors (Sivonen et al., 1995; Rapala et al., 1997; Sivonen and Jones, 1999; Kaebernick and Neilan, 2001). Most cyanotoxins are intracellular toxins and only marginally excreted. Their predominant release in the environment naturally occurs during cell scenescence, death and lysis. Thus, as long as toxigenic blooms are healthy, extracellular toxin concentrations remain low until its decay or artificial lysis after application of algicides (e.g. copper sulphate) (Jones and Orr, 1994; Kuiper- Goodman et al., 1999; Sivonen and Jones, 1999). 10

11 Chapter I General Introduction Cyanotoxins are usually classified according to their chemical structure, toxicity or organ specifity. Since the latter two often appear to have quite some diversity depending on the respective toxin and the route of administration the following classification of the most important cyanotoxins uses a structural approach, whereas the main focus lies on those that were investigated in this study Oligopeptides Oligopeptides represent the major part of cyanobacterial secondary metabolites (Welker and von Dohren, 2006). In cyanobacterial blooms of fresh and brackish waters the cyclic microcystins and nodularins are globally the most frequently occuring cyanotoxins of this class and turned out to be the most toxic at the same time (Sivonen and Jones, 1999; Spoof, 2005). Further oligopeptides frequently produced by bloom-forming cyanobacteria include cyanopeptolins, anabaenopeptins, microviridins, microginins and aeroginosins. However, these inhibitors of serine proteases, serine/threoninespecific protein phosphatases and other enzymes revealed to be far less toxic (Kaya et al., 1996; Namikoshi and Rinehart, 1996; Sano et al., 2001; Hastie et al., 2005; Ersmark et al., 2008; Sedmak et al., 2008) Microcystins Microcystins (MCs) are cyclic heptapeptides that have first been isolated from their eponymous producer Microcystis aeruginosa (Bishop et al., 1959; Konst et al., 1965; Carmichael et al., 1988a). Further species of the genus Microcystis, as well as Anabaena, Planktothrix, Oscillatoria, Nostoc, Anabaenopsis, Radiocystis, Arthrospira and Hapalosiphon have been reported to produce MCs (Sivonen and Jones, 1999; Spoof, 2005). Structure The general structure of MCs (Fig. 1.1) is cyclo(-d-ala 1 -L-X 2 -D-erythro-βmethylAsp 3 -L-Z 4 -Adda 5 -D-Glu 6 -N-methyldehydro-Ala 7 ) in which Adda stands for the unique D-amino acid 3-amino-9-methoxy-2,6,8-trimethyl-10-phenyldeca-4,6- dienoic acid and X and Z for variable L-amino acid residues (Botes et al., 1984; Botes et al., 1985; Rinehart et al., 1988; Rinehart et al., 1994). Substitutions in 11

12 Chapter I General Introduction those two positions constitute the main structural variations and are therefore used for the nomenclature of MCs (e.g. MCLR is a microcystin congener with L- leucine and L-arginine in position 2 and 7, respectively) (Carmichael et al., 1988a). However, further variations (e.g. demethylation of D-erythro-βmethylaspartic acid (D-MeAsp) and N-methyldehydroalanine (Mdha)) may occur in any of the seven amino acids leading to more than 80 structural analogues with molecular weights ranging from 900 to 1100 Da (Sivonen and Jones, 1999; Spoof, 2005; Zurawell et al., 2005; Humpage, 2008). Fig. 1.1: General structure of microcystins. Synthesis MCs are synthesized non-ribosomally by peptide synthetases and polyketide synthases that are combined in a multi-enzyme complex. A single gene cluster with ten open reading frames encodes for modules (mcya - mcyj) which compose the enzymes for MC synthesis. Nine modules have a synthesizing function, whereas one (McyH), a transmembrane protein belonging to the ATPbinding cassette transporter family, is putatively responsible for toxin transport and/or localization (Moore et al., 1991; Rinehart et al., 1994; Tillett et al., 2000; Falconer, 2005a; Welker and von Dohren, 2006). However, McyH seemed to be additionally involved in the microcystin biosynthesis pathway, since its deletion led to a complete halt in MC production (Pearson et al., 2004). Moreover, a multiplex polymerase chain reaction has been developed that can be used to identify contamination with microcystin producing cyanobacteria in 12

13 Chapter I General Introduction cyanobacterial dietary supplements and possibly other food products by amplification of genes like mcya of the microcystin synthetase gene cluster (Saker et al., 2005; Saker et al., 2007). In general, several MC congeners are produced at the same time (Harada et al., 1991b; Sivonen et al., 1992; Luukkainen et al., 1993; Namikoshi et al., 1995; Edwards et al., 1996; Lawton et al., 1999; Spoof, 2005; Welker and von Dohren, 2006; Pegram et al., 2008). Degradation The cyclic structure of MCs appears to be extremely stable and insusceptible towards temperature, ph, chemical hydrolysis and oxidation, especially under natural environmental conditions (Harada et al., 1996a; Harada and Tsuji, 1998). Hence, MCs may persist and remain toxic in waterbodies for weeks until slow photochemical degradation which is significantly accelerated in the presence of pigments (Jones et al., 1994; Tsuji et al., 1995; Lahti et al., 1997; Sivonen and Jones, 1999). Besides, certain heterotrophic aquatic bacteria of different genera (e.g. Sphingomonas, Pseudomonas and Paucibacter) are capable of decomposing MCs after an initial lag phase of several days. Depending on different environmental factors and MC concentrations complete or major degradation has been shown to occur within 2-10 days (Rapala et al., 1994; Bourne et al., 1996; Takenaka and Watanabe, 1997; Park et al., 2001; Christoffersen et al., 2002; Ishii et al., 2004). Toxicity and Molecular Mode of Action The majority of cyanobacterial poisonings of animals and humans are attributed to MCs. In mammals they predominantly affect the liver and hence are generally referred to as hepatotoxins (Carmichael, 1997; Kuiper-Goodman et al., 1999; Codd et al., 2005). MC toxicity predominantly relies on the very potent inhibition (at nanomolar concentrations) of serine/threonine-specific protein phosphatases (PPs) PP1 and PP2A, as well as PP3 to PP6, whereas inhibition of PP2B, PP2C and PP7 revealed to be ineffective (MacKintosh, 1993; Honkanen et al., 1994; Runnegar et al., 1995a; Toivola et al., 1997; Hastie et al., 2005). The causality between phosphatase inhibition and in vivo toxicity was demonstrated in mice (i.p.) by concomitant measurement of liver PP1 and PP2A activity and examination of 13

14 Chapter I General Introduction the clinical symptoms. Phosphatase inhibition was dose-dependent and proportional to the severity of the liver demage (Runnegar et al., 1993). The interaction between MCs and phosphatases comprises a two-step mechanism in which the first step already mediates the inhibition: An initial noncovalent, hence, reversible binding is formed within minutes by the alignment of the Adda side chain into a hydrophobic groove adjacent to the phosphatase s catalytic site and the formation of a hydrogen bond between the carboxyl group of D-glutamic acid and the binuclear metal ion catalytic centre of the phosphatase (Goldberg et al., 1995; Craig et al., 1996). An additional ionic interaction occurs between the carboxyl group of D-erythro-β-methylaspartic acid and arginine 96 and tyrosine 134 of the phosphatase s catalytic subunit (PPc) (Bagu et al., 1997; Maynes et al., 2006). In the second step that lasts several hours the methyl group of N-methyldehydroalanine is linked covalently to cysteine of the catalytic subunit of the phosphatase (cysteine 273 of PP1c and cysteine 266 of PP2Ac), which renders the binding irreversible, however, does not increase the inhibitory activity (MacKintosh et al., 1995; Runnegar et al., 1995a; Craig et al., 1996; Bagu et al., 1997; Maynes et al., 2006). A few MC congeners, as well as the closely related nodularins (see and Fig. 1.2), in which N-methyldehydroalanine is substituted by N- methyldehydrobutyrine are unable to form this covalent linkage to the phosphatases (Bagu et al., 1997; Hastie et al., 2005). This does also apply for dihydromicrocystins, whose double bond of N-methyldehydroalanine is reduced (MacKintosh et al., 1995; Craig et al., 1996). However, these modifications do not (Sano et al., 2004) or only moderately (5- to 50-fold) decrease the inhibitory potential (MacKintosh et al., 1995; Hoeger et al., 2007) and hence toxicity as shown in mice (Rinehart et al., 1994; Sivonen and Jones, 1999). On the contrary, the Adda-glutamate moiety was found to be crucial for the inhibitory capacity: MCLR and MCRR inhibited PP2A 100-times stronger than their geometrical isomers, [6(Z)-Adda 5 ]MCLR and [6(Z)-Adda 5 ]MCRR (Nishiwaki- Matsushima et al., 1991). Indeed, any structural modifications of either Adda (e.g. isomerization of its diene from 6(E) to 6(Z)) or D-glutamate (e.g. acetylation or esterification) have been reported to dramatically decrease or abolish the toxicity of MCs in mice (Harada et al., 1990a; Harada et al., 1990b; Nishiwaki-Matsushima et al., 1991; Namikoshi et al., 1992; Stotts et al., 1993; 14

15 Chapter I General Introduction Rinehart et al., 1994; Harada, 1996b). On the other hand, isolated Adda neither elicited inhibitory action on PP1 even at 10 µm (see also chapter III) nor toxicity at concentrations up to 10 mg/kg body weight (mouse, i.p.), while MCLR caused typical concentration-response effects with an IC 50 (concentration that inhibits 50% of the enzyme s activity) of 2 nm demonstrating the relevance of the remaining structural units for the biological activity (Harada et al., 2004). In general, since the Adda-glutamate moiety is very conserved variation in toxicity is low amongst most MC congeners with LD 50 values (dose of toxin that kills 50% of the exposed animals) ranging from µg/kg body weight (mouse, i.p.) (Rinehart et al., 1994; Sivonen and Jones, 1999). An exception is MCRR whose LD 50 (600 µg/kg body weight (mouse, i.p.)) appeared to be approximately one order of magnitude higher than MCLR (Krishnamurthy et al., 1986; Watanabe et al., 1988). Surprisingly, the inhibitory activity of both congeners on PP2A revealed to be in the same range: the IC 50 of MCRR and MCLR were 3.4 nm and 1.6 nm, respectively (Yoshizawa et al., 1990; Fujiki et al., 1996). Consequences of Phosphatase Inhibition PPs catalyze the dephosphorylation of intracellular phosphoproteins, thus represent the antagonists of protein kinases. Their interplay allows for concerted regulation of enzymes and other proteins which in turn regulate or control a vast variety of cellular functions and processes. PPs of type 1 and type 2 occur in all eukaryotes where they are responsible for the dephosphorylation of serine and threonine residues, thus a plethora of target proteins. E.g. PP1 and PP2A play pivotal roles in the regulation of cell growth and division, metabolism (e.g. glycogen metabolism), muscle contraction, intracellular transport, gene expression and protein synthesis (Cohen and Cohen, 1989a; Cohen, 1989b; Bollen and Stalmans, 1992; MacKintosh, 1993; Mumby and Walter, 1993; Cohen, 2002). Consequently, the inhibition of these PPs by MCs, which is also referred to as an activation of the corresponding protein kinases, results in the perturbance and disregulation of the listed cellular functions. In general, the eqilibrium between dephosphorylation and phosphorylation displaces, leading to an overall increase in phosphorylated cytosolic and cytoskeletal phosphoproteins as observed in isolated rat hepatocytes (Yoshizawa et al., 1990; Eriksson et al., 15

16 Chapter I General Introduction 1990a; Falconer and Yeung, 1992a). This MC-induced hyperphosphorylation was observed in all cytoskeletal components, i.e. microfilaments, microtubules and intermediate filaments (especially keratin 8 and 18) and resulted in their rapid reorganization and loss of cell integrity (Ohta et al., 1992; Wickstrom et al., 1995; Toivola et al., 1997; Toivola et al., 1998). Batista et al. (Batista et al., 2003) made similar observations with primary human hepatocytes, whose actin mesh collapsed into the centre of the cell following treatment with MCLR. MC-induced Apoptosis Depending on dose and time, either necrosis (at high concentrations) or apoptosis (at lower concentrations) has been observed subsequent to the disruption of the cytoskeleton (Hooser et al., 1991; McDermott et al., 1998; Hooser, 2000; Batista et al., 2003). However, the molecular mechanisms of MC-induced apoptosis are not entirely elucidated, although PP inhibition appeared to be crucial in triggering or executing programmed cell death (Fladmark et al., 2002). An additional role has been attributed to MC-induced generation of reactive oxygen species (ROS) causing mitochondrial permeability transition, a critical event in the progression of apoptotic cell death (Ding et al., 2000; Ding and Nam Ong, 2003; Gehringer, 2004a; Weng et al., 2007). Mikahailov et al. (Mikhailov et al., 2003) identified the ATP-synthase beta subunit as a further yet less important molecular target of MCs. They hypothesized that the adduct formation with MCs at high concentrations might play a mechanistic role in MC-induced apoptotic signalling by causing mitochondrial damage, i.e. loss of mitochondrial membrane potential and perturbance of mitochondrial functions. Acute and Subacute Effects The effects and symptoms of intoxications with MCs are as manifold as the consequences of PP inhibition may suggest. However, their severity depends on many factors like dose and duration of the exposure, as well as the route of intoxication and may vary among different species, gender and age (Dietrich and Hoeger, 2005; Fournie and Hilborn, 2008). Acute exposure to high doses of MCs causes sinusoid disruption, hepatocyte deformation and necrosis followed by rapid death (1-3 hours in mice) from liver haemorrhage or from liver failure (Falconer et al., 1981; Runnegar and 16

17 Chapter I General Introduction Falconer, 1982; Runnegar et al., 1986; Theiss et al., 1988; Hooser et al., 1989; Beasley et al., 2000). Hepatic and endothelial lesions are thereby accompanied by an increase in liver weight and size, as well as serum liver enzyme levels. Further pathological and ultrastructural features diagnosed in the liver are centrilobular hepatic necrosis, cessation of bile flow, loss of microvilli, bleb formation and induction of apoptosis in hepatocytes (Runnegar et al., 1995b; Wickstrom et al., 1996; Ito et al., 1997; Yoshida et al., 1997). Other organs affected, albeit less severely, include stomach, intestine, kidneys and lungs (Runnegar et al., 1986; Hooser et al., 1989; Falconer et al., 1992b; Falconer and Humpage, 1996; Ito et al., 1997). Furthermore, oral MC toxicity has also been shown to depend on the nutritional state of the exposed animals: fed rats were 1.7-fold less susceptible than fasted rats (25-h i.p. LD 50 of 72 µg/kg bw) (Miura et al., 1991). The authors suggested the higher susceptibility to either stem from the additional depletion of the already exhausted glycogen stores in fasted rats by activation of phosphorylase a as a result of MC mediated PP inhibition impairing the animal s energy reserves or the decreased respiratory capacity in fasted rats leading to a more advanced mitochondrial damage. A multitude of symptoms have been documented from acute human intoxications: In 1996, in a haemodialysis unit in Caruaru, Brazil, water contaminated with MC (and possibly cylindrospermopsin) was used for dialysis. 116 out of 131 patients developed symptoms of acute neuro- and hepatotoxicity including visual disturbances, vertigo, headaches, nausea and vomiting, muscle weakness and myalgia, painful huge hepatomegaly, liver plate disruption, liver cell deformity, necrosis and apoptosis, as well as death from liver failure. Biochemical investigations showed elevated liver enzyme activities, severe hypertriglyceridaemia and hyperbilirubinaemia. 52 patients succumbed to the so-called Caruaru syndrome (Jochimsen et al., 1998; Pouria et al., 1998; Kuiper-Goodman et al., 1999; Carmichael et al., 2001; Azevedo et al., 2002). In 1988, an incident of acute oral intoxication via contaminated drinking water led to a severe gastro-enteritis epidemic (about 2,000 cases) in the area of the Itaparica Dam, Bahia, Brazil which resulted in 88 deaths of predominantly children under the age of 5. The symptoms most commonly observed included 17

18 Chapter I General Introduction diarrhoea, colic-like abdominal pain, vomiting and fever. The newly flooded dam accomodated an immense cyanobacterial bloom of the genera Anabaena and Microcystis and the cases of gastro-enteritis were restricted to areas in which the dam served as a drinking water source (Teixera et al., 1993). Subacute oral intoxications with MCs at lower concentrations are characterized by diarrhoea, vomiting, weakness, pallor and elevated levels of hepatic enzymes in plasma which indicate toxic liver injury (Falconer et al., 1983; Bell and Codd, 1994). Chronic Effects and Tumour Promotion Chronic exposure to low doses of MCs has been shown to promote tumours in humans and animals. Falconer et al. (Falconer et al., 1988) examined the effects of exposure of mice to a toxic extract of Microcystis aeruginosa via drinking water over a period of 1 year. At high concentrations 4 out of 71 mice developed tumours, in contrast to only 2 out of 223 mice at lower concentrations. An epidemiological study on the incidence of primary liver cancer (PLC) in China, which is one of the highest worldwide with 24 mortalities per 100,000 population, revealed strongest correlations with hepatitis B incidence, followed by aflatoxins in the diet and MC contaminated drinking water from ponds and ditches. All three factors are considered to act together in promoting PLC (Yu, 1989; Yu, 1995; Falconer et al., 1999; Kuiper-Goodman et al., 1999). Since phosphatases may act as tumour suppressors, tumour promotion is possibly a result of phosphatase inhibition leading to MAPK signaling which in turn stimulates proliferation and inhibits apoptosis (Toivola and Eriksson, 1999; Gehringer, 2004a). Indeed, several tumour-promoting toxins like okadaic acid, calyculins, tautomycin, as well as nodularins and MCs are known to act via inhibition of PP1 and PP2A (MacKintosh, 1993). Evidence for the tumourpromoting and -initiating activity of MCs have been provided by several in vitro and in vivo studies. Suppression of apoptosis and stimulation of cytokinesis has been reported at lower MC concentrations (pm range) in polyploid hepatocytes in vitro (Humpage and Falconer, 1999). In a two-stage carcinogenesis study MCLR dose-dependently increased the occurence of positive foci of the placental form of glutathione S-transferase in rat liver initiated with diethylnitrosamine (Nishiwaki-Matsushima et al., 1992). Without initiator 18

19 Chapter I General Introduction neoplastic nodules formed in mice liver after repeated (100 times) i.p. injections of a sublethal dose (20 µg/kg bw) of MCLR. However, neither nodule formation nor liver damage was observed when MCLR (80 µg/kg bw) was orally administered (Ito et al., 1997). The genotoxic potential of MCs has been furthermore assessed in several genotoxicity assays. In human HepG2 cells MCLR dose- and time-dependently induced DNA strand breaks (Zegura et al., 2003; Zegura et al., 2004), whereby this effect could be reduced by different ROS scavengers. The authors therefore concluded that MCLR causes DNA damage by inducing the formation of ROS. Carcinogenic effects were also supported by Sano et al. (Sano et al., 2004): They reported the development of spontaneous liver tumour in 15 out of 22 mice i.p. injected with MCLR (12.5 or 25 µg/kg bw) once a week for 14 months. Tumour incidences thereby correlated with the generation of 8- hydroxydeoxyguanosine, a biomarker for oxidative stress, in the liver of the mice. In addition, in the in vitro cytokinesis-block micronucleus (CBMN) assay, a test that detects both chromosome loss and chromosome breakage, MCLR failed to induce significant alterations of DNA in contrast to nodularin and okadaic acid (Fessard et al., 2004). Thus, MCs appear not to be directly genotoxic, but indirectly by generating ROS at moderate to high concentrations. Moreover, for liver tissue damage a no observed adverse effect level (NOAEL; the highest concentration that fails to elicit signs of adverse effects) of 40 µg MCLR/kg bw per day was estimated from a subchronic study in which mice were orally gavaged with pure MCLR over a period of 13 weeks (Fawell et al., 1999). This NOAEL was employed with additional uncertainty factors (10 a total of 1,000) to derive a provisional tolerable daily intake (TDI) of 0.04 µg MCLR/kg bw (Falconer et al., 1999; Kuiper-Goodman et al., 1999; Dietrich and Hoeger, 2005), which has been used as a basis for risk assessments and calculations of guideline values including drinking water (1 µg/l; WHO, 1998) and cyanobacterial dietary supplements (1 µg/g dw; Gilroy et al., 2000). Extract Toxicity and Synergistic effects Surprisingly, the toxicity of cyanobacterial extracts often exceeds the toxic potential that would have been expected from the contained amount of toxins. This phenomenon probably relies on unknown or unnoticed active compounds 19

20 Chapter I General Introduction that are additionally contained in the extracts or on toxins acting synergistically as recently discussed (Dietrich et al., 2008; Pegram et al., 2008). Indeed, Fitzgeorge et al. (Fitzgeorge et al., 1994) determined an intranasal LD 50 of 2000 µg/kg bw for anatoxin-a (see ) in mice that was lowered to 500 µg/kg bw when a sublethal dose of 31.3 µg MCLR/kg bw (NOAEL for liver weight increase) was administered 30 minutes prior to anatoxin-a. By contrast, this synergism failed to recur by oral application of the toxins which was suggested to be due to the different route of administration (Rogers et al., 2005). Routes of Intoxication Intoxications with cyanobacterial toxins may occur via different routes of exposure as previously described (Falconer et al., 1999; Dietrich and Hoeger, 2005; Dietrich et al., 2008). Such scenarios include: exposure via contaminated drinking water exposure via contaminated food as a result of bioaccumulation in the food chain, irrigation with contaminated water or toxic blooms in rice fields exposure from recreational use of water exposure from contaminated cyanobacterial dietary supplements exposure via renal dialysis. The case studies on human poisonings described above demonstrate both the high symptomatic diversity of MC intoxications and their dependency on the respective route of intoxication that was elucidated in various animal studies as well: In mice the i.p. LD 50 of MCLR or Microcystis extracts appeared to be approximately a factor lower than the oral LD 50 (Falconer, 1991; Kotak et al., 1993; Yoshida et al., 1997; Fawell et al., 1999). In contrast, other routes of exposure of mice to MCLR corresponded well to the lethal dose by i.p. application with LD 50 s between 50 and 100 µg/kg bw for intratracheal (Ito et al., 2001), 43 µg/kg bw for intranasal and 67 µg/kg bw for i.v. application (Creasia, 1990). Differences in the LD 50 s and the symptoms elicited reflect varying bioavailability from the respective route of administration as a result of the chemical and biochemical characteristics of MCs and their toxicokinetics, i.e. transport and distribution in the exposed organism as specified in the following. 20

21 Chapter I General Introduction Organotropism, Uptake, Distribution and Excretion Cellular trafficking of MCs requires active transport, since they are rather hydrophilic molecules which precludes passive diffusion through cell membranes. The selective uptake into hepatocytes via the bile acid transport system has been demonstrated by its inhibition and by coincubation with bile salts and further substrates of this transport system, which reduced MC uptake and toxicity (Eriksson et al., 1990b; Runnegar et al., 1991; Runnegar et al., 1995c). Indeed, Fischer et al. (Fischer et al., 2005) identified members of the multispecific organic anion transporting polypeptides [human: OATPs/SLCOs; animals: Oatps/Slcos; (protein name/gene symbol) (Hagenbuch and Meier, 2004)], which are part of the bile acid transport system, as being capable of transporting [ 3 H]-dihydro-MCLR. Those members included OATP1B1, OATP1B3 and Oatp1b2 (rat), all located at the basolateral (sinusoidal) membrane of hepatocytes, as well as OATP1A2, located in liver, kidney and at the blood-brain-barrier (Hagenbuch and Meier, 2003; Hagenbuch and Meier, 2004; Bronger et al., 2005; Ho and Kim, 2005; Nies, 2007). Exchange with anions (e.g. bicarbonate) or efflux of glutathione and/or glutathione-s-conjugates is assumed to be the driving force for OATP/Oatpmediated transport as demonstrated in rat Oatp1a1 and -1a4 (Satlin et al., 1997; Li et al., 1998; Hagenbuch and Meier, 2003; Hagenbuch and Meier, 2004; Ho and Kim, 2005). Monks et al. (Monks et al., 2007) and Komatsu et al. (Komatsu et al., 2007) confirmed the uptake of non-labeled MCLR and further congeners via OATP1B1 and OATP1B3. Therefore, it has been suggested that OATP1B1, OATP1B3 and OATP1A2 are at least involved in the observed MC-mediated hepato- and neurotoxicity (Fischer et al., 2005; Dietrich et al., 2008). The organotropism was additionally corroborated by various investigations using radiolabeled MCs and different application routes as summarized by Dietrich and Hoeger (Dietrich and Hoeger, 2005). Briefly, radioactivity was predominantly detected in the liver, followed by the gastro-intestinal tract, the kidneys, the brain, the lungs and other organs. Recovery of radiolabeled MC after oral administration was thereby tremendously reduced (factor 80) in comparison to i.p. or i.v. injection indicating reduced bioavailability from oral application. Indeed, following oral ingestion it has been shown in mice that a 21

22 Chapter I General Introduction significant portion of free MC is not absorbed, but remains in the gastrointestinal tract and thus is likely to be excreted via faeces (Fujiki et al., 1996; Ito et al., 2000). Absorption was demonstrated in animal studies to mainly take place in the small intestine, especially in the ileum, but also to a small extend in the stomach (Dahlem et al., 1989; Stotts et al., 1997a; Stotts et al., 1997b; Ito et al., 2000). Subsequently, MC enters the venous blood stream, following the portal vein to the liver, where it rapidly accumulates (Stotts et al., 1997a; Stotts et al., 1997b) due to the first pass effect and selective uptake via the bile acid transport system, which was also observed after i.v. or i.p. injection (Falconer et al., 1986; Brooks and Codd, 1987; Robinson et al., 1989; Meriluoto et al., 1990; Robinson et al., 1991; Lin and Chu, 1994). Although the vast majority of MC is retained in the liver, it was also detectable in the bile (Runnegar et al., 1986; Stotts et al., 1997a; Stotts et al., 1997b), as well as in intestines and feces following i.v. and i.p. injection (Robinson et al., 1989; Robinson et al., 1991; Stotts et al., 1997a; Wang et al., 2008) providing evidence for enterohepatic circulation. Besides biliary excretion, elimination of MCs also occurs via urine, albeit to a much lesser extent (Falconer et al., 1986; Runnegar et al., 1986; Robinson et al., 1989; Robinson et al., 1991). Metabolization and Detoxification MCs are not degraded in the mammalian digestive tract due to the predominant presence of D-amino acids and the cyclic structure, which renders them resistant to enzymatic hydrolysis by eukaryotic peptidases (e.g. trypsin) (Runnegar and Falconer, 1981; Harada and Tsuji, 1998). However, strong evidence for the detoxification of MCs via the glutathione (GSH) pathway was provided by Hermansky et al. (Hermansky et al., 1991): pretreatment with GSH protected mice (no mortalities were observed) against a lethal dosage of MCLR. Indeed, Kondo et al. (Kondo et al., 1996) detected several MC metabolites among which two were identified as GSH and cysteine conjugates in mice and rats. The conjugation sites were shown to be the thiols of GSH and cysteine that bind nucleophilically to the Mdha residue of MCs, identical to the PPs (Kondo et al., 1992). 22

23 Chapter I General Introduction Under physiological conditions this reaction is catalyzed by the glutahtione S- transferase (GST) as demonstrated by GST assays (Pflugmacher et al., 1998; Takenaka, 2001) and increased GST activity in mice that corresponded with increased GST transcription following treatment with MCLR (Gehringer et al., 2004b). Further evidence for the detoxification of MCs by GSH/cysteine conjugation were given by Ito et al. (Ito et al., 2002): Both MCLR-GSH and MCLR-cysteine conjugates, administered intratracheally to mice, exhibited 12-fold reduced toxicity compared to native MCLR, although they inhibited PP1 and PP2A nearly equipotently in vitro. However, immunostaining revealed the highest signals for both conjugates in the kidneys and intestines of the mice, whereas the characteristic accumulation and damage in the liver failed to occur. The authors suggested that either the uptake of the conjugates into the liver is impeded or that they are effectively exported, e.g. by the ATP-dependent glutathione S-conjugate export (GS-X/MRP1), officially known as ABCC1 according to the HUGO Gene Nomenclature Committee (HGNC). The potential involvement of members of the ATP Binding Cassette superfamily (ABC transporters), especially multidrug resistance-associated proteins (MRPs) and multidrug resistance proteins (MDRs), in the export of MCs has recently been discussed by Dietrich et al. (Dietrich et al., 2008) Nodularins The first report on fatal animal poisonings by toxic Nodularia spumugena, the eponymous producer of nodularins (NODs), has been published by Francis in the late 19 th century (Francis, 1878). The general structure of NODs (Fig. 1.2; MW = 824 Da), cyclo(-d-erythro-βmethylasp 1 -L-Z 2 -Adda 3 -D-Glu 4-2-(methylamino)-2-dehydrobutyric acid 5 ) (Rinehart et al., 1988; Carmichael et al., 1988b; Sivonen and Jones, 1999; Spoof, 2005), obviously demonstrates their close relatedness to MCs. Besides being composed of five instead of seven amino acids, i.e. D-alanine and the adjacent variable L-amino acid are missing, NODs further differ from MCs in the substitution of Mdha by methyldehydrobutyric acid (Mdhb). The variable L- amino acid Z either represents arginine in nodularin-r (often simply referred to as nodularin (Rinehart et al., 1988)) or valine in nodularin-v (trivially named 23

24 Chapter I General Introduction motuporin) which was isolated from the marine sponge Theonella swinhoei (de Silva et al., 1992), however, is likely to originate from a cyanobacterial symbiont. Aside from these two analogues, only a few further variants have been found including two demethylated variants, [D-Asp 1 ]nodularin and [DMAdda 3 ]nodularin and the non-toxic [(6Z)-Adda 3 ]nodularin (Namikoshi et al., 1994; Rinehart et al., 1994). The relatedness of NODs and MCs reflects in nearly every aspect: NODs are also synthesized non-ribosomically by peptide synthetases and polyketide synthases. Their biosynthesis gene cluster is homologous to the mcy-cluster, however, consists of only nine open reading frames (ndaa - ndai) and lacks the two modules in the MC gene cluster that are responsible for the synthesis of D- alanine and the adjacent variable L-amino acid (Moffitt and Neilan, 2004; Rantala et al., 2004; Welker and von Dohren, 2006). Fig. 1.2: General structure of nodularins. Likewise MCs the molecular mode of action of NODs is based on the inhibition of serine/threonine-specific PPs, thus, the toxic effects elicited by both cyanopeptides are similar and comparably severe (Eriksson et al., 1988; Runnegar et al., 1988; Carmichael et al., 1988b; Yoshizawa et al., 1990; Ohta et al., 1994). However, as already mentioned (see ), NODs are unable to bind covalently to the phophatases (Craig et al., 1996; Bagu et al., 1997). Nevertheless, MCLR and NOD (both NOD-R and NOD-V) equipotently inhibit 24

25 Chapter I General Introduction PP1 and PP2A (Yoshizawa et al., 1990; Honkanen et al., 1991; de Silva et al., 1992). Furthermore, NOD revealed to be a stronger tumour promoter than MCLR as it more effectively increased the occurence of positive foci of the placental form of glutathione S-transferase in rat liver initiated with diethylnitrosamine in a twostage carcinogenesis experiment. In contrast to MCLR, NOD also induced positive foci without prior initiation which classifies it as a carcinogen (Ohta et al., 1994; Fujiki et al., 1996) Alkaloids The collective term alkaloid stands for basic, nitrogenous, heterocyclic compounds that are naturally produced in the secondary metabolism of predominantly plants (Vollhardt and Schore, 1995; Nultsch, 2001; Bruice, 2007). As broad as this definition is the structural and toxicological diversity of the cyanobacterial alkaloid toxins. They comprise dermatotoxic (e.g. lyngbyatoxin-a, aplysiatoxins), neurotoxic (e.g. saxitoxins and anatoxins) and cytotoxic (e.g. cylindrospermopsins) compounds that are regularly reponsible for severe, sometimes even fatal, human and animal poisonings (Kuiper-Goodman et al., 1999; Sivonen and Jones, 1999) Cylindrospermopsins In 1979 on Palm Island located off the coast of Queensland, Australia an algal bloom occured in Solomon Dam, the main drinking water reservoir of that island, causing discolouration and bad taste and odour of the water. In order to antagonize the bloom, it was treated with copper sulphate. Subsequently, among the local Aboriginal population 139 children and 10 adults developed severe hepatoenteritis. No cases of this illness were reported from a small group of the island s population that drew their drinking water from an alternative source. Since the identification of causative pathogenes, toxins or chemicals failed, the illness was entitled Palm Island Mystery Disease (Byth, 1980; Bourke et al., 1983). Investigations on following algal blooms in Solomon Dam identified the cyanobacterium Cylindrospermopsis raciborskii as the bloom forming species 25

26 Chapter I General Introduction and the most plausible causative of the disease. Extracts of this species elicited dose-dependent damage to the livers of mice following i.p. injection. At lower concentrations (10.5 mg/kg mouse bw) hepatocyte necrosis was mainly restricted to centrilobular regions, whereas at high concentrations (168 mg/kg mouse bw) all hepatocytes appeared to be affected. An LD 50 of 64 ± 5 mg/kg bw was determined at 24 hours after administration. Although the extracts were found to be primarily hepatotoxic, lungs, kidneys and the small intestine were also affected (Hawkins et al., 1985). From this extract Ohtani et al. (Ohtani et al., 1992) isolated and characterized an unusual alkaloid with a molecular weight of 415 Da that elicited the same symptoms: cylindrospermopsin (CYN; Fig. 1.3). This highly water-soluble zwitterion consists of a sulphated and methylated tricyclic guanidino moiety that is linked to uracil via a hydroxylated carbon (Ohtani et al., 1992; Falconer, 2005a). Currently, only two naturally occuring structural variants have been identified: 7-deoxycylindrospermopsin (Norris et al., 1999) and 7- epicylindrospermopsin (Banker et al., 2000). Fig. 1.3: Structure of cylindrospermopsin. Pure CYN appeared to be relatively stable at extreme temperatures, ph and sunlight, whereas in algal extracts it degrades rapidly when exposed to sunlight probably due to several attendant pigments (Chiswell et al., 1999). Besides C. raciborskii CYN has been further isolated from Umezakia natans (Harada et al., 1994), Aphanizomenon ovalisporum (Banker et al., 1997), Raphidiopsis curvata (Li et al., 2001), Anabaena bergii (Schembri et al., 2001), Aphanizomenon flos-aquae (Preussel et al., 2006) and Aphanizomenon gracile (Rücker et al., 2007; Wiedner et al., 2008). 26

27 Chapter I General Introduction Moreover, C. raciborskii was reported to be invasive, spreading from tropical and subtropical regions into more temperate climate (Padisak, 1997; Fastner et al., 2003; Neilan et al., 2003; Falconer and Humpage, 2006). Indeed, Fastner et al. (Fastner et al., 2007) detected CYN in 50% of 127 lakes and samples investigated from north-east Germany. Concentrations were recorded up to 73.2 µg CYN/g dw. Inter alia, this is hypothesized to be due to climate changes, especially global warming, that are predicted to favour toxic cyanobacterial blooms in terms of abundance, extension of their season and distribution (Shaw et al., 2001; Pearl and Huisman, 2009). Hence, this cyanotoxin is considered to be an increasing threat for human and animal health. CYN has been classified as a hepatotoxin, since it predominantly affects the liver (Ohtani et al., 1992). However, in vivo studies with extracts of CYN producing C. raciborskii or purified CYN revealed its organotropism to also comprise kidneys, lungs, thymus, heart, stomach and small intestine as indicated in the case study described above (Hawkins et al., 1985; Ohtani et al., 1992; Terao et al., 1994; Hawkins et al., 1997; Falconer et al., 1999; Seawright et al., 1999). The purified toxin administered i.p. to mice yielded LD 50 values of 2.1 and 0.2 mg CYN/kg bw at 24 hours and 5-6 days, respectively (Ohtani et al., 1992). Conversely to the increased toxicity of MC containing extracts, extracts of C. raciborskii appeared to be less toxic than purified CYN with i.p. LD 50 values ranging from mg/kg bw at 24 hours and mg/kg bw at 7 days (Hawkins et al., 1997; Falconer et al., 1999). In turn similar to MCs, the oral toxicity of CYN and CYN containing extracts revealed to be distinctly lower than the i.p. toxicity: In mice purified CYN and an extract of C. raciborskii resulted in LD 50 s of approximately 6.0 mg/kg bw (Shaw et al., 2001) and mg/kg bw at 2-6 days (Seawright et al., 1999), respectively. Although the organotropsim of CYN shares similarities with MCs and NODs, its molecular mode of action relies on completely different and dose-dependent mechanisms because of which it is now classified as a cytotoxin. Several in vitro studies using primary mouse and rat hepatocytes, as well as a rabbit reticulocyte lysate translation system revealed the following two primary mechanisms: 27

28 Chapter I General Introduction Irreversible inhibition of protein and GSH synthesis at sub- or low micromolar concentrations (Terao et al., 1994; Runnegar et al., 1995d; Froscio et al., 2001; Runnegar et al., 2002; Froscio et al., 2003). Cytochrome P450-mediated cytotoxicity at acute concentrations (Terao et al., 1994; Runnegar et al., 1995d; Froscio et al., 2003; Humpage et al., 2005). The latter was confirmed in an in vivo study in which mice were protected against CYN toxicity by preadministration of piperonyl butoxide, a P450 inhibitor (Norris et al., 2002). In addition, CYN has been shown to elicit genotoxic effects, i.e. strand breakages in livers of CYN treated mice (Shen et al., 2002) that were also observed in vitro in addition to loss of whole chromosmes as demonstrated using cytokinesis-blocked micronucleus and comet assays (Humpage et al., 2000; Humpage et al., 2005). The dependency on metabolic CYP450 activity was thereby indicated by a lack of strand breakages in CHO-K1 cells, in which metabolizing enzyme activities are low (Fessard and Bernard, 2003), and corroborated via pretreatment with inhibitors of CYP450 that prevented CYN induced DNA fragmentation in metabilically active primary rat hepatocytes (Humpage et al., 2005) Saxitoxins The neurotoxic saxitoxins (STXs) are the causative toxins for paralytic shellfish poisoning (PSP), one of four known symptoms of shellfish poisonings (the others being diarrhetic, neurotoxic and amnestic shellfish poisoning), and are therefore often referred to as PSP toxins (Kuiper-Goodman et al., 1999; Sivonen and Jones, 1999; Lehane, 2000). STXs are produced by certain dinoflagellates (e.g. some species of the genus Alexandrium, Gymnodinium catenatum and Pyrodinium bahamense var. compressum) and the following cyanobacteria: Anabaena circinalis, Aphanizomenon flos-aquae, Aphanizomenon gracile, Cylindrospermopsis raciborskii, Lyngbya wollei, Aphanizomenon issatschenkoi (Sawyer et al., 1968; Mahmood and Carmichael, 1986a; Negri et al., 1995b; Carmichael et al., 1997; Lagos et al., 1999; Pereira et al., 2000; Ferreira et al., 2001; Nogueira et al., 2004; Pereira et al., 2004b). 28

29 Chapter I General Introduction The general structure of the highly water-soluble STXs (Fig. 1.4; MW = Da) is tricyclic containing hydropurine rings, a carbamate group and five variable positions. Fig. 1.4: Structure of saxitoxins. They comprise three classes of derivates: the nonsulphated STXs, the singly sulphated gonyautoxins (GTXs) and the doubly sulphated C-toxins. Further variants are decarbamoyl derivatives and several Lyngbya-wollei toxins (LWTXs) (Sivonen and Jones, 1999; Lehane, 2000; van Apeldoorn et al., 2007). STXs act as voltage-gated sodium channel antagonists with varying potency (Kuiper-Goodman et al., 1999; Sivonen and Jones, 1999; Lehane, 2000; Briand et al., 2003; van Apeldoorn et al., 2007): Parental STX (i.p. LD 50 of 10 µg/kg bw in mice) appeared to be more than 160 times more toxic than C-toxin 1 (Oshima, 1995). This potent neurotoxicity poses a serious threat to human and animal health, especially via consumption of bivalve molluscs (e.g. mussels, oysters and clams) in which STXs are known to accumulate by filter feeding of toxic algae or cyanobacteria (Carmichael and Falconer, 1993; Anderson, 1994; Negri and Jones, 1995a; Negri et al., 1995b; Kuiper-Goodman et al., 1999; Sivonen and Jones, 1999; Lehane, 2000; Pereira et al., 2004a). In fact, 153 fatalities among 2235 cases of human PSPs have been reported worldwide from 1900 to 2002 (Batoréu et al., 2005). Typical PSP symptoms in mammals range from slight tingling and numbness of lips and tongue, nausea and vomitting to muscle 29

30 Chapter I General Introduction paralysis and death caused by cardio-respiratory arrest (Kuiper-Goodman et al., 1999; Batoréu et al., 2005) Anatoxins In contrast to saxitoxins, anatoxins have been isolated from cyanobacteria only: Anatoxin-a from Anabaena, Aphanizomenon, Cylindrospermum, Microcystis, Oscillatoria, Planktothrix and Raphidiopsis, homoanatoxin-a from Anabaena, Oscillatoria, Phormidium and Raphidiopsis and anatoxin-a(s) from Anabaena (Sivonen and Jones, 1999; van Apeldoorn et al., 2007). Anatoxin-a (Fig. 1.5 (A)) is a bicyclic secondary amine with a molecular weight of 165 Da (Devlin et al., 1977). Its homologue homoanatoxin-a (MW = 179 Da) carries a propionyl instead of the acetyl group at C-2 (Skulberg et al., 1992). Both mimic the effect of acetylcholine, however, cannot be enzymatically cleavaged by acetylcholinesterase. This causes prolonged depolarization and blockade of further electrical transmission (Carmichael et al., 1975; Soliakov et al., 1995; Carmichael, 1997). Although suggested by the name, anatoxin-a(s) is unrelated to anatoxin-a (S means salivation factor). It is a unique N-hydroxyguanidine methyl phosphate ester (Fig. 1.5 (B); MW = 252 Da) that irreversibly inhibits acethylcholinesterase similar to organophosphate insecticides (Mahmood and Carmichael, 1986b; Mahmood and Carmichael, 1987; Cook et al., 1989; Matsunaga et al., 1989; Hyde and Carmichael, 1991). A B Fig. 1.5: Structures of (A) anatoxin-a and (B) anatoxin-a(s). 30

31 Chapter I General Introduction Both extreme depolarization and inhibition of acethylcholinesterase lead to muscle paralysis and death by respiratory failure in mammals (Kuiper-Goodman et al., 1999; Briand et al., 2003; van Apeldoorn et al., 2007; Aráoz et al., 2009). Additional characteristic symptoms of anatoxin-a(s) are salivation and lacrimation (Mahmood and Carmichael, 1986b; Mahmood and Carmichael, 1987; Matsunaga et al., 1989). The i.p. LD 50 s in mice are 20 µg/kg bw for anatoxin-a(s) (Carmichael et al., 1990) and µg/kg bw for anatoxin-a and homoanatoxin-a (Devlin et al., 1977; Carmichael et al., 1990; Skulberg et al., 1992) Other Cyanobacterial Toxins β-n-methylamino-l-alanine The non-protein amino acid β-n-methylamino-l-alanine (BMAA; Fig. 1.6) has initially been isolated from extracts of cycad seeds (Cycas circinalis) from Guam, an island in the western Pacific Ocean (Vega and Bell, 1967; Vega et al., 1968). Recently Cox et al. (Cox et al., 2003; Cox et al., 2005) found that BMAA is of cyanobacterial origin and accumulates in cycad seeds as a result of a symbiosis between the cycad coralloid roots and a BMAA producing Nostoc species. The authors furthermore reported that BMAA increasingly accumulates in higher trophic levels, i.e. flying foxes (Pteropus mariannus) and the Chamorro, the indigenous population of Guam that consume both cycad seeds and flying foxes as part of their traditional diet. Fig. 1.6: Structure of β-n-methylamino-l-alanine (BMAA). 31

32 Chapter I General Introduction BMAA was found to induce a neurological disorder (i.e. corticomoto-neuronal dysfunction, chromatolytic and degenerative changes of motor neurons and symptoms similar to Parkinson s) in macaques following oral administration (Spencer et al., 1987a). Therefore, it has been hypothesized to be the causative agent for the increased incidence of amyotrophic lateral sclerosis/parkinsonismdementia complex (ALS/PDC) among the Chamorro, as well as the indigenous Auyu of Irian Jaya, Indonesia and the Japanese residents of the Kii peninsula of Honshu island, where cycad seeds are also part of the traditional diet or used in topical medicine (Spencer et al., 1987b; Spencer et al., 1987c). ALS/PDC is a severe tauopathy that shares similarities to amyotrophic lateral sclerosis, Parkinson s disease and Alzheimer s (Steele, 2005). Indeed, BMAA has been shown to be neuro- and excitotoxic on cultured mouse cortical neurons by acting as an agonist of glutamate receptors at relatively high concentrations and in dependence upon the presence of physiological concentrations of bicarbonate ions (Weiss et al., 1989a; Weiss et al., 1989b). Bicarbonate serves as a cofactor for the reversible reaction of BMAA with dissolved carbon dioxide to β-carbamate that mimics the effect of glutamate (Myers and Nelson, 1990). Moreover, in rat brain cells BMAA dose-dependently elevates intracellular calcium levels in the presence of bicarbonate ions (Brownson et al., 2002), an effect that is known to potentially induce cell death and neurodiseases. As suggested by the aforementioned macaque study (Spencer et al., 1987a) BMAA reaches the brain after oral application. In fact, % of the p.o. administered dose becomes bioavailable as shown in another macaque experiment (Duncan et al., 1992). Smith et al. (Smith et al., 1992) identified the cerebrovascular large neutral amino acid carrier as being responsible for the uptake of BMAA across the blood-brain-barrier in rats. Furthermore, BMAA not only exists as a free amino acid, but it also occurs in a protein-bound form that usually exceeds the former 10- to 240-fold (Murch et al., 2004b). In addition, Murch et al. concluded that protein-bound BMAA may form an endogenous reservoir from which free BMAA is slowly released by protein metabolism. They further hypthesized that this slow release might cause continuous damage to the brain providing an explanation for the long latency 32

33 Chapter I General Introduction period of ALS/PDC from years to decades (Spencer et al., 1991a; Kisby et al., 1992) Lipopolysaccharides A common component of the outer cell membrane of gram-negative prokaryotes, including cyanobacteria, are lipopolysaccharides (LPS). As their name indicates LPS consist of carbohydrates (core polysaccharides and an outer polysaccharide chain) and lipids (lipid A) whose composition is very variable among bacteria in general, but also among cyanobacteria (Sivonen and Jones, 1999; Briand et al., 2003; Wiegand and Pflugmacher, 2005). In contrast to the aforementioned cyanotoxins, LPS are endotoxins that may elicit irritant, pyrogenic, allergic and toxic effects predominantly by the fatty acid component (Weckesser et al., 1979; Kuiper-Goodman et al., 1999; Sivonen and Jones, 1999; Briand et al., 2003). However, cyanobacterial LPS seem to be less toxic compared to LPS from pathogenic gram-negative bacteria, e.g. Salmonella (Kuiper-Goodman et al., 1999; Briand et al., 2003; Wiegand and Pflugmacher, 2005). 33

34 Chapter II Objectives 2 Objectives The primary molecular mechanism underlying the toxicity of MCs and NODs (i.e. PP inhibtion) has been extensively investigated and is well comprehended. As mentioned in the previous chapter, the Adda moiety is of crucial importance for the inhibition of PPs and thus toxicity. However, isolated Adda was demonstrated to lack inhibitory activity on PP2A (Harada et al., 2004). In order to complete this finding and to exclude potential differences in the effects of Adda on PP1 and 2A, colorimetric PP inhibition assays were conducted in this study with both phosphatases (chapter III). Although numerous studies focused on the organotropism of MCs, only little information exists on the transporters that mediate the cellular uptake and excretion of MCs, as well as their role in congener-specific toxicity. The latter appeared to significantly vary among some congeners in vivo despite similar PP inhibitory capacities (e.g. MCLR and MCRR). Therefore, the role of human liver OATPs in congener-specific in vitro toxicity of four different MCs (MCLR (leucine, arginine), MCRR (arginine, arginine), MCLW (leucine, tryptophan) and MCLF (leucine, phenylalanine)) was examined using stably OATP1B1- and OATP1B3-transfected HEK293 cells and primary human hepatocytes (chapter IV). As a prerequisite for the comparison of both, the congeners and the different cell types, the toxicodynamics of these four MC congeners, i.e. their inhibitory capacity on recombinant and endogenous serine/threonine-specific PPs, were assessed. Furthermore, the risk emanating from exposure to CYN predominantly via drinking water is increasing, along with the abundance of its producers. Especially Cylindrospermopsis raciborskii, has been reported to be invasive and to spread from tropical and subtropical regions into more temperate climate (Padisak, 1997; Fastner et al., 2003; Neilan et al., 2003; Falconer and Humpage, 2006). Despite intensive research the molecular mechanisms of CYN are not completely understood and human toxicity has barely been adressed. Most in vitro studies on the toxicity of CYN were carried out on 34

35 Chapter II Objectives permanent cell lines and primary hepatocytes of mice and rats (Runnegar et al., 1995d; Shaw et al., 2000; Chong et al., 2002; Froscio et al., 2003; Humpage et al., 2005). In addition, the findings of Chong et al. (Chong et al., 2002) suggest the involvement of the bile acid transport system in facilitating the uptake of CYN. Thus, a further scope of this study was to determine the cytotoxicity of CYN along with the involvement of liver OATPs in primary human hepatocytes and OATP1B1- and OATP1B3-expressing HEK293 cells (chapter V). Contact to cyanotoxins may occur via several routes of exposure as summarized in the previous chapter. Of major concern are contaminations of food and drinking water, whereas the latter especially poses a threat for poorer countries where water treatment is limited or absent (Falconer, 1993; Dietrich and Hoeger, 2005; Falconer, 2005a; Falconer and Humpage, 2005b). However, in industrialized countries cyanobacterial dietary supplements (blue-green algae supplements (BGAS)) that are consumed for their putative beneficial health effects (i.e. increased alertness and energy, detoxification, efficacy against various viral infections, cancer and mental disorders like depression or attention-deficit disorders) represent an exceptional source for cyanotoxin exposure, in particular MCs (Gilroy et al., 2000; Kuiper-Goodman et al., 2000; Lawrence et al., 2001; Dietrich and Hoeger, 2005; Saker et al., 2005). Furthermore, as BMAA was not only detected in brain tissues from Chamorros who died from ALS/PDC, but also in brain tissues from Alzheimer patients from Canada (Cox et al., 2003; Murch et al., 2004a), which raises the question about the corresponding source of BMAA. According to the findings of Cox et al. (Cox et al., 2005) BMAA may be produced by all known groups of cyanobacteria. Therefore, besides assessing the health risk of a potential contamination of different BGAS samples with MCs using different analytical methods, a further aim was to analyze these supplements for contamination with BMAA (chapter VI). 35

36 Chapter III Inhibitory capacity of Adda on PP1 and 2A 3 Inhibitory capacity of Adda on protein phosphatase 1 and 2A Fischer A 1 and Dietrich DR 1 1 Human and Environmental Toxicology, University of Konstanz, Konstanz, Germany in preparation 3.1 Abstract The unusual D-amino acid 3-amino-9-methoxy-2,6,8-trimethyl-10-phenyldeca- 4,6-dienoic acid, abbreviated Adda, represents part of the toxic moiety of the cyclic cyanobacterial peptides microcystins (MCs) and nodularins (NODs), potent inhibitors of several serine/threonine-specific phophatases (PP). We examined the inhibitory potential of isolated Adda on recombinant PP1 and PP2A in comparison with one of its well-studied parental molecules MCLR. Adda had no effect on the activity of neither PP1 nor PP2A in concentrations up to 55.7 µm, whereas MCLR concentration-dependently inhibited both phosphatases in the low nanomolecular range. These findings clearly demonstrate the relevance of the remaining structural units of the MC molecule for its biological activity. Keywords: cyanobacteria, microcystin, Adda, protein phosphatase 3.2 Introduction Microcystins (MCs) and the closely related nodularins (NODs) are toxic cyanobacterial oligopeptides that share three amino acids: D-methylaspartic acid, D-glutamic acid acid and the unusal D-amino acid 3-amino-9-methoxy- 2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid (Adda). Their general structures are cyclo(-d-ala 1 -L-X 2 -D-erythro-β-methylAsp 3 -L-Z 4 -Adda 5 -D-Glu 6 -N-methyl- 36

37 Chapter III Inhibitory capacity of Adda on PP1 and 2A dehydro-ala 7 ) and cyclo(-d-erythro-β-methylasp 1 -L-Z 2 -Adda 3 -D-Glu 4 -Nmethyldehydrobutyrine 5 ), respectively. In NODs Z either stands for L-arginine (nodularin-r) or L-valine (nodularin-v or motuporin) contrary to MCs, where X and Y are variable L-amino acid residues that mainly contribute to the high diversity of structural analogs (Botes et al., 1984; Botes et al., 1985; Rinehart et al., 1988; Rinehart et al., 1994) and are therefore used for the nomenclature of MCs (e.g. MCLR is a microcystin congener with L-leucine and L-arginine in position 2 and 7, respectively) (Carmichael et al., 1988a). The toxicity of MCs and NODs predominantly stems from the very potent inhibition of several serine/threonine-specific protein phosphatases (PPs), especially PP1 and PP2A (MacKintosh, 1993; Honkanen et al., 1994; Runnegar et al., 1995a; Toivola et al., 1997; Hastie et al., 2005). The IC 50 s (concentration that inhibits 50% of the enzyme s activity) of both toxins for PP1 and PP2A were found to be in the low nanomolar range. The inhibition relies on the noncovalent binding of the Adda-glutamate domain that represents the toxic moiety of both oligopeptides. In contrast to NODs most MC congeners are capable of forming an additional covalent, thus, irreversible binding via the N- methyldehydroalanine residue that is linked to a cysteine residue of the PP (MacKintosh et al., 1995; Runnegar et al., 1995a; Craig et al., 1996; Bagu et al., 1997; Hastie et al., 2005; Maynes et al., 2006). Structural alterations in Adda (e.g. isomerization of its diene) or D-glutamate (e.g. acetylation or esterification) have been shown to drastically decrease or abolish the inhibitory potential and hence toxicity of MCs and NODs indicating the importance of this moiety for bioactivity (Harada et al., 1990a; Harada et al., 1990b; Nishiwaki-Matsushima et al., 1991; Namikoshi et al., 1992; Stotts et al., 1993; Rinehart et al., 1994; Harada, 1996b). The purpose of this study was to examine the inhibitory potential of isolated Adda on recombinant PP1 and PP2A in comparison with MCLR, the most intensively studied MC congener, using a colorimetric PP inhibition assay. 37

38 Chapter III Inhibitory capacity of Adda on PP1 and 2A 3.3 Material & Methods Chemicals and reagents All chemicals were of the highest analytical grade commercially available. Adda was synthesized as previously reported (Fischer et al., 2001). Microcystin-LR (MCLR) was obtained from Alexis (Switzerland) and dissolved in 75% MeOH. The concentrations of the stock and the working solutions were confirmed photometrically using the molar absorption coefficient of MCLR (39800 mol l -1 cm -1 ) published by Harada et al. (Harada et al., 1990a). This coefficient refers to MCLR dissolved in 100% MeOH, however, was shown to be applicable for 75% MeOH as well (Meriluoto et al., 2004; Meriluoto and Spoof, 2005). Additionally, MCLR concentrations were confirmed by HPLC-DAD analysis according to Lawton et al. (Lawton et al., 1994). MCLR and Adda were sterilized by filtration using a 0.22 µm filter (Millex-GV, sterile; Millipore, Ireland). Colorimetric protein phosphatase inhibition (cppia) assay with recombinant PP1 and 2A Adda and MCLR were diluted serially (1:3) from 55.7 µm to 0.21 nm and 500 nm to 7.63 pm, rspectively, and analyzed at least four times in duplicates. The assay was carried out as described by Herersztyn and Nicholson (Heresztyn and Nicholson, 2001) using PP 1 (rabbit skeletal muscle, recombinant (E. coli), New England Biolabs, USA) and PP 2A (isolated from human red blood cells, Promega, USA) in final concentrations of 3 units/ml and 1.5 units/ml, respectively. P-nitrophenylphosphate from Acros Organics, USA, was used as substrate for the PPs. Statistics Colorimetric protein phosphatase inhibition assays were carried out at least four times in duplicates. The mean values of each duplicate yielded the values for calculation of the standard deviation (n>3). For calculation of the respective R 2, EC 50 values and statistical analysis GraphPad Prism 4.03 software was used. Briefly, the respective mean values were logarithmized and normalized. Resulting curves were fitted by nonlinear regression. An F-test (P<0.05) was employed for for the comparison of the EC 50 values, hillslopes and curves. 38

39 Chapter III Inhibitory capacity of Adda on PP1 and 2A 3.4 Results The results obtained from the cppias clearly demonstrate that Adda neither reduced phosphatase activity in PP1 (Fig. 3.1 (A)) nor PP2A (Fig. 3.1 (B)) at the concentrations applied. On the contrary, MCLR concentration-dependently inhibited both phosphatases resulting in IC 50 values of 1.3 nm ( nm; R 2 = 0.78) for PP1 (Fig. 3.1 (A)) and 0.29 nm ( nm; R 2 = 0.95) for PP2A (Fig. 3.1 (B)). A % phosphatase activity [nm] MC-LR ADDA B % phosphatase activity [nm] MC-LR ADDA Fig 3.1: Inhibition of the catalytic subunit of recombinant PP1 (A) and PP2A (B) by Adda and MCLR. Values represent mean ± standard error of the mean of at least four independent analyses. 39

40 Chapter III Inhibitory capacity of Adda on PP1 and 2A 3.5 Discussion The capability of MCs and NODs of potently inhibiting serine/threonine-specific PPs, especially PP1 and PP2A has led to numerous, in part lethal, poisonings of animals and humans (Kuiper-Goodman et al., 1999). The i.p. LD 50 s (dose of toxin that kills 50% of the exposed animals) of both MCLR, the most prominent MC congener, and NOD were found to be 50 µg/kg mouse (Krishnamurthy et al., 1986; Eriksson et al., 1988). However, some MC and NOD congeners revealed to be less or even non-toxic as determined by mouse bioassays (i.p. administration) (Rinehart et al., 1994; Sivonen and Jones, 1999). Substitutions in the variable L-amino acids of MCs appeared to have only little effects on the toxicity except for MCRR whose LD 50 was found to be about one order of magnitude higher than the corresponding MCLR congener (Krishnamurthy et al., 1986; Watanabe et al., 1988). On the other hand, modifications in the Adda residue such as isomerization of the diene from 6(E) to 6(Z) rendered MCs and NODs non-toxic (Harada et al., 1990a; Harada et al., 1990b; Namikoshi et al., 1994). Nishiwaki-Matsushima et al. (Nishiwaki- Matsushima et al., 1991) reported the inhibitory activity of 6(Z)-Adda-MCLR and -MCRR on PP2A to be 100-times weaker than of the corresponding maternal MCLR and MCRR. The authors concluded that the Adda moiety is not only crucial for toxicity, but for phosphatase inhibition as well. We therefore examined the inhibitory activity of isolated Adda on PP1 and PP2A in comparison with MCLR. Our results clearly demonstrate that Adda neither reduced activity of PP1 nor PP2A even at µm. By contrast, MCLR concentration-dependently inhibited both phosphatases in the low nanomolecular range as expected. This is supported by the findings of Harada et al. (Harada et al., 2004): isolated Adda showed no inhibitory effects on PP1 (at 10 µm) and revealed to be non-toxic in a mouse bioassay at concentrations up to 10 mg/kg body weight (i.p.) in contrast to the native toxin MCLR. In conclusion, the inhibitory activity and hence toxicity of MCs and NODs relies not on Adda itself, but requires the remaining structural units of the maternal toxins. However, even though Adda is ineffective in PP inhibition, it remains to be elucidated wether it may induce any other effects in exposed organisms. 40

41 Chapter III Inhibitory capacity of Adda on PP1 and 2A Acknowledgements We would like to acknowledge the Arthur & Aenne Feindt Foundation (Hamburg, Germany) for kindly funding parts of this study. 41

42 Chapter IV OATP-mediated MC Congener Toxicity 4 The role of organic anion transporting polypeptides (OATPs/SLCOs) in the toxicity of different microcystin congeners in vitro: A comparison of primary human hepatocytes and OATP-transfected HEK293 cells Fischer A 1, Hoeger SJ 1, Stemmer K 1, Feurstein D 1, Knobeloch D 2, Nussler A 3, Dietrich DR 1 1 Human and Environmental Toxicology, University of Konstanz, Konstanz, Germany 2 Department of General, Visceral, and Transplant Surgery, Charité Campus Virchow, Berlin, Germany 3 Technical University Munich, Department of Traumatology, Munich, Germany Published in Toxicology and Applied Pharmacology (2010) 4.1 Abstract Cellular uptake of microcystins (MCs), a family of cyclic cyanobacterial heptapeptide toxins, occurs via specific organic anion transporting polypeptides (OATPs), where MCs inhibit serine/threonine specific protein phosphatase (PP). Despite comparable PP-inhibitory capacity, MCs differ greatly in their acute toxicity, thus raising the question whether this discrepancy results from MCspecific toxikokinetic rather than toxicodynamic differences. OATP mediated uptake of MC congeners MCLR, -RR, -LW and -LF was compared in primary human hepatocytes and HEK293 cells stably expressing recombinant human OATP1B1/SLCO1B1 and OATP1B3/SLCO1B3 in presence/absence of OATP substrates taurocholate (TC) and bromosulfophthalein (BSP) and measuring PP-inhibition and cytotoxicity. Control vector expressing HEK293 were resistant to MC cytotoxicity, while TC and BSP competition experiments reduced MC 42

43 Chapter IV OATP-mediated MC Congener Toxicity cytotoxicity in HEK293-OATP transfectants, thus confirming the requirement of OATPs for trans-membrane transport. Despite comparable PP-inhibiting capabilities, MCLW and -LF elicited cytotoxic effects at lower equimolar concentrations than MCLR and MCRR, hence suggesting congener selective transport into HEK293-OATP transfectants and primary human hepatocytes. Primary human hepatocytes appeared one order of magnitude more sensitive to MC congeners than the corresponding HEK293-OATP transfectants. Although the latter maybe due to a much lower level of PPs in primary human hepatocytes, the presence of OATPs other than 1B1 or 1B3 may have added to an increased uptake of MCs. In view of the high sensitivity of human hepatocytes and currently MCLR-only based risk calculations, the actual risk of human MC-intoxication and ensuing liver damage could be underestimated in freshwater cyanobacterial blooms where MCLW and -LF predominate. Keywords: cyanobacteria, microcystin, OATP 4.2 Introduction Microcystins (MCs) are toxic metabolites produced by several cyanobacterial (blue green algae) species found in almost every environment worldwide. They are cyclic heptapeptides consisting of >80 structural congeners with a size between Dalton (Sivonen and Jones, 1999; Spoof, 2005; Zurawell et al., 2005; Humpage, 2008). Their general structure is cyclo(-d-ala 1 -L-X 2 -D-erythro-β-methylAsp 3 -L-Z 4 - Adda 5 -D-Glu 6 -N-methyldehydro-Ala 7 ) in which Adda stands for the unique D- amino acid 3-amino-9-methoxy-2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid and X and Z for variable L-amino acid residues (Botes et al., 1984; Botes et al., 1985; Rinehart et al., 1988; Rinehart et al., 1994). The MC congeners differ primarily in these two L-amino acids, a trait also used for their nomenclature (Carmichael et al., 1988a), and secondarily in slight modifications of the amino acids of the cyclic heptapeptide backbone e.g. absence or additional methyl groups etc. 43

44 Chapter IV OATP-mediated MC Congener Toxicity MCs are very potent inhibitors for serine/threonine-specific protein phosphatases (PPs), especially PP1 and PP2A, as well as PP3 - PP6, whereas no inhibition of PP2B, PP2C and PP7 was observed (MacKintosh, 1993; Honkanen et al., 1994; Runnegar et al., 1995a; Toivola et al., 1997; Hastie et al., 2005). MC-mediated inhibition of PPs results in hyperphosphorylation of numerous phosphate-regulated enzymes and subsequent deregulation of fundamental cellular processes, e.g. disruption of the cytoskeleton (Yoshizawa et al., 1990; Eriksson et al., 1990a; Ohta et al., 1992; Wickstrom et al., 1995; Batista et al., 2003). The inhibitory capacity of single MC congeners on PP1 and PP2a in vitro are comparable, with IC 50 values in the sub- and lower nanomolar range (Robillot and Hennion, 2004; Hoeger et al., 2007; Monks et al., 2007), suggesting a highly conserved molecular mode of action. Surprisingly, the LD 50 values (mice, i.p.) for various MC congeners differ distinctly from one another, as summarized in Sivonen and Jones (Sivonen and Jones, 1999). However, the overt discrepancy between similar in vitro PP inhibition (in situ toxicodynamics) and differing in vivo (i.p.) acute toxicity remains unresolved to date. This led to the initial hypothesis of this study that the toxicity of single MCs is primarily a result of their individual toxicokinetics (absorption and distribution) and not due to differences in toxicodynamics (PP-inhibition). Indeed, as a result of their structure and amino acid composition, MCs are rather hydrophilic and spatially large molecules. Hence, they apparently are incapable of crossing cell membranes via passive diffusion, but rather require active transport via specific transporters, as suggested by indirect experiments (Eriksson et al., 1990b; Runnegar et al., 1995c) employing substrates for and inhibitors of organic anion transporting polypeptides (OATPs). Fischer et al. (Fischer et al., 2005) demonstrated that human OATP1B1, OATP1B3, OATP1A2 and rat Oatp1b2 [human: OATPs/SLCOs; animals: Oatps/Slcos; (protein name/gene symbol) (Hagenbuch and Meier, 2004)] are capable of transporting [ 3 H]-dihydromicrocystin-LR. The active uptake of native nonlabeled MCLR and other MC congeners via OATPs was confirmed by using cells stably expressing OATP1B1 and OATP1B3, respectively (Komatsu et al., 2007; Monks et al., 2007). Recently, Lu et al. (Lu et al., 2008) demonstrated a strongly decreased MCLR mediated hepatotoxicity in Oatp1b2-null mice suggesting this knock-out model as useful for interpreting the importance of this 44

45 Chapter IV OATP-mediated MC Congener Toxicity transporter s human orthologs OATP1B1 and OATP1B3 in the distribution of MCs and their ensuing potential toxicity. Indeed, members of the multispecific OATP family can be detected in nearly all tissues of humans, rodents and other animals. They play an important role in the absorption, distribution and excretion of numerous xenobiotica (Hagenbuch and Meier, 2003; Hagenbuch and Meier, 2004). While the human MCs-transporting OATP1B1 and OATP1B3, are exclusively located at the sinusoidal (basolateral) membrane of hepatocytes, OATP1A2 is located in the membrane of liver- and kidney-cells, as well as the blood-brain-barrier (Meier and Stieger, 2002; Hagenbuch and Meier, 2003; van Montfoort et al., 2003; Hagenbuch and Meier, 2004; Mikkaichi et al., 2004). Therefore, using HEK293 cells stably expressing liver specific OATP1B1 and OATP1B3 and primary human hepatocytes, this study determined whether the toxicity of single MCs (MCLR (L-leucine, L-arginine), MCRR (L-arginine, L- arginine), MCLW (L-leucine, L-tryptophan) and MCLF (L-leucine, L- phenylalanine)) is a result of their individual toxicokinetics. To determine the toxicodynamic characteristics (PP inhibition) of the four MC congeners, colorimetric protein phosphatase inhibition assays with recombinant PP1 and PP2A and homogenates of OATP1B1 and -1B3 expressing HEK293 cells were employed. Okadaic acid (OA), known to inhibit serine/threonine-specific protein phosphatases with similar potency as MCs (MacKintosh, 1993; Honkanen et al., 1994; Runnegar et al., 1995a; Toivola et al., 1997; Hastie et al., 2005), yet capable of transport independent trans-membrane diffusion, was employed as a toxicodynamic and -kinetic positive control. 4.3 Material & Methods Chemicals and reagents All chemicals were of the highest analytical grade commercially available. Standards of MCLR, MCRR, MCLW and MCLF were obtained from Alexis (Switzerland) and dissolved in 75% MeOH. The concentrations of the stock and the working solutions were confirmed photometrically using the molar absorption coefficient of MCLR and MCRR (39800 mol l -1 cm - 1 ) published by 45

46 Chapter IV OATP-mediated MC Congener Toxicity Harada et al. (Harada et al., 1990a). Although this is the molar absorption coefficient for MCLR/-RR dissolved in 100% MeOH it turned out to be applicable for 75% MeOH as well (Meriluoto et al., 2004; Meriluoto and Spoof, 2005). In the absence of published molar absorption coefficients for MCLW and MCLF, the molar absorption coefficient for MCLR and MCRR was applied. Additionally, concentrations were confirmed by HPLC-DAD analysis according to Lawton et al. (Lawton et al., 1994). Okadaic acid (OA) (Sigma, Germany) was dissolved in 100% H 2 O, diluted to the stock and working concentrations according to the manufacturer s specifications. All toxins were sterile filtrated using a 0.22 µm filter (Millex-GV, sterile; Millipore, Ireland). Cell systems Human primary hepatocytes were isolated in agreement with the ethical review board and after the patients written consent by a standard operating procedure and cultured as described previously (Nussler et al., 2009). All four donors were females. Donor 1: partial liver resection due to cholangiocarcinoma (CCC), born 1961; donor 2: partial liver resection due to liver metastasis of primary mamma carcinoma, born 1965; donor 3: partial liver resection due to liver metastasis of primary colorectal carcinoma, born 1962; and donor 4: partial liver resection due to liver metastasis of primary colorectal carcinoma, born Human embryonic kidney cells (HEK293) stably transfected with recombinant human organic anion transporting polypeptides 1B1 (HEK293-OATP1B1) and 1B3 (HEK293-OATP1B3), or control vector (HEK293-CV) were kindly provided by Prof. Dietrich Keppler (Division of Tumor Biochemistry, German Cancer Research Centre, Heidelberg, Germany). HEK293 cells were cultured in minimal essential medium (MEM) with Earle s Salts and L-Glutamine supplemented with 10% FBS, 100 units/ml penicillin, 100 mg/ml streptomycin and 400 µg/ml G418-sulphate at 37 C and 5% CO 2. All cell culture media and supplements were purchased from PAA Laboratories GmbH (Austria). Toxin exposure Passages 3-8 of transfected HEK293 cells were used for toxin exposure experiments. HEK293 cells were seeded in 96-well plates in MEM (supplemented as described above) at a density of 3 x 10 5 cells/ml and

47 Chapter IV OATP-mediated MC Congener Toxicity µl/well (6 x 10 4 cells/well). Prior to seeding 96-well plates were coated with poly- L-lysine (5 mg/ml). After 4-5 h the medium was decanted and cells incubated in 200 µl MEM (supplemented as described above, albeit with 1% FBS). Cells were incubated with serially diluted toxins (1:3); MCLR and -RR ranging from 5000 nm to 2.29 nm, MCLW and -LF ranging from 200 nm to 0.09 nm and OA ranging from 93 nm to 1.15 nm, respectively. For toxin exposure experiments human primary hepatocytes of three different donors were used. Primary hepatocytes from donors 1 and 2 were employed to assess donor-specific differences in MC mediated cytotoxicity (see below). Cells were seeded in collagen-coated 96-well plates (Price, 1975) at a density of 3 x 10 5 cells/ml and 200 µl/well (6 x 10 4 cells/well). After arrival the cells were incubated at 37 C and 5% CO 2 for 4-5 h. Prior to exposure the medium was decanted and replaced by 200 µl RPMI 1640 containing 100 units/ml penicillin and 100 mg/ml streptomycin. Cells were incubated with serially diluted toxins (1:3); MCLR and -RR ranging from 5000 nm to 2.29 nm, MCLW, -LF and OA ranging from 200 nm to 0.09 nm. Coincubation studies with bromosulfophthalein and taurocholate Co-incubation studies of MCLR and taurocholate (TC) or bromosulfophthalein (BSP) in human primary hepatocytes of donor 3 and HEK293 transfectants were performed under identical conditions as described for single MC congener exposures (vide supra: toxin exposure) with the following modifications: Primary human hepatocytes of donor 3 were seeded at a density of 2.25 x 10 6 cells/ml and 200 µl/well (4.5 x 10 5 cells/well) and exposed 4-5 h subsequent to seeding. HEK293 cells were exposed 4-5 h subsequent to seeding. Primary hepatocytes and HEK293 cells were exposed to concentrations of TC and BSP at 500 µm and µm and 100 µm and 33.3 µm, respectively, thus serving as OATP substrate controls. Competition experiments in primary hepatocytes and HEK293 cells were carried out with 5 µm MCLR in combination with 500 µm, 50 µm and 5 µm TC or 100 µm, 50 µm and 5 µm BSP. Based on the MCLR EC 50 values obtained in HEK293 transfectants, additional MCLR concentrations were used for TC and BSP competition experiments: 214 nm MCLR (EC 50 for HEK293-OATP1B1) for HEK293 cells transfected with OATP1B1 and nm (EC 50 for HEK293-47

48 Chapter IV OATP-mediated MC Congener Toxicity OATP1B3) for HEK293 cells transfected with OATP1B3. Since no EC 50 could be determined for CV-transfected HEK293 cells, they were exposed to 214 and nm MCLR in combination with TC or BSP. MTT reduction assay Following 48 h exposure of HEK293 cells or primary human hepatocytes to toxin, 20 µl MTT (5mg/ml) solution was added to each well. Following incubation of the cells with the MTT solution for 1.5 h, medium was carefully removed by pipetting. To re-dissolve MTT- formazan, 100µl solubilization buffer (95% (v/v) isopropanol, 5% (v/v) formic acid) was added to each well. After a minimum of 15 min careful shaking, absorption was determined at 550 nm in a microtitre plate reader (Tecan, microplate reader, infinite M200; Austria). As positive control (0% survival) cells were incubated with 1% TWEEN (8 wells/96-well plate). Cells incubated with medium only (16 wells/96-well plate) were taken as negative control (100% survival). The highest MeOH and water concentration in the assay was <2% and used as solvent control (16 wells/96- well plate). No differences in viability, condition or growth rate could be identified between solvent and negative control (data not shown). Colorimetric protein phosphatase inhibition (cppia) assay The assay was carried out with MCLR, -RR, -LW, -LF and okadaic acid (OA) in concentrations diluted serially (1:3) from 600 nm to 0.03 nm and 400 nm to 0.02 nm, respectively. MC congener-specific protein phosphatase inhibition in PP1 (rabbit skeletal muscle, recombinant (E. coli), New England Biolabs, USA) and PP2A (isolated from human red blood cells, Promega, USA) were determined as described by Herersztyn and Nicholson (Heresztyn and Nicholson, 2001) with the following minor modifications: both PP2A and PP1 were employed in stock concentrations of 500 units/ml and 2,500 units/ml, respectively, which resulted in final assay concentrations of 1.5 units/ml for PP2A and 3 units/ml for PP1. P-nitrophenylphosphate from Acros Organics, USA, was used as substrate for the protein phosphatases. The absorption was measured at 405 nm in a microtitre plate reader (Tecan, microplate reader, infinite M200; Austria). 48

49 Chapter IV OATP-mediated MC Congener Toxicity The assay was further modified when PPs of human primary hepatocytes homogenates (a mixture of donor 1 and 2 hepatocytes) or homogenates of HEK293-CV, -OATP1B1 and -OATP1B3 were used. After washing three times with PBS, cells were extracted in modified enzyme solution, containing H 2 O instead of BSA, for 10 min on ice while being re-suspended with a syringe and/or a pipette. Cell extract protein concentration was determined by the method of Bradford (Bradford, 1976) (Bio-Rad Protein Assay; Bio-Rad, Germany). Briefly, in a preliminary assay, homogenates of human primary hepatocytes, HEK293-OATP1B1, HEK293-OATP1B3 and HEK293-CV were taken through a dilution series with enzyme solution in order to approach the highest protein phosphatase activity applicable which is restricted by the maximal optical density. For the final optimized inhibition assay 3 µg/µl of total protein of human primary hepatocytes and 0.4 μg/µl of total protein of each respective HEK293 transfectant were employed. Homogenates were incubated with the respective inhibitors for 30 min at 37 C. Homogenates of the HEK293 cells were incubated with all four MC congeners and OA using the same concentration ranges as described above. Similarly, homogenates of human primary hepatocytes were incubated with MCLR and OA only and in concentrations diluted serially (1:3) from 600 nm to 2.47 nm and 400 nm to 1.65 nm, respectively, due to limited availability of primary hepatocytes from donors 1 and 2. Immunoblot detection of PP1 and PP2A Immunoblot detection was carried out as described previously (Feurstein et al., 2009) using 60 µg of total protein of primary hepatocytes from donor 4, as well as with total protein from a mixture of donors 1 and 2 (see Fig. 1S, supplemental material). Monoclonal mouse antibodies anti-pp1α (P7607; Sigma, Germany) and anti-pp2a (05-545; Millipore, Germany) were diluted 1:2000 and 1:500, respectively. The secondary HRP-labeled antibody rabbit anti-mouse (P0260; DakoCytomation, Germany) was applied in a dilution of 1:1000. GAPDH served as house-keeping protein control and was detected with a 1:200 dilution of rabbit polyclonal anti-gapdh (sc-25778, Santa Cruz Biotechnology Inc., USA). The secondary HRP-labeled antibody goat anti-rabbit (A0545; Sigma, Germany) was applied in a dilution of 1:

50 Chapter IV OATP-mediated MC Congener Toxicity Immunofluorescence detection of OATP1B1 and OATP1B3 Primary human hepatocytes (mixture of donor 1 and 2) were seeded onto collagen-coated cover slips (Price, 1975) in 24-well-plates at a density of 0.5 x 10 6 cells/well. HEK293 cells were seeded on poly-l-lysin coated cover slips (5 mg poly-l-lysin/ml) in 6-well-plates at the same density. After three washes with PBS cells were fixed by incubating with ice-cold Acetone/MeOH (1:1) for 5 min. Following removal of the fixative cells were dried on ice for another 5 min and subsequently stored at -20 C until further treatment. Prior to incubation with the polyclonal primary antibodies rabbit anti- OATP1B1 (König et al., 2000a) and rabbit anti -OATP1B3 (König et al., 2000b), both kindly provided by Dietrich Keppler (Division of Tumor Biochemistry, German Cancer Research Centre, Heidelberg, Germany), for 1 h at room temperature cells were washed three times with PBS for 10 min each. These washing steps were repeated prior to and again after incubation with the TRITC-labeled secondary antibody goat anti-rabbit IgG (T6778; Sigma, Germany) in the dark for another hour at room temperature. Finally, cells were washed once with H 2 O, mounted with Fluorescent Mounting Medium (Dako, Germany) onto a slide and kept dark at 4 C until visualizing using a confocal laser microscope (LSM 510 META, Zeiss, Germany). Statistics Colorimetric PP-inhibition assays and cytotoxicity studies were carried out 3 times in duplicates. Mean values and standard deviations were calculated based on the mean values of the replicates (n 3). For calculation of the respective R 2, EC 50 values and statistical analysis GraphPad Prism 5 software was used. Briefly, the respective mean values were log-transformed and normalized. The resulting curves were fitted by nonlinear regression. An F-test (P<0.05) was employed for the comparison of the EC 50 values, hillslopes and curves. Significant differences in cytotoxicity of individual microcystin-congener concentrations in comparison to concurrent controls were determined via Oneway ANOVA followed by Dunnett s Multiple Comparison Test (p<0.05). Co-incubation studies were performed three to seven times in duplicates. The mean values of each duplicate yielded the values for calculation of the standard deviation (n 3). Significance of effect was determined via One-way ANOVA 50

51 Chapter IV OATP-mediated MC Congener Toxicity followed by Tukey's Multiple Comparison Test was employed to assess significant differences between all groups (controls and exposures with/without competitor). 4.4 Results Comparable toxicodynamics of different microcystin congeners The MC congener IC 50 values obtained with PP1A (Tab. 4.1; Fig. 4.1 (A)) suggested a significant, albeit very small, difference in the mean phosphatase inhibiting capacity of the 4 different congeners, MCLR having highest activity followed by -RR, -LW and -LF, as shown by the non-overlapping 95% confidence intervals of the IC 50 values of the individual congeners (Tab. 4.1). In contrast, the four different congeners did not demonstrate significantly different PP-inhibiting capabilities in recombinant PP2A (Tab. 4.1; Fig. 4.1 (B)). However, all four MC congeners inhibited the recombinant PP1 and PP2A half-maximally between nm, suggesting overall a comparable PP-inhibiting capability, thus toxicodynamic property, irrespective of the MC congener and PP employed. A % phosphatase activity c [nm] B % phosphatase activity MC-LR MC-RR MC-LW MC-LF c [nm] Fig. 4.1: Inhibition of the catalytic subunit of recombinant PP1 (A) and PP2A (B) by MCLR, -RR, -LW and -LF. Values represent mean ± standard error of the mean of 3 independent analyses. 51

52 Chapter IV OATP-mediated MC Congener Toxicity Tab. 4.1: IC 50, EC 50 values with 95% confidence intervals in brackets and LD 50 (i.p. mouse) of the investigated MC congeners. IC 50 [nm] EC 50 [nm] LD 50 [nmol/kg] PP1 PP2A primary human hepatocytes (Donor 1) primary human hepatocytes (Donor 2) HEK293- OATP1B1 HEK293- OATP1B3 mouse (i.p.) MCLR 1.22 ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = y1 MCRR 1.48 ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = > ( ) R 2 = y2 MCLW 1.86 ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = nda MCLF 1.84 ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = ( ) R 2 = nda nda, no data available y 1, (Krishnamurthy et al., 1986; Lovell et al., 1989; Stoner et al., 1989) y 2, (Watanabe et al., 1988; Stoner et al., 1989) 52

53 Chapter IV OATP-mediated MC Congener Toxicity Similarly no MC congener-dependent differences in PP- inhibition could be observed in the respective homogenates of HEK293-CV, HEK293-OATP1B1, and HEK293-OATP1B3 (Fig. 4.2 (A - D)), thus corroborating the findings observed in the recombinant PPs. A comparable PP-inhibition curve was obtained with the positive control, OA (Fig. 4.2 (E)). Of the total intracellular PP activity (100%) present in the homogenates of HEK293 transfectants and primary human hepatocytes (a hepatocyte mixture of donor 1 and 2), MC congeners were capable of reducing PP activity to approx. 70% and to less than 90%, respectively, thus suggesting a dramatically lower amount of serine/threonine-specific PPs in the primary human hepatocytes (Fig. 4.2 (F)). A limited quantity of serine/threonine-specific PPs in the human hepatocytes is also suggested by the fact that no dose-response was obtained, but rather already the smallest concentrations applied already reduced PP activity maximally (to <90% of the total PP activity present). Above interpretation is supported by PP1 and PP2A immunoblot analysis in homogenates of primary human hepatocytes and HEK293 transfectants. Whereas, HEK293 transfectants presented with comparable bands of PP1 and PP2A (supported also by densitometric analysis using the PP:GAPDH density ratio), only faint PP1 and marginal PP2A could be detected in the homogenates of the primary human hepatocytes of donor 4 (Fig. 4.3). Indeed, densitometric analyses suggested that quantities of PP1 and PP2A in the primary hepatocytes of donor 4 were only 35% and 8% of the values observed in the HEK293 transfectants (Tab. 4.1 (S), supplemental material). 53

54 Chapter IV OATP-mediated MC Congener Toxicity A % phoshphatase activity MCLR [nm] B % phosphatase activity CV OATP1B1 OATP1B MCRR [nm] C % phosphatase activity MCLW [nm] D % phosphatase activity CV OATP1B1 OATP1B MCLF [nm] E % phosphatase activity CV OATP1B1 OATP1B OA [nm] F % phosphatase activity MCLR OA c [nm] Fig. 4.2: Inhibition of intracellular serine/threonine-specific PPs of HEK293-CV, - OATP1B1 and -OATP1B3 cell homogenates by MCLR (A), MCRR (B), MCLW (C), MCLF (D) and OA (E). Inhibition of intracellular serine/threonine-specific PPs of primary human hepatocyte cell homogenates (mixture of cells from donor 1 and 2) by MCLR and OA (F). Values represent mean ± standard error of the mean of 3 independent replicate analyses. 54

55 Chapter IV OATP-mediated MC Congener Toxicity Fig. 4.3: Immunoblot detection of PP1, PP2A and GAPDH in primary human hepatocytes (donor 4), HEK293 cells transfected with a control vector, OATP1B1 and OATP1B3. Equal amounts of total protein (60 µg/lane) were applied. Molecular weights were estimated by comparison with marker proteins. Due to the low detection level of PP1 and PP2A in primary human hepatocytes the respective images were overexposed. Microcystin congener dependent cellular uptake in HEK293-transfectants The presence of OATP1B1 and OATP1B3 was confirmed immunocytochemically in plasma membranes of primary human hepatocytes (Donor 2; Fig. 4.4 (A) and (B)) and HEK293 transfectants (Fig. 4.4 (C) and (D)). However, the immunofluorescence signal appeared to be more intensive in the OATP transfected HEK293 cells than in the primary human hepatocytes. No cross-reactivity of the 2 OATP-antibodies was observed (data not shown). As expected no OATP-immunopositive signal was detected in HEK293 cells transfected with the control vector (Fig. 4.4 (E) and (F)). Cells transfected with control vector showed no cytotoxicity irrespective of the MC congener or concentrations employed, whereas the cell-permeant positive control, OA, demonstrated the expected cytotoxicity (Fig. 4.5 (A)). Indeed, nearly identical dose-response curves were obtained with OA in all HEK293 transfectants (Fig. 4.5 (A - C)), although the EC 50 values varied slightly with 9.5 nm ( nm; R 2 = ) for HEK293-CV, 10.5 nm ( nm; R 2 = ) for HEK293- OATP1B1 and 7.6 nm ( nm; R 2 = ) for HEK293-OATP1B3. 55

56 Chapter IV OATP-mediated MC Congener Toxicity anti-oatp1b1 anti-oatp1b3 A: primary human hepatocytes B: primary human hepatocytes C: HEK293-OATP1B1 D: HEK293-OATP1B3 F: HEK293-CV E: HEK293-CV Fig. 4.4: Immunofluorescence detection of OATP1B1 in plasma membranes of primary human hepatocytes of donor 2 (A), HEK293-OATP1B1 (C) and HEK293-CV (E) and of OATP1B3 in plasma membranes of primary human hepatocytes of donor 2 (B), HEK293-OATP1B3 (D) and HEK293-CV (F). Note the absence of staining in HEK293-CV when stained with OATP1B1 and OATP1B3 antibody. 56

57 Chapter IV OATP-mediated MC Congener Toxicity In contrast, MC congener dependent concentration-response curves were observed in OATP1B1-, as well as OATP1B3-transfected cells (Fig. 4.5 (B) and (C)). The EC 50 values for HEK293-OATP1B1 ranged from 10.4 nm for MCLW to beyond 5000 nm for MCRR. For HEK293-OATP1B3, the EC 50 values ranged from 3.7 nm for MCLF to 1267 nm for MCRR (Tab. 4.1). The EC 50 values of the more hydrophobic MC congeners LW and LF were comparable amongst the transfectants, whereas the EC 50 values of MCLR and -RR differed distinctly from each other. Although, MCLR gave a comparable dose-response curve in both HEK293-OATP-transfectants, HEK293-OATP1B1 cells appeared to be less susceptible to MCs than HEK293-OATP1B3. Indeed, MCLW and -LF are taken up much more rapidly or with higher affinity (up to 2 orders of magnitude) by HEK293-OATP1B1 and HEK293-OATP1B3 cells than MCLR or RR. HEK293-OATP1B1 cells do not appear to transport MC-RR, whereas HEK293- OATP1B3 cells do transport MC-RR at approximately one order of magnitude less efficiently than MCLR and up to 4 orders of magnitude less efficiently than MCLW and -LF. While, MCLR had no effect on HEK293-CV-transfectants, high concentrations of TC (500 µm) in absence/presence of MCLR lead to significant but limited cytotoxicity (Fig. 4.6 (A)). In contrast, none of the HEK293-OATP transfectants were sensitive to TC cytotoxicity, whereas HEK293-OATP1B1 were susceptible to 100 µm BSP and HEK293-OATP1B3 were susceptible to 100 and 50 µm BSP (Fig. 4.6 (C - F)). A concentration of 33.3 µm BSP did not reduce viability in either OATP transfectants, albeit the HEK293-OATP1B3 cells appeared more sensitive towards BSP mediated cytotoxicity than the OATP1B1 transfectants. MCLR competition experiments with the known OATP substrates TC and BSP demonstrated a reduction of MCLR cytotoxicity in both HEK293-OATPtransfectants. However, TC afforded protection from MCLR mediated cytotoxicity at low concentrations of MCLR only. BSP, on the other hand, was protective in 5µM MCLR competition incubations at 100 and 50µM BSP in HEK293-OATP1B1 (Fig. 4.6 (D)), while a 5 µm BSP concentration was no longer protective. Similarly BSP was protective in HEK293-OATP1B1 at lower MCLR concentrations. In contrast, due to the high cytotoxicity of BSP in 57

58 Chapter IV OATP-mediated MC Congener Toxicity HEK293-OATP1B3 cells, only the non-cytotoxic 5 µm BSP concentrations afforded protection from nm MCLR mediated cytotoxicity (Fig. 4.6 (F)). A viability [% negative control] B concentration [nm] MCLR MCRR MCLW MCLF OA viability [% negative control] concentration [nm] MCLR MCRR MCLW MCLF OA C viability [% negative control] concentration [nm] MCLR MCRR MCLW MCLF OA Fig. 4.5: Cytotoxicity of MCLR, -RR, -LW, -LF and OA in HEK293-CV (A), HEK293- OATP1B1 (B) and HEK293-OATP1B3 (C). Negative control = medium control. Values represent mean ± standard error of the mean of 3 independent experiments. 58

59 Chapter IV OATP-mediated MC Congener Toxicity Fig. 4.6: Competitive inhibition of MCLR uptake in HEK293-CV (A and B), HEK293- OATP1B1 (C and D) and HEK293-OATP1B3 (E and F) following co-incubation with 59

60 Chapter IV OATP-mediated MC Congener Toxicity TC and BSP. Negative control = medium control. Values represent mean ± standard error of the mean of 7 independent experiments. Statistics: One-way ANOVA followed by Tukey s Multiple Comparison Test was employed to assess significant differences between all groups (controls and exposures with/without TC or BSP). A. a: significantly different (p<0.001) from the negative control, b: significantly different (p<0.001) from 5µM MCLR, c: significantly different (p<0.05) from 214 nm MCLR + 5 µm TC; B. a: significantly different (p<0.05) from the negative control; C. a: significantly different (p<0.05) from the negative control and TC-only groups, b: significantly different (p<0.001) from 5 µm MCLR, c: significantly different (p<0.001) from all 214 nm MCLR + TC groups, d: significantly different (p<0.01) from the negative control and the 50 and 5 µm TC-only groups; D. a: significantly different (p<0.01) from the negative control and BSP-only groups, b: significantly different (p<0.05) from 50, 33, and 5 µm BSP-only groups, c: significantly different (p<0.001) from 5 µm MCLR, d: significantly different (p<0.001) from 5 µm MCLR + 5 µm BSP;, e: significantly different (p<0.001) from the 214 nm MCLR and the 214 nm MCLR µm BSP groups; E. a: significantly different (p<0.05) from the negative control and TC-only groups, b: significantly different (p<0.001) from nm MCLR; F. a: significantly different (p<0.05) from the negative control and BSPonly groups, b: significantly different (p<0.01) from nm MCLR µm BSP. Microcystin congener dependent cellular uptake in primary human hepatocytes As observed in the HEK293-transfectants, MC congener dependent cytotoxicity was observed in the primary hepatocytes of both donors (Fig. 4.7), whereby MCLW and -LF were the most, MCLR intermediate, and MCRR the least cytotoxic of the congeners (Tab. 4.1). Differences were observed with regard to the relative sensitivity of the donor 1 and 2 hepatocytes in that hepatocytes of donor 2 were approximately one order of magnitude less sensitive to MCLR, - RR and OA (Tab. 4.1 and Fig. 4.7 (A) and (B)). Indeed exposure to the cellpermeant OA yielded EC 50 values of 3.0 nm ( nm; R 2 = ) and 35.1 nm ( nm; R 2 = ) in donors 1 and 2, respectively. Similar to the observations made with HEK293-OATP transfectants, no protective effect of TC was found in primary human hepatocytes of donor 3 when co-exposed to 5 µm MCLR (Fig. 4.8 (A)). TC itself proved to be cytotoxic to primary hepatocytes at concentrations µm. Moreover, BSP at concentrations 33.3 µm was cytotoxic to primary human hepatocytes (Fig. 4.8 (B)). However, despite the observed BSP inherent cytotoxicity, BSP afforded some protection against 5 µm MCLR induced toxicity, albeit only at 100 µm BSP. 60

61 Chapter IV OATP-mediated MC Congener Toxicity A viability [% negative control] B concentration [nm] MCLR MCRR MCLW MCLF OA viability [% negative control] concentration [nm] MCLR MCRR MCLW MCLF OA Fig. 4.7: Cytotoxicity of MCLR, -RR, -LW, -LF and OA in primary human hepatocytes of donor 1 (A) and donor 2 (B). Negative control = medium control. Values represent mean ± standard error of the mean of 3 independent experiments. 61

62 Chapter IV OATP-mediated MC Congener Toxicity Fig. 4.8: Competitive inhibition of MCLR uptake in primary human hepatocytes of donor 3 following co-incubation with TC (A) and BSP (B). Negative control = medium control. Values represent mean ± standard error of the mean of four independent experiments. Statistics: One-way ANOVA followed by Tukey s Multiple Comparison Test was employed to assess significant differences between all groups (controls and exposures with/without TC or BSP). A. a: significantly different (p<0.05) from the negative control, b: significantly different (p<0.001) from the 500 and µm TC group, c: significantly different (p<0.05) from µm TC-only group; B. a: significantly different (p<0.001) from the negative control; b: significantly different (p<0.001) from the 100 µm BSP group group, c: significantly different (p<0.001) from the 33.3 µm BSP group, d: significantly different (p<0.001) from 5 µm MCLR µm BSP group. 4.5 Discussion Both, human primary hepatocytes and OATP-transfected HEK293 cells provided for decisive advantages over the more routinely employed HepG2 or Mz-Hep-1 cells. Indeed, within the first 96 hours of seeding human primary hepatocytes retain most if not all transporters (Jigorel et al., 2005) and drug metabolizing enzymes at levels comparable with those in situ (reviewed in (Li, 2001; Gomez-Lechon et al., 2003; Donato et al., 2008)). Hence, results obtained from human primary cells are of greater physiological relevance than those obtained from immortal or immortalized cell lines. Despite the latter, differences amongst donors 1-3 (age, genetic differences, health and dietary 62

63 Chapter IV OATP-mediated MC Congener Toxicity history etc.) can provide for data variability. In view of the latter, the same gender and comparable age groups of the donors, with the exception of donor 4 who belonged to an older age group, were employed for the experiments presented here. In contrast, the advantage of using OATP expressing cell models, e.g. the OATP transfected HEK293 cells, is the reproducibility and limited variability of results obtained as well as the absence of other possibly interfering transporters, thereby guaranteeing causality of transporter associated effects. Indeed, Keppler et al. (Keppler, 2006; personal communication) tested HEK293 cells for the presence of several membrane transporters. Neither background of endogenous OATP1B1 and OATP1B3 or of other endogenous OATPs was detected. Consequently the comparison of primary human hepatocytes with HEK293 cells expressing functional hepatic OATPs would allow understanding of the potential differences in MC congener mediated apical cytotoxicity. However, one of the prerequisites for such a comparison is that MC congeners are comparable in their capability of inhibiting endogenous serine/threonine- PPs. Consequently, apical cytotoxicity was solely a resultant of the amount of PP inhibiting MC being transported within the experimental exposure timeframe. Comparable PP inhibiting capability of the four MC congeners tested was indeed demonstrated for recombinant PP1 and PP2A (Tab. 4.1, Fig. 4.1) and for homogenates of all HEK293 cells (Fig. 4.2), thus corroborating earlier findings by Robillot et al. (Robillot and Hennion, 2004), Hoeger et al. (Hoeger et al., 2007) and, with the exception of MCRR, by Monks et al. (Monks et al., 2007). A limited supply of primary human hepatocytes from one and the same donor prevented a comparable MC congener comparison as carried out with the HEK293 OATP transfectants. However, a PP inhibition assay with a mixture of homogenates of donors 1 and 2 and MC and OA demonstrated PP inhibition albeit at much lower levels than observed in the HEK293 cells (Fig. 4.2 (F)), possibly suggesting lower amounts of ser/thr-pps in the primary human hepatocytes. Indeed, MCs and OA maximally reduced total PP activity by approximately 25-35% and 15%, respectively (Fig. 4.2 (A - E)) in HEK293 cells. In contrast, MC-LR and OA led to a maximum reduction of total PP activity of 10-15% in the primary human hepatocytes tested (Fig. 4.2 (F)). Given that 63

64 Chapter IV OATP-mediated MC Congener Toxicity MCs and OA are specific inhibitors of distinct ser/thr-specific PPs, whereas the employed p-nitrophenylphosphate (pnpp) is substrate for all PPs, these results would suggest that the ratios of ser/thr-pps to total PP activity was different in the HEK293 and the primary human hepatocytes employed. The latter interpretation is supported by the fact that: 1.) Total PP activity of primary human hepatocytes determined by the colorimetric protein phosphatase inhibition assays (cppias) was comparatively low, since 7.5 times more total protein was needed to get a similar pnpp turnover as in the HEK293 cells, suggesting decreased PP activities. 2.) Immunoblot detection of PP1 and especially PP2A in both cell types revealed distinctly weaker bands for primary hepatocytes of donor 4 (Fig. 4.3) and of the donor 1 and 2 mixture (see Fig. 4.1 (S), supplemental material), although equal protein amounts were employed. 3.) Densitometric analyses using PP to GAPDH ratios, suggested that PP1 and PP2A levels in primary human hepatocytes of donor 4 were only 35% and 8%, respectively, of the levels found in HEK293 transfectants (see Tab. 4.1 (S), supplemental material). Although the latter may also be due to differences in the protein phosphataseepitopes in HEK293 cells and primary hepatocytes, thereby altering primary antibody epitope recognition and binding specificities (Michalski et al., 2002; Letschert et al., 2004; Lee et al., 2005), the principle finding of the cppias strongly supports the finding that MC congeners have comparable PP inhibiting capacities in human cell lines (HEK293) as well as in primary human hepatocytes of donors 1 and 2. Thus differences in apical cytotoxicity observed amongst MC congeners or amongst cell lines or cell types appear to be primarily driven by the type and expression levels of OATPs, albeit differences in protein phosphatase type and expression levels cannot be ignored. Since the HEK293 were originally transfected by subcloning the OATP cdnas into the expression vector pcdna3.1 (+) (König et al., 2000a; König et al., 2000b), thus being under control of the same promoter (cytomegally virus), it may be assumed that the HEK293 cells were stably expressing comparable levels of recombinant human OATP1B1/SLCO1B1 and OATP1B3/SLCO1B8. The latter assumption was at least qualitatively supported by the 64

65 Chapter IV OATP-mediated MC Congener Toxicity immunofluorescence staining for the two OATP transporters (Fig. 4.4), thus allowing the assumption that any apical MC congener induced cytotoxicity is not the result of differences in PP and/or OATP expression amongst the two HEK293 transfectants. Indeed, the levels of PP expression following densitometric analyses appeared to be comparable amongst all HEK293 transfectants (Tab. 4.1 (S), supplemental material). MC cytotoxicity was congener-dependent (Tab. 4.1; Fig. 4.5) in both OATP-expressing HEK293 cells. Since no differences in intracellular PP inhibition among MC congeners were observed, it can be assumed that cytotoxicity directly reflects varying affinities and transporting capacities of the OATP expressed for the respective MCs. The more hydrophobic MC congeners LW and LF elicited the highest effects with similar 48-h EC 50 values each in the low nanomolar range. Remarkably, MCLR appeared to be approximately 20-times and 60-times less toxic than MCLW and -LF in OATP1B1- and OATP1B3-expressing HEK293 cells, respectively. MCRR was the least cytotoxic congener and elicited no effects in HEK293-OATP1B1 at all concentrations applied. HEK293-OATP1B3 were more than 300-times less susceptible to MCRR than to MCLW and -LF. No cytotoxicity was observed in control vector-transfected HEK293 cells, thus confirming the assumed OATP transport-dependent uptake with ensuing cytotoxicity for all four MC congeners. The latter findings confirmed the general trend observed in earlier cytotoxicity data in HeLa cells transiently transfected with OATP1B1 and OATP1B3 (Monks et al., 2007), albeit the 72-h EC 50 s obtained for MC-LW and -LF in the transiently transfected HeLa cells were in the subnanomolar range. The overt differences in EC 50 s obtained with the transiently transfected HeLa when compared to the stably transfected HEK293 cells reported here, most likely are a reflection of a.) different expression levels of the respective OATPs, b.) differences in PP type and expression levels, c.) duration of exposure and d) metabolic (Phase I and II) differences (Grisham et al., 1978; Hugo-Wissemann et al., 1991; Nyberg et al., 1994; Stange et al., 1995; Donato et al., 2008) Most importantly, the finding of transport affinity-dependent MC toxicity was confirmed by the results obtained from exposure of primary human hepatocytes to single MC congeners (Fig. 4.7). The primary hepatocytes of donors 1 and 2 demonstrated a higher susceptibility towards the more hydrophobic MC 65

66 Chapter IV OATP-mediated MC Congener Toxicity congeners: MCLW and MCLF appeared to be about 7 to 60-times and 7 to 39- times more toxic than the corresponding MCLR, respectively, whereas a roughly 37 to 49-times lower susceptibility towards MCRR than to MCLR was observed. In general, primary human hepatocytes of donors 1 and 2 appeared to be up to two orders of magnitude more susceptible to MCs induced cytotoxicity than the corresponding HEK293 OATP transfectants (Tab. 4.1). Moreover, vast differences in MCLR and -RR cytotoxicity were observed amongst the two donors tested. While the former could stem from a difference in metabolic capability between HEK293 cells and primary human hepatocytes as well as amongst donors (donors 1-3) of the primary hepatocytes, the latter could result from a difference in OATP expression level as well as other factors amongst the donors of the primary hepatocytes, despite that care was taken to ensure that all donors were female and that donors 1-3 were from the same age group. Some of these factors, beyond age and gender, are condition and metabolic capability of the respective donor s hepatocytes, i.e. largely depend on the donor s constitution, disease history, or to metabolic alteration during culturing (Bissell et al., 1978; Castell and Gomez-Lechon, 2009). Metabolic conversion, e.g. glutathione conjugation of MC, renders the MC-GSH conjugate much less toxic (Kondo et al., 1992). Hence, cells with lower levels of GSH-transferase or depleted of GSH will be more susceptible to MC induced PP inhibition and thus cytotoxicity. However, given that above interpretation holds true, it would remain to be proven that the GSH conjugation of MC congeners LR, RR, LW and LF is congener dependent. The expression levels of OATPs in primary human hepatocytes appear to vary within different zones of the liver. König et al. reported stronger OATP1B3- immunofluorescence staining near the central vein than close to the portal vein suggesting a lobular zoning of OATP expression (König et al., 2000b). Hence, the differences in MC congener susceptibility in the primary human hepatocytes observed amongst donors in this study could also be a reflection of differences in hepatic zonal origin introduced during liver biopsy of donors 1 and 2. The direct proof of OATP mediated uptake of MCs via use of radiolabelled substrates was not possible due to the unavailability of radiolabelled MCs. As an alternative co-incubation studies with the known OATP substrates TC and 66

67 Chapter IV OATP-mediated MC Congener Toxicity BSP, as competitor of MCLR, were employed. These uptake competition studies should provide for indirect evidence of MC uptake into HEK293 transfectants (Fig. 4.6) and primary human hepatocytes of donor 3 (Fig. 4.8), and thus support the findings of the cytotoxicity studies. Protection against MCLR-mediated cytotoxicity by co-incubation with TC and BSP could be observed in HEK293-OATP1B1 and -OATP1B3, hence, providing further evidence for OATP-mediated uptake of MCs. None of the applied concentrations of TC provided for protection against 5 µm MCLR neither in primary human hepatocytes nor in OATP-transfected HEK293 cells. In contrast, BSP appeared to afford some protection from MCLR mediated cytotoxicity in primary human hepatocytes (100 µm BSP) and in HEK293-OATP1B1 (100 µm and 50 µm BSP). Notably the lack and limited protection afforded by TC and BSP, respectively, in primary human hepatocytes (Fig. 4.8) may be explained by the fact that the primary human hepatocytes have very low levels of ser/thr PPs (Fig. 4.3) whereby already minute concentrations of MCLR entering the primary hepatocytes will suffice to induce cytotoxicity. The latter is also confirmed by the results of the cytotoxicity experiments (Fig. 4.7, Tab. 4.1) demonstrating the higher susceptibility of primary human hepatocytes to MC toxicity than the corresponding HEK293 transfectants (Fig. 4.5, Tab. 4.1) containing higher quantities of ser/thr PPs (Fig. 4.3). Generally increased protection against MCLR cytotoxicity was observed in HEK293 OATP-transfectants when co-incubating TC and BSP with lower concentrations of MCLR. Indeed, the higher the concentration of TC, the higher the cytoprotective effect towards MCLR cytotoxicity. In contrast, 5 µm BSP provided highest protection against MCLR in HEK293-OATP1B3 cells, while concentrations 50 µm BSP appeared to be cytotoxic as well. HEK293- OATP1B1 cells were less susceptible to BSP cytotoxicity, BSP thus affording a much higher protective effect towards MCLR cytotoxicity (Fig. 4.6). The observed reduced MC mediated cytotoxicity upon co-incubation MCLR, - LW and -LF with TC and BSP confirms earlier findings by Feurstein et al. (Feurstein et al., 2009) in primary murine whole brain cells, known to express moatp1b2, of the OATP1B family (including rat roatp1b2, human OATP1B1 and OATP1B3; (Hagenbuch and Meier, 2004)). 67

68 Chapter IV OATP-mediated MC Congener Toxicity As the ser/thr-specific protein phosphatase inhibitory capacity of the four different MC congeners were comparable in vitro (comparable toxicodynamics, Tab. 4.1), the observed overt differences in MC congener apical toxicity in vivo (i.p. LD 50 values) had been at the center of discussion for the last two decades. Certainly, a possible central role of toxicokinetics and thus the superfamily of OATPs has been assumed by Fischer et al. (Fischer et al., 2005), Dietrich and Hoeger (Dietrich and Hoeger, 2005), Dietrich et al. (Dietrich et al., 2008) and Komatsu et al. (Komatsu et al., 2007). However, supporting scientific evidence has so far remained scarce. The cytotoxicity results presented in this study using HEK293 OATP-transfectants (Fig. 4.5), as well as primary human hepatocytes (Fig. 4.7) thus emphasize the importance of toxicokinetics for understanding of the factors underlying the observed in vivo differences in toxicity amongst various MC congeners. Indeed, the 48-h EC 50 value for MCRR using OATP1B3-transfected HEK293 cells was approximately 5-fold higher than MCLR, whereas OATP1B1-transfected cells were resistant to MCRR cytotoxicity. In primary human hepatocytes the viability EC 50 values for MCLR and MCRR differed by a factor of 35 to 50 depending on the donor (1 or 2). Distinct differences between MCLR and -RR toxicity have also been reported from other in vitro studies: Fastner et al. (Fastner et al., 1995) and Eriksson et al. (Eriksson et al., 1990a) demonstrated that primary rat hepatocytes are less susceptible to MCRR (EC nm) than to MCLR (EC nm). Thus, irrespective of the cells used (primary hepatocytes or transfected HEK293), MCRR was always 1-2 orders of magnitude less cytotoxic than MCLR. This difference in cytotoxicity compares well with the approximately 1- order of magnitude difference (i.p. LD 50 values) observed in acute mouse in vivo toxicity between MCRR and MCLR (Krishnamurthy et al., 1986; Watanabe et al., 1988; Lovell et al., 1989; Stoner et al., 1989). Conclusion and possible implication for present risk assessments The results derived from the OATP-transfected HEK293 cells and the primary human hepatocytes presented here provide strong evidence that the more hydrophobic MCLW and -LF are distinctly more toxic than MCLR in vitro and thus potentially also in vivo. The higher toxicity of MCLF and -LW was also 68

69 Chapter IV OATP-mediated MC Congener Toxicity reported for primary murine whole brain cells (Feurstein et al., 2009) and for the protozoa Tetrahymena pyriformis (Ward and Codd, 1999). Given that all in vitro experiments with MCRR and -LR provided a similar toxicity ratio as the mouse in vivo i.p LD 50 assays (Tab. 4.1), one can assume that an in vivo mouse i.p. exposure to MCLF and -LW would also provide for a higher toxicity of MCLF and -LW than what is reported for MCLR and -RR. As all present microcystin risk assessments and evolving guidelines values are based on data derived from in vivo exposures to MCLR, a more in-depth analysis of the risk posed by other MC congeners, e.g. MCLF and -LW, may be warranted. Moreover, the distribution and abundance of MCLW and -LF in environmental samples have to be taken into consideration with regard to their importance for risk assessments. Unfortunately, only little information exist on their occurrence in natural samples like bloom material. Azevedo et al. made the first report on MCLF isolated from a bloom of M. aeruginosa from a Brazilian water supply (Azevedo et al., 1994). Spoof et al. investigated 93 samples from freshwater and brackish water locations in Finland, whereby 34 samples contained concentrations of MC or nodularin 0.2 µg/l (Spoof et al., 2003). However, only in one sample MCLF and traces of MCLW were detected. Interestingly, an earlier analysis of two natural bloom samples of M. aeruginosa in Scottland, and the rumen contents from a lamb that died after ingestion of scum from one of those samples, revealed a similar MC profile (including MCLW and -LF) as the strain M. aeruginosa PCC 7813 (Lawton et al., 1995) in which MCLW and -LF have repeatedly been detected (Lawton et al., 1995; Ward and Codd, 1999; Metcalf and Codd, 2000). The authors therefore consider a much more widespread occurrence of MCLW and -LF variants than previously assumed. Thus, MCLW and -LF might pose an unknown, yet serious threat to human and animal health. In conclusion, OATP-dependent cytotoxicity underlines the importance of identifying and characterizing further peptides responsible for the cell trafficking of different MC congeners using both transfected cell line models and primary cells. It must be assumed, that more OATPs and most likely other membrane carriers are capable to transport MC into and out of various cell types of organisms. Indeed, to date only 4 out of 11 known human OATPs have been tested for uptake of MC, namely OATP1A2, -1B1, -1B3 and -2B1 (this study; 69

70 Chapter IV OATP-mediated MC Congener Toxicity (Fischer et al., 2005; Komatsu et al., 2007; Monks et al., 2007). Since only OATP2B1 failed in transporting MC it appears very likely that some untested OATPs are capable of microcystin transport. Knowledge of MC congener specific transport is crucial for the understanding of MC toxicity, including organ distribution, and hence, the improvement of human risk assessments and establishment of guideline values. 4.6 Supplemental information Tab. 4.1 (S): Ratios of PP1 and PP2A versus GAPDH in HEK293 transfectants and primary human hepatocytes following densitometric evalutation of the western blots. Density PP1/GAPDH PP2A/GAPDH HEK293-CV 1,12 1,11 HEK293-OATP1B1 1,08 1,33 HEK293-OATP1B3 1,19 1,27 human hepatocytes 0,35 0,08 Mean PP1/GAPDH PP2A/GAPDH HEK293-CV 1,12 1,11 HEK293-OATP1B1 1,08 1,32 HEK293-OATP1B3 1,19 1,27 human hepatocytes 0,35 0,08 Volume PP1/GAPDH PP2A/GAPDH HEK293-CV 1,12 1,11 HEK293-OATP1B1 1,08 1,33 HEK293-OATP1B3 1,19 1,27 human hepatocytes 0,35 0,08 70

71 Chapter IV OATP-mediated MC Congener Toxicity PP1 PP2A HEK293-CV HEK293-OATP1B1 HEK293-OATP1B3 primary human hepatocytes Fig. 4.1 (S): Immunoblot detection of PP1 and PP2A in primary human hepatocytes (mixture between donor 1 and 2), HEK293 cells transfected with a control vector (CV), OATP1B1 and OATP1B3. Equal amounts of total protein (60 µg/lane) were applied. Molecular weights were estimated by comparison with marker proteins. Due to limited availability of primary human hepatocytes the detection could not be normalized against a housekeeping protein. Acknowledgements We would like to acknowledge the Arthur & Aenne Feindt Foundation (Hamburg, Germany), the BMBF (01GG0732-AKN), as well as the European Union (PEPCY QLRT ) for kindly funding parts of this study. We would also like to thank Prof. Dr. Dietrich Keppler (Division of Tumor Biochemistry, German Cancer Research Centre, Heidelberg, Germany) for kindly providing the transfected HEK293 cells, Dr. Elisa May, Christine Strasser and Daniela Hermann (Bio Imaging Centre, University of Konstanz, Germany) for the introduction and assistance to laser confocal microscopy, as well as Alicja Panas, Jasmin Leyhausen, Julia Kleinteich and Isabelle Eisele for practical assistance. 71

72 Chapter V CYN Uptake and Toxicity 5 In vitro toxicity of cylindrospermopsin in primary human hepatocytes and OATPexpressing HEK293 cells Fischer A 1, Feurstein D 1, Knobeloch D 2, Nussler A 3, Dietrich DR 1 1 Human and Environmental Toxicology, University of Konstanz, Konstanz, Germany 2 Department of General, Visceral, and Transplant Surgery, Charité Campus Virchow, Berlin, Germany 3 Technical University Munich, Department of Traumatology, Munich, Germany in preparation 5.1 Abstract The cyto- and genotoxic cyanobacterial alkaloid cylindrospermopsin (CYN) is presumed to be the causative toxin for the severe cases of hepatoenteritis on Palm Island in In vivo studies revealed the liver to be the organ predominantly affected following exposure to CYN. Its toxicity was demonstrated to stem from the inhibition of protein synthesis at low concentrations (maximum inhibition at 0.5 µm) and P450-mediated cytotoxicity at high concentrations (1-5 µm) using primary mouse hepatocytes. Despite the hydrophilic character of CYN its low molecular weight of only 415 Da might enable passive diffusion. However, evidence for the involvement of the bile acid transport system has been given by in vitro studies using primary rat hepatocytes. Therefore, we compared the cytotoxicities of CYN with OATP transportdependent MCLR and cell-permeant OA in primary human hepatocytes of two different donors at 48 h and liver OATP1B1- and OATP1B3-transfected, as well as mock-transfected HEK293 cells at 48, 72 and 96 h using the MTT reduction assay. In primary human hepatocytes CYN, MCLR and OA elicited typical doseresponse effects on viability with 48-h EC 50 s ranging from nm, 72

73 Chapter V CYN Uptake and Toxicity nm and nm, respectively. On the contrary, in OATPexpressing HEK293 cells we estimated a time-independent EC 50 value of approximately 5000 nm (highest concentration applied), whereas viability in mock-transfected HEK293 cells remained above 50%. This supports the involvement of liver OATPs in CYN uptake, even though OATP-mediated CYN toxicity was not as evident as for MCLR. In addition, coincubation of CYN with OATP substrates bromosulphophthalein and taurocholate failed to protect HEK293 cells against CYN toxicity. On the other hand, CYN appeared to protect OATP-expressing HEK293 cells against toxic concentrations of BSP. The results suggest some involvement of OATPs in transporting CYN, but further investigations are needed to clearify their importance and contribution to CYN toxicity. Keywords: cyanobacteria, cylindrospermopsin, primary human hepatocytes, OATP 5.2 Introduction In 1992 Ohtani et al. (Ohtani et al., 1992) identified cylindrospermopsin (CYN) from its eponymous producer Cylindrospermopsis raciborskii which is considered as the causative organism of the Palm Island Mystery Disease, a major outbreak of severe hepatoenteritis on Palm Island near Queensland, Australia in 1979 (Bourke et al., 1983; Hawkins et al., 1985). CYN is a cyanobacterial alkaloid with a molecular weight of 415 Da. It is a highly watersoluble zwitterion that is composed of a sulphated and methylated tricyclic ring structure containing a guanidine moiety that is linked to a hydroxymethyluracil (Ohtani et al., 1992; Falconer, 2005a). To date only two structural variants, 7- epicylindrospermopsin (Banker et al., 2000) and 7-deoxycylindrospermopsin (Norris et al., 1999) have been found to occur naturally. CYN predominantly targets the liver, causing dose-dependent necrosis and haemorrhages, however, other organs, such as kidneys, lungs, thymus, heart, stomach and small intestine were found to be affected as well, as demonstrated by various exposure studies in which mice were treated either with extracts of CYN producing C. raciborskii or purified toxin by oral or i.p. administration 73

74 Chapter V CYN Uptake and Toxicity (Hawkins et al., 1985; Ohtani et al., 1992; Terao et al., 1994; Hawkins et al., 1997; Falconer et al., 1999; Seawright et al., 1999). For the purified toxin i.p. LD 50 values of 2100 and 200 µg/kg bw were determined at 24 hours and 5-6 days, respectively (Ohtani et al., 1992). A provisional guidline value of 1 µg/l for drinking water has been suggested based on no-observed-adverse-effect levels (NOAELs) and several uncertainty factors (Shaw et al., 2000; Humpage and Falconer, 2003). The toxicity of CYN relies on the inhibition of protein biosynthesis as shown in reticulocyte lysates (Terao et al., 1994; Froscio et al., 2001; Runnegar et al., 2002) and primary rat hepatocytes, in which inhibition of GSH synthesis was additionally reported (Runnegar et al., 1995d; Runnegar et al., 2002). Moreover, evidence for the involvement of CYN metabolization by cytochrome P450 enzymes (Terao et al., 1994) that generate (an) active metabolite(s) have been provided in vitro (Runnegar et al., 1995d; Froscio et al., 2003; Humpage et al., 2005) and in vivo (Norris et al., 2002). In fact, Froscio et al. (Froscio et al., 2003) distinguished between two events in CYN toxicity using primary mouse hepatocytes: the inhibition of protein synthesis that appeared to be irreversible and an early response at lower concentrations (maximum inhibition at 0.5 µm) and P450-mediated cytotoxicity at acute concentrations (1-5 µm). In addition, studies on DNA alteration provided evidence of CYN to be genotoxic forming DNA adducts (Shaw et al., 2000) and causing strandbreaks in hepatocytes of CYN treated mice (Shen et al., 2002), as well as a loss of whole chromosomes in a human lymphoblastoid cell line (Humpage et al., 2000; Humpage et al., 2005). Fessard et al. (Fessard and Bernard, 2003) suggested an active CYN metabolite as being responsible for the reported genotoxicity, since strandbreaks failed to appear in CHO-K1 cells in which metabolizing enzyme activities are low. Indeed, inhibitors of CYP450 prevented CYN induced DNA fragmentation in primary rat hepatocytes (Humpage et al., 2005). In general, permanent cell lines appeared to be less susceptible to CYN than primary hepatocytes (Shaw et al., 2000; Chong et al., 2002). This might be due to a lower CYP450 activity in the cell lines. E.g. although a decrease in mrna contents of most CYPs has been reported in primary hepatocytes after 24 hours of culturing, enzyme activities retained and levels were still 400 times higher than in hepatoma cells (Rodriguez-Antona et al., 2002). In addition, reduced susceptibility might also stem from a lack of membrane transporters, as it is 74

75 Chapter V CYN Uptake and Toxicity commonly observed in cell lines derived for example from liver and gut like McArdle RH-7777 (Torchia et al., 1996) and HepG2 (Marchegiano et al., 1992; Torchia et al., 1996; Boaru et al., 2006) that might mediate the uptake of CYN. Due to the hydrophilic character of CYN, Runnegar et al. (Runnegar et al., 2002) considered passive diffusion through cell membranes unlikely, suggesting the requirement of transporter-mediated uptake into cells and members of the solute-carrier family as possible candidates. However, they found that the inhibtion of protein synthesis was independent of transport in vitro. Chong et al. (Chong et al., 2002) demonstrated a dose-dependent increase in viability of primary hepatocytes exposed to lethal concentrations of CYN (800 ng/ml) at 48 hours when coincubating with the bile acids, cholate and taurocholate. In addition, the authors reported that KB cells devoid of the bile acid transport system were approximately 8 times less susceptible to CYN than the primary hepatocytes and suggested passive diffusion or further transport systems as additional uptake mechanisms. To our knowledge no experiments on CYN toxicity and uptake into human cells have been conducted. Therefore, the purpose of this study was 1.) to determine CYN cytotoxicity in primary human hepatocytes and 2.) to asses the potential role of human liver OATPs on the uptake of CYN by comparing its cytotoxic effects in HEK293 cells stably expressing OATP1B1 and OATP1B3 in the presence or absence of the known OATP substrates taurocholate (TC) and bromosulphophthalein (BSP) (Meier and Stieger, 2002; Hagenbuch and Meier, 2003; Hagenbuch and Meier, 2004). Both substrates have already been used to competitively inhibit the uptake of MCLR, a cyanobacterial serine/threoninespecific protein phosphatase inhibitor (Honkanen et al., 1990; MacKintosh et al.; MacKintosh, 1993; Runnegar et al., 1993; Runnegar et al., 1995a; Hastie et al., 2005), by different OATPs (Fischer et al., 2005; Komatsu et al., 2007; Monks et al., 2007; Feurstein et al., 2009; Fischer et al., 2010). The experiments were carried out with MCLR and OA, a cell-permeant serine/threonine-specific protein phosphatase inhibitor produced by different genera of dinoflagellates (Haystead et al., 1989; Cohen et al., 1990; Hardie, 1993), as control for OATP-mediated uptake and transporter-independent uptake, respectively. 75

76 Chapter V CYN Uptake and Toxicity 5.3 Material & Methods Chemicals and reagents All chemicals were of the highest analytical grade commercially available. Standards of cylindrospermopsin (CYN), microcystin-lr (MCLR) and okadaic acid (OA) were obtained from Alexis (Switzerland) and Sigma (Germany), respectively. Dried purified CYN was additionally provided by Dr. Andrew R. Humpage (Department of Clinical and Experimental Pharmacology, University of Adelaide, Adelaide, Australia). CYN was dissolved in pure water, diluted to the stock and working concentrations according to the providers specifications. Concentrations, stability and purity were repeatedly confirmed at the beginning and throughout the experiments by reversed-phase HPLC-DAD. The analysis was carried out according to the method of Lawton et al. (Lawton et al., 1994) for the analysis of MCs, however, with detection performed at a wavelength of 262 nm. A system of Shimadzu consisting of a Shimadzu system controler SCL- 10Avp, Shimadzu auto injector SIL-10Advp, two Shimadzu pumps LC-10ATvp, Shimadzu diode array detector SPD-M10Avp, Shimadzu degasser DGU-14A and a Grom C 18 reversed-phase column Grom-Sil 120 ODS-4 HE, 4.6 x 250 mm, 5 µm particle size was employed. MCLR was dissolved in 75% MeOH. The concentrations of the stock and the working solutions were confirmed photometrically using the molar absorption coefficient of MCLR (39800 mol l -1 cm -1 ) published by Harada et al. (Harada et al., 1990a). This coefficient refers to MCLR dissolved in 100% MeOH, however, was shown to be applicable for 75% MeOH as well (Meriluoto et al., 2004; Meriluoto and Spoof, 2005). Additionally, MCLR concentrations were confirmed by HPLC-DAD analysis according to Lawton et al. (Lawton et al., 1994) using the same system described for CYN analysis. Okadaic acid (OA) (Sigma, Germany) was dissolved in 100% H 2 O, diluted to the stock and working concentrations according to the manufacturer s specifications. All toxins were sterilized by filtration using a 0.22 µm filter (Millex- GV, sterile; Millipore, Ireland). Cell systems Human primary hepatocytes were isolated in agreement with the ethical review board and after the patients written consent by a standard operating procedure 76

77 Chapter V CYN Uptake and Toxicity and cultured as described previously (Nussler et al., 2009). Both donors were females. Donor 1: partial liver resection due to cholangiocarcinoma (CCC), born 1961 and donor 2: partial liver resection due to liver metastasis of primary mamma carcinoma, born Human embryonic kidney cells (HEK293) stably transfected with recombinant human organic anion transporting polypeptides 1B1 (HEK293-OATP1B1) and 1B3 (HEK293-OATP1B3), as well as a control vector (HEK293-CV) were kindly provided by Prof. Dietrich Keppler (Division of Tumor Biochemistry, German Cancer Research Centre, Heidelberg, Germany). HEK293 cells were cultured as described in chapter IV. CYN cytotoxicity in primary human hepatocytes and OATP- and control vector-transfected HEK293cells For cytotoxicity experiments human primary hepatocytes of two different donors were sent seeded in collagen-coated 96-well plates (Price, 1975) at a density of 3 x 10 5 cells/ml and 200 µl/well (6 x 10 4 cells/well). Subsequent to arrival the cells were incubated at 37 C and 5% CO 2 for 4-5 h. Prior to exposure the medium was decanted and replaced by 200 µl RPMI 1640 containing 100 units/ml penicillin and 100 mg/ml streptomycin. Passages 3-8 of transfected HEK293 cells were used for toxin exposure experiments. HEK293 cells were seeded in 96-well plates (Greiner Bio-One GmbH, Frickenhausen, Germany), coated with poly-l-lysine (5 mg/ml), in MEM (supplemented as described above) at the same density as hepatocytes. After 4-5 h the medium was decanted and cells were incubated in 200 µl MEM (supplemented as described above, but with 1% FBS). Primary hepatocytes and HEK293 cells were incubated with CYN serially diluted (1:3) from 5000 nm to 2.29 nm for 48 h and 48, 72 and 96 h, respectively. HEK293 cells were additionally exposed to MCLR and OA in serial dilutions (1/3) from 5000 nm to 2.29 nm and 100 nm to nm, respectively. Cytotoxicity data of MCLR and OA on primary human hepatocytes have been collected alongside with CYN, but will be published elsewhere (Fischer et al., 2010). Coincubation studies on CYN with bromosulfophthalein and taurocholate using OATP- and control vector-transfected HEK293cells 77

78 Chapter V CYN Uptake and Toxicity Coincubation studies of CYN and taurocholate (TC) or bromosulfophthalein (BSP) on HEK293 cells were performed under identical conditions as described for single toxin exposures. Cells were exposed 4-5 h after seeding to 5 µm CYN coincubated with concentrations of 500 µm, 50 µm and 5 µm TC and 100 µm, 50 µm and 5 µm BSP, respectively, for 72 h and 96 h. As controls and for comparison, cells were additionally exposed to CYN and both OATP substrates without coincubation at the following concentrations: CYN serially diluted (1:3) from 5000 nm to 2.29 nm, BSP at 100 µm and 33.3 µm and TC at 500 µm and µm. MTT reduction assay Following exposure of HEK293 cells or human hepatocytes, 20 µl MTT (5 mg/ml) solution was added to each well and cells were incubated for 1.5 h. Subsequently the medium was removed carefully by pipette. The formed insoluble MTT-formazan was redissolved by adding 100 µl solubilization buffer (95% (v/v) isopropanol, 5% (v/v) formic acid) to each well and gentle shaking of the plate for at least 15 min. Finally, absorption was measured at 550 nm in a microtitre plate reader (Tecan, microplate reader, infinite M200; Austria). As positive control (0% survival) cells were incubated with 1% TWEEN (8 wells/96- well plate). The highest MeOH and water concentrations in the assays were <2% and 12.5%, respectively, and were used as the corresponding solvent controls (8 wells/96-well plate). The solvent controls served as negative controls (100% survival) and were compared to cells incubated with medium only (8 wells/96-well plate). No differences in viability, condition or growth rate could be identified between solvent and negative control (data not shown). Statistics Cytotoxicity studies were carried out at least three times in duplicates. The mean values of each duplicate yielded the values for calculation of the standard deviation (n 3). For calculation of the respective R 2, EC 50 values and statistical analysis GraphPad Prism 4.03 software was used. Briefly, the respective mean values were log-transformed and normalized. The resulting curves were fitted by nonlinear regression. An F-test (P<0.05) was employed for for the comparison of the EC 50 values, hillslopes and curves. One-way ANOVA 78

79 Chapter V CYN Uptake and Toxicity followed by Tukey's Multiple Comparison Test was employed to compare the effects of the highest concentration of CYN (5 µm) on transfected HEK293 cells at the respective time of exposure. Coincubation studies were performed at least three times in duplicates. The mean values of each duplicate yielded the values for calculation of the standard deviation (n 3). Data were statistically analyzed by One-way ANOVA followed by Dunnett s Multiple Comparison Test, whereas the respective negative control (medium control) was used as reference. One-way ANOVA followed by Tukey's Multiple Comparison Test was employed to assess significant differences between the data of the coincubation studies. 5.4 Results Cytotoxicity of CYN in primary human hepatocytes and liver OATPexpressing HEK293 cells CYN elicited typical concentration-response effects in primary human hepatocytes (Tab. 5.1; Fig. 5.1), however, with donor-dependent differences in susceptibility resulting in 48 h EC 50 values of nm (donor 1) and nm (donor 2). In general, donor 1 appeared to be more susceptible showing increased sensitivity towards MCLR and OA (see chapter IV; Tab. 4.1) as well. HEK293 cells appeared to be far less susceptible towards CYN than hepatocytes (Tab. 5.1; Fig. 5.2 (A-C)). After 48 h a decrease of approximately 50% in viability was only measured in HEK293-OATP1B3 at the highest CYN concentration (5 µm). 79

80 Chapter V CYN Uptake and Toxicity Tab. 5.1: 48 h EC 50 values with 95% confidence intervals in brackets of CYN, MCLR and OA in primary human hepatocytes. CYN MCLR OA primary human hepatocytes (Donor 1) nm ( ) R 2 = *3.4 nm ( ) R 2 = *3.0 nm ( ) R 2 = primary human hepatocytes (Donor 2) nm ( ) R 2 = *24.6 nm ( ) R 2 = *35.1 nm ( ) R 2 = ) *, Fischer et al., 2010 viability [% negative control] CYN [nm] Donor I Donor II Fig. 5.1: 48 h cytotoxicity of CYN in primary human hepatocytes of two different donors. Negative control = solvent control. Values represent mean ± standard error of the mean of at least three independent experiments. 80

81 Chapter V CYN Uptake and Toxicity Tab. 5.2: 48, 72 and 96 h EC 50 values with 95% confidence intervals in brackets of CYN, MCLR and OA in primary human hepatocytes and transfected HEK293 cells. HEK293-CV HEK293-OATP1B1 HEK293-OATP1B3 48 h 72 h 96 h 48 h 72 h 96 h 48 h 72 h 96 h CYN > 5000 nm > 5000 nm > 5000 nm > 5000 nm ~ 5000 nm ~ 5000 nm ~ 5000 nm ~ 5000 nm ~ 5000 nm MCLR n.d. n.d. n.d nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = OA 9.41 nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = nm ( ) R 2 = n.d., not determinable 81

82 Chapter V CYN Uptake and Toxicity This value was reached by HEK293-OATP1B1 not until 72 h, whereas viability of HEK293-CV remained above 50% and at the same level during any time of exposure. However, statistical analysis revealed no significant time dependency of CYN toxicity in any of the transfected HEK293 cells (P>0.05; Tukey's Multiple Comparison Test). A B viability [% negative control] c [nm] viability [% negative control] c [nm] C viability [% negative control] c [nm] CYN (48 h) CYN (72 h) CYN (96 h) Fig. 5.2: Cytotoxicity of CYN in HEK293-CV (A), HEK293-OATP1B1 (B) and HEK293-OATP1B3 (C) at 48, 72 and 96 h. Negative control = solvent control. Values represent mean ± standard error of the mean of at least five independent experiments. On the contrary, MCLR cytotoxicity was concentration- and time-dependent in both OATP-expressing HEK293 cells (Tab. 5.2; Fig. 5.3 (B and C)): 48, 72 and 96 h EC 50 values of nm, nm and nm and nm, nm and nm were calculated for HEK293-OATP1B1 and HEK293-OATP1B3, respectively. However, differences in the EC 50 s of HEK293-OATP1B3 were statistically insignificant (P>0.05; F-test), hence, data analysis resulted in a 82

83 Chapter V CYN Uptake and Toxicity single curve (Fig. 5.3 (C)). HEK293-CV remained insusceptible towards MCLR at any concentration and time of exposure (Tab. 5.2; Fig. 5.3 (A)). Cell-permeant OA caused concentration-dependent effects in all HEK293 transfectants (Tab. 5.2; Fig. 5.4 (A-C)) with comparable EC 50 s in the low nanomolar range. However, the respective EC 50 values indicate no time dependency of OA cytotoxicity: in HEK293-CV 48, 72 and 96 h EC 50 s were 9.41 nm, nm and 9.25 nm, in HEK293-OATP1B nm, 9.09 nm and 6.27 nm and in HEK293-OATP1B nm, 7.08 nm and 5.56 nm, respectively. A B viability [% negative control] c [nm] viability [% negative control] c [nm] C viability [% negative control] c [nm] MCLR (48 h) MCLR (72 h) MCLR (96 h) Fig. 5.3: Cytotoxicity of MCLR in HEK293-CV (A), HEK293-OATP1B1 (B) and HEK293-OATP1B3 (C) at 48, 72 and 96 h. Negative control = solvent control. Values represent mean ± standard error of the mean of at least four independent experiments. 83

84 Chapter V CYN Uptake and Toxicity A B viability [% negative control] c [nm] viability [% negative control] c [nm] C viability [% negative control] c [nm] OA (48 h) OA (72 h) OA (96 h) Fig. 5.4: Cytotoxicity of OA in HEK293-CV (A), HEK293-OATP1B1 (B) and HEK293-OATP1B3 (C) at 48, 72 and 96 h. Negative control = solvent control. Values represent mean ± standard error of the mean of at least four independent experiments. Coincubation studies on CYN with bromosulfophthalein and taurocholate using OATP- and control vector-transfected HEK293cells For coincubation studies HEK293 cells were exposed only for 72 and 96 h to reduce sample size and improve menageability. Moreover, cytotoxicity data suggested differences to be rather significant at longer times of exposure. However, neither TC nor BSP lead to a significant protection against CYN toxicity at 72 and 96 h in any HEK293 transfectant (Fig. 5.5 (A-F) and Fig.5.6 (A-F); P>0.05; Tukey's Multiple Comparison Test). Contrarily, 5 µm CYN elicited significant effects only in OATP-expressing HEK293 cells, whereas reduction of viability of HEK293-CV was insignificant (Fig. 5.5 (A and B) and Fig. 5.6 (A and B)). 84

85 Chapter V CYN Uptake and Toxicity A B viability [% negative control] negative control 5 µm CYN 500 µm TC µm TC 5 µm CYN µm TC 5 µm CYN + 50 µm TC 5 µm CYN + 5 µm TC viability [% negative control] negative control 5 µm CYN 100 µm BSP 33.3 µm BSP 5 µm CYN µm BSP 5 µm CYN + 50 µm BSP 5 µm CYN + 5 µm BSP C D viability [% negative control] negative control ** ** ** ** 5 µm CYN 500 µm TC µm TC 5 µm CYN µm TC 5 µm CYN + 50 µm TC 5 µm CYN + 5 µm TC viability [% negative control] negative control * ** ** ** ** 5 µm CYN 100 µm BSP 33.3 µm BSP 5 µm CYN µm BSP 5 µm CYN + 50 µm BSP 5 µm CYN + 5 µm BSP E F viability [% negative control] negative control ** ** ** ** 5 µm CYN 500 µm TC µm TC 5 µm CYN µm TC 5 µm CYN + 50 µm TC 5 µm CYN + 5 µm TC viability [% negative control] negative control ** * ** ** ** 5 µm CYN 100 µm BSP 33.3 µm BSP 5 µm CYN µm BSP 5 µm CYN + 50 µm BSP 5 µm CYN + 5 µm BSP Fig. 5.5: Coincubation study (72 h). Cytotoxic effects in HEK293-CV (A and B), HEK293-OATP1B1 (C and D) and HEK293-OATP1B3 (E and F) following coincubation of CYN with TC or BSP. Negative control = solvent control. Values represent mean ± standard error of the mean of three independent experiments. Each respective negative control was used as reference for comparison (One-way ANOVA followed by Dunnet s Multiple Comparison Test; * P<0.05; ** P<0.01). 85

86 Chapter V CYN Uptake and Toxicity A B viability [% negative control] negative control * ** ** 5 µm CYN 500 µm TC µm TC 5 µm CYN µm TC 5 µm CYN + 50 µm TC 5 µm CYN + 5 µm TC viability [% negative control] negative control 5 µm CYN 100 µm BSP 33.3 µm BSP 5 µm CYN µm BSP 5 µm CYN + 50 µm BSP 5 µm CYN + 5 µm BSP C D viability [% negative control] negative control * 5 µm CYN 500 µm TC µm TC 5 µm CYN µm TC 5 µm CYN + 50 µm TC 5 µm CYN + 5 µm TC viability [% negative control] negative control * ** 5 µm CYN 100 µm BSP 33.3 µm BSP 5 µm CYN µm BSP 5 µm CYN + 50 µm BSP 5 µm CYN + 5 µm BSP E F viability [% negative control] negative control 5 µm CYN ** ** ** ** 500 µm TC µm TC 5 µm CYN µm TC 5 µm CYN + 5 µm TC 5 µm CYN + 50 µm TC Fig. 5.6: Coincubation study (96 h). Cytotoxic effects in HEK293-CV (A and B), HEK293-OATP1B1 (C and D) and HEK293-OATP1B3 (E and F) following coincubation of CYN with TC or BSP. Negative control = solvent control. Values represent mean ± standard error of the mean of three independent experiments. Each respective negative control was used as reference for comparison (One-way ANOVA followed by Dunnet s Multiple Comparison Test; * P<0.05; ** P<0.01). 86 viability [% negative control] negative control ** 5 µm CYN ** 100 µm BSP ** 33.3 µm BSP 5 µm CYN µm BSP ** 5 µm CYN + 5 µm BSP 5 µm CYN + 50 µm BSP

87 Chapter V CYN Uptake and Toxicity In addition, similar to the results obtained from CYN exposure experiments on HEK293 cells (Fig. 5.2 (A-C)) no distinct time dependency was observed within the coincubation studies. Furthermore, high concentrations of BSP (100 µm) significantly reduced viability of OATP-expressing HEK293 cells at 72 and 96 h, whereas HEK293-CV remained unaffected. This reduction was diminished by coincubation with 5 µm CYN in both OATP-expressing HEK293 cells at 72 and 96 h (Fig. 5.5 (D and F) and Fig. 5.6 (D and F)). While this effect appeared to be independent of the BSP concentration coincubated in OATP1B1-expressing cells, viability of OATP1B3-expressing cells further concentration-dependently increased by lower concentrations of coincubated BSP. All HEK293 transfectants remained unaffected by any concentration of TC applied (Fig. 5.5 (A, C and E) and Fig. 5.6 (A, C and E)). However, although CYN had no significant effects on susceptibility of HEK293-CV, viability of these cells decreased significantly compared to negative control after 96 h of exposure (Fig. 5.6 (A)). This effect failed to appear at 72 h of exposure. 5.5 Discussion CYN cytotoxicity in primary human hepatocytes and OATP- and control vector-transfected HEK293 cells The typical concentration-response effects caused by CYN in primary hepatocytes, failed to appear in OATP-expressing, as well as control vectorexpressing HEK293 cells at the concentrations and times of exposure applied. This is most likely due to metabolic differences between those cell types. While primary human hepatocytes retain their drug metabolizing activity including CYP450 enzymes at levels comparable to those in situ (Li, 2001; Gomez- Lechon et al., 2003)), it is known that hepatoma cell lines and other immortalized cell lines, e.g. HepG2 or Mz-Hep-1, are lacking these properties or exhibit reduced activity (Nyberg et al., 1994; Stange et al., 1995; Rodriguez- Antona et al., 2002; Donato et al., 2008). Since it was found that CYP450 enzymes are involved in CYN metabolization (Terao et al., 1994) resulting in the formation of (an) active metabolite(s) in vitro (Runnegar et al., 1995d; Froscio et 87

88 Chapter V CYN Uptake and Toxicity al., 2003; Humpage et al., 2005) and in vivo (Norris et al., 2002) we suggest a lack of CYP450 activity to be responsible for the difference in susceptibility of primary human hepatocytes and HEK293 cells. This assumption is supported by the following findings: Permanent cell lines, KB and Hela, appeared to be less susceptible to CYN than primary rat hepatocytes (Shaw et al., 2000; Chong et al., 2002) and CYN toxicity in primary mouse hepatocytes was diminished when CYP450 activity was inhibited (Froscio et al., 2003). However, Chong et al. (Chong et al., 2002) suggested the lack of a bile acid transport system to be responsible for the reduced CYN toxicity in the KB cells. Indeed, comparison of the cytotoxic effects of CYN on control vector- and both OATP-expressing HEK293 cells indicates a role of OATP1B1 and OATP1B3 in the uptake of CYN (Fig. 5.2), even though this uptake was not as obvious as for MCLR, that provided for typical concentration-response effects in both OATP-expressing HEK293 cells (Fig. 5.4 (C and C)), whereas no effects at all in HEK293-CV at the concentrations applied (Fig. 5.4 (A)). Furthermore, our results revealed CYN cytotoxicity in HEK293 cells to be independent of the duration of the exposure. Besides its CYP450-mediated toxicity, CYN acts as a protein synthesis inhibitor as shown in reticulocyte lysates (Terao et al., 1994; Froscio et al., 2001; Runnegar et al., 2002) and primary rat hepatocytes (Runnegar et al., 1995d; Runnegar et al., 2002). Froscio et al. (Froscio et al., 2003) demonstrated this inhibition to be irreversible and an early response (maximum inhibition after 4 h) at lower concentrations (maximum inhibition at 0.5 µm) in primary mouse hepatocytes. Since we determined viability of HEK293 cells exposed to acute concentrations of CYN (up to 5 µm) not before 48 h protein synthesis might have already been completely and irreversibly inhibited. Assuming a concomitant lack of CYP450 metabolizing activity and, hence, no CYP450-mediated toxicity this may provide for a possible explanation for the observed ineffectiveness of elongated exposure times. Effect of TC and BSP on CYN cytotoxicity in OATP- and control vectortransfected HEK293 cells Further indications for the involvement of OATP1B1 and OATP1B3 are provided by the results obtained from the coincubation studies: The CYN control (5 µm), 88

89 Chapter V CYN Uptake and Toxicity applied for comparison of the coincubation, lead to an insignificant decrease in viability compared to the negative control in HEK293-CV (Fig. 5.5 (A and B) and Fig. 5.6 (A and B)), whereas this decrease was significant in both HEK293- OATP1B1 and HEK293-OATP1B3 (Fig. 5.5 (C-F) and Fig. 5.6 (C-F)). On the other hand, despite both TC and BSP were coincubated in excess (up to 500 µm and 100 µm, respectively) they failed to significantly protect the OATPexpressing HEK293 cells from CYN cytotoxicity (Fig. 5.5 (C-F) and Fig. 5.6 (C- F)). This is in part contrary to the finding of Chong et al. (Chong et al., 2002) who reported a concentration-dependent protection of primary rat hepatocytes against a lethal dose of CYN (800 ng/ml) by cholate and TC at 48 h, whereas no protection occurred at shorter (24 h) or longer (72 h) times of exposure. They concluded a further less effective uptake route, possibly passive diffusion, to become important when bile acid transport capacity is reduced in the presence of bile acids, still leading to death of the hepatocytes at 72 h. In our HEK293 cells a background of endogenous bile acid transporters was excluded, because no hints for a transport directed into the cells were observed (Keppler, 2006; personal communication). In this regard the tendency in the reduction of viability of HEK293-CV by CYN might be explained by passive diffusion, which is further supported by the unexpected finding that no timedependent toxicity was measured for the cell-permeant OA either: Its EC 50 s on the respective HEK293 transfectants remained in the same low nanomolar range at all times of exposure (Tab. 5.2; Fig. 5.4). However, in HEK293-CV a time-dependent increase in susceptibility from 72 h to 96 h was observed when CYN was coincubated with any concentration of TC (Fig. 5.5 (A) and Fig. 5.6 (A)). A possible explanation might be the increased osmotic potential leading to an increase in diffusion of CYN into these cells, which in turn would also diminish the protective effect by BSP and TC against CYN in OATP-expressing HEK293 cells. Nonetheless, metabolic inactivity combined with irreversible protein synthesis inhibition as discussed above appears more feasible to cause the lack of protection against CYN toxicity by the coincubated OATP substrates in these cells. The involvement of OATP1B1 and OATP1B3 in the uptake of CYN is also supported by its significant protection of both OATP-expressing HEK293 cells against the toxicity of high concentrations of BSP (100 µm) (Fig. 5.5 (D and F) and Fig. 5.6 (D and F); P>0.05; Tukey's Multiple Comparison 89

90 Chapter V CYN Uptake and Toxicity Test). Hence, both OATPs appear to have very high affinity for CYN leading to an unchanged decrease in viability despite the excess of BSP and TC. Conclusion Even though this study provides evidence for the involvement of human liver OATP1B1 and OATP1B3 in the uptake of CYN their role in human primary hepatocytes needs to be further elucidated, e.g. by the use of substrate competitors like BSP and TC, as well as shorter and longer times of exposure. However, since availability of primary human hepatocytes is usually highly limited and implemented by donor-specific variability the additional employment of immortalized cell lines as applied in this study is also of great importance. Furthermore, to allow comparison between results obtained from different cell types it appears crucial that such studies are conducted in conjunction with a determination of the respective CYN metabolizing activity. In addition, identification and knowledge on the mode of action of (an) active CYN metabolite(s) will certainly further improve the interpretation of results on the uptake of CYN. Acknowledgements We would like to acknowledge the Arthur & Aenne Feindt Foundation (Hamburg, Germany) for kindly funding parts of this study, Dr. Andrew R. Humpage (Department of Clinical and Experimental Pharmacology, University of Adelaide, Adelaide, Australia) for kindly providing purified CYN and Prof. Dr. Dietrich Keppler (Division of Tumor Biochemistry, German Cancer Research Centre, Heidelberg, Germany) for kindly providing the transfected HEK293 cells. 90

91 Chapter VI MC and BMAA in BGAS 6 Detection of microcystins and β-nmethylamino-l-alanine in blue-green algae supplements Fischer A 1, Hoeger SJ 1, Fastner J 2, Robertson A 3, Dietrich DR 1 1 Human and Environmental Toxicology, University of Konstanz, Konstanz, Germany 2 Federal Environmental Agency, Section II Drinking-water resources and treatment, Berlin, Germany 3 NOAA, Northwest Fisheries Science Center, Environmental Conservation Division, Harmful Algal Blooms Program, Seattle, Wisconsin, USA in preparation 6.1 Abstract Blue-green algae dietary supplements (BGAS) are still consumed at a large scale for their putative beneficial health effects, despite the publicized fact that they might be contaminated with cyanobacterial hepato- and neurotoxins. Microcysin-assigned intoxications of consumers of BGAS products, which are based on Aphanizomenon flos-aquae harvested in Upper Klamath Lake in Oregon, have been reported and, hence, warnings against this hepatotoxic cyanotoxin have been proclaimed. Recent studies detected the neurotoxic excitatory amino acid β-n-methylamino-l-alanine (BMAA) in the vast majority (95%) of the cyanobacterial genera tested and also unexpectedly in the brain tissue of Alzheimer s patients from Canada. This may indicate an unknown source of exposure to this cyanobacterial excitotoxin, since it has been associated with the increased incidents of an extensively endemic neurodegenerative disease among the indigenous population of Guam. In this study we therefore examined the presence of microcystins and BMAA in different blue-green algae products by colorimetric protein phophatase inhibition 91

92 Chapter VI MC and BMAA in BGAS assay, ELISA and LC-MS/MS, respectively. While we detected none of these products positively for BMAA, 10 out of 11 contained microcystins with 5 exceeding the maximum acceptable concentration of 1 µg/g dw defined by the Oregon Health Division. Keywords: blue-green algae supplements, microcystin, BMAA 6.2 Introduction Cyanobacteria are cosmopolitans, existing in nearly every environment imaginable. Their worldwide distribution led to versatile human exposure. While some of them, primarily Spirulina, Arthrospira, and Nostoc species, served men as aliments for centuries (Gao, 1998; Abdulqader et al., 2000; Carmichael et al., 2000) others are infamous for producing a variety of toxic secondary metabolites, usually classified into hepatotoxins and neurotoxins. Especially the hepatotoxic microcystins (MCs) have been responsible for severe human (Teixera et al., 1993; Jochimsen et al., 1998) and animal poisonings (Odriozola et al., 1984; Carbis et al., 1995; Ernst et al., 2001). These cyclic heptapeptides comprise a family of more than 80 congeners (Spoof, 2005; Zurawell et al., 2005; Humpage, 2008) with microcystin-lr (MCLR) being the most intensively investigated. The main variations in the general structure cyclo(-d-ala 1 -L-X 2 -D-methylAsp 3 -L-Z 4 -Adda 5 -D-Glu 6 -N-methyldehydro-Ala 7 ) predominantly occur in two of the seven amino acids (X and Z at position 2 and 4, respectively) and hence are used for the nomenclature of MCs (Carmichael et al., 1988a) according to the one-letter code for amino acids. Their primary molecular mode of action is the inhibition of several serine/threonine-specific protein phosphatases (PPs), especially PP1 and PP2A (Honkanen et al., 1990; MacKintosh, 1993; Runnegar et al., 1993). Besides hepatotoxins, some cyanobacteria are capable of producing potent neurotoxic alkaloids like anatoxins and saxitoxins (PSPs) that also have been associated with severe human (Falconer, 1993; Garcia et al., 2004) and animal poisonings (Mahmood et al., 1988; Negri et al., 1995b). Anatoxins mimic the effect of acetyl choline (Aronstam and Witkop, 1981; Swanson et al., 1986), 92

93 Chapter VI MC and BMAA in BGAS whereas saxitoxins block sodium channels of nerve cells (Catterall, 1980; Kuiper-Goodman et al., 1999), hence, both are leading to paralysis, respiratory depression and respiratory failure at sufficiently high doses (Carmichael et al., 1975; Kao, 1993; Carmichael, 1997). Due to eutrophication of surface waters, mass development of potentially toxic cyanobacteria is widespread along with the hazard of intoxications, especially in poorer countries with limited or no drinking water treatment at all (Herath, 1995; de Figueiredo et al., 2004; Hoeger et al., 2004). However, in industrialized countries that are less threatened by contaminated drinking water or different toxin-accumulated food items, an exceptional source of potential exposure to cyanotoxins is on the rise since the early 1980 s. This relatively new threat is posed by voluntarily ingested dietary supplements, mostly based on Spirulina (Arthrospira) spp. or Aphanizomenon flos-aquae. These blue-green algae supplements (BGAS), as they are commonly referred to, are consumed for their advertised putative beneficial health effects, such as increased alertness and energy, detoxification, weight loss, efficacy against various viral infections, including herpes and influenza, and even against cancer and mental disorders like depression or attention-deficit disorders (Gilroy et al., 2000; Lukassowitz, 2002; Saker et al., 2005). In 1999 the number of consumers was estimated to be over one million in North America alone (Falconer et al., 1999). While Aphanizomenon flos-aquae has been reported to be a potential producer of neurotoxins, like anatoxin-a (Rapala et al., 1993; Carmichael, 1997) and various PSP toxins (Jackim and Gentile, 1968; Ikawa et al., 1982; Mahmood and Carmichael, 1986a; Carmichael, 1997; Pereira et al., 2000; Ferreira et al., 2001; Liu et al., 2006), the strain Aphanizomenon flos-aquae Ralfs ex Born. & Flah. var. flos-aquae from Upper Klamath Lake in Oregon, USA, which is harvested for processing into the supplements usually marketed as AFA, turned out to be non-toxic with regard to neuro- and hepatotoxicity (Carmichael et al., 2000). However, potentially toxin producing cyanobacteria coexist in Lake Klamath, including Microcystis aeruginosa, Anabaena flos-aquae and Oscillatoria spp. (Carmichael et al., 2000). Indeed, contamination of A. flosaquae products with the hepatotoxic microcystins (MCs) have been reported (Gilroy et al., 2000; Kuiper-Goodman et al., 2000; Lawrence et al., 2001; Saker et al., 2005; Saker et al., 2007) and a maximum acceptable concentration 93

94 Chapter VI MC and BMAA in BGAS (MAC) of 1 µg MC/g dw was enforced by the Oregon Health Division and the Oregon Department of Agriculture (Gilroy et al., 2000). In general, Spirulina is considered a non-toxic genus (Salazar et al., 1996; Salazar et al., 1998), although anatoxins have been detected in supplements labeled Spirulina (Draisci et al., 2001; Osswald et al., 2008). Moreover, contaminations with microcystins occur less frequently in these products due to controlled cultivation conditions in ponds (Gilroy et al., 2000; Kuiper-Goodman et al., 2000). Contrarily, Arthrospira fusiformis is a potential producer of microcystins itself (Ballot et al., 2004; Ballot et al., 2005) and a case of hepatotoxicty that has been assigned to consumption of Arthrospira has been reported (Iwasa et al., 2002). Moreover, this species may also produce the neurotoxic alkaloid anatoxin-a (Ballot et al., 2004; Ballot et al., 2005). However, due to morphological similarities Spirulina and Arthrospira are easily confused, hence, neuro- and hepatotoxin detection in Spirulina might possibly be attributed either to mislabelling of the products or to a contamination with another toxic cyanobacterial species. Recently, it was found that 20 out of 21 cyanobacterial genera and 29 out of 30 cyanobacterial strains tested, including a marine strain of Aphanizomenon flosaquae, produce the excitatoric amino acid β-n-methylamino-l-alanine (BMAA) (Cox et al., 2005). BMAA is an excitotoxic amino acid that acts as an agonist of animal glutamate receptors (Weiss et al., 1989a; Weiss et al., 1989b; Myers and Nelson, 1990) and elevates intracellular calcium levels (Brownson et al., 2002). It is considered as the causative agent in one hypothesis for the increased incidents of amyotrophic lateral sclerosis/parkinsonism-dementia complex (ALS/PDC) among the Chamorro, the indigenous population of Guam, as a result of bioaccumulation in their traditional diet, i.e. cycad seeds and flying foxes with endosymbiotic cyanobacteria of the genus Nostoc being the primary source (Spencer et al., 1987a; Cox and Sacks, 2002; Bannack and Cox, 2003; Cox et al., 2003). This controversial hypothesis and BMAA toxicity are reviewed by various authors (Ince and Codd, 2005; Papapetropoulos, 2007; Steele and McGeer, 2008). The neurodegenerative disease elicits symptoms similar to amyotrophic lateral sclerosis, Parkinson s disease and Alzheimer s. Besides Guam, there are two further high-incidence foci of ALS/PDC in Irian Jaya, West 94

95 Chapter VI MC and BMAA in BGAS New Guinea and on Kii peninsula of Japan where cycad seeds are part of the traditional diet as well (Spencer et al., 1987b; Spencer et al., 1987c). BMAA appeared to occur in a free and protein-bound form. The latter is hypothesized to build an endogenous neurotoxic reservoir from which free BMAA might slowly be released causing chronic damage to the brain over years or even decades (Murch et al., 2004b). Interestingly, BMAA was not only detected in the brain tissue of ALS/PDC patients from Guam, but also in those of Alzheimer s patients from Canada (Cox et al., 2003; Murch et al., 2004a) suggesting a further source for BMAA intoxication that is not restricted to Guam. Indeed, Johnson et al. (Johnson et al., 2008) positively analyzed BMAA in all 21 samples of Nostoc commune that is part of the traditional diet of the indigenous people in the mountains of Peru, by HPLC-FD, UPLC-UV, UPLC/MS, and LC/MS/MS. The averaged concentrations ranged from 6.18 to µg/g. In addition, Metcalf et al. (Metcalf et al., 2008) detected free and protein-bound BMAA in all 12 environmental samples tested, including blooms, scums, and mats from waterbodies throughout the UK confirming the widespread distribution of BMAA in cyanobacteria. The predominant genera were identified as Microcystis, Planktothrix, Anabaena, Aphanizomenon, Gomphosphaeria, Nodularia, Pseudanabaena and Oscillatoria. Besides BMAA, they also reported the presence of other cyanotoxins in all but 2 of the samples: predominantly MCs, but also nodularin, anatoxin-a and saxitoxin, as well as an unidentified substance that elicited acute neurotoxicity in a bioassay. Therefore, the scope of this study was to examine different BGAS in order to assess 1) the known health risk of a potential contamination with MCs by colorimetric protein phosphatase inhibition assay, Adda-ELISA and LC-MS/MS and 2) the possible additional health hazard caused by potentially contained BMAA using LC-MS/MS. 6.3 Material & Methods Chemicals and reagents All chemicals were of the highest analytical grade commercially available. Standards of microcystins (MC) MCRR, MCYR, MCLR, MCLA, MCLW and 95

96 Chapter VI MC and BMAA in BGAS MCLF were obtained from Alexis (Switzerland). They were dissolved in 75% MeOH and the concentrations of the stock and the working solutions were proven photometrically by using the molar absorption coefficient of MCLR and MCRR (39800 mol l -1 cm -1 ) published by Harada et al. (Harada et al., 1990b). This coefficient was applied for all microcystins, since no molar absorption coefficients are published for the other congeners. Although this is the molar absorption coefficient for MCLR/-RR dissolved in 100% MeOH it turned out to be applicable for 75% MeOH as well (Meriluoto and Spoof, 2005). Additionally, concentrations were confirmed by HPLC-DAD analysis according to Lawton et al. (Lawton et al., 1994). L-BMAA hydrochloride was purchased from Sigma (Germany) and dissolved in 100% H 2 O, diluted to the stock and working concentrations according to the manufacturer s specifications. Methanolic extraction of microcystins from blue green algae supplements 11 products of different brands containing Aphanizomenon flos-aquae from Lake Klamath (Code: Aph # 1-11) were examined. Three (Aph # 2-11) or four (Aph # 1) subsamples of each product were examined and extracted separately. 375 mg of each subsample were dissolved in 75% MeOH by vigorous shaking and ultrasonicated for 30 min (15 min in the following steps) in an ice-cold ultrasonic bath. After centrifugation at 4,000 x g for another 30 min the resulting supernatants were collected and stored on ice. The pellets were resuspended in 75% MeOH as described above. This methanolic extraction was repeated 3 times and the respective supernatants were pooled. The total volume of ~45 ml/sample was evaporated passively to a small residue on ice, which was in turn filled up to 15 ml with purified water, followed again by vigorous shaking and ultrasonication for 15 min in an ice-cold ultrasonic bath. Purification of microcystins from aqueous extracts A solid phase extraction using C18 silica cartridges (Waters, Sep-Pak Vac 6cc (1g)) was performed to purify and concentrate microcystins potentially present in the aqueous extracts. The cartridges were preconditioned with 2x 5 ml MeOH (100%) and equilibrated with 2x 3 ml H 2 O prior to adding the respective extracts. After elution of the extracts the cartridges were washed three times with 4 ml H 2 O. MCs potentially bound to the cartridges were eluted with 3 x

97 Chapter VI MC and BMAA in BGAS ml MeOH (100%). These eluates were collected and evaporated passively over night to a small residue on ice prior to being completely dried in by vacuum centrifugation. Subsequently, these dried extracts were redissolved in 600 µl MeOH (100%) by vigorous shaking and untrasonification for 7.5 min. Then 2,400 µl H 2 O were added leading to a final concentration of 20% methanol, followed again by vigorous shaking and ultrasonification for 7.5 min. A last centrifugation step at 13,000 x g for 20 min was applied to remove particles. The resulting supernatants were used for MC analysis. Colorimetric protein phosphatase inhibition (cppia) assay with recombinant PP1 All three (four) extracts of the 11 A. flos-aquae samples (Aph # 1-11) were diluted serially (1:3) and analyzed three times in duplicates. In every assay MCLR served as internal standard for quantification and was applied diluted serially (1:3) at concentrations from nm to 0.02 nm on every plate. The concentration of microcystin in the samples was calculated from the obtained standard curve and expressed in MCLR equivalents/g dw. The assay was carried out as described by Herersztyn and Nicholson with one modification (Heresztyn and Nicholson, 2001): instead of PP2A we used PP1 (rabbit skeletal muscle, recombinant (E. coli); New England Biolabs (USA)) in a stock concentrations of 2,500 units/ml, which resulted in a concentration of 3 units/ml in the final assay. P-nitrophenylphosphate from Acros Organics (USA) was used as substrate. Adda-Enzyme Linked Immunosorbent Assay (Adda-ELISA) All A. flos-aque samples, but sample Aph # 1 were analyzed by a competitive indirect ELISA in a previous work (Hoeger et al., 2003), using a commercially available kit (Microcystins (Adda specific) ELISA Kit, Abraxis (USA)) according to the manufacturer s specifications. Therefore, only three separately generated extracts of BGAS sample Aph # 1 and as a positive control one extract of each BGAS sample Aph # 8 and # 9 were analyzed in duplicates in different dilutions on two separate plates of the ELISA kit according to the manufacturer s specifications. Thus, the data of sample Aph # 1 is composed of three sample replicates and two technical replicates. MC concentrations were 97

98 Chapter VI MC and BMAA in BGAS calculated from the obtained standard curve using the MCLR standards provided in the kit, hence, are given as MCLR equivalents. The range of this ELISA is between 0.15 nm and 5 nm (approximately 0.15 µg/l and 5 µg/l). Liquid Chromatography Tandem Mass Spectrometry (LC-MS/MS) analysis of microcystins The LC-MS/MS analyses of the BGAS extracts (three extracts of AFA # 1-11) were carried out on an Agilent 1100 series HPLC system (Agilent Technologies, Waldbronn, Germany) coupled to a API 4000 triple quadrupole mass spectrometer (Applied Biosystems/MDS Sciex, Framingham, MA) equipped with a turbo-ionspray interface. The extract was separated using a Purospher STAR RP-18 endcapped column (30 x 4 mm, 3 µm particle size, Merck, Germany) at 30 C. The mobile phase consisted of 0.5% formic acid (A) and acetonitrile with 0.5% formic acid (B) at a flow rate of 0.5 ml min - 1 with the following gradient programme: 0 min 25% B, 10 min 70% B, 11 min 70% B (Spoof et al., 2003). The injection volume was 10 µl. The mass spectrometer was operated in the multiple reaction monitoring mode (MRM) for the confirmation of microcystins listed in Table 6.1. The ion dwell time was 200 ms. Quantitation of microcystins was achieved using the MRM mode. Standard curves were established for MCRR, -YR, -LR, -LW, -LF and - LA (Alexis chemicals, Switzerland) and analyzed in a line with the unknowns (one calibration curve after 10 unknowns). In the MRM mode the limit of detection (LOD) for the single MC congeners were in the range of pg o.c. The samples were also analyzed for the presence of other MC congeners for which no standards are commercially available using the precursor ion mode as described previously by Hiller et al. (Hiller et al., 2007). Though applied for qualitative purposes only, the analyses of samples with known microcystin concentrations by this method gave higher LODs (~1 µg/g dw) than with the MRM mode. 98

99 Chapter VI MC and BMAA in BGAS Tab. 6.1: Transitions and parameters for the detection of microcystins by LC-MS/MS in the MRM mode; DP: declustering potential, CE: collision energy, CXP: cell exit potential; *, [M+2H] 2+ Toxin Transitions m/z [M+H] + m/z DP CE CXP LOD (pg o.c.) Microcystin-RR 519.7* Microcystin-LA Microcystin-LF ,5 Microcystin-LR Microcystin-LW Microcystin-YR Extraction of BMAA from blue green algae supplements Three of the A. flos-aquae products analyzed for MC contamination (Code: Aph # 1, # 8 and # 9) and two additional Spirulina products (Code: SP # 1 and # 2) were examined. Of each product two samples composed of 3 subsamples were extracted separately. The samples were extracted in 0.1 M trichloroacetic acid (TCA) and sonicated with a probe sonicator (600 watts; 30 khz) for 2 min on ice. Subsequently, they were centrifuged at 15,800 x g for 5 min at 4 C. The supernatants, consituting the free fraction of potentially contained BMAA, were dried by vacuum centrifugation and stored at -20 C until further processing. The pellets were washed three times with 1 ml acetone (-20 C) prior to acidic hydrolysis with 6 M HCl for 24 h at 110 C. Afterwards, the hydrolysates, constituting the protein-bound fraction of potentially contained BMAA, were dried and stored as described for the supernatants. All fractions were reconstituted in 10% ACN/H 2 O, followed by centrifugal filtration at 10,000 x g using centrifugal filter units (Millipore; ultrafree-mc GV 0.22 µm). The filtrates were transferred into UPLC-vials (Waters; glass maximum recovery vials) and stored at -20 C until analysis. Two subsamples per replicate were spiked with 10 ng BMAA: both during extraction, one prior to hydrolysis and again both prior to final analysis. 99

100 Chapter VI MC and BMAA in BGAS Liquid Chromatography Tandem Mass Spectrometry (LC-MS/MS) analysis of BMAA Extracts of the blue-green algal supplements were sent to Dr. Alison Robertson at the National Oceanic and Atmospheric Administration, Northwest Fisheries Science Center for confirmatory mass spectrometric analysis of BMAA. Sample extracts were shipped on dry ice and stored at -80 C upon arrival. All solvents and water used for our analyses were of HPLC grade and purchased from JT Baker Laboratory Chemicals (Phillipsburg, NJ). Samples were analyzed by liquid chromatography tandem mass spectrometry (LC-MS/MS) in positive ion mode using multiple reaction monitoring (MRM) for the confirmation of BMAA. All MS experiments were performed using a Waters UPLC system coupled to a Quattro micro triple quadrupole tandem mass spectrometer (MicroMass, Waters, US) fitted with an electrospray ionization interface (capillary: +2.6 kv, cone: +45 V). Normal phase analyte separation was achieved using hydrophilic interaction with a TosoTSKgel Amide-80 analytical column (5 µm, 250 mm x 2.1 mm, Waters, MA) with a guard column (2 mm x 1 mm), both maintained at 40 C. A flow rate of 0.25 ml/min resulted in a stable backpressure of approx psi and was used for all analyses. Samples (10 µl injection volume) were separated using a linear gradient from 90-30% B for 10 min, followed by a 1 min hold at 30% B, and then reequilibration to 90% B for 5 min. Mobile phase consisted of water (A) and acetonitrile (B) both containing 50 mm formic acid. Method optimization, calibration, retention time verification, matrix effects and characteristic MS/MS fragmentation was determined using a standard solution of L-BMAA hydrochloride from Sigma ( 97% purity; B107; Lot 087K47032) at the same concentration as used for spiking. Identification of suspect toxin peaks was undertaken using multiple reaction monitoring (MRM) for three characteristic transition ions of the protonated BMAA including; 119>102 (collision: 10 ev), 119>88 (collision: 15 ev), and 119>74 (collision: 10 ev). The relative ion ratios for these three transitions was also recorded and used for additional verification of BMAA. All MS/MS parameters were optimized prior to analysis using full MS and MS/MS scans and quantification was determined using an 8-point calibration curve (50 pg/ml - 5 µg/ml DA). Matrix effects were assessed by spiking and standard addition experiments and samples were 100

101 Chapter VI MC and BMAA in BGAS subsequently diluted as required to eliminate any observed effects such as ionization suppression and BMAA retention shifts. Data analysis including peak integration and quantitation was performed using MassLynx software (version 4.1; Waters Laboratory Informatics). Mass spectrometer source and analyzer parameters were as follows: Extractor 2 V; RF Lens 0.2 V; source temperature 130 C; desolvation temperature 385 C; cone gas flow 200 L/Hr; desolvation gas flow 500L/Hr; collision cell entrance 50 and exit -8; and multiplier at 650 V. Statistics In order to compare the different methods of MC analysis the corresponding mean values (MCLR equivalents and total MC concentration) of BGAS samples Aph # 1, Aph # 8 and Aph # 9 were statistically analyzed by One-way ANOVA followed by Tukey's Multiple Comparison Test using GrAphPad Prism 4.03 software. 6.4 Results Colorimetric protein phosphatase inhibition (cppia) assay with recombinant PP1 The inhibitory capacities of 11 A. flos-aquae extracts (Aph # 1-11) on recombinant PP1 were tested in three independent colorimetric protein phosphatase inhibition assays. The concentrations determined by comparing the inhibition of the BGAS samples with the inhibiton of MCLR standard dilutions ranged from µg MCLR equivalents/g dw (Tab. 6.2). Every sample elicited inhibitiory effects on PP1, whereas the MAC of 1 µg/g dw defined by the Oregon Health Division and the Oregon Department of Agriculture was exceeded by 7 samples. Adda-Enzyme Linked Immunosorbent Assay (Adda-ELISA) Three of the 11 BGAS extracts were investigated by the Adda-specific ELISA. The assay revealed MC contaminations in all three BGAS samples (Aph # 1, # 8 and # 9) analyzed. However, the MAC of 1 µg/g dw was only exceeded by 101

102 Chapter VI MC and BMAA in BGAS BGAS samples Aph # 8 and # 9 (Tab. 6.2). Moreover, the MC concentrations in these two samples were in the same range as deteced by Hoeger et al. (Hoeger et al., 2003). Tab. 6.2: MC concentrations in BGAS as determined by cppia, Adda-ELISA (expressed as MCLR equivalents) and LC-MS/MS; values of cppia and ELISA represent mean ± standard deviation of at least three independent replicate analyses of three (Aph # 2-11) or four (Aph # 1) separately extracted samples per BGAS; values of LC-MS/MS represent mean, maximum and minimum (in brackets) of one analysis of three (Aph # 2-11) or four (Aph # 1) separately extracted samples per BGAS. Aph cppia Adda-ELISA LC-MS/MS MCLR equivalents [µg/g] MCLR [µg/g] MCLA [µg/g] total MC [µg/g] # ± ± 0.02 n.d. n.d. n.d. # ± (0.25; 0.27) 0.03 (0.03; 0.03) 0.29 (0.28; 0.29) # ± (0.20; 0.21) 0.25 (0.24; 0.26) 0.46 (0.45; 0.46) # ± (1.26; 1.31) 0.21 (0.20; 0.21) 1.49 (1.46; 1.52) # ± (1.42; 1.56) 0.03 (0.03; 0.04) 1.53 (1.46; 1.59) # ± (0.08; 0.08) n.d (0.08; 0.08) # ± ± 0.30 * 1.14 (1.09; 1.18) 0.03 (0.03; 0.03) 1.17 (1.12; 1.20) # ± ± ± 0.50 * 5.42 (5.29; 5.57) 0.28 (0.28; 0.28) 5.69 (5.57; 5.85) # ± ± ± 0.08 * 1.08 (0.95; 1.32) 0.04 (0.04; 0.04) 1.12 (0.99; 1.36) # ± ± 0.16 * 0.36 (0.35; 0.36) 0.12 (0.11; 0.12) 0.47 (0.47; 0.48) # ± (0.03; 0.03) 0.10 (0.10; 0.10) 0.13 (0.13; 0.13) *, (Hoeger et al., 2003) -, not determined in this study n.d., none of the commercially available standard MCs detectable. 102

103 Chapter VI MC and BMAA in BGAS LC-MS/MS analysis of MCs The 11 different A. flos-aquae products (Aph # 1-11) were additionally analyzed for MCs by LC-MS/MS. The MC congeners contained in the respective samples and the means of their calculated concentrations using the corresponding MC standards are shown in Table 6.2. None of the commercially available congeners MCRR, -YR, -LR, -LA, -LW and -LF (Fig. 6.1 (A)) were detected in BGAS sample Aph # 1 using the sensitive MRM modus (Tab. 6.1). The other BGAS products (Aph # 2-11) were found to be contaminated with different MC congeners in concentrations ranging from µg/g dw (total MC content). According to this analysis five of the 11 BGAS samples exceeded the MAC of 1 µg/g dw. Besides MCLR, which appeared to be present in all samples but Aph # 1, MCLA was also commonly detected (except for Aph # 1 and # 6). Representatively, the chromatogram of BGAS sample Aph # 8 is shown in Fig. 6.1 (B). Furthermore, in sample Aph # 8 traces of MCRR were found. Fig. 6.1: Reconstructed LC-MS/MS chromatograms of a mixture of standard microcystins (A) and of an extract of BGAS sample Aph # 8 (B) using the MRM mode (Tab. 6.1). 103

104 Chapter VI MC and BMAA in BGAS Comparison of the different MC analyzing methods The three detection methods employed resulted in distinct differences in the determination of especially low MC contaminations of some of the BGAS as exemplified in Fig. 6.2 (Aph # 1 and # 9). On the contrary, high contaminations revealed no significant differences in the analyses (Fig. 6.2 (Aph # 8)). However, with increasing contamination the standard deviations increased at the same time, when detected by cppia or ELISA (Tab. 6.2). In general LC- MS/MS and cppia analysis yielded the lowest and highest concentrations, respectively, whereas the results obtained by the Adda-ELISA appeared to be in between the other two methods of detection except for BGAS sample # 1 (Tab. 6.2). MCLR equivalents [µg/g dw] *** n.d. * cppia ELISA LC-MS/MS cppia ELISA LC-MS/MS cppia ELISA LC-MS/MS Aph # 1 Aph # 8 Aph # 9 Fig. 6.2: Comparison of the methods employed for MC analysis of three different BGAS samples (Aph # 1, # 8 and # 9); each sample was analyzed by cppia, Adda- ELISA and LC-MS/MS; bars represent means ± standard deviation of three (cppia), two (Adda-ELISA) or one analysis of three (Aph # 8 and # 9) or four (Aph # 1) separately prepared BGAS extracts; n.d., none of the standard MCs (commercially available standards) detectable. The different methods were compared for each BGAS by one-way ANOVA followed by Tukey's Multiple Comparison Test; * P<0.05; *** P< LC-MS/MS analysis of BMAA The LC-MS/MS analysis revealed that neither the three A. flos-aquae products (Aph # 1, # 8 and # 9) nor the two Spirulina products (SP # 1 and # 2) analyzed 104

105 Chapter VI MC and BMAA in BGAS contained any BMAA. The excitotoxic amino acid was positively detected in the spiked extracts only as representatively exemplified in Fig. 6.3 and 6.4 with Aph # 1 and # 9, respectively. Matrix peaks were also present, but could be delineated from true BMAA peaks due to different retention times and absence in one or two of the characteristic BMAA ion transitions as confirmed by the spiked samples. Fig. 6.3: LC-MS/MS chromatograms of BGAS sample Aph # 1 with three characteristic ion transitions for BMAA: 119>102 (black), 119>88 (blue), and 119>74 (red). A: non-spiked free BMAA fraction; B: spiked free BMAA fraction; C: protein hydrolysate (protein-bound BMAA fraction); D: spiked protein hydrolysate (proteinbound BMAA fraction); E: BMAA standard (at same level as spike). BMAA was confirmed when verified in at least 3 ion transitions and when the retention time (or spike peak) was concordant with the standard. 105

106 Chapter VI MC and BMAA in BGAS Fig. 6.4: LC-MS/MS chromatograms of BGAS sample Aph # 9 with three characteristic ion transitions for BMAA: 119>102 (black), 119>88 (blue), and 119>74 (red). A: non-spiked free BMAA fraction; B: spiked free BMAA fraction; C: protein hydrolysate (protein-bound BMAA fraction); D: spiked protein hydrolysate (proteinbound BMAA fraction); E: BMAA standard (at same level as spike). BMAA was confirmed when verified in at least 3 ion transitions and when the retention time (or spike peak) was concordant with the standard. 106

107 Chapter VI MC and BMAA in BGAS 6.5 Discussion Detection of microcystins in blue green algae supplements The 11 BGAS products of different brands containing Aphanizomenon flosaquae from Lake Klamath examined for contamination with MC were all tested positively by cppia. This was confirmed by the LC-MS/MS analysis except for sample Aph # 1 in which none of the MC congeners commercially available could be detected. MC standards are a prerequisite for the sensitive MRM modus, which allows the detection of concentrations in the range of pg o.c. The precursor ion mode was applied for the detection of those MC congeners without standard available for comparison. However, the limit of detection for this technique was higher (approximately 1 µg/g dw). This is one possible explanation for the higher MC concentrations that were determined by cppia, as well as by Adda-ELISA. In addition, since the cppia measures the total inhibition of PPs, in this case PP1, it has to be taken into consideration that different MC congeners may not inhibit PPs equally, even though the MC congeners commercially available appeared to be equipotent or at least in the same range with regard to PP inhibition (Rivasseau et al., 1999; Robillot and Hennion, 2004; Fischer et al., 2010). Indeed, Hoeger et al. (Hoeger et al., 2007) found significant differences in protein phosphatase inhibition between MC-RR and [Asp 3,Dhb 7 ]MC-RR: The PP1- and PP2A-IC 50 of the latter revealed to be 22-fold and 50-fold higher, respectively. Besides, 6(Z)-Adda isomers of MCLR and MCRR also showed a dramatically reduced inhibitory action (100-fold) on PP2A (Nishiwaki-Matsushima et al., 1991). Also, other substances with inhibitory capacity on PP1 might have been present in the extracts and hence would have interfered false-positively with the actual MC contamination. Furthermore, the complex matrices of the extracts might false-positively contribute to the increased levels of contamination observed in the cppias and Adda-ELISAs as also indicated by the relatively high variability of the results. Besides the analysis itself, this variability also appears to depend on the method of extraction, as well as the kind of standards used for quantification (Fastner et al., 2002). However, despite these impairments and although no certified reference material (CRM) was employed the concentrations determined by the Adda-ELISA are in accordance with Hoeger et al. (Hoeger et al., 2003) as 107

108 Chapter VI MC and BMAA in BGAS shown in Table 6.1. And nonetheless, despite the differences in the MC contamination determined by these detection methods, the respective results still support each other as they are in the same range. At the same time, they emphasize the importance of combining at least two different methods to analyze MC contamination as concluded by various authors (Hawkins et al., 2005; Spoof, 2005). The number of BGAS samples that exceeded the MAC of 1 µg/g dw appeared to be dependent on the method of analysis. While 6 samples fell below this guideline when analyzed by LC-MS/MS, this was correct for only 4 samples when analyzed by cppia. Thereby, the highest concentrations determined exceeded the MAC 5-8 times. In addition, all of the samples were positive for contamination with MCs. Hence, these findings clearly demonstrate the serious health threat for the consumers of these dietary supplements. Especially, when taking into consideration that the determination of the TDI, of which the guideline value for BGAS has been derived, was calculated for a 60 kg adult, it becomes obvious that particularly children are at high risk. Batches of BGAS that undercut the value of 1 µg/g dw are officially considered safe for human consumption and thus are released for the market. However, a MC contamination of half of this value equals the MAC for a child with a body weight of 30 kg. In fact, these supplements are also marketed for the use by children, e.g. as a substitute for the pharmacological therapy of the Attention Deficit Hyperactivity Disorder (ADHD) (Linderman, 1995; Simonsohn, 2000; Simonsohn, 2006). The reasonability of the guideline value and the risk posed by BGAS consumption has already been reviewed and discussed in detail (Dietrich and Hoeger, 2005; Dietrich et al., 2008). Moreover, in 2006 the WHO / International Agency for Research on Cancer (IARC) made the adjudication to categorize MCs as a possible carcinogen in humans (Grosse et al., 2006). Under this aspect even contaminations below the MAC might pose an unassessed risk for adult consumers as well. Detection of BMAA in blue green algae supplements None of the three A. flos-aquae products (Aph # 1, # 8 and # 9) or two Spirulina products (SP # 1 and # 2) analyzed for BMAA contamination using LC-MS/MS 108

109 Chapter VI MC and BMAA in BGAS were tested positively. However, we were able to detect BMAA in the spiked samples, clearly demonstrating the effectiveness of this method of detection. These findings suggest that cyanobacterial dietary supplements are unlikely to be a source for BMAA despite its apparent common occurrence in 95% of the cyanobacterial genera tested (Cox et al., 2005). However, more BGAS products have to be examined in order to evaluate this hypothesized risk properly. Furthermore, the different methods of detection and quantification applied for BMAA appear to be conflictive. A frequently employed method is the derivatization of BMAA with 9-fluorenylmethyl chloroformate (FMOC) (Kisby et al., 1988; Kisby et al., 1992; Montine et al., 2005) or 6-aminoquinolyl-Nhydroxysuccinimidyl carbamate (AQC) (Banack and Cox, 2003; Cox et al., 2003; Murch et al., 2004a; Murch et al., 2004b; Cox et al., 2005; Banack et al., 2006; Metcalf et al., 2008) prior to its separation by HPLC with subsequent fluorescence detection. Even though LC-MS or LC-MS/MS analysis was additionally applied in the studies that used AQC as derivative to verify or further analyze the obtained results, this method was challenged by various authors: Robertson et al. (Robertson et al., 2007) examined five different cyanobacterial species from four different genera. All species, except for one, were previously analyzed positively for BMAA using aforesaid method (Cox et al., 2005). However, Robertson et al. were unable to confirm this finding using Hydrophilic Interaction Liquid Chromatography / Mass Spectrometry (HILIC-MS) (Robertson et al., 2007). They criticized that the reports from Cox et al. relied upon LC-fluorescence data and that no clear LC-MS confirmation was provided (Cox et al., 2005). This had also been stated by Duncan and Marini (Duncan and Marini, 2006). In order to conclusively demonstrate the production of BMAA by the marine laboratory strain Nostoc sp. CMMED-01, one of the strains that has been tested positively for BMAA by Cox et al. (Cox et al., 2005), Banack et al. (Banack et al., 2007) confirmed its presence by HPLC-FD, UPLC-UV, Amino Acid Analyzer, LC-MS, and Triple Quadrupole LC-MS/MS. Moreover, Montine et al. (Montine et al., 2005) examined brain tissue from controls, Alzheimer s disease and Chamorros with PDC according to the method of Kisby et al. (Kisby et al., 1988). In contrast to others (Cox et al., 2003; Murch et al., 2004a; Murch et al., 2004b) they failed to detect free BMAA in any but the spiked samples. They speculated that their inability to confirm 109

110 Chapter VI MC and BMAA in BGAS these studies might be due to the similar quantitative method the other authors used along with an unsufficient sample size and improper fixation and preservation techniques. Duncan and Marini (Duncan and Marini, 2006) confirmed this conclusion by declaring the fixation and storage of the samples as inappropriate and the measured data as likely being artefacts. Conclusion The analyses of the dietary supplements containing A. flos-aquae from Lake Klamath presented in this study clearly demonstrate a serious health threat posed by contamination with MCs that frequently exceed the MAC of 1 µg microcystin-lr/g dw. Therefore, these BGAS products should be more thoroughly examined for the presence of MCs and sales controlled or even restricted if contaminations exceeding the MAC persist. In contrast, our data suggest that BMAA apparently doesn t contribute to the potential hazard of BGAS consumption, although more samples are needed to be examined to entirely exclude this possibility. However, anatoxin-a, dihydrohomoanatoxin-a and a novel isomer of epoxyanatoxin-a were detected in BGAS based on Spirulina (Draisci et al., 2001; Osswald et al., 2008). Recently, Rawn et al. (Rawn et al., 2007) tested several Spirulina and A. flos-aquae products for anatoxin-a and its metabolites. While they detected dihydroanatoxin-a and epoxyanatoxin-a in 3 A. flos-aquae products using HPLC with fluorescence detection, they failed to confirm this result by the more sensitive LC-MS/MS analysis and therefore considered the BGAS free of anatoxins. Nevertheless, these findings indicate a further potential threat for consumers of these dietary supplements, especially with regard to a potential cocontamination with MCs. A concomitant occurrence of MCs and anatoxins might elicit synergistic effects as demonstrated by Fitzgeorge et al. (Fitzgeorge et al., 1994), who compared their single toxicities with the toxicity of both toxins administered simultaneously. They determined an intranasal LD 50 of 2000 µg/kg bw for anatoxin-a and a non-lethal dose of 31.3 µg/kg bw for MC-LR in mice. This non-lethal dose was administered 30 minutes prior to anatoxin-a and lead to a reduction of the LD 50 for anatoxin-a to one fourth of the initial value. 110

111 Chapter VI MC and BMAA in BGAS Given the tremendous amount of consumers, the frequent contamination with MCs, and the lack of knowledge on possible contaminations with further cyanotoxins, as well as on potential additive effects in case of co-occurences, it appears obvious that further investigations on these concerns are crucial in order to assess the health risk posed by BGAS properly. Acknowledgements We would like to acknowledge the Arthur & Aenne Feindt Foundation (Hamburg, Germany) as well as the European Union (PEPCY QLRT ) for kindly funding parts of this study. 111

112 Chapter VII General Discussion 7 General Discussion Fossils of the oldest life forms on earth were dated back to 3.3 to 3.5 billion years ago and identified as cyanobacteria (Schopf and Packer, 1987). Thenceforward these organisms were able to develop the required skills to populate nearly any given habitat on earth and to assert themselves against competitiors and planktivores. In the course of this time a vast amount of cyanobacterial secondary metabolites, complex molecules that are not necessary for the primary metabolism, emerged whose functions predominantly remained unknown. However, some of them elicit adverse effects in invertebrates and vertebrates including humans (Kuiper-Goodman et al., 1999; Sivonen and Jones, 1999; Dow and Swoboda, 2000; Kaebernick and Neilan, 2001; Welker and von Dohren, 2006). Such compounds are therefore referred to as cyanotoxins and comprise, inter alia, single amino acids, peptides and alkaloids. The role of the Adda moiety in PP inhibition The cyclic heptapeptidic MCs attracted major attention as they are the most frequently found cyanotoxins in fresh and brackish waters and in addition are responsible for the majority of cyanobacterial poisonings of animals and humans. The inhibition of various serine/threonine-specific PPs, especially PP1 and PP2A, has been identified as the primary molecular mechanism underlying the toxicity of MCs (see chapter I). The structural unmodified Adda moiety has thereby been shown to be of utmost importance for the inhibitory activity and hence toxicity of MCs and NODs (Harada et al., 1990a; Harada et al., 1990b; Nishiwaki-Matsushima et al., 1991; Namikoshi et al., 1994). However, isolated Adda failed to reduce PP1 activity and turned out to be non-toxic when injected intraperitoneally to mice (Harada et al., 2004). The results presented in chapter III confirm the lack of inhibitory action of Adda on PP1 and in addition on PP2A (Fig. 3.1). Thus, MC- and consequently NOD-mediated PP inhibition not only requires a structural unmodified Adda moiety, but also the remaining structural units of the maternal toxins. 112

113 Chapter VII General Discussion Organotropism and cell-trafficking of MCs As described in chapter I, MCs are predominantly organotropic for the liver, but may also be distributed to other organs. Cell trafficking via passive diffusion is thereby precluded due to their rather hydrophilic character and spatially large size. Thus, cellular uptake and excretion require active transport via transmembrane carriers. Obviously, the identification of such carriers is of utmost importance for an improved understanding of the toxicodynamics of MCs and might also serve as an approach for medication against intoxications with these cyanotoxins (e.g. administration of therapeutic agents that compete with MCs for cellular uptake). The involvement of the bile acid transport system in MC cell trafficking has been demonstrated in earlier studies (Eriksson et al., 1990b; Runnegar et al., 1991; Runnegar et al., 1995c). Indeed, more recent investigations confirmed certain members of the sodium-independent OATP/SLCO superfamily as being capable of mediating the uptake of MCs (Fischer et al., 2005; Meier-Abt et al., 2006; Komatsu et al., 2007; Monks et al., 2007; Lu et al., 2008; Feurstein et al., 2009). The presented cytotoxicity and competitive uptake experiments on human liver OATP1B1- and OATP1B3-expressing HEK293 cells, as well as primary human hepatocytes (chapter IV) unambiguously corroborate these findings and support the involvement of these OATPs in the organotropism of MCs for the liver (i.e. uptake from the portal vein into the hepatocytes). However, only one further human OATP, namely OATP1A2, has been identified to date as being capable of mediating the uptake of MC (Fischer et al., 2005). Although this carrier might contribute to the organotropism for kidneys (i.e. excretion from the renal tubular cells into the tubules and resorption from the tubules into the renal tubular cells), brain (i.e. uptake from the cerebral blood into the capillary endothelial cells and across the blood-brain-barrier) and liver (i.e. uptake from the portal vein into the hepatocytes), as it is expressed in these organs (Hagenbuch and Meier, 2003; Hagenbuch and Meier, 2004; Bronger et al., 2005; Ho and Kim, 2005; Lee et al., 2005; Nies, 2007), the identity of the transporters that are responsible for the uptake into other MC-affected tissues (e.g. from the intestine into the enterocytes and from the afferent arterioles into the renal tubular cells) remains to be uncovered. 113

114 Chapter VII General Discussion So far 11 human OATPs are known and have been found in nearly all tissues, where they play important roles in the absorption, distribution and excretion of numerous endogenous and exogenous substrates (Hagenbuch and Meier, 2003; Hagenbuch and Meier, 2004). Since out of the 4 human OATPs tested by Fischer et al. (Fischer et al., 2005) (OATP1A2, -2B1, -1B1 and -1B3) only OATP2B1, detected in multiple tissues including liver, placenta, intestine and brain (Hagenbuch and Meier, 2003; Hagenbuch and Meier, 2004; Bronger et al., 2005; Ho and Kim, 2005; Nies, 2007), failed to transport [ 3 H]-dihydro-MCLR, it appears very likely that further members of the OATP/Oatp superfamily are capable of mediating uptake of MCs. Table 7.1 summarizes known and possible MC transporting OATPs/Oatps in a selection of tissues that are predominantly affected by MC intoxications. There is also the fact that transport via OATPs/Oatps is generally directed into cells. Although bidirectional transport was reported for some members including OATP1A2, -1B1 and -1B3 (Shi et al., 1995; Li et al., 2000; Hagenbuch and Meier, 2003; Hagenbuch and Meier, 2004; Ho and Kim, 2005; Mahagita et al., 2007), it cannot provide for an explanation for the observed cell trafficking in most tissues, since OATPs are usually located at certain membrane sites only. For example, MC-transporting OATP1A2, -1B1 and -1B3 are located at the basolateral membrane of hepatocytes. However, no OATPs/Oatps have been detected in the apical membranes of hepatocytes (Kullak-Ublick et al., 1994; Eckhardt et al., 1999; Kakyo et al., 1999; König et al., 2000a; König et al., 2000b; Hagenbuch and Meier, 2004; Ho and Kim, 2005). Nevertheless, MC have been shown to pass this barrier into the bile canaliculi and to follow the enterohepatic circle (Falconer et al., 1986; Runnegar et al., 1986; Sahin et al., 1996; Stotts et al., 1997a; Stotts et al., 1997b). Hence, other transporters must also be involved in MC cell trafficking, especially exporters that are responsible for the excretion from the enterocytes into the portal vein and from the hepatocytes into the bile. Multidrug resistance proteins (MDRs/ABCBs) and multidrug resistance associated proteins (MRPs/ABCCs), both members of the ATP Binding Cassette superfamily (ABC transporters), are suspected to play a role in MC organotropism and detoxification by excretion (Ito et al., 2002; Dietrich et al., 2008) for various reasons: 114

115 Chapter VII General Discussion Endogenous and xenobiotic substances that are taken up by OATPs/Oatps and biotransformed (phase 1 and 2 reactions) by cytochrome P450- mediated oxidation and/or conjugation with glucuronate, sulfate, or GSH are commonly excreted by MRPs/ABCCs or MDRs/ABCBs (König et al., 1999a; Faber et al., 2003; Keppler, 2005). Metabolization of MCs has been shown to occur via conjugation to GSH (Kondo et al., 1996; Pflugmacher et al., 1998; Takenaka, 2001). MRPs and MDRs mediate transport in the opposite direction than OATPs irrespective wether they are located at opposite membrane sites than OATPs, i.e. at the apical membranes of hepatocytes (MDR1/ABCB1, MDR3/ABCB3, MRP2/ABCC2) and basolateral membranes of enterocytes (MRP1/ABCC1, MRP3/ABCC3) or at the same membrane site, i.e. MRPs and MDRs at the apical membranes of renal tubular cells (MDR1/ABCB1, MRP2/ABCC2, MRP4/ABCC4), enterocytes (MDR1/ABCB1, MRP2/ABCC2), capillary endothelial cells of the brain (MDR1/ABCB1, MRP1/ABCC1, MRP4/ABCC4, MRP5/ABCC5) and at basolateral membranes of hepatocytes (MRP1/ABCC1, MRP3/ABCC3, MRP4/ABCC4) (Cordon-Cardo et al., 1989; Cordon-Cardo et al., 1990; König et al., 1999a; König et al., 1999b; Borst and Elferink, 2002; Faber et al., 2003; van Montfoort et al., 2003; Fromm, 2004; Ho and Kim, 2005; Rius et al., 2006; Nies, 2007). The molecular weight of MCs (900 to 1100 Da) is within the range of MDRs and MRPs substrates (König et al., 1999a; Borst et al., 2000; Faber et al., 2003; Schinkel and Jonker, 2003). Phosphatase inhibition was found to decrease the uptake of MCs into isolated rat hepatocytes suggesting the involvement of phosphorylationcontrolled transporters in MC cell trafficking (Runnegar et al., 1995c). Confirmation of one or several members of these multispecific ABC carriers would tremendously improve the understanding of MC organotropism. 115

116 Chapter VII General Discussion Tab. 7.1: Organ specifity of known and possible MC transporting organic anion transporting polypeptides [human: OATPs/Slcos; animals: Oatps/Slcos; (protein name/gene symbol)]. organs human rat/mouse/skate references gastrointestinal tract OATP2A1 OATP2B1 - OATP3A1 OATP4A1 Oatp1a5 (Hagenbuch and Meier, 2004; Cheng et al., 2005; Ho and Kim, 2005) OATP1A2 x OATP1B1 x,* Oatp1a1 - (Kullak-Ublick et al., 2001; Faber et al., OATP1B3 x,* Oatp1a4-2003; Hagenbuch and Meier, 2003; van liver OATP2A1 Oatp1b2 x,y Montfoort et al., 2003; Hagenbuch and OATP2B1 - Oatp1d1 z Meier, 2004; Cheng et al., 2005; Ho and OATP3A1 Oatp2b1 Kim, 2005; Lee et al., 2005) OATP4A1 OATP1A2 x Oatp1a1 - (Tamai et al., 2000; Hagenbuch and OATP2A1 Oatp1a3 Meier, 2003; van Montfoort et al., 2003; kidney OATP3A1 Oatp1a6 Hagenbuch and Meier, 2004; Cheng et OATP4A1 Oatp3a1 al., 2005; Gao et al., 2005; Ho and Kim, OATP4C1 Oatp4c1 2005; Lee et al., 2005) blood-brainbarrier OATP1A2 x OATP1C1 OATP2A1 OATP3A1 OATP4A1 OATP2B1 - Oatp1a4 Oatp1c1 (Gao et al., 2000; Pizzagalli et al., 2002; Hagenbuch and Meier, 2003; van Montfoort et al., 2003; Hagenbuch and Meier, 2004; Bronger et al., 2005; Cheng et al., 2005; Gao et al., 2005; Ho and Kim, 2005; Lee et al., 2005; Nies, 2007) - no dihydro-mclr transport demonstrated by Fischer et al. (Fischer et al., 2005) x dihydro-mclr transport demonstrated by Fischer et al. (Fischer et al., 2005) y MCLR transport demonstrated by Lu et al. (Lu et al., 2008) z MCLR transport demonstrated by Maier-Abt et al. (Meier-Abt et al., 2006) * MCLR, -RR (OATP1B3 only), -LW and -LF transport demonstrated in the present study 116

117 Chapter VII General Discussion The role of liver OATPs in MC congener-specific toxicity Another crucial finding of the studies presented in chapter IV is that the four MC congeners investigated (MCLR, -RR, -LW and -LF) differ dramatically with regard to toxicokinetics (Fig. 4.5 and 4.7), i.e. congener-selective uptake by OATPs, but not with regard to toxicodynamics, as they equipotently inhibited recombinant as well as intracellular serine/threonine-specific PPs (Fig. 4.1 and 4.2). This strongly suggests that MC toxicity is congener-specific and primarily mediated by the uptake of OATPs. Despite these convincing results, further research is needed to exclude congener-dependent differences in the metabolization of MCs that might also contribute to differences in bioavailability and, hence, toxicity. On the other hand, differences in metabolization, i.e. conjugation of MC to GSH, are rather unlikely as the four congeners investigated are all from the same group of MCs according to the classification of Kaya et al. (Kaya et al., 2001): They all posess a Mdha residue and offer therefore a binding possibility for both GSH and PPs via the α-, β-unsaturated carbonyl of the Mdha moiety (Hermansky et al., 1991; Kondo et al., 1992). In addition, as all four congeners nearly equipotently inhibited PP1 and 2A, it might be assumed that GSH conjugation to these four MCs is similarly congener-independent. Thus, the differences in cytotoxicity observed in OATP-expressing HEK293 cells and primary human hepatocytes most likely reflect varying affinities and transporting capacities of OATP1B1 and -1B3 for these MCs which, in turn, provides a very plausible explanation for the discrepancy between the divergent in vivo toxicity (mouse, i.p. (Krishnamurthy et al., 1986; Watanabe et al., 1988)) and similar phosphatase inhibition of MCLR and MCRR (this study; Yoshizawa et al., 1990; Fujiki et al., 1996). Unexpectedly, the more hydrophobic congeners LW and LF revealed to be the most toxic in both OATP-expressing HEK293 cells and primary human hepatocytes (Tab. 4.1; Fig. 4.5 and 4.7). Hitherto MCLR has been assumed to be the most potent and abundant congener (Rinehart et al., 1994; Sivonen and Jones, 1999). That is why risk assessments and guideline values are based on the toxicity of this congener only (Falconer et al., 1994; Fawell et al., 1994; WHO, 1998; Fawell et al., 1999; Gilroy et al., 2000). Unfortunately, no data exist 117

118 Chapter VII General Discussion on the toxicity of MCLW and -LF in mammals. However, if these congeners prove to be more toxic than MCLR in vivo as well, current risk assessments and guidline values might have to be revised. Thereby, studies should not only focus on the acute in vivo toxicity, but also on tumour promotion caused by chronic exposure. The observed differences in toxicokinetics (i.e. varying transporter affinities) will most certainly lead to differences in bioavailability in vivo and, hence, might result in a congener-dependent tumour promoting potency, despite similar toxicodynamics. Accordingly, this would also have implications on the monitoring of toxic cyanobacterial blooms: Commonly used screening methods like the functional colorimetric protein phosphatase inhibition assay (cppia) or the structural Adda-ELISA indicate toxicodynamics and recognize Adda-containing molecules (i.e. MCs and NODs), respectively. However, neither of them gives information on the composition of MC congeners in a bloom sample. For such a purpose expensive and time-consuming detection systems like HPLC or LC-MS/MS are needed. Consequently, the threat posed by a MCLW-/MCLF-producing bloom might be underestimated. On the other hand, there is an information shortage on the distribution and abundance of blooms that naturally produce MCLW and -LF (see chapter IV). Therefore, the actual threat that strongly depends on these two factors, too, remains to be assessed. Comparison of primary hepatocytes and OATP-expressing HEK293 cells The employment of primary human hepatocytes in the investigation of the toxicodynamics of MCs is of obvious importance: First of all hepatocytes are the primary target of intoxications with these cyanotoxins. Moreover, within the first 96 hours after their isolation primary hepatocytes still maintain the in situ expression pattern of their transmembrane carriers (Jigorel et al., 2005) as well as their drug metabilizing enzymes (Li, 2001; Gomez-Lechon et al., 2003; Donato et al., 2008). Hence, the use of these primary cells may provide physiologically highly relevant results, yet implies some weak points: Factors like age, state of health, medication/xenobiotic treatment are donorspecific, hence, add to data variability (see chapter IV). Availability of primary tissue is usually highly limited. 118

119 Chapter VII General Discussion Prolonged culturing leads to a loss of transporters and alters the metabolic activity (vide supra). Uptake of MCs can be measured as a reduction in viability and is inhibitable by coincubation with OATP substrates (Fig. 4.7 and 4.8). However, this does not allow for a perspicuous identification of the responsible transporters (i.e. the distinct member of bile acid transport system). In order to attribute uptake-dependent MC toxicity to certain carriers, i.e. to proof transporter associated effects, liver OATP-expressing HEK293 cells were included in the investigations. Thereby, causality between the expression of the respective OATP and cytotoxic effects was guaranteed by: the inclusion of control vector-transfected HEK293 cells that served as a negative control for the transporter-dependent toxicity of MCs (Fig. 4.5). the use of the cell-permeant PP inhibitor okadaic acid as a transporterindependent positive control (Fig. 4.5). coincubation of MCLR and OATP substrates BSP and TC that lead to a reduction in susceptibility against MCLR (Fig. 4.6). the absence of endogenous possibly interfering OATPs that was corroborated by missing hints for a transport directed into the parental HEK293 cells (Keppler, 2006; personal communication). Moreover, the use of transfected immortalized cell lines implies additional advantages over primary cells: cell lines allow for a permanent and lowmaintenance culture, posess a stable metabolism and express the transfected protein at constant levels. Hence, such cell models enhance reproducibility and minimize variability of the results at the same time. The results presented in chapter IV not only confirmed the uptake of MCs into both human hepatocytes and OATP-expressing HEK293 cells, but also revealed that hepatocytes are distinctly more sensitive (Tab. 4.1) even though the immunocytochemical staining of OATP1B1 and -1B3 appeared to be obviously lower compared to the HEK293 cells (Fig. 4.4). This is possibly due to the observed lower levels of serine/threonine-specific PPs (Fig. 4.3), but might also stem from a decrease in metabolic, hence, detoxicating activity and/or an increased uptake by interfering OATPs (e.g. OATP1A2). In any case, the tremendous differences in the susceptibility highlight the importance to employ a cell model expressing the carrier of interest in 119

120 Chapter VII General Discussion combination with primary cells of target organs/tissues in order to obtain both informative and physiologically relevant data. The role of liver OATPs in CYN toxicity Most of the severe human intoxications caused by cyanotoxins can be either attributed to MCs or CYNs. In fact, CYN was probably to some extend involved in the infamous Caruaru incident (see chapter I; Carmichael et al., 2001) and is considered the causative agent in the Palm Island Mystery Disease (see chapter I). Currently, these cyto- and genotoxic alkaloids pose a serious threat to human health as high concentrations have been repeatedly reported from drinking water sources in the USA and Australia (Falconer and Humpage, 2006). However, this threat is not only restricted to those countries: CYN was found to be produced by a range of cyanbacterial species from different genera throughout the world. On top of that, some of these species, especially C. raciborskii, were not only found to be widely distributed, but also to be invasive (see chapter I). Hence, within the last 15 to 20 years CYN moved into the focus of cyanotoxin research. However, in contrast to MCs the knowledge on CYN toxicity including mechanism(s), effects on human primary cells and toxicokinetics is limited and insufficiently understood (see chapter I; Humpage et al., 2008). The data presented in chapter V clearly demonstrate typical dose-response effects of CYN on primary human hepatocytes of two different donors (Fig. 5.1) with 48-h EC 50 s of and nm (donor 1 and 2, respectively (Tab. 5.1). Therewith, primary human hepatocytes were approximately 13 to 80 times less susceptible to CYN than to MCLR (48-h EC 50 : 3.4 and 24.6 nm (donor 1 and 2, respectively (Tab. 5.1)). Despite different exposure times, this ratio roughly coincides with the reported 72-h EC 50 values of both toxins for primary rat hepatocytes: 96.3 nm for CYN and 8 nm MCLR (Chong et al., 2002). Moreover, Chong et al. (Chong et al., 2002) demonstrated a dose-dependent protection of rat hepatocytes against lethal doses of CYN by coincubation with cholate and taurocholate providing hints for a contribution of the bile acid transport system in facilitating the uptake of CYN into rat hepatocytes and further suggested passive diffusion to also be involved. 120

121 Chapter VII General Discussion Indeed, viability of both OATP1B1- and OATP1B3-expressing HEK293 cells was significantly reduced by exposure to 5 µm CYN in contrast to control vector-transfected HEK293 (Tab. 5.2; Fig. 5.2). Nevertheless, a clear trend in the reduction of viability could also be observed in the latter (Fig. 5.2) supporting the assumption of Chong et al. that passive diffusion might also be involved in the uptake of CYN (Chong et al., 2002). However, contrary to the findings of Chong et al. neither TC nor BSP reduced the cytotoxic effects of CYN on the OATP-expressing HEK293 cells (Fig. 5.5 and 5.6). This can most plausibly be explained by differences in the metabolic activity of primary cells and immortalized cell lines as discussed above and by the fact that CYN toxicity relies on two different mechanims: The irreversible inhibition of protein synthesis and the generation of (a) toxic CYN metabolite(s) by CYP450 metabolization (see chapter I; Froscio et al., 2003). The latter mechanism is considered the primary cause of cell death at higher concentrations and over an acute time frame in primary hepatocytes, whereas protein synthesis inhibition might be more relevant in cells with reduced CYP450 enzyme activity (Froscio et al., 2003) which is generally the case in immortalized cell lines (Li, 2001; Rodriguez-Antona et al., 2002; Gomez-Lechon et al., 2003; Donato et al., 2008). Hence, this might explain the observed lack of protection by BSP and TC in OATP-expressing HEK293 cells as well as the tremendously reduced susceptibility of these cells in comparison to primary human hepatocytes. This hypothesis is additionally supported by the reduced cytotoxicity of CYN in KB cells in comparison to primary rat hepatocytes (Shaw et al., 2000; Chong et al., 2002) that Chong et al. (Chong et al., 2002) attributed to the absence of the bile acid transport system. Unfortunately, a preliminary attempt to identify the member of the CYP450 family that is responsible for the formation of the toxic CYN metabolite(s) in primary human hepatocytes by using member-specific inducers and inhibitors failed to provide reproducable results (data not shown). An identification would tremendously improve the understanding of CYN toxicity and would furthermore facilitate the evaluation of the potential role of OATPs in mediating the uptake of CYN. 121

122 Chapter VII General Discussion Cyanotoxin contamination of blue-green algae supplements While the studies discussed above primarily focused on the toxicodynamics and -kinetics of MCs and CYN, the purpose of the analyses presented in chapter VI were aimed at the evaluation of the risk presented by possible contaminations of cyanobacterial dietary supplements with MCs and the excitotoxic amino acid BMAA. Thereby, none of the products analyzed for BMAA (two based on Spirulina and three based on Aphanizomenon flos-aquae) were tested positively (Fig. 6.3 and 6.4). This finding disproves the hypothesis that blue green algae supplements (BGAS) are a potential source for BMAA (see chapter VI) which is at the same time contradictory to the assumption of Cox et al. that BMAA is produced by the vast majority of cyanobacteria (Cox et al., 2005). However, as only a total of five different products was tested, further analyses need to be conducted in order to reliably preclude BMAA contamination of BGAS. In contrast to BMAA, MCs turned out to be common contaminants of dietary supplements containing A. flos-aquae from Upper Klamath Lake in Oregon, USA: All products were tested positively for MCs by at least two different analytical methods with concentrations of total MC ranging from µg/g dw (Tab. 6.2). Thereby, the maximum acceptable concentration (MAC) of 1 µg MC/g dw was exceeded by approximately half of the samples (5-7 out of 11 samples depending on the method of analysis). The variability between the different analyses (Fig. 6.2) demonstrates the importance of employing at least two different analytical methods in order to obtain reliable results. In addition, the potential risk emanating from these BGAS is further highlighted by the identification of MCLR and MCLA as the most common contaminants (Tab. 6.2; Fig. 6.1), as both rank among to the most potent congeners with an LD 50 value of 50 µg/g bw (mouse, i.p.) each (Botes et al., 1982; Botes et al., 1984; Krishnamurthy et al., 1986; Lovell et al., 1989; Stoner et al., 1989; Kaya and Watanabe, 1990). This finding is supported by the analyses of Kuiper- Goodman et al. (Kuiper-Goodman et al., 2000). Furthermore, the provisional guideline value of 1 µg/g dw for BGAS was based on the tolerable daily intake (0.04 µg MCLR/kg bw; see chapter I and VI) that includes several uncertainty factors (Gilroy et al., 2000). However, this guideline value does not take additional sources for MC exposure into account: e.g. 122

123 Chapter VII General Discussion drinking water, food, recreational use of MC contaminated water bodies etc. (see chapter I). In addition, the guideline was calculated for the international unit adult with a weight of 60 or 70 kg. However, A. flos-aquae products are often specifically marketed for consumption by children that, hence, would be exposed to three to six times the MC concentration per day (Dietrich and Hoeger, 2005). Hence, the reasonability of such guidelines becomes equivocal, especially with regard to the tumour promoting capacity of MCs (no adequate studies on chronic exposure have been conducted), the fact that all calculations are based on MCLR toxicity only (see the discussion above on the increased in vitro toxicity of MCLW and -LF) and the unassessed threat of possible synergistic effects caused by co-contamination with different cyanotoxins as already discussed by Dietrich and Hoeger (Dietrich and Hoeger, 2005) and Dietrich et al. (Dietrich et al., 2008). Indeed, BGAS based on Spirulina/Arthrospira have been reported to contain anatoxins (Draisci et al., 2001; Osswald et al., 2008) as well as MCs (Gilroy et al., 2000; Kuiper-Goodman et al., 2000; Iwasa et al., 2002) which have been demonstrated to act synergistically (see chapter I and VI; Fitzgeorge et al., 1994). This hazard is potentially present in BGAS containing A. flos-aquae as well: Although the A. flos-aquae strain (Ralfs ex Born. & Flah. var. flos-aquae) harvested for BGAS production from Upper Klamath Lake is considered nontoxic itself (Carmichael et al., 2000) other strains of this species have been reported to produce anatoxin-a (Rapala et al., 1993; Carmichael, 1997) and various PSP toxins (Jackim and Gentile, 1968; Ikawa et al., 1982; Mahmood and Carmichael, 1986a; Carmichael, 1997; Pereira et al., 2000; Ferreira et al., 2001; Liu et al., 2006). In addition, the coexisting cyanobacterial species in Upper Klamath Lake Microcystis aeruginosa, Anabaena flos-aquae and Oscillatoria spp. (Carmichael et al., 2000) might not only be responsible for the observed MC contaminations, but may also be potential producers of anatoxins and PSPs (Devlin et al., 1977; Mahmood and Carmichael, 1987; Park et al., 1993; Humpage et al., 1994; Sivonen and Jones, 1999; Carmichael et al., 2000; Wiegand and Pflugmacher, 2005; van Apeldoorn et al., 2007). Furthermore, Preußel et al. concluded to expand the surveillance of A. flos-aquae-based 123

124 Chapter VII General Discussion BGAS to CYN as they detected substantial amounts of this toxin in A. flosaquae strains from Germany (Preussel et al., 2006). In conclusion, the consumption of A. flos-aquae containing dietary supplements implies an actual risk emanating from frequently detected contaminations with MCs (present study; Gilroy et al., 2000; Hoeger et al., 2003; Kuiper-Goodman et al., 2000; Lawrence et al., 2001; Saker et al., 2005; Saker et al., 2007) as well as a potential risk posed by contaminations with further cyanotoxins. 124

125 Summary 8 Summary Within their 3.3 to 3.5 billion years of existance cyanobacteria became the most diverse and widespread group of prokaryotes and adapted to nearly any given habitat throughout the world. During this period of time they developed an enormous variety of secondary metabolites with predominantly unknown physiological function. Some of these chemically and structurally very diverse compounds have been in the focus of international research ever since their toxic potential on other organisms became known. Especially the MCs and more recently the CYNs have drawn major attention due to their frequent and widespread occurence and numerous in part fatal incidents of animal and human poisonings. MCs are cyclic heptapeptides comprising more than 80 congeners, whereas CYNs are classified as alkaloids. They consist of a sulphated and methylated tricyclic guanidino moiety linked to a hydroxymethyluracil. The primary mechanism underlying the toxicity of MCs emerged to be the potent inhibition of serine/threonine-specific PPs, in particular PP1 and 2A. The unique amino acid Adda has thereby been identified as the toxic moiety of the MC molecule. However, isolated Adda was shown to be inhibitory ineffective on PP1 activity. Corresponding effects on PP2A were not determined. Therefore, the first part of this study served to provide the lacking information: As expected, isolated Adda also failed to inhibit the activity of PP2A which corroborates the common assumption that PP inhibition by MCs and, hence, by the structural closely related NODs requires both the Adda moiety and the remaining structural units of the maternal toxins. In contrast to MCs, the molecular mode of action of the geno- and cytotoxic CYNs (i.e. inhibition of protein synthesis and generation of (a) toxic metabolite(s) by CYP450 metabolization) is not completely understood and no information has so far been provided on its effects in human primary cells. Furthermore, both toxins are known to primarily target the liver, but have been found to additionally affect other organs as well. Some members of the superfamily of organic anion transporting polypeptides (OATPs/SLCOs), that have been found to be expressed in nearly every organ/tissue, have been 125

126 Summary identified to be jointly responsible for the observed organotropism of the cell impermeant MCs (more precisely tritiated dihydro-mclr and MCLR). Besides passive diffusion, these carriers or other members of the bile acid transport system have been suggested to also be involved in mediating the uptake of CYN. Therefore, two further aims of the present study were: to assess possible differences in the toxicokinetics (i.e. the uptake-mediated toxicity) of four different MCs (LR, RR, LW and LF) and as a prerequisite to compare the toxicodynamics (i.e. inhibition of serine/threonine-specific PPs) of these congeners and to determine the cytotoxicity of CYN in primary human hepatocytes and liver OATP-expressing HEK293 cells to evaluate possible transportdependent effects. The experiments revealed that the four MC congeners investigated equipotently inhibit recombinant and cellular serine/threonine-specific PPs, but differed tremendously with regard to cytotoxicity in liver OATP-expressing HEK293 as well as primary human hepatocytes. Hence, these findings strongly suggest that MCs are taken up by OATP1B1 and OATP1B3 with varying affinities and lead to the assumption that the vast differences in the in vivo toxicity of MCs (e.g. MCLR and MCRR) can rather be attributed to congener-specific toxicokinetics than toxicodynamics. In addition, primary hepatocytes appeared to be up to two orders of magnitude more susceptible to MCs induced cytotoxicity than the corresponding HEK293 OATP transfectants which highlights the importance to additionally employ primary cells in order to obtain physiologically more relevant results. A further crucial finding was that MCLW and -LF elicited much higher toxicity in both primary human hepatocytes and OATP-expressing HEK293 cells than MCLR, the congener commonly considered as the most toxic. If this is found to be true for their in vivo toxicities as well, the current risk assessments and guideline values might have to be revised since those are based on MCLR toxicity only. In contrast, OATP-associated toxicity was found to be far less relevant for CYN than for MCs. Cytotoxicity of CYN appeared to rather depend on the metabolic condition of the exposed cell types since the immortalized, hence, metabolically 126

127 Summary less active, OATP-expressing HEK293 cells were distinctly less susceptible than primary human hepatocytes. This suggests that the second mechanism in CYN totoxicity, i.e. the generation of (a) toxic metabolite(s) by CYP450 metabolization, has a higher impact at acute concentrations than the inhibition of protein synthesis which is supported by previous findings in primary mouse hepatocytes. Moreover, the present data indicate that passive diffusion is indeed a mechanism in the uptake of CYN as the viability of control vector-transfected HEK293 cells was only marginally less reduced than that of the respective OATP transfectants. The final scope of this study was to analyze cyanobacterial dietary supplements for contaminations with MCs and the excitotoxic amino acid BMAA in order to assess the risk emanating from this potential source of intoxication. Since BMAA could neither be detected in Spirulina nor Aphanizomenon flosaquae based supplements these products unlikely constitute a threat with regard to this toxin that has been suggested to be produced by the vast majority of cyanobacteria. On the contrary, the analyses revealed MCs to be common contaminants of Aphanizomenon flos-aquae containing supplements. All of the samples were tested positively and approximately half of them exceeded the maximal acceptable concentration of 1 µg MCLR equivalents/g dw defined by the Oregon Health Division. In conclusion, this study provides crucial findings on the toxicity of MCs that might have considerable implications on present risk assessments and guideline values and gives further insight on the uptake and toxicity of CYN. Moreover, it clearly demonstrates that MCs pose a serious health risk for consumers of cyanobacterial dietary supplements based on A. flos-aquae. 127

128 Zusammenfassung 9 Zusammenfassung Als Ergebnis ihrer 3,5-3,6 Millarden Jahre langen Evolution entwickelten sich Cyanobakterien zu den vielfältigsten und weitverbreitetsten Prokaryoten und eroberten fast jeden erdenklichen Lebensraum auf der ganzen Welt. Während dieser Zeit brachten sie eine enorme Vielfalt an Sekundärmetaboliten hervor, deren physiologische Funktionen weitgehend unerforscht sind. Einige dieser chemisch und strukturell äußerst unterschiedlichen Substanzen erlangten das Interesse der internationalen Forschung, seit ihre Toxizität auf andere Organismen bekannt wurde. Unter ihnen erregten vor allem die Microcystine (MCs) und seit kürzerem auch die Cylindrospermopsine (CYNs) die größte Aufmerksamkeit aufgrund ihres häufigen und weitverbreiteten Auftretens, sowie zahlreichen, zum Teil tödlichen Vergiftungsfällen von Menschen und Tieren. MCs sind zyklische Heptapeptide, von denen mehr als 80 Kongenere bekannt sind, während CYNs zu den Alkaloiden zählen. Sie bestehen aus einer sulfatierten und methylierten trizyklischen Guanidingruppe, die mit einem Hydroxymethyluracil verbunden ist. Als primärer Wirkmechanismus der Toxizität von MCs stellte sich die starke Hemmung Serin/Threonin-spezifischer Proteinphosphatasen (PPs), insbesondere PP1 and 2A, heraus. Die einzigartige Aminosäure Adda wurde dabei als toxische Komponente innerhalb des MC-Moleküls identifiziert. Isoliertes Adda zeigte hingegen keinerlei hemmende Wirkung auf die Aktivität von PP1. Entsprechende Versuche mit PP2A wurden bisher nicht durchgeführt. An diese fehlende Information richtete sich der erste Teil der vorliegenden Studie. Wie erwartet rief Adda auch in PP2A keine inhibitorischen Effekte hervor. Dies bestätigt die Annahme, dass die Hemmung von PPs durch MCs, und folglich auch durch die strukturell eng verwandten Nodularine, sowohl Adda, als auch die verbleibenden strukturellen Einheiten des nativen Moleküls benötigt. Im Gegensatz zu MCs ist der molekulare Wirkmechanismus von CYNs (d.h. die Hemmung der Proteinsynthese und die Bildung eines oder mehrerer toxischer Metabolite durch CYP450) nicht vollständig geklärt. Darüber hinaus existieren keine Informationen über die Wirkung auf humane Primärzellen. 128

129 Zusammenfassung Bekanntermaßen gilt die Leber als primäres Zielorgan beider Toxine, wobei zusätzlich auch weitere Organe betroffen sein können. Einige Vertreter aus der Superfamilie der Organic anion transporting polypeptides (OATPs/SLCOs), die in nahezu allen Organen/Geweben nachgewiesen wurden, identifizierte man als mitverantwortlich für die Organverteilung der zellundurchlässigen MCs (genauer gesagt von tritiiertem Dihydro-MCLR und MCLR). Bestehende Versuche deuten darauf hin, dass diese Transporter oder andere Vertreter des Gallensalztransportsystems, neben passiver Diffusion, an der Aufnahme von CYN beteiligt sind. Aufgrund dessen waren zwei weitere Ziele dieser Doktorarbeit: die Ermittlung möglicher Unterschiede in der Toxikokinetik (d.h. der aufnahmevermittelten Toxizität) verschiedener MCs (LR, RR, LW and LF) und als Voraussetzung hierfür der Vergleich der Toxikodynamik (d.h. der Hemmung Serin/Threonin-spezifischer PPs) dieser Kongenere und die Bestimmung der Zytotoxizität von CYN auf primäre humane Hepatozyten und hepatische OATPs exprimierende HEK293-Zellen, um mögliche transportabhängige Effekte zu evaluieren. Die Experimente ergaben, dass die vier untersuchten MC-Kongenere rekombinante, sowie zelluläre Serin/Threonin-spezifische PPs gleichermaßen wirksam hemmen. Andererseits unterschieden sie sich enorm bezüglich ihrer Zytotoxizität auf OATP exprimierende HEK293-Zellen, als auch primäre humane Hepatozyten. Demnach weisen diese Ergebnisse stark darauf hin, dass MCs von OATP1B1 und -1B3 mit unterschiedlicher Affinität transportiert werden und führen zu der Annahme, dass die gewaltigen Unterschiede in der in vivo Toxizität von MCs (z.b. MCLR und -RR) eher kongenerspezifischer Toxikokinetik, als Toxikodynamik zugeschrieben werden können. Darüber hinaus stellten sich primäre humane Hepatozyten im Vergleich zu den entsprechenden mit OATPs transfizierten HEK293-Zellen als um bis zu zwei Größenordnungen anfälliger gegenüber MCs heraus. Diese Feststellung hebt die Bedeutung des Einsatzes von Primärzellen hervor, um physiologisch relevantere Ergebnisse zu erzielen. Ein weiteres äußerst wichtiges Resultat ist, dass MCLW und -LF sowohl in OATP exprimierenden HEK293-Zellen, als auch in primären humanen 129

130 Zusammenfassung Hepatozyten weit höhere Toxizität auslösten als MCLR, das gemeinhin als das toxischste Kongener gilt. Vorausgesetzt dies trifft auch auf die in vivo Toxizität zu, könnte es sein, dass die bestehenden Risikoabschätzungen und Richtwerte überarbeitet werden müssen, da diese ausschließlich auf der Toxizität von MCLR beruhen. Im Gegensatz dazu stellte sich heraus, dass die OATP assoziierte Toxizität für CYN weit weniger relevant ist, als für MCs. Die Zytotoxizität von CYN schien eher vom Stoffwechselzustand der exponierten Zelltypen abzuhängen, da die immortalisierten und folglich metabolisch weniger aktiven HEK293-Zellen deutlich weniger empfindlich waren, als die primären humanen Hepatozyten. Dies deutet darauf hin, dass der zweite toxische Wirkmechanismus von CYN, also die Bildung eines oder mehrerer toxischer Metabolite durch CYP450, einen größeren Einfluss bei akuten Konzentrationen hat, als die Hemmung der Proteinsynthese. Bekräftigt wird diese Annahme durch frühere Ergebnisse mit primären murinen Hepatozyten. Des Weiteren weisen die vorliegenden Daten darauf hin, dass passive Diffusion in der Tat ein Aufnahmeweg von CYN ist, da die Lebensfähigkeit kontrollvektortransfizierter HEK293-Zellen, verglichen mit den jeweiligen OATP- Transfektanden, nur geringfügig verringert war. Das letzte Ziel dieser Doktorarbeit war die Analyse cyanobakterieller Nahrungsergänzungsmittel auf Kontaminationen mit MCs und der exzitotoxischen Aminosäure BMAA, um das Risiko abzuschätzen, das von dieser potentiellen Vergiftungsquelle ausgeht. Von BMAA wird angenommen, dass es von der überwiegenden Mehrzahl der Cyanobakterien gebildet werden kann. Es konnte jedoch weder in Nahrungsergänzungsmitteln nachgewiesen werden, die auf Spirulina, noch auf Aphanizomenon flos-aquae basieren. Aufgrund dessen geht von diesen Produkten wahrscheinlich keine Gefahr bezüglich BMAA aus. Hingegen ergaben die Analysen, dass Produkte, die A. flos-aquae enthalten, gemeinhin mit MCs kontaminiert sind. Alle Proben wurden positiv getestet, wobei etwa die Hälfte die maximal akzeptierbare Konzentration von 1 µg MCLR-Äquvalente/g Trockengewicht, festgelegt von der Oregon Health Division, überschritt. 130

131 Zusammenfassung Schlussfolgernd liefert die vorliegende Doktorarbeit äußerst wichtige Ergebnisse über die Toxizität von MCs, die beachtliche Auswirkungen auf bestehende Risikoabschätzungen und Richtwerte haben könnten, und verbessert das Verständnis über die Aufnahme und Toxizität von CYN. Darüber hinaus belegt sie eindeutig, dass MCs ein ernstes Gesundheitsrisiko für Konsumenten von Nahrungsergänzungsmitteln, die auf A. flos-aquae basieren, darstellen. 131

132 Abbreviations 10 Abbreviations ABCB ABCC Adda ANOVA ATP BGAS BMAA BSA bw cppia CYN Da DAD D-MeAsp DNA dw EC 50 ELISA FBS FD GST GTX HPLC IC IC 50 i.p. i.v. LC-MS LC-MS/MS ATP-binding cassette, sub-family B ATP-binding cassette, sub-family C 3-amino-9-methoxy-2,6,8,-trimethyl-10-phenyl-4,6,- decadienoic acid analysis of variance adenosine triphosphate blue-green algae supplement Beta-N-methylamino-L-alanine bovine serum albumin body weight colorimetric protein phosphatase inhibition assay cylindrospermopsin Dalton diode-array detection D-erythro-β-methylaspartic acid deoxyribonucleic acid dry weight half maximal effective concentration enzyme linked immunosorbent assay fetal bovine serum fluorescence detection glutathione S-transferase gonyautoxin high performance liquid chromatography inhibitory concentration half maximal inhibitory concentration intraperitoneal intravenous liquid chromatography mass spectrometry liquid chromatography-tandem mass spectrometry 132

133 Abbreviations LD LD 50 LOD LPS MAC MAPK MC MCLA MCLF MCLR MCLW MCYR MCRR Mdha Mdhb MDR MeOH MRP MTT MW n.d. nda NOD OA OATP o.c. PBS PCC PLC pnpp PP PSP lethal dose median lethal dose limit of detection lipopolysaccharide maximum acceptable concentration mitogen-activated protein kinase microcystin microcystin-la (leucine and alanine) microcystin-lf (leucine and phenylalanine) microcystin-lr (leucine and arginine) microcystin-lw (leucine and tryptophan) microcystin-yr (tyrosine and arginine) microcystin-lr (arginine and arginine) N-methyldehydroalanine 2-(methylamino)-2-dehydrobutyric acid multidrug resistance protein methanol multidrug resistance-associated protein 3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide molecular weight not determinable; none of the standard MCs detectable no data available nodularin okadaic acid organic anion transporting polypeptide opus citatum phosphate buffered saline Pasteur culture collection primary liver cancer p(ara)-nitrophenylphosphate protein phosphatase paralytic shellfish poisoning 133

134 Abbreviations R 2 ROS SDS-PAGE SLCO SPE MRM STX TDI UPLC UV WHO coefficient of determination reactive oxygen species sodium dodecyl sulfate polyacrylamide gel electrophoresis solute carrier organic anion transporter family solid phase extraction selected reaction monitoring mode saxitoxin tolerable daily intake Ultra High Performance Liquid Chromatography ultraviolet World Health Organization 134

135 References 11 References Abdulqader, G., Barsanti, L., and Tredici, M. R. (2000). Harvest of Arthrospira platensis from Lake Kossorom (Chad) and its household usage among the Kanembu. Journal of Applied Phycology 12, Anderson, D. M. (1994). Red tides. Sci Am 271, Aráoz, R., Molgo, J., and Tandeau de Marsac, N. (2009). Neurotoxic cyanobacterial toxins. Toxicon. Aronstam, R. S., and Witkop, B. (1981). Anatoxin-a interactions with cholinergic synaptic molecules. Proc Natl Acad Sci U S A 78, Azevedo, S. M., Carmichael, W. W., Jochimsen, E. M., Rinehart, K. L., Lau, S., Shaw, G. R., and Eaglesham, G. K. (2002). Human intoxication by microcystins during renal dialysis treatment in Caruaru-Brazil. Toxicology , Azevedo, S. M. F. O., Evans, W. R., Carmichael, W. W., and Namikoshi, M. (1994). First report of microcystins from a Brazilian isolate of the cyanobacterium Microcystis aeruginosa. Journal of Applied Phycology 6, Bagu, J. R., Sykes, B. D., Craig, M. M., and Holmes, C. (1997). A molecular basis for different interactions of marine toxins with protein phosphatase-1 - Molecular models for bound motuporin, microcystins, okadaic acid, and calyculin A. Journal of Biological Chemistry 272, Ballot, A., Krienitz, L., Kotut, K., Wiegand, C., Metcalf, J. S., Codd, G. A., and Pflugmacher, S. (2004). Cyanobacteria and cyanobacterial toxins in three alkaline rift valley lakes of Kenya - Lakes Bogoria, Nakuru and Elmenteita. Journal of Plankton Research 26, Ballot, A., Krienitz, L., Kotut, K., Wiegand, C., and Pflugmacher, S. (2005). Cyanobacteria and cyanobacterial toxins in the alkaline crater lakes Sonachi and Simbi, Kenya. Harmful Algae 4, Banack, S. A., and Cox, P. A. (2003). Biomagnification of cycad neurotoxins in flying foxes: implications for ALS-PDC in Guam. Neurology 61, Banack, S. A., Johnson, H. E., Cheng, R., and Cox, P. A. (2007). Production of the Neurotoxin BMAA by a Marine Cyanobacterium. Mar Drugs 5, Banack, S. A., Murch, S. J., and Cox, P. A. (2006). Neurotoxic flying foxes as dietary items for the Chamorro people, Marianas Islands. J Ethnopharmacol 106, Banker, R., Carmeli, S., Hadas, O., Teltsch, B., Porat, R., and Sukenik, A. (1997). Identification of cylindrospermopsin in Aphanizomenon ovalisporum (Cyanophyceae) isolated from Lake Kinneret, Israel. Journal of Phycology 33, Banker, R., Teltsch, B., Sukenik, A., and Carmeli, S. (2000). 7- Epicylindrospermopsin, a toxic minor metabolite of the cyanobacterium Aphanizomenon ovalisporum from lake Kinneret, Israel. J Nat Prod 63, Bannack, S. A., and Cox, P. A. (2003). Biomagnification of cycad neurotoxins in flying foxes: implications for ALS-PDC in Guam. Neurology 61,

136 References Bartram, J., Carmichael, W. W., Chorus, I., Jones, G., and Skulberg, O. M. (1999). Introduction. In Toxic Cyanobacteria in Water: A Guide to their Public Health Consequences, Monitoring and Management (I. Chorus, and J. Bartram, Eds.), pp E & FN Spon, London. Batista, T., de Sousa, G., Suput, J. S., Rahmani, R., and Suput, D. (2003). Microcystin-LR causes the collapse of actin filaments in primary human hepatocytes. Aquatic Toxicology 65, Batoréu, M. C. C., Dias, E., Pereira, P., and Franca, S. (2005). Risk of human exposure to paralytic toxins of algal origin. Environ Toxicol Pharmacol 19, Beasley, V. R., Lovell, R. A., Holmes, K. R., Walcott, H. E., Schaeffer, D. J., Hoffmann, W. E., and Carmichael, W. W. (2000). Microcystin-LR decreases hepatic and renal perfusion, and causes circulatory shock, severe hypoglycemia, and terminal hyperkalemia in intravascularly dosed swine. J Toxicol Environ Health A 61, Bell, S., and Codd, G. (1994). Cyanobacterial toxins and human health. Reviews in Medical Microbiology 5, Bishop, C. T., Anet, E. F., and Gorham, P. R. (1959). Isolation and identification of the fast-death factor in Microcystis aeruginosa NRC-1. Can J Biochem Physiol 37, Bissell, D. M., Levine, G. A., and Bissell, M. J. (1978). Glucose metabolism by adult hepatocytes in primary culture and by cell lines from rat liver. Am J Physiol 234, C Boaru, D. A., Dragos, N., and Schirmer, K. (2006). Microcystin-LR induced cellular effects in mammalian and fish primary hepatocyte cultures and cell lines: a comparative study. Toxicology 218, Bollen, M., and Stalmans, W. (1992). The structure, role, and regulation of type 1 protein phosphatases. Critical Review of Biochemistry and Molecular Biology 27, Borst, P., and Elferink, R. O. (2002). Mammalian ABC transporters in health and disease. Annual Reviews of Biochemistry 71, Borst, P., Evers, R., Kool, M., and Wijnholds, J. (2000). A family of drug transporters: the multidrug resistance-associated proteins. J Natl Cancer Inst 92, Botes, D. P., Kruger, H., and Viljoen, C. C. (1982). Isolation and characterization of four toxins from the blue-green alga, Microcystis aeruginosa. Toxicon 20, Botes, D. P., Tuinman, A. A., Wessels, P. L., Viljoen, C. C., Kruger, H., Williams, D. H., Santikarn, S., Smith, R. J., and Hammond, S. J. (1984). The structure of cyanoginosin-la, a cyclic heptapeptide toxin from the cyanobacterium Microcystis aeruginosa. Journal of Chemical Society, Perkin Transactions 1, Botes, D. P., Wessels, P. L., Kruger, H., Runnegar, M. T. C., Santikarn, S., Smith, R. J., Barna, J. C. J., and Williams, D. H. (1985). Structural studies on cyanoginosins-lr, -YR, -YA, and -YM, peptide toxins from Microcystis aeruginosa. Journal of the Chemical Society, Perkin Transactions 1, Bourke, A. T. C., Hawes, R. B., Neilson, A., and Stallman, N. D. (1983). An outbreak of hepato-enteritis (the Palm Island Mystery Disease) possibly caused by algal intoxication. Toxicon 3,

137 References Bourne, D. G., Jones, G. J., Blakeley, R. L., Jones, A., Negri, A. P., and Riddles, P. (1996). Enzymatic pathway for the bacterial degradation of the cyanobacterial cyclic peptide toxin microcystin LR. Applied and Environmental Microbiology 62, Bradford, M. M. (1976). A rapid and sensitive method for the quantification of mirogram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, Briand, J. F., Jacquet, S., Bernard, C., and Humbert, J. F. (2003). Health hazards for terrestrial vertebrates from toxic cyanobacteria in surface water ecosystems. Vet Res 34, Bronger, H., Konig, J., Kopplow, K., Steiner, H. H., Ahmadi, R., Herold-Mende, C., Keppler, D., and Nies, A. T. (2005). ABCC drug efflux pumps and organic anion uptake transporters in human gliomas and the blood-tumor barrier. Cancer Res 65, Brooks, W. P., and Codd, G. A. (1987). Distribution of Microcystis aeruginosa peptide toxin and interactions with hepatic microsomes in mice. Pharmacol Toxicol 60, Brownson, D. M., Mabry, T. J., and Leslie, S. W. (2002). The cycad neurotoxic amino acid, beta-n-methylamino-l-alanine (BMAA), elevates intracellular calcium levels in dissociated rat brain cells. J Ethnopharmacol 82, Bruice, P. Y. (2007). Organische Chemie. Pearson Studium, München. Byth, S. (1980). Palm Island mystery disease. Med J Aust 2, 40, 42. Carbis, C. R., Waldron, D. L., Mitchell, G. F., Anderson, J. W., and McCauley, I. (1995). Recovery of hepatic function and latent mortalities in sheep exposed to the blue-green alga Microcystis aeruginosa. Vet. Record 137, Carmichael, W., and Falconer, I. (1993). Diseases related to freshwater bluegreen algal toxins, and control measures. In Algal toxins in seafood and drinking water (I. Falconer, Ed.), pp Academic Press, London. Carmichael, W. W. (1997). The Cyanotoxins. Advances in Botanical Research 27, Carmichael, W. W., Azevedo, S. M., An, J. S., Molica, R. J., Jochimsen, E. M., Lau, S., Rinehart, K. L., Shaw, G. R., and Eaglesham, G. K. (2001). Human fatalities from cyanobacteria: chemical and biological evidence for cyanotoxins. Environ Health Perspect 109, Carmichael, W. W., Beasley, V., Bunner, D. L., Eloff, J. N., Falconer, I., Gorham, P., Harada, K., Krishnamurthy, T., Yu, M. J., Moore, R. E., Rinehart, K. L., Runnegar, M., Skulberg, O. M., and Watanabe, M. (1988a). Naming of cyclic heptapeptide toxins of cyanobacteria (bluegreen algae). Toxicon 26, Carmichael, W. W., Biggs, D. F., and Gorham, P. R. (1975). Toxicology and pharmacological action of anabaena flos-aquae toxin. Science 187, Carmichael, W. W., Drapeau, C., and Anderson, D. M. (2000). Harvesting of Aphanizomenon flos-aquae Ralfs ex Born. & Flah. var. flos-aquae (Cyanobacteria) from Klamath Lake for human dietary use. Journal of Applied Phycology 12, Carmichael, W. W., Eschedor, J. T., Patterson, G. M., and Moore, R. E. (1988b). Toxicity and partial structure of a hepatotoxic peptide produced 137

138 References by the cyanobacterium Nodularia spumigena Mertens emend. L575 from New Zealand. Appl Environ Microbiol 54, Carmichael, W. W., Evans, W. R., Yin, Q. Q., Bell, P., and Moczydlowski, E. (1997). Evidence for paralytic shellfish poisons in the freshwater cyanobacterium Lyngbya wollei (Farlow ex Gomont) comb. nov. Appl Environ Microbiol 63, Carmichael, W. W., Mahmood, N. A., and Hyde, E. G. (1990). Natural toxins from cyanobacteria (blue-green algae). In Marine Toxins, Origin, Structure and Molecular Pharmacology (S. Hall, and G. Strichartz, Eds.), pp American Chemical Society - Symposium Series, Washington. Castell, J. V., and Gomez-Lechon, M. J. (2009). Liver cell culture techniques. Methods Mol Biol 481, Castenholz, R. W., and Waterbury, J. B. (1989). Group 1. Cyanobacteria. Preface. In Bergey's Manual of Systematic Bacteriology (J. T. Stanley, M. P. Bryant, N. Pfennig, and J. G. Holt, Eds.), pp Lippincott Williams & Wilkins, Baltimore. Catterall, W. A. (1980). Neurotoxins that act on voltage-sensitive sodium channels in excitable membranes. Annu Rev Pharmacol Toxicol 20, Cheng, X., Maher, J., Chen, C., and Klaassen, C. D. (2005). Tissue distribution and ontogeny of mouse organic anion transporting polypeptides (oatps). Drug Metab Dispos 33, Chiswell, R. K., Shaw, G. R., Eaglesham, G., Smith, M. J., Norris, R. L., Seawright, A. A., and Moore, M. R. (1999). Stability of cylindrospermopsin, the toxin from the cyanobacterium, Cylindrospermopsis raciborskii: Effect of ph, temperature, and sunlight on decomposition. Environmental Toxicology 14, Chong, M. W., Wong, B. S., Lam, P. K., Shaw, G. R., and Seawright, A. A. (2002). Toxicity and uptake mechanism of cylindrospermopsin and lophyrotomin in primary rat hepatocytes. Toxicon 40, Chorus, I. (2001). Cyanotoxin occurrence in freshwaters - A summary of survey results from different countries. In Cyanotoxins - Occurrence, causes, consequences (I. Chorus, Ed.), pp Springer Verlag, Berlin. Chorus, I. (2001). Cyanotoxins: occurence, causes, consequences. Springer. Christoffersen, K., Lyck, S., and Winding, A. (2002). Microbial activity and bacterial community structure during degradation of microcystins. Aquatic Microbial Ecology 27, Codd, G. (1995). Cyanobacterial toxins: occurrence, properties and biological significance. Water Science & Technology 32, Codd, G. A., Morrison, L. F., and Metcalf, J. S. (2005). Cyanobacterial toxins: risk management for health protection. Toxicol Appl Pharmacol 203, Cohen, P. (1989b). The structure and regulation of protein phosphatases. Annual Review of Biochemistry 58, Cohen, P., and Cohen, P. T. W. (1989a). Protein phosphatases come of age. Journal of Biological Chemistry 264, Cohen, P., Holmes, C. F., and Tsukitani, Y. (1990). Okadaic acid: a new probe for the study of cellular regulation. Trends Biochem Sci 15, Cohen, P. T. (2002). Protein phosphatase 1--targeted in many directions. J Cell Sci 115, Cook, W. O., Beasley, V. R., Lovell, R. A., Dahlem, A. M., Hooser, S. B., Mahmood, N. A., and Carmichael, W. W. (1989). Consistent inhibition of 138

139 References peripheral cholinesterases by neurotoxins from the freshwater cyanobacterium Anabaena flos-aquae: Studies of ducks, swine, mice and a steer. Environmental Toxicology and Chemistry 8, Cordon-Cardo, C., O'Brien, J. P., Boccia, J., Casals, D., Bertino, J. R., and Melamed, M. R. (1990). Expression of the multidrug resistance gene product (P-glycoprotein) in human normal and tumor tissues. J Histochem Cytochem 38, Cordon-Cardo, C., O'Brien, J. P., Casals, D., Rittman-Grauer, L., Biedler, J. L., Melamed, M. R., and Bertino, J. R. (1989). Multidrug-resistance gene (Pglycoprotein) is expressed by endothelial cells at blood-brain barrier sites. Proc Natl Acad Sci U S A 86, Cox, P. A., Banack, S. A., and Murch, S. J. (2003). Biomagnification of cyanobacterial neurotoxins and neurodegenerative disease among the Chamorro people of Guam. Proc Natl Acad Sci U S A 100, Cox, P. A., Banack, S. A., Murch, S. J., Rasmussen, U., Tien, G., Bidigare, R. R., Metcalf, J. S., Morrison, L. F., Codd, G. A., and Bergman, B. (2005). Diverse taxa of cyanobacteria produce beta-n-methylamino-l-alanine, a neurotoxic amino acid. Proc Natl Acad Sci U S A 102, Cox, P. A., and Sacks, O. W. (2002). Cycad neurotoxins, consumption of flying foxes, and ALS-PDC disease in Guam. Neurology 58, Craig, M., Luu, H. A., McCready, T. L., Williams, D., Andersen, R. J., and Holmes, C. F. (1996). Molecular mechanisms underlying the interaction of motuporin and microcystins with type-1 and type-2a protein phosphatases. Biochem Cell Biol 74, Creasia, D. (1990). Acute inhalation toxicity of microcystin-lr. Toxicon 28, 605. Dahlem, A. M., Hassan, A. S., Swanson, S. P., Carmichael, W. W., and Beasley, V. R. (1989). A model system for studying the bioavailability of intestinally administered microcystin-lr, a hepatotoxic peptide from the cyanobacterium Microcystis aeruginosa. Pharmacol Toxicol 64, de Figueiredo, D. R., Azeiteiro, U. M., Esteves, S. M., Goncalves, F. J., and Pereira, M. J. (2004). Microcystin-producing blooms--a serious global public health issue. Ecotoxicol Environ Saf 59, de Silva, E. D., Williams, D. E., and Andersen, R. J. (1992). Motuporin, a potent protein phosphatase inhibitor isolated from the Papua New Guinea sponge Theonella swinhoei Gray. Tetrahedron Letters 33, Devlin, J. P., Edwards, O. E., Gorham, P. R., Hunter, N. R., Pike, R. K., and Stavric, B. (1977). Anatoxin-a, a toxic alkaloid from Anabaena flos-aquae NRC-44h. CAN. J. CHEM. 55, Dietrich, D., and Hoeger, S. (2005). Guidance values for microcystins in water and cyanobacterial supplement products (blue-green algal supplements): a reasonable or misguided approach? Toxicol Appl Pharmacol 203, Dietrich, D. R., Fischer, A., Michel, C., and Hoeger, S. J. (2008). Toxin mixture in cyanobacterial blooms--a critical comparison of reality with current procedures employed in human health risk assessment. Adv Exp Med Biol 619, Ding, W. X., and Nam Ong, C. (2003). Role of oxidative stress and mitochondrial changes in cyanobacteria-induced apoptosis and hepatotoxicity. FEMS Microbiol Lett 220,

140 References Ding, W. X., Shen, H. M., and Ong, C. N. (2000). Critical role of reactive oxygen species and mitochondrial permeability transition in microcystin-induced rapid apoptosis in rat hepatocytes. Hepatology 32, Donato, M. T., Lahoz, A., Castell, J. V., and Gomez-Lechon, M. J. (2008). Cell lines: a tool for in vitro drug metabolism studies. Curr Drug Metab 9, Dow, C. S., and Swoboda, U. K. (2000). Cyanotoxins. In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. Draisci, R., Ferretti, E., Palleschi, L., and Marchiafava, C. (2001). Identification of anatoxins in blue-green algae food supplements using liquid chromatography-tandem mass spectrometry. Food Additives and Contaminants 18, Duncan, M. W., and Marini, A. M. (2006). Debating the cause of a neurological disorder. Science 313, Duncan, M. W., Markey, S. P., Weick, B. G., Pearson, P. G., Ziffer, H., Hu, Y., and Kopin, I. J. (1992). 2-Amino-3-(methylamino)propanoic acid (BMAA) bioavailability in the primate. Neurobiol Aging 13, Eckhardt, U., Schroeder, A., Stieger, B., Hochli, M., Landmann, L., Tynes, R., Meier, P. J., and Hagenbuch, B. (1999). Polyspecific substrate uptake by the hepatic organic anion transporter Oatp1 in stably transfected CHO cells. American Journal of Physiology 276, G Edwards, C., Lawton, L. A., Coyle, S. M., and Ross, P. (1996). Laboratory-scale purification of microcystins using flash chromatography and reversedphase high-performance liquid chromatography. J Chromatogr A 734, Eriksson, J. E., Gronberg, L., Nygard, S., Slotte, J. P., and Meriluoto, J. A. (1990b). Hepatocellular uptake of 3H-dihydromicrocystin-LR, a cyclic peptide toxin. Biochim Biophys Acta 1025, Eriksson, J. E., Meriluoto, J. A., Kujari, H. P., Osterlund, K., Fagerlund, K., and Hallbom, L. (1988). Preliminary characterization of a toxin isolated from the cyanobacterium Nodularia spumigena. Toxicon 26, Eriksson, J. E., Toivola, D., Meriluoto, J. A., Karaki, H., Han, Y. G., and Hartshorne, D. (1990a). Hepatocyte deformation induced by cyanobacterial toxins reflects inhibition of protein phosphatases. Biochem Biophys Res Commun 173, Ernst, B., Hitzfeld, B., and Dietrich, D. (2001). Presence of Planktothrix sp. and cyanobacterial toxins in Lake Ammersee, Germany and their impact on whitefish (Coregonus lavaretus L.). Environmental Toxicology 16, Ersmark, K., Del Valle, J. R., and Hanessian, S. (2008). Chemistry and biology of the aeruginosin family of serine protease inhibitors. Angew Chem Int Ed Engl 47, Faber, K. N., Muller, M., and Jansen, P. L. (2003). Drug transport proteins in the liver. Adv Drug Deliv Rev 55, Falconer, I., Bartram, J., Chorus, I., Kuiper-Goodman, T., Utkilen, H., Burch, M., and Codd, G. A. (1999). Safe levels and safe practices. In Toxic cyanobacteria in water. A guide to their public health consequences, monitoring and management (I. Chorus, and J. Bartram, Eds.), pp E & FN SPON, London and New York. 140

141 References Falconer, I. R. (1991). Tumor promotion and liver injury caused by oral consumption of cyanobacteria. Environmental Toxicology & Water Quality 6, Falconer, I. R. (1993). Algal Toxins in Seafood and Drinking Water. Academic Press, London. Falconer, I. R. (2005a). Cyanobacterial Toxins of Drinking Water Supplies: Cylindrospermopsins and Microcystins. CRC Press, Boca Raton, Florida. Falconer, I. R., Beresford, A. M., and Runnegar, M. T. (1983). Evidence of liver damage by toxin from a bloom of the blue-green alga, Microcystis aeruginosa. Med J Aust 1, Falconer, I. R., Buckley, T., and Runnegar, M. T. (1986). Biological half-life, organ distribution and excretion of 125-I-labelled toxic peptide from the blue-green alga Microcystis aeruginosa. Aust J Biol Sci 39, Falconer, I. R., Burch, M. D., Steffensen, D. A., Choice, M., and Coverdale, O. R. (1994). Toxicity of the blue-green alga (cyanobacterium) Microcystis aeruginosa in drinking water to growing pigs, as an animal model for human injury and risk assessment. Environmental Toxicology & Water Quality 9, Falconer, I. R., Dornbusch, M., Moran, G., and Yeung, S. K. (1992b). Effect of the cyanobacterial (blue-green algal) toxins from Microcystis aeruginosa on isolated enterocytes from the chicken small intestine. Toxicon 30, Falconer, I. R., Hardy, S. J., Humpage, A. R., Froscio, S. M., Tozer, G. J., and Hawkins, P. R. (1999). Hepatic and renal toxicity of the blue-green alga (cyanobacterium) Cylindrospermopsis raciborskii in male Swiss albino mice. Environmental Toxicology 14, Falconer, I. R., and Humpage, A., R. (1996). Tumour promotion by cyanobacterial toxins. Phycologia 35, Falconer, I. R., and Humpage, A. R. (2005b). Health risk assessment of cyanobacterial (blue-green algal) toxins in drinking water. Int J Environ Res Public Health 2, Falconer, I. R., and Humpage, A. R. (2006). Cyanobacterial (blue-green algal) toxins in water supplies: Cylindrospermopsins. Environ Toxicol 21, Falconer, I. R., Jackson, A. R. B., Langley, J., and Runnegar, M. T. C. (1981). Liver pathology in mice in poisoning by the blue-green alga Microcystis aeruginosa. Australian Journal of Biological Sciences 34, Falconer, I. R., Smith, J. V., Jackson, A. R., Jones, A., and Runnegar, M. T. (1988). Oral toxicity of a bloom of the Cyanobacterium Microcystis aeruginosa administered to mice over periods up to 1 year. Journal of Toxicology and Environmental Health 24, Falconer, I. R., and Yeung, D. S. (1992a). Cytoskeletal changes in hepatocytes induced by Microcystis toxins and their relation to hyperphosphorylation of cell proteins. Chem Biol Interact 81, Fastner, J., Codd, G. A., Metcalf, J. S., Woitke, P., Wiedner, C., and Utkilen, H. (2002). An international intercomparison exercise for the determination of purified microcystin-lr and microcystins in cyanobacterial field material. Analytical and Bioanalytical Chemistry 374, Fastner, J., Heinze, R., and Chorus, I. (1995). Microcystin-content, hepatotoxicity and cytotoxicity of cyanobacteria in some German water bodies. Water Science and Technology 32,

142 References Fastner, J., Heinze, R., Humpage, A. R., Mischke, U., Eaglesham, G. K., and Chorus, I. (2003). Cylindrospermopsin occurrence in two German lakes and preliminary assessment of toxicity and toxin production of Cylindrospermopsis raciborskii (Cyanobacteria) isolates. Toxicon 42, Fastner, J., Rucker, J., Stuken, A., Preussel, K., Nixdorf, B., Chorus, I., Kohler, A., and Wiedner, C. (2007). Occurrence of the cyanobacterial toxin cylindrospermopsin in northeast Germany. Environ Toxicol 22, Fawell, J., James, C., and James, H. (1994). Toxins from blue-green algae: Toxicological assessment of Microcystin-LR and a method for its determination in water. Foundation for Water Research, Marlow. Fawell, J. K., Mitchell, R. E., Everett, D. J., and Hill, R. E. (1999). The toxicity of cyanobacterial toxins in the mouse: I microcystin-lr. Hum Exp Toxicol 18, Ferreira, F. M. B., Soler, J. M. F., Fidalgo, M. L., and Fernandez-Vila, P. (2001). PSP toxins from Aphanizomenon flos-aquae (cyanobacteria) collected in the Crestuma-Lever reservoir (Douro river, northern Portugal). Toxicon 39, Fessard, V., and Bernard, C. (2003). Cell alterations but no DNA strand breaks induced in vitro by cylindrospermopsin in CHO K1 cells. Environ Toxicol 18, Fessard, V., Hégarat, L., and Mourot, A. (2004). Comparison of the genotoxic results obtained from the in vitro cytokinesis-block micronucleus assay with various toxins inhibitors of protein phosphatases: okadaic acid, nodularin and microcystin-lr. 6th International Conference on Toxic Cyanobacteria. Feurstein, D., Holst, K., Fischer, A., and Dietrich, D. R. (2009). Oatp-associated uptake and toxicity of microcystins in primary murine whole brain cells. Toxicol Appl Pharmacol 234, Fischer, A., Hoeger, S. J., Feurstein, D., Stemmer, K., Knobeloch, D., Nussler, A., and Dietrich, D. (2010). The role of organic anion transporting polypeptides (OATPs/SLCOs) in the toxicity of different microcystin congeners in vitro: A comparison of primary human hepatocytes and OATP-transfected HEK293 cells. Toxicol Appl Pharmacol, in press. Fischer, W. J., Altheimer, S., Cattori, V., Meier, P. J., Dietrich, D. R., and Hagenbuch, B. (2005). Organic anion transporting polypeptides expressed in liver and brain mediate uptake of microcystin. Toxicol Appl Pharmacol 203, Fischer, W. J., Garthwaite, I., Miles, C. O., Ross, K. M., Aggen, J. B., Chamberlin, A. R., Towers, N. R., and Dietrich, D. R. (2001). Congenerindependent immunoassay for microcystins and nodularins. Environmental Science & Technology 35, Fitzgeorge, R. B., Clark, S. A., and Keevil, C. W. (1994). Routes of intoxication. In Detection methods for cyanobacterial toxins (G. A. Codd, T. M. Jefferies, C. W. Keevil, and E. Potter, Eds.). Royal Society of Chemistry, Cambridge, UK. Fladmark, K. E., Brustugun, O. T., Mellgren, G., Krakstad, C., Boe, R., Vintermyr, O. K., Schulman, H., and Doskeland, S. O. (2002). Ca2+/calmodulin-dependent protein kinase II is required for microcystininduced apoptosis. J Biol Chem 277,

143 References Fournie, J. W., and Hilborn, E. D. (2008). Ecosystem effects workgroup report. Adv Exp Med Biol 619, Francis, G. (1878). Poisonous Australian lake. Nature 18, Fromm, M. F. (2004). Importance of P-glycoprotein at blood-tissue barriers. Trends Pharmacol Sci 25, Froscio, S. M., Humpage, A. R., Burcham, P. C., and Falconer, I. R. (2001). Cell-free protein synthesis inhibition assay for the cyanobacterial toxin cylindrospermopsin. Environ Toxicol 16, Froscio, S. M., Humpage, A. R., Burcham, P. C., and Falconer, I. R. (2003). Cylindrospermopsin-induced protein synthesis inhibition and its dissociation from acute toxicity in mouse hepatocytes. Environ Toxicol 18, Fujiki, H., Sueoka, E., and Suganuma, M. (1996). Carcinogenesis of Microcystins. In Toxic Microcystis (M. F. Watanabe, K. Harada, W. W. Carmichael, and H. Fujiki, Eds.), pp CRC Press Inc., Boca Raton, New York, London, Tokyo. Gao, B., Hagenbuch, B., Kullak-Ublick, G. A., Benke, D., Aguzzi, A., and Meier, P. J. (2000). Organic anion-transporting polypeptides mediate transport of opioid peptides across blood-brain barrier. J Pharmacol Exp Ther 294, Gao, B., Huber, R. D., Wenzel, A., Vavricka, S. R., Ismair, M. G., Reme, C., and Meier, P. J. (2005). Localization of organic anion transporting polypeptides in the rat and human ciliary body epithelium. Exp Eye Res 80, Gao, K. (1998). Chinese studies on the edible blue-green alga, Nostoc flagelliforme: a review. Journal of Applied Phycology 10, Garcia, C., del Carmen Bravo, M., Lagos, M., and Lagos, N. (2004). Paralytic shellfish poisoning: post-mortem analysis of tissue and body fluid samples from human victims in the Patagonia fjords. Toxicon 43, Gehringer, M. M. (2004a). Microcystin-LR and okadaic acid-induced cellular effects: a dualistic response. FEBS Letters 557, 1-8. Gehringer, M. M., Shephard, E. G., Downing, T. G., Wiegand, C., and Neilan, B. A. (2004b). An Investigation into the detoxification of microcystin-lr by the glutathion pathway in Balb/c mice. The International Journal of Biochemistry & Cell Biology 36, Gilroy, D. J., Kauffman, K. W., Hall, R. A., Huang, X., and Chu, F. S. (2000). Assessing potential health risks from microcystin toxins in blue-green algae dietary supplements. Environmental Health Perspectives 108, Goldberg, J., Huang, H. B., Kwon, Y. G., Greengard, P., Nairn, A. C., and Kuriyan, J. (1995). Three-dimensional structure of the catalytic subunit of protein serine/threonine phosphatase-1. Nature 376, Gomez-Lechon, M. J., Donato, M. T., Castell, J. V., and Jover, R. (2003). Human hepatocytes as a tool for studying toxicity and drug metabolism. Curr Drug Metab 4, Graham, L. E., Graham, J. M., and Wilcox, L. W. (2009). Algae. Benjamin Cummings, San Francisco. Grisham, J. W., Charlton, R. K., and Kaufman, D. G. (1978). In vitro assay of cytotoxicity with cultured liver: accomplishments and possibilities. Environ Health Perspect 25,

144 References Grosse, Y., Baan, R., Straif, K., Secretan, B., El Ghissassi, F., and Cogliano, V. (2006). Carcinogenicity of nitrate, nitrite, and cyanobacterial peptide toxins. Lancet Oncol 7, Hagenbuch, B., and Meier, P. J. (2003). The superfamily of organic anion transporting polypeptides. Biochim Biophys Acta 1609, Hagenbuch, B., and Meier, P. J. (2004). Organic anion transporting polypeptides of the OATP/ SLC21 family: phylogenetic classification as OATP/ SLCO superfamily, new nomenclature and molecular/functional properties. Pflugers Arch 447, Harada, K.-I., and Tsuji, K. (1998). Persistence and decomposition of hepatotoxic microcystins produced by cyanobacteria in natural environment. Journal of Toxicology-Toxin Reviews 17, Harada, K.-I., Tsuji, K., Watanabe, M. F., and Kondo, F. (1996a). Stability of microcystins from cyanobacteria: III. Effect of ph and temperature. Phycologia 35, Harada, K. (1996b). Chemistry and detection of microcystins. In Toxic Microcystis (M. Watanabe, K. Harada, W. Carmichael, and H. Fujiki, Eds.), pp CRC Press Inc., Boca Raton. Harada, K., Imanishi, S., Kato, H., Mizuno, M., Ito, E., and Tsuji, K. (2004). Isolation of Adda from microcystin-lr by microbial degradation. Toxicon 44, Harada, K., Matsuura, K., Suzuki, M., Watanabe, M. F., Oishi, S., Dahlem, A. M., Beasley, V. R., and Carmichael, W. W. (1990a). Isolation and characterization of the minor components associated with microcystins LR and RR in the cyanobacterium (blue-green algae). Toxicon 28, Harada, K., Mayumi, T., Shimada, T., Fujii, K., Kondo, F., Park, H. D., and Watanabe, M. F. (2001). Co-production of microcystins and aeruginopeptins by natural cyanobacterial bloom. Environmental Toxicology 16, Harada, K., Ogawa, K., Kimura, Y., Murata, H., Suzuki, M., Thorn, P., Evans, W. R., and Carmichael, W. W. (1991a). Microcystins from Anabaena flosaquae NRC Chem Res Toxicol 4, Harada, K., Ogawa, K., Matsuura, K., Murata, H., Suzuki, M., Watanabe, M. F., Itezono, Y., and Nakayama, N. (1990b). Structural determination of geometrical isomers of microcystins LR and RR from cyanobacteria by two-dimensional NMR spectroscopic techniques. Chemical Research in Toxicology 3, Harada, K., Ogawa, K., Matsuura, K., Nagai, H., Murata, H., Suzuki, M., Itezono, Y., Nakayama, N., Shirai, M., and Nakano, M. (1991b). Isolation of two toxic heptapeptide microcystins from an axenic strain of Microcystis aeruginosa, K-139. Toxicon 29, Harada, K. I., Ohtani, I., Iwamoto, K., Suzuki, M., Watanabe, M. F., Watanabe, M., and Terao, K. (1994). Isolation of cylindrospermopsin from a cyanobacterium Umezakia natans and its screening method. Toxicon 32, Hardie, D. G. (1993). Use of protein phosphatase inhibitors in intact cells. In Protein Phosphorylation: A Practical Approach (D. G. Hardie, Ed.), pp Oxford University Press, Oxford. Hastie, C. J., Borthwick, E. B., Morrison, L. F., Codd, G. A., and Cohen, P. T. (2005). Inhibition of several protein phosphatases by a non-covalently 144

145 References interacting microcystin and a novel cyanobacterial peptide, nostocyclin. Biochim Biophys Acta 1726, Hawkins, P. R., Chandrasena, N. R., Jones, G. J., Humpage, A. R., and Falconer, I. R. (1997). Isolation and toxicity of Cylindrospermopsis raciborskii from an ornamental lake. Toxicon 35, Hawkins, P. R., Novic, S., Cox, P., Neilan, B. A., Burns, B. P., Shaw, G., Wickramasinghe, W., Peerapornpisal, Y., Ruangyuttikarn, W., Itayama, T., Saitou, T., Mizuochi, M., and Inamori, Y. (2005). A review of analytical methods for assessing the public health risk from microcystin in the aquatic environment. Journal of Water Supply: Research and Technology - Aqua 54, Hawkins, P. R., Runnegar, M. T., Jackson, A. R., and Falconer, I. R. (1985). Severe hepatotoxicity caused by the tropical cyanobacterium (blue-green alga) Cylindrospermopsis raciborskii (Woloszynska) Seenaya and Subba Raju isolated from a domestic water supply reservoir. Applied & Environmental Microbiology 50, Haystead, T. A., Sim, A. T., Carling, D., Honnor, R. C., Tsukitani, Y., Cohen, P., and Hardie, D. G. (1989). Effects of the tumour promoter okadaic acid on intracellular protein phosphorylation and metabolism. Nature 337, Herath, G. (1995). The algal bloom problem in Australian waterways: an economic appraisal. Review of Marketing and Agricultural Economics 63, Heresztyn, T., and Nicholson, B. C. (2001). Determination of cyanobacterial hepatotoxins directly in water using a protein phosphatase inhibition assay. Water Research 35, Hermansky, S. J., Stohs, S. J., Eldeen, Z. M., Roche, V. F., and Mereish, K. A. (1991). Evaluation of potential chemoprotectants against microcystin-lr hepatotoxicity in mice. J Appl Toxicol 11, Hiller, S., Krock, B., Cembella, A., and Luckas, B. (2007). Rapid detection of cyanobacterial toxins in precursor ion mode by liquid chromatography tandem mass spectrometry. J Mass Spectrom 42, Hitzfeld, B., Lampert, C., Späth, N., Mountfort, D., Kaspar, H., and Dietrich, D. (2000). Toxin production in cyanobacterial mats from ponds on the McMurdo Ice Shelf, Antarctica. Toxicon 38, Ho, R. H., and Kim, R. B. (2005). Transporters and drug therapy: implications for drug disposition and disease. Clin Pharmacol Ther 78, Hoeger, S. J., Haarscheidt, D., and Dietrich, D. (2003). Toxische Sekundärmetabolite in cyanobakteriellen Nahrungsergänzungsmitteln. Federal Office of Public Health FOPH, Switzerland. Hoeger, S. J., Schmid, D., Blom, J. F., Ernst, B., and Dietrich, D. R. (2007). Analytical and functional characterization of microcystins [Asp3]MC-RR and [Asp3,Dhb7]MC-RR: consequences for risk assessment? Environ Sci Technol 41, Hoeger, S. J., Shaw, G., Hitzfeld, B. C., and Dietrich, D. R. (2004). Occurrence and elimination of cyanobacterial toxins in two Australian drinking water treatment plants. Toxicon 43, Honkanen, R. E., Codispoti, B. A., Tse, K., Boynton, A. L., and Honkanan, R. E. (1994). Characterization of natural toxins with inhibitory activity against serine/threonine protein phosphatases. Toxicon 32,

146 References Honkanen, R. E., Dukelow, M., Zwiller, J., Moore, R. E., Khatra, B. S., and Boynton, A. L. (1991). Cyanobacterial nodularin is a potent inhibitor of type 1 and type 2A protein phosphatases. Mol Pharmacol 40, Honkanen, R. E., Zwiller, J., Moore, R. E., Daily, S. L., Khatra, B. S., Dukelow, M., and Boynton, A. L. (1990). Characterization of microcystin-lr, a potent inhibitor of type 1 and type 2A protein phosphatases. J Biol Chem 265, Hooser, S. B. (2000). Fulminant hepatocyte apoptosis in vivo following microcystin-lr administration to rats. Toxicologic Pathology 28, Hooser, S. B., Beasley, V. R., Lovell, R. A., Carmichael, W. W., and Haschek, W. M. (1989). Toxicity of microcystin LR, a cyclic heptapeptide hepatotoxin from Microcystis aeruginosa, to rats and mice. Vet Pathol 26, Hooser, S. B., Beasley, V. R., Waite, L. L., Kuhlenschmidt, M. S., Carmichael, W. W., and Haschek, W. M. (1991). Actin filament alterations in rat hepatocytes induced in vivo and in vitro by microcystin-lr, a hepatotoxin from the blue-green alga, Microcystis aeruginosa. Vet Pathol 28, Hugo-Wissemann, D., Anundi, I., Lauchart, W., Viebahn, R., and de Groot, H. (1991). Differences in glycolytic capacity and hypoxia tolerance between hepatoma cells and hepatocytes. Hepatology 13, Humpage, A. (2008). Toxin types, toxicokinetics and toxicodynamics. Adv Exp Med Biol 619, Humpage, A., Rositano, J., Bretag, A., Brown, R., Baker, P., Nicholson, B., and Steffensen, D. (1994). Paralytic shellfish poisons from Australian cyanobacterial blooms. Australian Journal of Marine and Freshwater Research 45, Humpage, A. R., and Falconer, I. R. (1999). Microcystin-LR and liver tumor promotion: Effects on cytokinesis, ploidy, and apoptosis in cultured hepatocytes. Environmental Toxicology 14, Humpage, A. R., and Falconer, I. R. (2003). Oral toxicity of the cyanobacterial toxin cylindrospermopsin in male Swiss albino mice: determination of no observed adverse effect level for deriving a drinking water guideline value. Environ Toxicol 18, Humpage, A. R., Fenech, M., Thomas, P., and Falconer, I. R. (2000). Micronucleus induction and chromosome loss in transformed human white cells indicate clastogenic and aneugenic action of the cyanobacterial toxin, cylindrospermopsin. Mutation Research/DNA Repair 472, Humpage, A. R., Fontaine, F., Froscio, S., Burcham, P., and Falconer, I. R. (2005). Cylindrospermopsin genotoxicity and cytotoxicity: role of cytochrome P-450 and oxidative stress. J Toxicol Environ Health A 68, Hyde, E. G., and Carmichael, W. W. (1991). Anatoxin-a(s), a naturally occurring organophosphate, is an irreversible active site-directed inhibitor of acetylcholinesterase (EC ). J Biochem Toxicol 6, Ikawa, M., Wegener, K., Foxall, T. L., and Sasner, J. J., Jr. (1982). Comparison of the toxins of the blue-green alga Aphanizomenon flos-aquae with the Gonyaulax toxins. Toxicon 20, Ince, P. G., and Codd, G. A. (2005). Return of the cycad hypothesis - does the amyotrophic lateral sclerosis/parkinsonism dementia complex (ALS/PDC) of Guam have new implications for global health? Neuropathol Appl Neurobiol 31,

147 References Ishii, H., Nishijima, M., and Abe, T. (2004). Characterization of degradation process of cyanobacterial hepatotoxins by a gram-negative aerobic bacterium. Water Research 38, Ito, E., Kondo, F., and Harada, K. (1997). Hepatic necrosis in aged mice by oral administration of microcystin-lr. Toxicon 35, Ito, E., Kondo, F., and Harada, K. (2000). First report on the distribution of orally administered microcystin-lr in mouse tissue using an immunostaining method. Toxicon 38, Ito, E., Kondo, F., and Harada, K. (2001). Intratracheal administration of microcystin-lr, and its distribution. Toxicon 39, Ito, E., Kondo, F., Terao, K., and Harada, K. (1997). Neoplastic nodular formation in mouse liver induced by repeated intraperitoneal injections of microcystin-lr. Toxicon 35, Ito, E., Takai, A., Kondo, F., Masui, H., Imanishi, S., and Harada, K. (2002). Comparison of protein phosphatase inhibitory activity and apparent toxicity of microcystins and related compounds. Toxicon 40, Iwasa, M., Yamamoto, M., Tanaka, Y., Kaito, M., and Adachi, Y. (2002). Spirulina-associated hepatotoxicity. The American Journal of Gastroenterology 97, Jackim, E., and Gentile, J. (1968). Toxins of a blue-green alga: similarity to saxitoxin. Science 162, Jigorel, E., Le Vee, M., Boursier-Neyret, C., Bertrand, M., and Fardel, O. (2005). Functional expression of sinusoidal drug transporters in primary human and rat hepatocytes. Drug Metab Dispos 33, Jochimsen, E. M., Carmichael, W. W., An, J. S., Cardo, D. M., Cookson, S. T., Holmes, C. E., Antunes, M. B., de Melo Filho, D. A., Lyra, T. M., Barreto, V. S., Azevedo, S. M., and Jarvis, W. R. (1998). Liver failure and death after exposure to microcystins at a hemodialysis center in Brazil. N Engl J Med 338, Johnson, H. E., King, S. R., Banack, S. A., Webster, C., Callanaupa, W. J., and Cox, P. A. (2008). Cyanobacteria (Nostoc commune) used as a dietary item in the Peruvian highlands produce the neurotoxic amino acid BMAA. J Ethnopharmacol 118, Jones, G. J., Bourne, D. G., Blakeley, R. L., and Doelle, H. (1994). Degradation of the cyanobacterial hepatotoxin microcystin by aquatic bacteria. Nat Toxins 2, Jones, G. J., and Orr, P. T. (1994). Release and degradation of microcystin following algicide treatment of a Microcystis aeruginosa bloom in a recreational lake, as determined by HPLC and protein phosphatase inhibition assay. Water Research 28, Kaebernick, M., and Neilan, B. A. (2001). Ecological and molecular investigations of cyanotoxin production. FEMS Microbiology Ecology 35, 1-9. Kakyo, M., Sakagami, H., Nishio, T., Nakai, D., Nakagomi, R., Tokui, T., Naitoh, T., Matsuno, S., Abe, T., and Yawo, H. (1999). Immunohistochemical distribution and functional characterization of an organic anion transporting polypeptide 2 (oatp2). FEBS Lett 445, Kao, C. Y. (1993). Paralytic Shellfish Poisoning. In Algal Toxins in Seafood and Drinking Water (I.Falconer, Ed.), pp Academic Press, London. 147

148 References Kaya, K., Sano, T., Beattie, K. A., and Codd, G. A. (1996). Nostocyclin, a novel 3-amino-6-hydroxy-2-piperidone-containing cyclic depsipeptide from the cyanobacterium Nostoc sp. Tetrahedron Letters 37, Kaya, K., Sano, T., Inoue, H., and Takagi, H. (2001). Selective determination of total normal microcystin by colorimetry, LC/UV detection and/or LC/MS. Analytica Chimica Acta 450, Kaya, K., and Watanabe, M. M. (1990). Microcystin composition of an axenic clonal strain of Microcystis viridis and Microcystis viridis-containing waterblooms in Japanese freshwaters. J. Appl. Phycol. 2, 173. Keppler, D. (2005). Uptake and efflux transporters for conjugates in human hepatocytes. Methods Enzymol 400, Kisby, G. E., Ellison, M., and Spencer, P. S. (1992). Content of the neurotoxins cycasin (methylazoxymethanol beta-d-glucoside) and BMAA (beta-nmethylamino-l-alanine) in cycad flour prepared by Guam Chamorros. Neurology 42, Kisby, G. E., Roy, D. N., and Spencer, P. S. (1988). Determination of beta-nmethylamino-l-alanine (BMAA) in plant (Cycas circinalis L.) and animal tissue by precolumn derivatization with 9-fluorenylmethyl chloroformate (FMOC) and reversed-phase high-performance liquid chromatography. J Neurosci Methods 26, Komatsu, M., Furukawa, T., Ikeda, R., Takumi, S., Nong, Q., Aoyama, K., Akiyama, S., Keppler, D., and Takeuchi, T. (2007). Involvement of mitogen-activated protein kinase signaling pathways in microcystin-lrinduced apoptosis after its selective uptake mediated by OATP1B1 and OATP1B3. Toxicol Sci 97, Kondo, F., Ikai, Y., Oka, H., Okumura, M., Ishikawa, N., Harada, K., Matsuura, K., Murata, H., and Suzuki, M. (1992). Formation, characterization, and toxicity of the glutathione and cysteine conjugates of toxic heptapeptide microcystins. Chem Res Toxicol 5, Kondo, F., Matsumoto, H., Yamada, S., Ishikawa, N., Ito, E., Nagata, S., Ueno, Y., Suzuki, M., and Harada, K. (1996). Detection and identification of metabolites of microcystins formed in vivo in mouse and rat livers. Chem Res Toxicol 9, König, J., Cui, Y., Nies, A. T., and Keppler, D. (2000a). A novel human organic anion transporting polypeptide localized to the basolateral hepatocyte membrane. Am J Physiol Gastrointest Liver Physiol 278, G König, J., Cui, Y., Nies, A. T., and Keppler, D. (2000b). Localization and genomic organization of a new hepatocellular organic anion transporting polypeptide. The journal of biological chemistry 275, König, J., Nies, A. T., Cui, Y., Leier, I., and Keppler, D. (1999a). Conjugate export pumps of the multidrug resistance protein (MRP) family: localization, substrate specificity, and MRP2-mediated drug resistance. Biochim Biophys Acta 1461, König, J., Rost, D., Cui, Y., and Keppler, D. (1999b). Characterization of the human multidrug resistance protein isoform MRP3 localized to the basolateral hepatocyte membrane. Hepatology 29, Konst, H., McKercher, P. D., Gorham, P. R., Robertson, A., and Howell, J. (1965). Symptoms and pathology produced by toxic Microcystis aeruginosa NRC-1 in laboratory and domestic animals. Can J Comp Med Vet Sci 29,

149 References Kotak, B. G., Kenefick, S. L., Fritz, D. L., Rousseaux, C. G., Prepas, E. E., and Hrudey, S. E. (1993). Occurrence and toxicololgical evaluation of cyanobacterial toxins in Alberta lakes and farm dugouts. Water Research 27, Krishnamurthy, T., Carmichael, W. W., and Sarver, E. W. (1986). Toxic peptides from freshwater cyanobacteria (blue-green algae). I. Isolation, purification and characterization of peptides from Microcystis aeruginosa and Anabaena flos-aquae. Toxicon 24, Kuiper-Goodman, T., Falconer, I. R., and Fitzgerald, D. J. (1999). Human Health Aspects. In Toxic Cyanobacteria in Water: A Guide to their Public Health Consequences, Monitoring and Management (I. Chorus, and J. Bartram, Eds.), pp E & FN Spon, London. Kuiper-Goodman, T., Lawrence, J. F., and Morrisey, M. (2000). Risk assessment of microcystins in blue-green algal health food products. X international IUPAC symposium on mycotoxins and phycotoxins, pp. 65. Kullak-Ublick, G.-A., Hagenbuch, B., Stieger, B., Wolkoff, A. W., and Meier, P. J. (1994). Functional characterization of the basolateral rat liver organic anion transporting polypeptide. Hepatology 20, Kullak-Ublick, G. A., Ismair, M. G., Stieger, B., Landmann, L., Huber, R., Pizzagalli, F., Fattinger, K., Meier, P. J., and Hagenbuch, B. (2001). Organic anion-transporting polypeptide B (OATP-B) and its functional comparison with three other OATPs of human liver. Gastroenterology 120, Lagos, N., Onodera, H., Zagatto, P. A., Andrinolo, D., Azevedo, S. M., and Oshima, Y. (1999). The first evidence of paralytic shellfish toxins in the fresh water cyanobacterium Cylindrospermopsis raciborskii, isolated from Brazil. Toxicon 37, Lahti, K., Rapala, J., Färdig, M., Niemelä, M., and Sivonen, K. (1997). Persistence of cyanobacterial hepatotoxin, microcystin-lr in particulate material and dissolved in lake water. Water Research 31, Lampert, W., and Sommer, U. (1999). Limnoökologie. Georg Thieme Verlag, Stuttgart. Lampert, W., and Sommer, U. (1999). Limnoökologie. Georg Thieme Verlag, Stuttgart. Lawrence, J. F., Niedzwiadek, B., Menard, C., Lau, B. P., Lewis, D., Kuiper- Goodman, T., Carbone, S., and Holmes, C. (2001). Comparison of liquid chromatography/mass spectrometry, ELISA, and phosphatase assay for the determination of microcystins in blue-green algae products. Journal of the AOAC International 84, Lawton, L., McElhiney, J., and Edwards, C. (1999). Purification of closely eluting hydrophobic microcystins (peptide cyanotoxins) by normal-phase and reversed-phase flash chromatography. Journal of Chromatography A 848, Lawton, L. A., and Codd, G. A. (1991). Cyanobacterial (blue-green algal) toxins and their significance in UK and European waters. Journal of the Institute for Water and Environmental Management 5, Lawton, L. A., Edwards, C., Beattie, K. A., Pleasance, S., Dear, G. J., and Codd, G. A. (1995). Isolation and characterization of microcystins from laboratory cultures and environmental samples of Microcystis aeruginosa and from an associated animal toxicosis. Nat Toxins 3,

150 References Lawton, L. A., Edwards, C., and Codd, G. A. (1994). Extraction and highperformance liquid-chromatographic method for the determination of microcystins in raw and treated waters. Analyst 119, Lee, W., Glaeser, H., Smith, L. H., Roberts, R. L., Moeckel, G. W., Gervasini, G., Leake, B. F., and Kim, R. B. (2005). Polymorphisms in human organic anion-transporting polypeptide 1A2 (OATP1A2): implications for altered drug disposition and central nervous system drug entry. J Biol Chem 280, Lehane, L. (2000). Paralytic Shellfish Poisoning: A review. National Office of Animal and Plant Health, Agriculture, Fisheries and Forestry - Australia, Canberra. Letschert, K., Keppler, D., and Konig, J. (2004). Mutations in the SLCO1B3 gene affecting the substrate specificity of the hepatocellular uptake transporter OATP1B3 (OATP8). Pharmacogenetics 14, Li, A. P. (2001). Screening for human ADME/Tox drug properties in drug discovery. Drug Discov Today 6, Li, L., Lee, T. K., Meier, P. J., and Ballatori, N. (1998). Identification of glutathione as a driving force and leukotriene C4 as a substrate for oatp1, the hepatic sinusoidal organic solute transporter. J Biol Chem 273, Li, L., Meier, P. J., and Ballatori, N. (2000). Oatp2 mediates bidirectional organic solute transport: a role for intracellular glutathione. Mol Pharmacol 58, Li, R. H., Carmichael, W. W., Brittain, S., Eaglesham, G. K., Shaw, G. R., Liu, Y. D., and Watanabe, M. M. (2001). First report of the cyanotoxins cylindrospermopsin and deoxycylindrospermopsin from Raphidiopsis curvata (cyanobacteria). J Phycol 37, Lin, J.-R., and Chu, F. S. (1994). Kinetics of distribution of microcystin LR in serum and liver cytosol of mice: an immunochemical analysis. Journal of Agriculture and Food Chemistry 42, Linderman, B. (1995). Complicated Child? Simple Options. Ransom Hill Press, Ramona, CA, USA. Liu, Y., Chen, W., Li, D., Shen, Y., Li, G., and Liu, Y. (2006). First report of aphantoxins in China--waterblooms of toxigenic Aphanizomenon flosaquae in Lake Dianchi. Ecotoxicol Environ Saf 65, Lovell, R. A., Schaeffer, D. J., Hooser, S. B., Haschek, W. M., Dahlem, A. M., Carmichael, W. W., and Beasley, V. R. (1989). Toxicity of intraperitoneal doses of microcystin-lr in two strains of male mice. J Environ Pathol Toxicol Oncol 9, Lu, H., Choudhuri, S., Ogura, K., Csanaky, I. L., Lei, X., Cheng, X., Song, P. Z., and Klaassen, C. D. (2008). Characterization of organic anion transporting polypeptide 1b2-null mice: essential role in hepatic uptake/toxicity of phalloidin and microcystin-lr. Toxicol Sci 103, Lukassowitz, I. (2002). BgVV and BfArM issue warning: Food supplements with AFA algae are no substitute for medical treatment. Bundesinstitut für gesundheitlichen Verbraucherschutz und Veterinärmedizin, Berlin, Germany. Luukkainen, R., Sivonen, K., Namikoshi, M., Fardig, M., Rinehart, K. L., and Niemela, S. I. (1993). Isolation and identification of eight microcystins from thirteen Oscillatoria agardhii strains and structure of a new microcystin. Appl Environ Microbiol 59,

151 References MacKintosh, C. (1993). Assay and purification of protein (serine/threonine) phosphatases. In Protein Phosphorylation: A Practical Approach (D. G. Hardie, Ed.), pp Oxford University Press, Oxford. MacKintosh, C., Beattie, K. A., Klumpp, S., Cohen, P., and Codd, G. A. (1990). Cyanobacterial microcystin-lr is a potent and specific inhibitor of protein phosphatases 1 and 2A from both mammals and higher plants. FEBS Lett 264, MacKintosh, R. W., Dalby, K. N., Campbell, D. G., Cohen, P. T., Cohen, P., and MacKintosh, C. (1995). The cyanobacterial toxin microcystin binds covalently to cysteine-273 on protein phosphatase 1. FEBS Letters 371, Mahagita, C., Grassl, S. M., Piyachaturawat, P., and Ballatori, N. (2007). Human organic anion transporter 1B1 and 1B3 function as bidirectional carriers and do not mediate GSH-bile acid cotransport. Am J Physiol Gastrointest Liver Physiol 293, G Mahmood, N. A., and Carmichael, W. W. (1986a). Paralytic shellfish poisons produced by the freshwater cyanobacterium Aphanizomenon flos-aquae NH-5. Toxicon 24, Mahmood, N. A., and Carmichael, W. W. (1986b). The pharmacology of anatoxin-a(s), a neurotoxin produced by the freshwater cyanobacterium Anabaena flos-aquae NRC Toxicon 24, Mahmood, N. A., and Carmichael, W. W. (1987). Anatoxin-a(s), an anticholinesterase from the cyanobacterium Anabaena flos-aquae NRC Toxicon 25, Mahmood, N. A., Carmichael, W. W., and Pfahler, D. (1988). Anticholinesterase poisonings in dogs from a cyanobacterial (blue-green algae) bloom dominated by Anabaena flos-aquae. Am J Vet Res 49, Marchegiano, P., Carubbi, F., Tiribelli, C., Amarri, S., Stebel, M., Lunazzi, G. C., Levy, D., and Bellentani, S. (1992). Transport of sulfobromophthalein and taurocholate in the HepG2 cell line in relation to the expression of membrane carrier proteins. Biochem Biophys Res Commun 183, Matsunaga, S., Moore, R., Niemczura, W., and Carmichael, W. (1989). Anatoxin-a(s), a potent anticholinesterase from Anabaena flos-aquae. Journal of the American Chemical Society 111, Maynes, J. T., Luu, H. A., Cherney, M. M., Andersen, R. J., Williams, D., Holmes, C. F., and James, M. N. (2006). Crystal structures of protein phosphatase-1 bound to motuporin and dihydromicrocystin-la: elucidation of the mechanism of enzyme inhibition by cyanobacterial toxins. J Mol Biol 356, McDermott, C. M., Nho, C. W., Howard, W., and Holton, B. (1998). The cyanobacterial toxin, microcystin-lr, can induce apoptosis in a variety of cell types. Toxicon 36, Meier-Abt, F., Hammann-Haenni, A., Stieger, B., Ballatori, N., and Boyer, J. L. (2006). The organic anion transport polypeptide 1d1 (Oatp1d1) mediates hepatocellular uptake of phalloidin and microcystin into skate liver. Toxicol Appl Pharmacol 218, Meier, P. J., and Stieger, B. (2002). Bile salt transporters. Annu Rev Physiol 64, Meriluoto, J., Karlsson, K., and Spoof, L. (2004). High-Troughput Screening of Ten Microcystins and Nodularins, Cyanobacterial Peptide Hepatotoxins, 151

152 References by Reverse-Phase Liquid Chromatography-Electrospry Ionisation Mass Spectrometry. Chromatographia 59, Meriluoto, J., and Spoof, L. (2005). SOP: Preparation of standard solutions of microcystin-lr for HPLC calibration. In TOXIC: Cyanobacterial Monitoring and Cyanotoxin Analysis (J. Meriluoto, and G. A. Codd, Eds.), pp Abo Akademi University Press, Abo. Meriluoto, J. A., Nygard, S. E., Dahlem, A. M., and Eriksson, J. E. (1990). Synthesis, organotropism and hepatocellular uptake of two tritium-labeled epimers of dihydromicrocystin-lr, a cyanobacterial peptide toxin analog. Toxicon 28, Metcalf, J. S., Banack, S. A., Lindsay, J., Morrison, L. F., Cox, P. A., and Codd, G. A. (2008). Co-occurrence of beta-n-methylamino-l-alanine, a neurotoxic amino acid with other cyanobacterial toxins in British waterbodies, Environ Microbiol 10, Metcalf, J. S., and Codd, G. A. (2000). Microwave oven and boiling waterbath extraction of hepatotoxins from cyanobacterial cells. FEMS Microbiol Lett 184, Michalski, C., Cui, Y., Nies, A. T., Nuessler, A. K., Neuhaus, P., Zanger, U. M., Klein, K., Eichelbaum, M., Keppler, D., and Konig, J. (2002). A naturally occurring mutation in the SLC21A6 gene causing impaired membrane localization of the hepatocyte uptake transporter. J Biol Chem 277, Mikhailov, A., Harmala-Brasken, A. S., Hellman, J., Meriluoto, J., and Eriksson, J. E. (2003). Identification of ATP-synthase as a novel intracellular target for microcystin-lr. Chemico-Biological Interactions 142, Mikkaichi, T., Suzuki, T., Tanemoto, M., Ito, S., and Abe, T. (2004). The organic anion transporter (OATP) family. Drug Metab Pharmacokinet 19, Miura, G. A., Robinson, N. A., Lawrence, W. B., and Pace, J. G. (1991). Hepatotoxicity of microcystin-lr in fed and fasted rats. Toxicon 29, Moffitt, M. C., and Neilan, B. A. (2004). Characterization of the nodularin synthetase gene cluster and proposed theory of the evolution of cyanobacterial hepatotoxins. Appl Environ Microbiol 70, Molica, R. J. R., Oliveira, E. J. A., Carvalho, P. V. V. C., Costa, A. N. S. F., Cunha, M. C. C., Melo, G. L., and Azevedo, S. M. F. O. (2005). Occurrence of saxitoxins and an anatoxin-a(s)-like anticholinesterase in a Brazilian drinking water supply. Harmful Algae 4, Monks, N. R., Liu, S., Xu, Y., Yu, H., Bendelow, A. S., and Moscow, J. A. (2007). Potent cytotoxicity of the phosphatase inhibitor microcystin LR and microcystin analogues in OATP1B1- and OATP1B3-expressing HeLa cells. Mol Cancer Ther 6, Montine, T. J., Li, K., Perl, D. P., and Galasko, D. (2005). Lack of betamethylamino-l-alanine in brain from controls, AD, or Chamorros with PDC. Neurology 65, Moore, R. E., Chen, J. L., Moore, B. S., and Patterson, G. M. L. (1991). Biosynthesis of microcystin-lr. Origin of the carbons in the ADDA and Masp units. Journal of the American Chemical Society 113, Moreno, I. M., Pereira, P., Franca, S., and Camean, A. (2004). Toxic cyanobacteria strains isolated from blooms in the Guadiana river (southwestern Spain). Biol Res 37,

153 References Mumby, M. C., and Walter, G. (1993). Protein serine/threonine phosphatases: structure, regulation, and functions in cell growth. Physiological Reviews 73, Mur, L. R., Skulberg, O. M., and Utkilen, H. (1999). Cyanobacteria in the environment. In Toxic Cyanobacteria in Water: A Guide to their Public Health Consequences, Monitoring and Management (I. Chorus, and J. Bartram, Eds.), pp E & FN Spon, London. Murch, S. J., Cox, P. A., and Banack, S. A. (2004b). A mechanism for slow release of biomagnified cyanobacterial neurotoxins and neurodegenerative disease in Guam. Proc Natl Acad Sci U S A 101, Murch, S. J., Cox, P. A., Banack, S. A., Steele, J. C., and Sacks, O. W. (2004a). Occurrence of beta-methylamino-l-alanine (BMAA) in ALS/PDC patients from Guam. Acta Neurol Scand 110, Myers, T. G., and Nelson, S. D. (1990). Neuroactive carbamate adducts of beta- N-methylamino-L-alanine and ethylenediamine. Detection and quantitation under physiological conditions by 13C NMR. J Biol Chem 265, Namikoshi, M., Choi, B. W., Sakai, R., Sun, F., Rinehart, K. L., Carmichael, W. W., Evans, W. R., Cruz, P., Munro, M. H. G., and Blunt, J. W. (1994). New nodularins: a general method for structure assignment. Journal of Organic Chemistry 59, Namikoshi, M., and Rinehart, K. (1996). Bioactive compounds produced by cyanobacteria. Journal of Industrial Microbiology 17, Namikoshi, M., Rinehart, K. L., Sakai, R., Scotts, R. R., Dahlem, A. M., Beasley, V. R., Carmichael, W. W., and Evans, W. R. (1992). Identification of 12 hepatotoxins from a Homer Lake bloom of the cyanobacteria Microcystis aeruginosa, Microcystis viridis and Microcystis wesenbergii: nine new microcystins. Journal of Organic Chemistry 57, Namikoshi, M., Sun, F., Choi, B. W., Rinehart, K. L., Carmichael, W. W., Evans, W. R., and Beasley, V. R. (1995). Seven more microcystins from Homer Lake cells: application of the general method for structure assignment of peptides containing alpha, beta-dehydroamino acid unit(s). Journal of Organic Chemistry 60, Negri, A. P., and Jones, G. J. (1995a). Bioaccumulation of paralytic shellfish poisoning (PSP) toxins from the cyanobacterium Anabaena circinalis by the freshwater mussel Alathyria condola. Toxicon 33, Negri, A. P., Jones, G. J., and Hindmarsh, M. (1995b). Sheep mortality associated with paralytic shellfish poisons from the cyanobacterium Anabaena circinalis. Toxicon 33, Neilan, B. A., Saker, M. L., Fastner, J., Torokne, A., and Burns, B. P. (2003). Phylogeography of the invasive cyanobacterium Cylindrospermopsis raciborskii. Molecular Ecology 12, Nies, A. T. (2007). The role of membrane transporters in drug delivery to brain tumors. Cancer Lett 254, Nishiwaki-Matsushima, R., Nishiwaki, S., Ohta, T., Yoshizawa, S., Suganuma, M., Harada, K.-I., Watanabe, M. F., and Fujiki, H. (1991). Structurefunction relationships of microcystins, liver tumor promoters, in interaction with protein phosphatase. Japanese Journal of Cancer Research 82, Nishiwaki-Matsushima, R., Ohta, T., Nishiwaki, S., Suganuma, M., Kohyama, K., Ishikawa, T., Carmichael, W. W., and Fujiki, H. (1992). Liver tumor 153

154 References promotion by the cyanobacterial cyclic peptide toxin microcystin-lr. J Cancer Res Clin Oncol 118, Nogueira, I. C., Pereira, P., Dias, E., Pflugmacher, S., Wiegand, C., Franca, S., and Vasconcelos, V. M. (2004). Accumulation of paralytic shellfish toxins (PST) from the cyanobacterium Aphanizomenon issatschenkoi by the cladoceran Daphnia magna. Toxicon 44, Norris, R. L., Eaglesham, G. K., Pierens, G., Shaw, G. R., Smith, M. J., Chiswell, R. K., Seawright, A. A., and Moore, M. R. (1999). Deoxycylindrospermopsin, an analog of cylindrospermopsin from Cylindrospermopsis raciborskii. Environmental Toxicology 14, Norris, R. L., Seawright, A. A., Shaw, G. R., Senogles, P., Eaglesham, G. K., Smith, M. J., Chiswell, R. K., and Moore, M. R. (2002). Hepatic xenobiotic metabolism of cylindrospermopsin in vivo in the mouse. Toxicon 40, Nultsch, W. (2001). Allgemeine Botanik. Georg Thieme Verlag, Stuttgart. Nussler, A. K., Nussler, N. C., Merk, V., Brulport, M., Schormann, W., Yao, P., and Hengstler, J. G. (2009). The Holy Grail of Hepatocyte Culturing and Therapeutic Use. In Strategies in Regenerative Medicine (M. Santin, Ed.), pp Springer New York, University of Brighton. Nyberg, S. L., Remmel, R. P., Mann, H. J., Peshwa, M. V., Hu, W. S., and Cerra, F. B. (1994). Primary hepatocytes outperform Hep G2 cells as the source of biotransformation functions in a bioartificial liver. Ann Surg 220, Odriozola, E., Ballabene, N., and Salamanco, A. (1984). [Poisoning in cattle caused by blue-green algae]. Rev Argent Microbiol 16, Ohta, T., Nishiwaki, R., Yatsunami, J., Komori, A., Suganuma, M., and Fujiki, H. (1992). Hyperphosphorylation of cytokeratins 8 and 18 by microcystin-lr, a new liver tumor promoter, in primary cultured rat hepatocytes. Carcinogenesis 13, Ohta, T., Sueoka, E., Iida, N., Komori, A., Suganuma, M., Nishiwaki, R., Tatematsu, M., Kim, S. J., Carmichael, W. W., and Fujiki, H. (1994). Nodularin, a potent inhibitor of protein phosphatases 1 and 2A, is a new environmental carcinogen in male F344 rat liver. Cancer Res 54, Ohtani, I., Moore, R. E., and Runnegar, M. T. C. (1992). Cylindrospermopsin: A Potent Hepatotoxin from the Blue-Green Alga Cylindrospermopsis raciborskii. Journal of the American Chemical Society 114, Oliver, R. L., and Ganf, G. G. (2000). Freshwater blooms. In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. Oren, A. (2000). Salts and Brines. In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. Oshima, Y. (1995). Postcolumn derivatization liquid chromatographic method for paralytic shellfish toxins. J. AOAC Int. 78, Osswald, J., Rellán, S., Saker, M., Gago-Martínez, A., and Vasconcelos, V. (2008). First detection of anatoxin-a in human and animal dietary supplements containing cyanobacteria (blue-green algae). poster presentation at SETAC Warsaw. 154

155 References Padisak, J. (1997). Cylindrospermopsis raciborskii (Woloszynska) Seenayya et Subba Raju, an expanding, highly adaptive cyanobacterium: worldwide distribution and review of its ecology. Archiv für Hydrobiologie 107, Paerl, H. W. (2000). Marine Plankton. In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. Papapetropoulos, S. (2007). Is there a role for naturally occurring cyanobacterial toxins in neurodegeneration? The beta-n-methylamino-lalanine (BMAA) paradigm. Neurochem Int 50, Park, H. D., Sasaki, Y., Maruyama, T., Yanagisawa, E., Hiraishi, A., and Kato, K. (2001). Degradation of the cyanobacterial hepatotoxin microcystin by a new bacterium isolated from a hypertrophic lake. Environmental Toxicology 16, Park, H. D., Watanabe, M. F., Harda, K., Nagai, H., Suzuki, M., Watanabe, M., and Hayashi, H. (1993). Hepatotoxin (microcystin) and neurotoxin (anatoxin-a) contained in natural blooms and strains of cyanobacteria from Japanese freshwaters. Nat Toxins 1, Paul, V. J. (2008). Global warming and cyanobacterial harmful algal blooms. Adv Exp Med Biol 619, Pearl, H. W., and Huisman, J. (2009). Climate change: a catalyst for global expansion of harmful cyanobacterial blooms. Environmental Microbiology Reports 1, Pearson, L. A., Hisbergues, M., Borner, T., Dittmann, E., and Neilan, B. A. (2004). Inactivation of an ABC transporter gene, mcyh, results in loss of microcystin production in the cyanobacterium Microcystis aeruginosa PCC Appl Environ Microbiol 70, Pegram, R. A., Nichols, T., Etheridge, S., Humpage, A., LeBlanc, S., Love, A., Neilan, B., Pflugmacher, S., Runnegar, M., and Thacker, R. (2008). Cyanotoxins Workgroup report. Adv Exp Med Biol 619, Pentecost, A., and Whitton, B. A. (2000). Limestones. In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. Pereira, P., Dias, E., Franca, S., Pereira, E., Carolino, M., and Vasconcelos, V. (2004a). Accumulation and depuration of cyanobacterial paralytic shellfish toxins by the freshwater mussel Anodonta cygnea. Aquat Toxicol 68, Pereira, P., Li, R., Carmichael, W. W., Dias, E., and Franca, S. (2004b). Taxonomy and production of paralytic shellfish toxins by the freshwater cyanobacterium Aphanizomenon gracile LMECYA40. Eur. J. Phycol. 39, Pereira, P., Onodera, H., Andrinolo, D., Franca, S., Araujo, F., Lagos, N., and Oshima, Y. (2000). Paralytic shellfish toxins in the freshwater cyanobacterium Aphanizomenon flos-aquae, isolated from Montargil reservoir, Portugal. Toxicon 38, Pflugmacher, S., Wiegand, C., Oberemm, A., Beattie, K. A., Krause, E., Codd, G. A., and Steinberg, C. E. (1998). Identification of an enzymatically formed glutathione conjugate of the cyanobacterial hepatotoxin microcystin-lr: the first step of detoxication. Biochim Biophys Acta 1425,

156 References Pizzagalli, F., Hagenbuch, B., Stieger, B., Klenk, U., Folkers, G., and Meier, P. J. (2002). Identification of a novel human organic anion transporting polypeptide as a high affinity thyroxine transporter. Mol Endocrinol 16, Pouria, S., de Andrade, A., Barbosa, J., Cavalcanti, R. L., Barreto, V. T., Ward, C. J., Preiser, W., Poon, G. K., Neild, G. H., and Codd, G. A. (1998). Fatal microcystin intoxication in haemodialysis unit in Caruaru, Brazil. Lancet 352, Preussel, K., Stuken, A., Wiedner, C., Chorus, I., and Fastner, J. (2006). First report on cylindrospermopsin producing Aphanizomenon flos-aquae (Cyanobacteria) isolated from two German lakes. Toxicon 47, Price, P. J. (1975). Preparation and use of rat-tail collagen. Methods in Cell Science 1, Rantala, A., Fewer, D. P., Hisbergues, M., Rouhiainen, L., Vaitomaa, J., Borner, T., and Sivonen, K. (2004). Phylogenetic evidence for the early evolution of microcystin synthesis. Proceedings of the National Academy of Sciences,USA 101, Rapala, J., Lahti, K., Sivonen, K., and Niemela, S. I. (1994). Biodegradability and adsorption on lake sediments of cyanobacterial hepatotoxins and anatoxin-a. Lett Appl Microbiol 19, Rapala, J., Sivonen, K., Luukkainen, R., and Niemela, S. I. (1993). Anatoxin-a concentration in Anabaena and Aphanizomenon under different environmental conditions and comparison of growth by toxic and non-toxic Anabaena-strains: A laboratory study. Journal of Applied Phycology 5, Rapala, J., Sivonen, K., Lyra, C., and Niemela, S. I. (1997). Variation of microcystins, cyanobacterial hepatotoxins, in Anabaena spp. as a function of growth stimuli. Appl Environ Microbiol 63, Rawn, D. F., Niedzwiadek, B., Lau, B. P., and Saker, M. (2007). Anatoxin-a and its metabolites in blue-green algae food supplements from Canada and Portugal. J Food Prot 70, Rinehart, K. L., Harada, K.-I., Namikoshi, M., Chen, C., Harvis, C. A., Munro, M. H. G., Blunt, J. W., Mulligan, P. E., Beasley, B. R., Dahlem, A. M., and Carmichael, W. W. (1988). Nodularin, Microcystin, and the Configuration of Adda. Journal of the American Chemical Society 110, Rinehart, K. L., Namikoshi, M., and Choi, B. W. (1994). Structure and biosynthesis of toxins from blue-green algae (cyanobacteria). Journal of Applied Phycology 6, Rius, M., Hummel-Eisenbeiss, J., Hofmann, A. F., and Keppler, D. (2006). Substrate specificity of human ABCC4 (MRP4)-mediated cotransport of bile acids and reduced glutathione. Am J Physiol Gastrointest Liver Physiol 290, G Rivasseau, C., Racaud, P., Deguin, A., and Hennion, M. C. (1999). Development of a bioanalytical phosphatase inhibition test for the monitoring of microcystins in environmental water samples. Analytica Chimica Acta 394, Robertson, A., Blay, P., Reeves, K., Thomas, K., and Quilliam, M. A. (2007). HILIC-MS Analysis of Excitotoxic Amino Acids in Marine and Freshwater Algae and Cyanobacteria. VII. International Conference on Toxic Cyanobacteria. 156

157 References Robillot, C., and Hennion, M. C. (2004). Issues arising when interpreting the results of the protein phosphatase 2A inhibition assay for the monitoring of microcystins. Analytics Chimica Acta 512, Robinson, N. A., Miura, G. A., Matson, C. F., Dinterman, R. E., and Pace, J. G. (1989). Characterization of chemically tritiated microcystin-lr and its distribution in mice. Toxicon 27, Robinson, N. A., Pace, J. G., Matson, C. F., Miura, G. A., and Lawrence, W. B. (1991). Tissue distribution, excretion and hepatic biotransformation of microcystin-lr in mice. J Pharmacol Exp Ther 256, Rodriguez-Antona, C., Donato, M. T., Boobis, A., Edwards, R. J., Watts, P. S., Castell, J. V., and Gomez-Lechon, M. J. (2002). Cytochrome P450 expression in human hepatocytes and hepatoma cell lines: molecular mechanisms that determine lower expression in cultured cells. Xenobiotica 32, Rogers, E. H., Hunter, E. S., 3rd, Moser, V. C., Phillips, P. M., Herkovits, J., Munoz, L., Hall, L. L., and Chernoff, N. (2005). Potential developmental toxicity of anatoxin-a, a cyanobacterial toxin. J Appl Toxicol 25, Rücker, J., Stuken, A., Nixdorf, B., Fastner, J., Chorus, I., and Wiedner, C. (2007). Concentrations of particulate and dissolved cylindrospermopsin in 21 Aphanizomenon-dominated temperate lakes. Toxicon 50, Runnegar, M., Berndt, N., and Kaplowitz, N. (1995c). Microcystin uptake and inhibition of protein phosphatases: effects of chemoprotectants and selfinhibition in relation to known hepatic transporters. Toxicol Appl Pharmacol 134, Runnegar, M., Berndt, N., Kong, S. M., Lee, E. Y., and Zhang, L. (1995a). In vivo and in vitro binding of microcystin to protein phosphatases 1 and 2A. Biochem Biophys Res Commun 216, Runnegar, M. T., Falconer, I. R., Buckley, T., and Jackson, A. R. (1986). Lethal potency and tissue distribution of 125I-labelled toxic peptides from the blue-green alga Microcystis aeruginosa. Toxicon 24, Runnegar, M. T., Gerdes, R. G., and Falconer, I. R. (1991). The uptake of the cyanobacterial hepatotoxin microcystin by isolated rat hepatocytes. Toxicon 29, Runnegar, M. T., Jackson, A. R., and Falconer, I. R. (1988). Toxicity of the cyanobacterium Nodularia spumigena Mertens. Toxicon 26, Runnegar, M. T., Kong, S., and Berndt, N. (1993). Protein phosphatase inhibition and in vivo hepatotoxicity of microcystins. Am J Physiol 265, G Runnegar, M. T., Kong, S. M., Zhong, Y. Z., and Lu, S. C. (1995d). Inhibition of reduced glutathione synthesis by cyanobacterial alkaloid cylindrospermopsin in cultured rat hepatocytes. Biochem Pharmacol 49, Runnegar, M. T., Maddatu, T., Deleve, L. D., Berndt, N., and Govindarajan, S. (1995b). Differential toxicity of the protein phosphatase inhibitors microcystin and calyculin A. J Pharmacol Exp Ther 273, Runnegar, M. T., Xie, C., Snider, B. B., Wallace, G. A., Weinreb, S. M., and Kuhlenkamp, J. (2002). In vitro hepatotoxicity of the cyanobacterial alkaloid cylindrospermopsin and related synthetic analogues. Toxicol Sci 67, Runnegar, M. T. C., and Falconer, I. R. (1981). Isolation, characterization, and pathology of the toxin from the blue-green alga Microcystis aeruginosa. In 157

158 References The Water Environment - Algal Toxins and Health (W. W. Carmichael, Ed.), pp Plenum Press, New York. Runnegar, M. T. C., and Falconer, I. R. (1982). The in vivo and in vitro biological effects of the peptide hepatotoxin from the blue-green alga Microcystis aeruginosa. South African Journal of Science 78, Sahin, A., Tencalla, F. G., Dietrich, D. R., and Naegeli, H. (1996). Biliary excretion of biochemically active cyanobacteria (blue-green algae) hepatotoxins in fish. Toxicology 106, Saker, M. L., Jungblut, A.-D., Neilan, B. A., Rawn, D. F. K., and Vasconcelos, V. M. (2005). Detection of microcystin synthetase genes in health food supplements containing the freshwater cyanobacterium Aphanizomenon flos-aquae. Toxicon 46, Saker, M. L., Jungblut, A. D., Neilan, B. A., Rawn, D. F., and Vasconcelos, V. M. (2005). Detection of microcystin synthetase genes in health food supplements containing the freshwater cyanobacterium Aphanizomenon flos-aquae. Toxicon 46, Saker, M. L., Welker, M., and Vasconcelos, V. M. (2007). Multiplex PCR for the detection of toxigenic cyanobacteria in dietary supplements produced for human consumption. Appl Microbiol Biotechnol 73, Salazar, M., Chamorro, G. A., Salazar, S., and Steele, C. E. (1996). Effect of Spirulina maxima consumption on reproduction and peri- and postnatal development in rats. Food Chem Toxicol 34, Salazar, M., Martinez, E., Madrigal, E., Ruiz, L. E., and Chamorro, G. A. (1998). Subchronic toxicity study in mice fed Spirulina maxima. Journal of Ethnopharmacology 62, Sano, T., Takagi, H., and Kaya, K. (2004). A Dhb-microcystin from the filamentous cyanobacterium Planktothrix rubescens. Phytochemistry 65, Sano, T., Takagi, H., Sadakane, K., Ichinose, T., Kawazato, H., and Kaya, K. (2004). Carcionogenic effects of microcystin-lr and Dhb-microcystin-LR on mice liver. 6th International Conference on Toxic Cyanobacteria. Sano, T., Usui, T., Ueda, K., Osada, H., and Kaya, K. (2001). Isolation of new protein phosphatase inhibitors from two cyanobacteria species, Planktothrix spp. J Nat Prod 64, Satlin, L. M., Amin, V., and Wolkoff, A. W. (1997). Organic anion transporting polypeptide mediates organic anion/hco3- exchange. J Biol Chem 272, Sawyer, P. J., Gentile, J. H., and Sasner, J. J., Jr. (1968). Demonstration of a toxin from Aphanizomenon flos-aquae (L.) Ralfs. Canadian Journal of Microbiology 14, Schembri, M. A., Neilan, B. A., and Saint, C. P. (2001). Identification of genes implicated in toxin production in the cyanobacterium Cylindrospermopsis raciborskii. Environ Toxicol 16, Schinkel, A. H., and Jonker, J. W. (2003). Mammalian drug efflux transporters of the ATP binding cassette (ABC) family: an overview. Adv Drug Deliv Rev 55, Schopf, J. W., and Packer, B. M. (1987). Early Archean (3.3-billion to 3.5-billionyear-old) microfossils from Warrawoona Group, Australia. Science 237, Seawright, A. A., Nolan, C. C., Shaw, G. R., Chiswell, R. K., Norris, R. L., Moore, M. R., and Smith, M. J. (1999). The oral toxicity for mice of the 158

159 References tropical cyanobacterium Cylindrospermopsis raciborskii (Woloszynska). Environmental Toxicology 14, Sedmak, B., Carmeli, S., and Elersek, T. (2008). "Non-toxic" cyclic peptides induce lysis of cyanobacteria-an effective cell population density control mechanism in cyanobacterial blooms. Microb Ecol 56, Shaw, G., Garnett, C., Moore, M. R., and Florian, P. (2001). The Predicted Impact of Climate Change on Toxic Algal (Cyanobacterial) Blooms and Toxin Production in Queensland. Environmental Health 1, Shaw, G. R., Seawright, A. A., and Moore, M. R. (2001). Toxicology and Human health Implications of the Cyanobacterial Toxin Cylindrospermopsin. In Mycotoxins and Phycotoxins in Perspective at the turn of the Millennium (W. J. de Koe, R. A. Samson, H. P. van Egmond, J. Gilbert, and M. Sabino, Eds.), pp de Koe, W.J., Wageningen, The Netherlands. Shaw, G. R., Seawright, A. A., Moore, M. R., and Lam, P. K. (2000). Cylindrospermopsin, a cyanobacterial alkaloid: evaluation of its toxicologic activity. Ther Drug Monit 22, Shen, X., Lam, P. K., Shaw, G. R., and Wickramasinghe, W. (2002). Genotoxicity investigation of a cyanobacterial toxin, cylindrospermopsin. Toxicon 40, Shi, X., Bai, S., Ford, A. C., Burk, R. D., Jacquemin, E., Hagenbuch, B., Meier, P. J., and Wolkoff, A. W. (1995). Stable inducible expression of a functional rat liver organic anion transport protein in HeLa cells. The Journal of Biochemical Chemistry 270, Simonsohn, B. (2000). Die Heilkraft der Afa-Alge. Goldmann, München. Simonsohn, B. (2006). Hyperaktivität: Warum Ritalin keine Lösung ist. Gesunde Strategien, die wirklich helfen. Goldmann, München. Sivonen, K., and Jones, G. (1999). Cyanobacterial toxins. In Toxic Cyanobacteria in Water: A Guide to their Public Health Consequences, Monitoring and Management (I. Chorus, and J. Bartram, Eds.), pp E & FN Spon, London. Sivonen, K., Namikoshi, M., Evans, W. R., Carmichael, W. W., Sun, F., Rouhiainen, L., Luukkainen, R., and Rinehart, K. L. (1992). Isolation and characterization of a variety of microcystins from seven strains of the cyanobacterial genus Anabaena. Appl Environ Microbiol 58, Sivonen, K., Namikoshi, M., Luukkainen, R., Färdig, M., Rouhiainen, L., Evans, W. R., Carmichael, W. W., Rinehart, K. L., and Niemelä, S. I. (1995). Variation of cyanobacterial hepatotoxins in Finland. In The Contaminants in the Nordic Ecosystem, Dynamics, Processes and Fate (M. Munawar, and M. Luotola, Eds.), pp SPB Academic Publishing, Amsterdam. Skulberg, O. M., Carmichael, W. W., Andersen, R. A., Matsunaga, S., Moore, R. E., and Skulberg, R. (1992). Investigations of a neurotoxic Oscillatorialean strain (cyanophyceae) and its toxin. Isolation and characterization of homoanatoxin-a. Env. Toxicol. Chem. 11, Skulberg, O. M., Carmichael, W. W., Codd, G. A., and Skulberg, R. (1993). Taxonomy of toxic cyanophyceae (Cyanobacteria). In Algal Toxins in Seafood and Drinking Water (I. R. Falconer, Ed.), pp Academic Press, London. Smith, Q. R., Nagura, H., Takada, Y., and Duncan, M. W. (1992). Facilitated transport of the neurotoxin, beta-n-methylamino-l-alanine, across the blood-brain barrier. J Neurochem 58,

160 References Soliakov, L., Gallagher, T., and Wonnacott, S. (1995). Anatoxin-a-evoked [3H]dopamine release from rat striatal synaptosomes. Neuropharmacology 34, Spencer, P. S., Kisby, G. E., and Ludolph, A. C. (1991a). Long-latency neurodegenerative disease in the western Pacific. Geriatrics 46 Suppl 1, Spencer, P. S., Nunn, P. B., Hugon, J., Ludolph, A. C., Ross, S. M., Roy, D. N., and Robertson, R. C. (1987a). Guam amyotrophic lateral sclerosisparkinsonism-dementia linked to a plant excitant neurotoxin. Science 237, Spencer, P. S., Ohta, M., and Palmer, V. S. (1987c). Cycad use and motor neurone disease in Kii peninsula of Japan. Lancet 2, Spencer, P. S., Palmer, V. S., Herman, A., and Asmedi, A. (1987b). Cycad use and motor neurone disease in Irian Jaya. Lancet 2, Spoof, L. (2005). Microcystins and nodularins. In TOXIC: cyanobacterial monitoring and cyanotoxin analysis (J. Meriluoto, and G. A. Codd, Eds.), pp Abo Akademi University Press. Spoof, L., Vesterkvist, P., Lindholm, T., and Meriluoto, J. (2003). Screening for cyanobacterial hepatotoxins, microcystins and nodularin in environmental water samples by reversed-phase liquid chromatography-electrospray ionisation mass spectrometry. J Chromatogr A 1020, Stange, J., Mitzner, S., Strauss, M., Fischer, U., Lindemann, S., Peters, E., Holtz, M., Drewelow, B., and Schmidt, R. (1995). Primary or established liver cells for a hybrid liver? Comparison of metabolic features. Asaio J 41, M Steele, J. C. (2005). Parkinsonism-dementia complex of Guam. Mov Disord 20 Suppl 12, S99-S107. Steele, J. C., and McGeer, P. L. (2008). The ALS/PDC syndrome of Guam and the cycad hypothesis. Neurology 70, Stoner, R. D., Adams, W. H., Slatkin, D. N., and Siegelman, H. W. (1989). The effects of single L-amino acid substitutions on the lethal potencies of the microcystins. Toxicon 27, Stotts, R. R., Namikoshi, M., Haschek, W. M., Rinehart, K. L., Carmichael, W. W., Dahlem, A. M., and Beasley, V. R. (1993). Structural modifications imparting reduced toxicity in microcystins from Microcystis spp. Toxicon 31, Stotts, R. R., Twardock, A. R., Haschek, W. M., Choi, B. W., Rinehart, K. L., and Beasley, V. R. (1997a). Distribution of tritiated dihydromicrocystin in swine. Toxicon 35, Stotts, R. R., Twardock, A. R., Koritz, G. D., Haschek, W. M., Manuel, R. K., Hollis, W. B., and Beasley, V. R. (1997b). Toxicokinetics of tritiated dihydromicrocystin-lr in swine. Toxicon 35, Swanson, K. L., Allen, C. N., Aronstam, R. S., Rapoport, H., and Albuquerque, E. X. (1986). Molecular mechanisms of the potent and stereospecific nicotinic receptor agonist (+)-anatoxin-a. Mol Pharmacol 29, Takenaka, S. (2001). Covalent glutathione conjugation to cyanobacterial hepatotoxin microcystin LR by F344 rat cytosolic and microsomal glutathione S-transferases. Environ Toxicol Pharmacol 9, Takenaka, S., and Watanabe, M. F. (1997). Microcystin LR degradation by Pseudomonas aeruginosa alkaline protease. Chemosphere 34,

161 References Tamai, I., Nezu, J., Uchino, H., Sai, Y., Oku, A., Shimane, M., and Tsuji, A. (2000). Molecular identification and characterization of novel members of the human organic anion transporter (OATP) family. Biochem Biophys Res Commun 273, Teixera, M., Costa, M., Carvalho, V., Pereira, M., and Hage, E. (1993). Gastroenteritis epidemic in the area of the Itaparica Dam, Bahia, Brazil. Bulletin of the Pan-American Health Organization 27, Terao, K., Ohmori, S., Igarashi, K., Ohtani, I., Watanabe, M. F., Harada, K. I., Ito, E., and Watanabe, M. (1994). Electron microscopic studies on experimental poisoning in mice induced by cylindrospermopsin isolated from blue-green alga Umezakia natans. Toxicon 32, Theiss, W. C., Carmichael, W. W., Wyman, J., and Bruner, R. (1988). Blood pressure and hepatocellular effects of the cyclic heptapeptide toxin produced by the freshwater cyanobacterium (blue-green alga) Microcystis aeruginosa strain PCC Toxicon 26, Tillett, D., Dittmann, E., Erhard, M., von Dohren, H., Borner, T., and Neilan, B. A. (2000). Structural organization of microcystin biosynthesis in Microcystis aeruginosa PCC7806: an integrated peptide-polyketide synthetase system. Chem Biol 7, Toivola, D. M., and Eriksson, J. E. (1999). Toxins affecting cell signalling and alteration of cytoskeletal structure. Toxicology in Vitro 13, Toivola, D. M., Goldman, R. D., Garrod, D. R., and Eriksson, J. E. (1997). Protein phosphatases maintain the organization and structural interactions of hepatic keratin intermediate filaments. Journal of Cell Science 110, Toivola, D. M., Omary, M. B., Ku, N. O., Peltola, O., Baribault, H., and Eriksson, J. E. (1998). Protein phosphatase inhibition in normal and keratin 8/18 assembly-incompetent mouse strains supports a functional role of keratin intermediate filaments in preserving hepatocyte integrity. Hepatology 28, Torchia, E. C., Shapiro, R. J., and Agellon, L. B. (1996). Reconstitution of bile acid transport in the rat hepatoma McArdle RH-7777 cell line. Hepatology 24, Tsuji, K., Watanuki, T., Kondo, F., Watanabe, M., Suzuki, S., Nakazawa, H., Suzuki, M., Uchida, H., and Harada, K.-I. (1995). Stability of microcystins from cyanobacteria-ii. Effect of UV light on decomposition and isomerization. Toxicon 33, van Apeldoorn, M. E., van Egmond, H. P., Speijers, G. J., and Bakker, G. J. (2007). Toxins of cyanobacteria. Mol Nutr Food Res 51, van den Hoek, C., Jahns, H. M., and Mann, D. G. (1993). Algen. Georg Thieme Verlag, Stuttgart. van Montfoort, J. E., Hagenbuch, B., Groothuis, G. M., Koepsell, H., Meier, P. J., and Meijer, D. K. (2003). Drug uptake systems in liver and kidney. Current Drug Metabolism 4, Vega, A., Bell, A., and Nunn, P. (1968). The preparation of L- and D α-amino β- methylaminopropionic acid and the identification of the compound isolated from Cycas circinalis as the L-isomer. Phytochemistry 7, Vega, A., and Bell, E. A. (1967). α-amino-β-methylaminopropionic acid, a new amino acid from seeds of Cycas circinalis. Phytochemistry 6, Vezie, C., Brient, L., Sivonen, K., Bertru, G., Lefeuvre, J., and Salkinoja- Salonen, M. (1998). Variation of Microcystin Content of Cyanobacterial 161

162 References Blooms and Isolated Strains in Lake Grand-Lieu (France). Microb Ecol 35, Vincent, W. F. (2000). Cyanobacterial Dominance in the Polar Regions. In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. Vollhardt, K. P. C., and Schore, N. E. (1995). Organische Chemie. VCH, Weinheim. Wang, Q., Xie, P., Chen, J., and Liang, G. (2008). Distribution of microcystins in various organs (heart, liver, intestine, gonad, brain, kidney and lung) of Wistar rat via intravenous injection. Toxicon 52, Ward, C. J., and Codd, G. A. (1999). Comparative toxicity of four microcystins of different hydrophobicities to the protozoan, Tetrahymena pyriformis. J Appl Microbiol 86, Ward, D. M., and Castenholz, R. W. (2000). Cyanobacteria in Geothermal Habitats. In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. Watanabe, M. F., Oishi, S., Harda, K., Matsuura, K., Kawai, H., and Suzuki, M. (1988). Toxins contained in Microcystis species of cyanobacteria (bluegreen algae). Toxicon 26, Watanabe, M. M., Kaya, K., and Takamura, N. (1992). Fate of the toxic cyclic heptapeptides, the microcystins, from blooms of Microcystis (cyanobacteria) in a hypertrophic lake. Journal of Phycology 28, Weckesser, J., Drews, G., and Mayer, H. (1979). Lipopolysaccharides of photosynthetic prokaryotes. Annu Rev Microbiol 33, Weiss, J. H., Christine, C. W., and Choi, D. W. (1989b). Bicarbonate dependence of glutamate receptor activation by beta-n-methylamino-lalanine: channel recording and study with related compounds. Neuron 3, Weiss, J. H., Koh, J. Y., and Choi, D. W. (1989a). Neurotoxicity of beta-nmethylamino-l-alanine (BMAA) and beta-n-oxalylamino-l-alanine (BOAA) on cultured cortical neurons. Brain Res 497, Welker, M., and von Dohren, H. (2006). Cyanobacterial peptides - nature's own combinatorial biosynthesis. FEMS Microbiol Rev 30, Weng, D., Lu, Y., Wei, Y., Liu, Y., and Shen, P. (2007). The role of ROS in microcystin-lr-induced hepatocyte apoptosis and liver injury in mice. Toxicology 232, Whitton, B. A., and Potts, M. (2000). Introduction to the Cyanobacteria. In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. WHO (1998). Cyanobacterial toxins: Microcystin-LR. In Guidelines for drinkingwater quality, pp World Health Organization, Geneva. Wickstrom, M., Haschek, W., Henningsen, G., Miller, L. A., Wyman, J., and Beasley, V. (1996). Sequential ultrastructural and biochemical changes induced by microcystin-lr in isolated perfused rat livers. Nat Toxins 4, Wickstrom, M. L., Khan, S. A., Haschek, W. M., Wyman, J. F., Eriksson, J. E., Schaeffer, D. J., and Beasley, V. R. (1995). Alterations in microtubules, 162

163 References intermediate filaments, and microfilaments induced by microcystin-lr in cultured cells. Toxicol Pathol 23, Wiedner, C., Rucker, J., Fastner, J., Chorus, I., and Nixdorf, B. (2008). Seasonal dynamics of cylindrospermopsin and cyanobacteria in two German lakes. Toxicon 52, Wiegand, C., and Pflugmacher, S. (2005). Ecotoxicological effects of selected cyanobacterial secondary metabolites: a short review. Toxicol Appl Pharmacol 203, Wynn-Williams, D. D. (2000). Cyanobacteria in the Deserts - Life at the Limit? In The Ecology of Cyanobacteria - Their Diversity in Time and Space (B. A. Whitton, and M. Potts, Eds.), pp Kluwer Academic Publishers, Dordrecht/London/Boston. Yoshida, T., Makita, Y., Nagata, S., Tsutsumi, T., Yoshida, F., Sekijima, M., Tamura, S., and Ueno, Y. (1997). Acute oral toxicity of microcystin-lr, a cyanobacterial hepatotoxin, in mice. Nat Toxins 5, Yoshizawa, S., Matsushima, R., Watanabe, M. F., Harada, K., Ichihara, A., Carmichael, W. W., and Fujiki, H. (1990). Inhibition of protein phosphatases by microcystins and nodularin associated with hepatotoxicity. J Cancer Res Clin Oncol 116, Yu, S.-Z. (1989). Drinking water and primary liver cancer. In Primary liver cancer (Z. Y. Tang, M. C. Wu, and S. S. Xia, Eds.), pp China Academic Publishers/Springer, New York. Yu, S.-Z. (1995). Primary prevention of hepatocellular carcinoma. Journal of Gastroenterology and Hepatology 10, Zegura, B., Lah, T. T., and Filipic, M. (2004). The role of reactive oxygen species in microcystin-lr-induced DNA damage. Toxicology 200, Zegura, B., Sedmak, B., and Filipic, M. (2003). Microcystin-LR induces oxidative DNA damage in human hepatoma cell line HepG2. Toxicon 41, Zurawell, R. W., Chen, H., Burke, J. M., and Prepas, E. E. (2005). Hepatotoxic cyanobacteria: a review of the biological importance of microcystins in freshwater environments. J Toxicol Environ Health B Crit Rev 8,

164 Appendix 12 Apendix 164

165 Appendix Erklärung Die vorliegende Arbeit wurde selbstständig und ohne unzulässige Hilfe Dritter, sowie keiner weiteren als der angegebenen Hilfsmittel verfasst. Die Quellen, die anderen Werken wörtlich oder sinngemäß entnommen sind, sind als solche kenntlich gemacht. Weitere Personen, insbesondere Promotionsberater, waren an der inhaltlich materiellen Erstellung dieser Arbeit nicht beteiligt. Die Arbeit wurde in gleicher oder ähnlicher Form weder im In- noch Ausland keiner anderen Prüfungsbehörde vorgelegt. Eigenabgrenzung / Kooperation Die LC-MS/MS -Analysen der Nahrungsergänzungsmittel (Kapitel VI) wurden im Rahmen einer Zusammenarbeit von Dr. Jutta Fastner (Microcystine) vom Umweltbundesamt, II.3.3 Trinkwasserressourcen und -aufbereitung, Berlin und Dr. Alison Robertson (BMAA) vom Northwest Fisheries Science Center, Environmental Conservation Division, Harmful Algal Blooms Program, Seattle, Wisconsin, USA durchgeführt. Erste Versuche zur kompetitiven Aufnahme von Microcystin mit Bromosulfophthalein und Taurocholat in transfizierte HEK293-Zellen (Kapitel IV) wurden von Julia Kleinteich und Isabelle Eisele im Rahmen eines Vertiefungskurses in der AG Human- und Umwelttoxikologie der Universität Konstanz unter meiner Betreuung durchgeführt. Alle weiteren Leistungen wurden von mir selbst erbracht. Konstanz, Juni 2010 Andreas Fischer 165

166 Appendix Danksagung An dieser Stelle möchte ich mich herzlich bei all denen bedanken, die mich während meiner Doktorarbeit unterstützt und zu deren Entstehung beigetragen haben. Besonders danken möchte ich:...prof. Dr. Daniel Dietrich für die Möglichkeit die Promotion in seiner Arbeitsgruppe und unter seiner Betreuung durchzuführen....prof. Dr. Karl-Otto Rothhaupt für seine Bereitschaft als Koreferent zu fungieren....der Arthur und Aenne Feindt Stiftung Hamburg für die Finanzierung und unkomplizierte Unterstützung dieser Arbeit....den Cyanos Dr. Stefan Hoeger, Dr. Bernhard Ernst und Dr. Daniel Feurstein für die vielen fachlichen Diskussionen, Anregungen und Hilfestellungen....Dr. Kerstin Stemmer, Dr. Jutta Fastner vom Umweltbundesamt, II.3.3 Trinkwasserressourcen und -aufbereitung, Berlin, Dr. Alison Robertson vom Northwest Fisheries Science Center, Environmental Conservation Division, Harmful Algal Blooms Program, Seattle, Wisconsin, USA, Prof. Dr. Andreas Nüssler vom Klinikum rechts der Isar der Technischen Universität München, Abteilung Unfallchirurgie und Dr. Daniel Knobeloch von der Klinik für Allgemein-, Viszeral- und Transplantationschirurgie, Berlin für ihren Beitrag und ihre konstruktive Kritik an den jeweiligen Projekten....Heiko Krieger für seine Unterstützung in allen außerfachlichen Belangen und seine Freundschaft....Dr. Susanne Huljic für die vier lustigen und schönen Jahre in unserem gemeinsamen Büro....meiner Familie für ihre Unterstützung und ihren Rat, auf die ich immer zählen konnte....und ganz besonders Sabine für ihren bedingungslosen Beistand zu jeder Zeit, ihre Aufmunterung und Liebe. 166