costs because of the use of universal fluorogenic reporters American Association for Clinical Chemistry

Transcription

1 Clinical Chemistry 58: (2012) Molecular Diagnostics and Genetics Mediator Probe PCR: A Novel Approach for Detection of Real-Time PCR Based on Label-Free Primary Probes and Standardized Secondary Universal Fluorogenic Reporters Bernd Faltin, 1 Simon Wadle, 1 Günter Roth, 1 Roland Zengerle, 1,2,3 and Felix von Stetten 1,3* BACKGROUND: The majority of established techniques for monitoring real-time PCR amplification involve individual target-specific fluorogenic probes. For analysis of numerous different targets the synthesis of these probes contributes to the overall cost during assay development. Sequence-dependent universal detection techniques overcome this drawback but are prone to detection of unspecific amplification products. We developed the mediator probe PCR as a solution to these problems. METHODS: A set of label-free sequence-specific primary probes (mediator probes), each comprising a targetspecific region and a standardized mediator tag, is cleaved upon annealing to its target sequence by the polymerases 5 nuclease activity. Release of a mediator triggers signal generation by cleavage of a complementary fluorogenic reporter probe. RESULTS: Real-time PCR amplification of human papillomavirus 18 (HPV18), Staphylococcus aureus, Escherichia coli, and Homo sapiens DNA dilution series showed exceptional linearity when detected either by novel mediator probes (r ) or state-ofthe-art hydrolysis probes (TaqMan probes) (r ). For amplification of HPV18 DNA the limits of detection were 78.3 and 85.1 copies per 10- L reaction when analyzed with the mediator probe and hydrolysis probe, respectively. Duplex amplification of HPV18 target DNA and internal standard had no effects on back calculation of target copy numbers when quantified with either the mediator probe PCR (r ) or the hydrolysis probe PCR (r ). CONCLUSIONS: The mediator probe PCR has equal performance to hydrolysis probe PCR and has reduced costs because of the use of universal fluorogenic reporters American Association for Clinical Chemistry Monitoring nucleic acid amplification is an indispensable tool in clinical diagnostic areas that include the discrimination of genotypes and accurate quantification of pathogen load in patient specimens (1, 2). In various amplification techniques fluorogenic molecules such as intercalating dyes (3) and modified oligonucleotides enable detection of minute amounts of nucleic acids (4). Although intercalating dyes are cost efficient, they may detect unspecific by-products, leading to false-positive results (5). In contrast, fluorogenic oligonucleotides have the advantage of sequence specificity and the disadvantage of higher synthesis costs. Hence, a universal method for real-time detection of amplification is required that combines sequence specificity with low cost. A number of such sequence-dependent universal detection techniques have been suggested (6 14). Typically, these methods allow flexible assay design with only one single fluorogenic probe for a variety of different assays. Although application of these sequence-dependent universal detection techniques is more cost efficient than the use of sequence-specific fluorogenic probes, these techniques still suffer from major shortcomings. With the use of bipartite primers (6 12) unspecific amplification products as well as primer dimers are still detected. Furthermore, the modification of thermocycling profiles by increase (10) or reduction of temperatures (15) may lead to lowered analytical sensitivity or an increase of unspecific byproducts caused by mispriming and side reactions (7). The analytical specificity can be increased by use of a 1 Laboratory for MEMS Applications, Department of Microsystems Engineering IMTEK, University of Freiburg, Freiburg, Germany, 2 BIOSS Centre for Biological Signaling Studies, University of Freiburg, Freiburg, Germany, 3 HSG- IMIT, Freiburg, Germany. * Address correspondence to this author at: Georges-Koehler-Allee 103, Department of Microsystems Engineering, IMTEK, University of Freiburg, Freiburg, Germany. Fax ; vstetten@imtek.de. Received April 24, 2012; accepted July 27, Previously published online at DOI: /clinchem

2 Mediator Probe PCR probe-based system (14), but the restriction to unfavorable singleplex reactions is in opposition to the demand of multiplexing required in clinical diagnosis (16). To fulfill the requirements stated above we developed the novel technique described here, the Mediator Probe PCR. Materials and Methods The principle of the mediator probe (MP) 4 PCR is illustrated in Fig. 1. PCR amplification of the target DNA is performed with usual oligonucleotide primers and Thermus aquaticus polymerase. Sequence-specific realtime detection is realized by a bifunctional oligonucleotide, the MP that is cleaved upon interaction with the target sequence, and thereafter initiates activation of a second oligonucleotide, the fluorogenic universal reporter (UR). Cleavage and activation is catalyzed by the polymerase. REQUIRED OLIGONUCLEOTIDES The bifunctional MP has a 3 region, designated as probe, which is complementary to the target, whereas its 5 region, designated as mediator, is a generically designed sequence tag that is noncomplementary to any expected target sequence. The UR acts as a self-contained target-independent signaling oligonucleotide. It exhibits a hairpin-shaped secondary structure and contains a fluorophore and a quencher in close proximity on opposite sides of the stem. This arrangement allows an efficient fluorescence resonance energy transfer (FRET) between the attached moieties (17). Its unpaired 3 stem contains the mediator hybridization site, which is complementary to the sequence of the mediator. REACTION SCHEME During the course of the MP PCR, target amplification and detection take place simultaneously in a concerted reaction. In the denaturation step the DNA template (Fig. 1A) is separated in single strands (Fig. 1B). During cooling down to annealing temperature the primers and the MP hybridize to the target DNA. It is noteworthy that the 5 region (i.e., the mediator moiety) of the MP does not hybridize to the target, and this situation results in a flap structure (Fig. 1C). On primer elongation the 5 flap of the MP is threaded into the nuclease domain of the polymerase and is cleaved off (18 21). 4 Nonstandard abbreviations: MP, mediator probe; UR, universal reporter; FRET, fluorescence resonance energy transfer; SISAR, serial invasive signal amplification reaction; HPV18, human papillomavirus 18; T m, melting temperature; E q, efficiency of quenching; LOD, limit of detection; DDQ, 2,3-dichloro-5,6-dicyano- 1,4-benzoquinone; BHQ, di-tert-butylhydroquinone; C q, cycle of quantification. The released fragment is referred to as the mediator and now exhibits a 3 -OH. The 3 region of the MP (i.e., the probe moiety) is digested during extension of the nascent nucleic acid chain (18). With duplication of each target molecule 1 mediator is released to the bulk solution (Fig. 1D). Subsequently, the activated mediator diffuses to the UR and is captured by the mediator hybridization site (Fig. 1E). The polymerase elongates the 3 end of the mediator (Fig. 1F) resulting in fluorescence dequenching. Two pathways for signal generation are proposed. Because of the polymerase s 5 nuclease activity, the 5 terminus of the UR is degraded and the quencher moiety is cleaved off (Fig. 1G). In some cases, the polymerase can destabilize the stem duplex and unfold the hairpin structure without digestion of the 5 terminus (Fig. 1H). Both pathways finally lead to dequenching of the fluorophore due to impeded FRET, and fluorescence emission accumulates with each successive amplification cycle. Both pathways can occur in parallel, because Taq polymerases are known to possess different levels of exonuclease activity (22) and may dissociate hairpin structures with only partial digestion of the 5 terminus (23). Although the reaction scheme is structurally related to the serial invasive signal amplification reaction [SISAR, Invader Squared (13)], the MP PCR benefits from target amplification, which allows for the analytically sensitive detection of the target analyte. Furthermore, SISAR is exclusively based on nucleolytic activity, whereas signal generation in the MP PCR requires both polymerization and nucleolytic activity of Taq polymerase. In contrast to SISAR (13), the hybridization of an uncleaved MP and its UR allows neither elongation nor structure-specific cleavage and prevents unspecific signal generation. Also, misprimed amplification products are not detected because the MP will not hybridize to any of these constructs. This circumvents false-positive results. The MP PCR is capable of detecting duplex PCRs by 2 URs with different fluorophores. In this respect the MP PCR is comparable to state-of-the-art techniques (24 26). SAMPLE MATERIAL The pbr322 plasmid containing the full-length human papillomavirus 18 (HPV18) genome was provided by GenoID (Budapest, Hungary). Staphylococcus aureus DNA samples were obtained from the Genomic Research Laboratory (Prof. Jacques Schrenzel, Geneva, Switzerland) and contained the genomic locus exfoliative toxin B (GenBank accession number AP003088). Escherichia coli K12 DH5 Z1 DNA (27) containing the genomic locus peptidoglycan-associated lipoprotein (GenBank accession X05123) was isolated by use of a magnetic bead based DNA isolation kit (AJ Innu- Clinical Chemistry 58:11 (2012) 1547

3 Fig. 1. Schematic illustration of the MP PCR. Oligonucleotides required for amplification and detection are shown in the box. Amplification and detection are shown in steps (A) to (H). The nucleic acid target (A) is denaturated at increased temperatures (B). (C), Annealing of MP and primer molecules. The 5 portion of the MP does not anneal to the target. (D), Primer elongation and cleavage of MP. With each target duplication 1 mediator is released to the bulk solution. Subsequently, the mediator anneals to the UR (E). Mediator elongation (F) leads to dequenching of the fluorophore induced either by sequential degradation of the 5 terminus and release of the quencher moiety (G) or displacement of the 5 terminus and unfolding of the hairpin (H). Both ways contribute to signal generation. All reaction steps take place within 1 thermocycle. screen). Human genomic DNA was isolated from whole blood with the QIAamp DNA Blood mini kit (Qiagen). For duplex PCR reactions commercially available human DNA (Roche Diagnostics) was used. DNA samples were diluted in 0.2 Tris-EDTA buffer. We added 10 ng/ L salmon sperm DNA (Invitrogen) 1548 Clinical Chemistry 58:11 (2012) to the dilution buffer to prevent unspecific adsorption of the target DNA to the reaction tubes. OLIGONUCLEOTIDES Oligonucleotides used in this work are listed in Table 1. Primer and hydrolysis probes were either ordered ac-

4 Mediator Probe PCR Table 1. List of oligonucleotide sequences. a Modification Target Description Sequence (5 3 ) 5 3 Length, nt Reference E. coli K12 peptidoglycan-associated lipoprotein (pal gene), GenBank accession no. X05123 S. aureus exfoliative toxin B, GenBank accession no. AP HPV18, GenBank accession no. NC_ H. sapiens ACTB, GenBank accession no. AC_ / HGNC:132 Universal reporter 01 b,c CCG CAG* A*A*G ATG AGA TC(dT-FAM) GCG GTG TTG GTC DABCYL C 6 NH 2 67 This work GTA GAG CCC AGA ACG ATT TTT TTT TTT TTT TTT TTT T Universal reporter 02 b,c CCG CAG* A*A*G ATG AGA TC(dT-Cy5) GCG GTG TTC ACT BHQ-2 C 6 NH 2 67 This work GAC CGA ACT GGA GCA TTT TTT TTT TTT TTT TTT TTT T Forward primer GGC AAT TGC GGC ATG TTC TTC C 22 (29) Reverse primer TGT TGC ATT TGC AGA CGA GCC T 22 Hydrolysis probe ATG CGA ACG GCG GCA ACG GCA ACA TGT 6-FAM BHQ-1 27 Mediator probe d AAA TCG TTC TGG GCT CTA CGC GAA CGG CGG CAA CGG CAA CAT GT PH 44 This work Forward primer AGA TGC ACG TAC TGC TGA AAT GAG 24 (28) Reverse primer AAT AAA GTA CGG ATC AAC AGC TAA AC 26 Hydrolysis probe CCG CCT ACT CCT GGA CCA GG 6-FAM BBQ 20 Mediator probe d AAA TCG TTC TGG GCT CTA CGG TAT TCA CAG TGG TAA AGG CGG ACA ACA PH 48 This work Forward primer GCT GGC AGC TCT AGA TTA TTA ACT G 25 GenoID Reverse primer GGT CAG GTA ACT GCA CCC TAA 21 Hydrolysis probe GGT TCC TGC AGG TGG TGG CA 6-FAM BHQ-1 20 Mediator probe d AAA TCG TTC TGG GCT CTA CGG TTC CTG CAG GTG GTG GCA PH 39 This work Forward primer TCA CCC ACA CTG TGC CCA TCT ACG A 25 (30) Reverse primer CAG CGG AAC CGC TCA TTG CCA ATG G 25 Hydrolysis probe 01 ATG CCC TCC CCC ATG CCA TCC TGC GT 6-FAM DDQ-1 26 Hydrolysis probe 02 ATG CCC TCC CCC ATG CCA TCC TGC GT Cy5 DDQ-2 26 Mediator probe 01 d AAA TCG TTC TGG GCT CTA CGC CCT CCC CCA TGC CAT CCT GCG T PH 47 This work Mediator probe 02 d ATG CTC CAG TTC GGT CAG TGC CCT CCC CCA TGC CAT CCT GCG T PH 47 This work a Sequences of universal reporter, primer molecules, hydrolysis probes, and mediator probes. b The self-complementary sequence stretches of the universal reporters are underlined. c The asterisk (*) indicates phosphothioates. d The mediator sequence of the mediator probe is depicted in italic and bold letters; the probe sequence is underlined. Clinical Chemistry 58:11 (2012) 1549

5 cording to previous studies (28 30) or designed in this work for the purpose of demonstrating feasibility of the MP PCR. Oligonucleotides for HPV18 amplification were kindly provided by GenoID (Budapest, Hungary). All modified oligonucleotides were purified by HPLC. DESIGN OF MEDIATOR PROBES The MP design is a 2-step process. The probe (3 region) and the mediator (5 region) region overlap by 1 nucleotide in their 5 and 3 terminus, respectively. Therefore, the 5 terminal nucleotide of the probe must be identical with the 3 terminal nucleotide of the mediator. In our assay, a guanosine nucleotide was required based on the sequence of the mediator. The probe region was designed according to guidelines recommended for the layout of hydrolysis probes [length: nt, probe melting temperature (T m probe ) 5 10 C higher than T m primer ] (31). If applicable, the sequence of validated hydrolysis probes could be used. The mediator was a sequence stretch (length: nt, T m mediator approximately equal to T m primer ) that was designed to exhibit no homology to the intended targets (see Table 1 in the Data Supplement that accompanies the online version of this article at To prevent elongation of the MP the 3 terminus was blocked with a phosphate group. UR DESIGN The UR oligonucleotide (Table 1) was designed in silico (32, 33) to obtain a hairpin-shaped structure with an unpaired single-stranded 3 stem. Secondary structure prediction was performed using RNAfold (32), and T m determination was calculated with the VisOMP (Visual Oligonucleotide Modeling Program) (33). For secondary structure analyses no dangling end energies, DNA settings, and 60 C were applied in the advanced folding options in contrast to default settings. T m of the stem (GC content 71%) is 71.4 C and allows refolding during the cooling step to 60 C within each thermocycle. The folded structure provides the FRET pair in close proximity within each strand of the stem. A FRET pair (Table 1) comprising the 5 terminal quencher and internal fluorophore is selected to achieve a potentially high quenching efficiency. The 3 unpaired stem (45 nt) contained the mediator hybridization site (20 nt), which was reverse complementary to the mediator sequence. To prevent elongation of the UR the 3 terminus was blocked with an amino group. For duplex PCR studies a second UR was designed with an identical sequence except for an altered mediator hybridization site and FRET pair (Table 1). EFFICIENCY OF FLUORESCENCE QUENCHING The selection of appropriate fluorophore dyes and quencher moieties was fundamental for high quenching efficiencies and analytically sensitive detection of minute amounts of nucleic acids (34). To determine the efficiency of quenching (E q ) for each dual-labeled hydrolysis probe and UR molecule the fluorescence emission was acquired with and without DNase I treatment (see online Supplemental Fig. 1). The E q is defined as: E q 1 I undigested /I digested 100, where I undigested is the fluorescence emission of the undigested sample and I digested is the fluorescence emission of DNase I treated samples. MP PCR AND HYDROLYSIS PROBE PCR ASSAYS The MP PCR reaction comprised 1 PCR buffer (GenoID, Budapest, Hungary), 0.1 U/ L HotStarTaq plus polymerase (Qiagen), 200 mol/l deoxyribonucleotides (Qiagen), 300 nmol/l UR (synthesis by IBA), a 300 nmol/l target-specific primer pair and 200 nmol/l MP (synthesis by biomers.net). Hydrolysis probe PCR reactions consisted of the same amount of listed reagents, except the MP was substituted by the hydrolysis probe (200 nmol/l; synthesis by biomers. net), and no UR was added. DNA template was added if appropriate and was compensated in NTC (no template controls) by the same amount of dih 2 O. Reaction volume was 10 L. All real-time PCR reactions were carried out in a Corbett Rotor-Gene 6000 (Corbett Research Pty., now Qiagen GmbH) with a universal thermocycling profile as follows: initial polymerase activation at 95 C for 5 min, followed by 45 cycles comprising denaturation at 95 C for 15 s and a combined annealing and elongation step at 60 C for 45 s if not stated otherwise. Fluorescence signals were acquired at the end of each elongation step. Data analysis was carried out with Rotor-Gene 6000 software (version ). STATISTICAL ANALYSIS The limit of detection (LOD) for HPV18 detection was determined by amplifying various DNA concentrations (10 4,10 3,5 10 2,10 2,5 10 1,10 1,10 0, and 10 1 copies per reaction) and no template controls in 10 replicates each. The fraction of positive amplifications per DNA concentration was determined. Probit analysis using SPSS (Statistical Package for Social Sciences, version 19; IBM) allowed prediction of the copy number per reaction that obtained a positive amplification with 95% probability (35) Clinical Chemistry 58:11 (2012)

6 Mediator Probe PCR Results EFFICIENCY OF FLUORESCENCE QUENCHING Fluorescence emissions of all fluorogenic molecules (Table 1) increased upon disintegration compared to undigested probes. Observed E q values for specific hydrolysis probes range from 54.5% (3.1%) [Cy5/2,3- dichloro-5,6-dicyano-1,4-benzoquinone 2 (DDQ-2)] to 92.7% (0.5%) [FAM/di-tert-butylhydroquinone 1 (BHQ-1)]. Quenching efficiencies for URs were 83.7% (1.4%) (Cy5/BHQ-2) and 90.9% (0.4%) (FAM/Dabcyl) (see online Supplemental Fig. 1). These results agree with the reported E q values for FAM/Dabcyl (80% 91%), FAM/BHQ-1 (88% 93%) and Cy5/ BHQ-2 (91% 96%) obtained under optimized conditions (34). MEDIATOR PROBE PCR VS HYDROLYSIS PROBE PCR In model assays the performance of the MP PCR was compared to the hydrolysis probe PCR. First, reaction efficiency, LOD, interassay variation, intraassay variation, and duplexing capabilities were analyzed. For these experiments, different concentrations of HPV18 DNA (10 2,10 3,10 4,10 5, and 10 6 copies per reaction if not stated otherwise) were amplified by use of both techniques in parallel. Second, different targets were amplified by use of both techniques in parallel. LIMIT OF DETECTION The LOD was determined as the DNA concentration deemed positive with 95% probability. Probit analysis yielded analytical sensitivities of 78.3 copies per reaction (95% CI: copies per reaction) for the MP PCR and 85.1 copies per reaction (95% CI: copies per reaction) for the hydrolysis probe PCR (Fig. 2A). INTRAASSAY IMPRECISION Five concentrations of a HPV18 DNA dilution series (10 2,10 3,10 4,10 5, and 10 6 copies per reaction) were amplified in 8 replicates. r 2 Values of (MP PCR) and (hydrolysis probe PCR) indicated excellent linearity (Fig. 2B). Percentage CVs for amplification of copies per reaction were 55.1% 9.9% (MP PCR) and 38.3% 10.7% (hydrolysis probe PCR). Accuracy ranged from 21.6% to 8.1% (MP PCR) and from 19.4% to 9.8% (hydrolysis probe PCR). Details are presented in online Supplemental Table 2. INTERASSAY IMPRECISION Five individually prepared batches of reaction mixes were used for amplification of 5 concentrations of an HPV18 DNA dilution series (10 2,10 3,10 4,10 5, and 10 6 copies per reaction). Each concentration was amplified in triplicates. Linearity of amplification was demonstrated for the MP PCR (r ) and hydrolysis probe PCR (r ) (Fig. 2C). Interassay imprecision for copy numbers of per reaction ranged from 25.0% to 8.7% (MP PCR) and from 34.7% to12.7% (hydrolysis probe PCR). Accuracy was 3.4% to 7.0% (MP PCR) and 2.0% to 12.4% (hydrolysis probe PCR) for copies per reaction. Details are presented in online Supplemental Table 3. DUPLEX AMPLIFICATION As a model assay a fragment of an HPV18 DNA containing plasmid (10 2,10 3,10 4,10 5, and 10 6 initial copies) was coamplified with 300 copies of the Homo sapiens genome. The individual reactions were carried out in triplicate. The hydrolysis probe for HPV18 was labeled with FAM/BHQ-1 and the probe for actin, beta (ACTB) 5 with Cy5/DDQ-2. For duplex PCR the UR UR01 was labeled with FAM/Dabcyl and UR02 possesses a Cy5/BHQ-2 pair. Fig. 2D shows the linearity of HPV18 amplification over different DNA concentrations for MP PCR (r ) and hydrolysis probe PCR (r ). Back calculation of ACTB was not valid because only 1 concentration was amplified in the duplex assays. However, cycle of quantification (C q ) values were obtained by setting the threshold to 0.02 in the red channel for both MP PCR and hydrolysis probe PCR. Mean C q values for coamplified ACTB and HPV18 DNA samples were 33.0 (0.5) and 31.8 (0.4) for the MP PCR and hydrolysis probe PCR, respectively. APPLICATION OF THE MP PCR AND HYDROLYSIS PROBE PCR TO DIFFERENT TARGETS The universal nature of the MP PCR was demonstrated by use in 4 clinically relevant targets. For comparison, the hydrolysis probe PCR was conducted for each target in parallel. Linearity of input and back-calculated output copy number was determined for each target and amplification technique (Fig. 3). The results for detection of the serial dilution series of the human papilloma virus-18 L1 (HPV18 L1) gene (MP PCR r /hydrolysis probe PCR r ), S. aureus exfoliative toxin B gene (S. aureus ExfB) (0.991/0.988), E. coli peptidoglycan-associated lipoprotein (E. coli pal) gene (0.996/0.988), and the human actin gene (0.991/ 0.993) indicated high agreement between the MP PCR and the established hydrolysis probe PCR (Table 2). 5 Genes: ACTB, actin, beta; HPV18 L1, human papilloma virus-18 L1; S. aureus ExfB, Staphylococcus aureus exfoliative toxin B; E. coli pal, Escherichia coli peptidoglycan-associated lipoprotein. Clinical Chemistry 58:11 (2012) 1551

7 Fig. 2. Comparative characterization of MP PCR and hydrolysis probe PCR. Different concentrations of HPV18 DNA were amplified. No template controls were included in all experiments. Back-calculated copy numbers of the MP PCR (abscissa) are plotted against results of the hydrolysis probe PCR (ordinate) (B D). (A), LOD. The probability for successful amplification (abscissa) of a given input copy number (ordinate) was predicted with Probit analysis for the MP PCR (black) and the hydrolysis probe PCR (gray). Upper and lower bounds represent 95% CI (dashed lines). (B), Intraassay imprecision was calculated for 5 different DNA concentrations with 8 replicates each. (C), Interassay imprecision was determined in 5 individual independent PCR runs per technique, with triplicate amplifications of 5 different DNA concentrations per run. (D) Duplex amplification of various HPV18 DNA concentrations and 300 copies of ACTB. The calculated copy numbers of HPV18 are plotted for the MP PCR (abscissa) and the hydrolysis probe PCR (ordinate). See text for C q values of ACTB. Discussion The striking feature of our assay is the decoupling of amplification and fluorescence detection, which allows the use of standardized fluorogenic UR oligonucleotides. The sequences of the mediator and URs were designed in silico and show no similarity to any target according to the BLAST (Basic Local Alignment Sequence Tool) search (see online Supplemental Table 1). The UR adopts a hairpin-shaped secondary structure, thus providing optimal conditions for efficient FRET quenching [ 90% (FAM/Dabcyl), 80% (Cy5/BHQ-2)]. We redesigned UR01 as follows: 5 -CACGCG*A*A*GATGAGATCGCG(dT-Cy5) GTGTTGGTCGTAGAGCCCAGAACGA-3, where 5 is BHQ-2, 3 is a C3 spacer, and the asterisks represent phosphothioates. The new UR01 has an improved quenching efficiency [mean (SD), 98.87% (0.46%)]. Better initial quenching increases sensitivity and thus 1552 Clinical Chemistry 58:11 (2012)

8 Mediator Probe PCR Fig. 3. Amplification of different targets with the MP PCR and hydrolysis probe PCR. DNA dilution series HPV18 (A), E. coli (B), S. aureus (C), and human beta actin (D) were amplified with the MP PCR and the state-of-the-art hydrolysis probe PCR. For each assay the back-calculated copy values for the MP PCR (abscissa) were plotted against values for the hydrolysis probe PCR (ordinate). improves MP PCR results. The close proximity of fluorophore and quencher within the hairpin structure results in high and constant quenching efficiency. Such strong suppression of the initial background signal is desirable for analytically sensitive target detection in any PCR assay. In contrast to our findings, FAMlabeled state-of-the-art hydrolysis probes have revealed various quenching efficiencies in the range of 60% to 93% due to diverse quenching moieties and deviating FRET distances between fluorescence donor and acceptor. The Cy5/DDQ-2 labeled hydrolysis probe showed a low E q value of 55%. The amplification of HPV18 DNA was selected as a model assay to compare the novel MP PCR to hydrolysis probe PCR, the gold standard for nucleic acid testing. The LOD of both techniques was determined with Probit analysis and was comparable for both methods (MP PCR: 78.3; hydrolysis probe PCR: 85.1 copies per reaction). Inter- and intraassay imprecision were within the same range for 10 2 to 10 6 copies per reaction (MP PCR 25.0% 8.7%, 55.1% 9.9%; hydrolysis probe PCR 34.7% 12.7%, 38.3% 10.7%), indicating reliable quantification over several orders of magnitude. Reducing the elongation time in different PCR assays from 50 to 6 s did not influence the validity for quantification (see Fig. 2 in the online Supplemental Data). These findings suggest that the MP PCR is suitable for the rapid cycling protocols achieved with the latest real-time thermocyclers. Clinical Chemistry 58:11 (2012) 1553

9 Table 2. Overview of calculated copy numbers. a Mediator probe PCR Hydrolysis probe PCR Target Input copy number, n Output, n SD % CV Output, n SD % CV HPV18 L E. coli pal S. aureus ExfB H. sapiens ACTB Coamplification HPV18 L H. sapiens ACTB b C q : C q : a Calculated copy numbers (no. output) of 4 targets amplified with mediator probe PCR and hydrolysis probe PCR. SD and imprecision (CV) were calculated for each target and copy number. b Quantification of copy number is not feasible. The threshold for ACTB was set to 0.02 and obtained C q values are presented. Two URs with different mediator hybridization sequences and FRET modifications were designed. These reporters should be capable of duplex detection of any target-gene combination with high potential for cost savings in routine diagnostics or assay development. Coamplification of various amounts of HPV18 DNA (target) and constant copy numbers of the internal control (ACTB) was successfully demonstrated. The assay was performed with differently labeled hydrolysis probes. Target gene amplification was linear over 5 orders of magnitude (r for both techniques), and even high concentrations did not affect monitoring of the internal control. Furthermore, to demonstrate the broad application of the novel MP PCR, 4 targets were amplified by either the MP PCR or the state-of-the-art hydrolysis probe PCR assay. The target genes of HPV18, S. aureus, E. coli, and H. sapiens were selected. The backcalculated output copy numbers showed high agreement with input copy numbers (MP PCR r ; hydrolysis probe PCR r ). The amplification of these targets was monitored with only one layout of a novel fluorogenic UR throughout multiple assays, whereas individual, cost-intensive, doubly modified hydrolysis probes had to be used for each of the targets. Use of the same fluorogenic UR in all analysis protocols leads to constant initial background fluorescence in all reaction wells. This feature allows fluorescence monitoring of different targets within the same run without under- and overestimation of arising signals as is typically observed for hydrolysis probes with various efficiencies of quenching. For all of the targets analyzed, a universal 2-step thermocycling protocol and consistent reagent concentrations for each 1554 Clinical Chemistry 58:11 (2012)

10 Mediator Probe PCR target were employed, allowing a straightforward assay design and user friendliness. The MP PCR requires only one single UR layout that can be used for real-time detection of virtually any target DNA. Therefore, this reporter can be synthesized in larger batches and at a lower price per unit than it is possible for individual sequence-specific fluorogenic probes, such as commonly used hydrolysis probes. In contrast, in the MP PCR the actual sequence-specific MP is label free and can be synthesized at a lower price than labeled probes, especially if small batch sizes are required. Cost estimation is dependent on individual and regional discounts. As an example, cost assessment of an international supplier revealed $245 per duallabeled hydrolysis probe, $55 per MP, and $600 per UR (catalogue prices for identical synthesis scales). Consequently, a set of 8 individual hydrolysis probes would cost $1960. A set of 1 UR and 8 MPs would be about $1400. This calculation considers a higher order quantity of the UR required for all reactions. We believe that the MP PCR takes an exceptional position in universal sequence-specific nucleic acid detection, overcoming the pitfalls of existing universal nucleic acid testing methods like detection of unspecific amplification products, altered thermocycling conditions, or proprietary reagent chemistry. The MP PCR might have future applications in molecular diagnostics. For example, a set of 2 URs in combination with allele-specific MPs may be involved in highly flexible mutation-detection screenings or broad-range References typing of single nucleotide polymorphisms. The MP concept opens up the scope of flexible assay designs at a reasonable cost and at constant detection conditions. Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article. Authors Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest: Employment or Leadership: None declared. Consultant or Advisory Role: None declared. Stock Ownership: None declared. Honoraria: None declared. Research Funding: EU FP7 project AutoCast (no ) to consortium partner University Freiburg. Expert Testimony: None declared. Patents: B. Faltin, DE ; S. Wadle, DE ; G. Roth, DE ; F. von Stetten, DE Role of Sponsor: No sponsor was declared. Acknowledgments: The authors acknowledge Jacques Schrenzel and Patrice Francois, GBRL Geneva, for providing S. aureus DNA samples. We also thank Csaba Jeney, GenoID, Budapest, for providing PCR buffer, HPV18 DNA samples, and corresponding oligonucleotide sequences. Stefanie Reinbold and Lucas Dreesen are gratefully acknowledged for technical assistance and Mark Karle is thanked for E. coli cultivation and DNA isolation. 1. Kaltenboeck B, Wang CM. Adv in real-time PCR: Application to clinical laboratory diagnostics. Adv Clin Chem 2005;40: Mackay IM, Arden KE, Nitsche A. Real-time PCR in virology. Nucleic Acids Res 2002;30: Gudnason H, Dufva M, Bang DD, Wolff A. Comparison of multiple DNA dyes for real-time PCR: effects of dye concentration and sequence composition on DNA amplification and melting temperature. Nucl Acid Res 2007; Juskowiak B. Nucleic acid-based fluorescent probes and their analytical potential. Anal Bioanal Chem 2011;399: Ririe KM, Rasmussen RP, Wittwer CT. Product differentiation by analysis of DNA melting curves during the polymerase chain reaction. Anal Biochem 1997;245: Li XM, Huang Y, Guan Y, Zhao MP, Li YZ. Universal molecular beacon-based tracer system for real-time polymerase chain reaction. Anal Chem 2006;78: Moser MJ, Marshall DJ, Grenier JK, Kieffer CD, Killeen AA, Ptacin JL, et al. Exploiting the enzymatic recognition of an unnatural base pair to develop a universal genetic analysis system. Clin Chem 2003;49: Nuovo GJ, Hohman RJ, Nardone GA, Nazarenko IA. In situ amplification using universal energy transfer-labeled primers. J Histochem Cytochem 1999;47: Rickert AM, Lehrach H, Sperling S. Multiplexed real-time PCR using universal reporters. Clin Chem 2004;50: Whitcombe D, Brownie J, Gillard HL, McKechnie D, Theaker J, Newton CR, Little S. A homogeneous fluorescence assay for PCR amplicons: its application to real-time, single-tube genotyping. Clin Chem 1998;44: Yang LT, Liang WQ, Jiang LX, Li WQ, Cao W, Wilson ZA, Zhang DB. A novel universal real-time PCR system using the attached universal duplex probes for quantitative analysis of nucleic acids. BMC Mol Biol 2008; Zhang YL, Zhang DB, Li WQ, Chen JQ, Peng YF, Cao W. A novel real-time quantitative PCR method using attached universal template probe. Nucl Acid Res 2003; Hall JG, Eis PS, Law SM, Reynaldo LP, Prudent JR, Marshall DJ, et al. Sensitive detection of DNA polymorphisms by the serial invasive signal amplification reaction. Proc Nat Acad Sci USA 2000; 97: Tani H, Miyata R, Ichikawa K, Morishita S, Kurata S, Nakamura K, et al. Universal quenching probe system: flexible, specific, and cost-effective realtime polymerase chain reaction method. Anal Chem 2009;81: Li XM, Huang Y, Song C, Zhao MP, Li YZ. Several concerns about the primer design in the universal molecular beacon real-time PCR assay and its application in HBV DNA detection. Anal Bioanal Chem 2007;388: Elnifro EM, Ashshi AM, Cooper RJ, Klapper PE. Multiplex PCR: optimization and application in diagnostic virology. Clin Microbiol Rev 2000;13: Didenko VV. DNA probes using fluorescence resonance energy transfer (FRET): designs and applications. Biotechniques 2001;31: , 1118, Holland PM, Abramson RD, Watson R, Gelfand DH. Detection of specific polymerase chain reaction product by utilizing the 5 3 exonuclease activity of Thermus aquaticus DNA polymerase. Proc Natl Acad Sci USA 1991;88: Longley MJ, Bennett SE, Mosbaugh DW. Characterization of the 5 to 3 exonuclease associated with Thermus-Aquaticus DNA-Polymerase. Nucl Acid Res 1990;18: Lyamichev V, Brow MA, Dahlberg JE. Structurespecific endonucleolytic cleavage of nucleic acids by eubacterial DNA polymerases. Science 1993; 260: Lyamichev VI, Brow MAD, Varvel VE, Dahlberg Clinical Chemistry 58:11 (2012) 1555

11 JE. Comparison of the 5 nuclease activities of Taq DNA polymerase and its isolated nuclease domain. Proc Natl Acad Sci USA 1999;96: Kreuzer KA, Bohn A, Lass U, Peters UR, Schmidt CA. Influence of DNA polymerases on quantitative PCR results using TaqMan (TM) probe format in the LightCycler (TM) instrument. Mol Cell Probes 2000;14: Kutyavin IV. New approach to real-time nucleic acids detection: folding polymerase chain reaction amplicons into a secondary structure to improve cleavage of Forster resonance energy transfer probes in 5 -nuclease assays. Nucl Acid Res 2010; Koppel R, Zimmerli F, Breitenmoser A. Heptaplex real-time PCR for the identification and quantification of DNA from beef, pork, chicken, turkey, horse meat, sheep (mutton) and goat. Eur Food Res Technol 2009;230: Lee LG, Livak KJ, Mullah B, Graham RJ, Vinayak RS, Woudenberg TM. Seven-color, homogeneous detection of six PCR products. Biotechniques 1999;27: Richardson JA, Gerowska M, Shelbourne M, French D, Brown T. Six-colour HyBeacon probes for multiplex genetic analysis. Chembiochem 2010;11: Lutz R, Bujard H. Independent and tight regulation of transcriptional units in Escherichia coli via the LacR/O, the TetR/O and AraC/I-1-I-2 regulatory elements. Nucl Acid Res 1997;25: Francois P, Harbarth S, Huyghe A, Renzi G, Bento M, Gervaix A, et al. Methicillin-resistant Staphylococcus aureus, Geneva, Switzerland, Emerg Infect Dis 2008;14: Karle M, Miwa J, Czilwik G, Auwärter V, Roth G, Zengerle R, von Stetten F. Continuous microfluidic DNA extraction using phase-transfer magnetophoresis. Lab Chip 2010;10: Tezak Z, Hoffman EP, Lutz JL, Fedczyna TO, Stephan D, Bremer EG, et al. Gene expression profiling in DQA1*0501( ) children with untreated dermatomyositis: a novel model of pathogenesis. J Immunol 2002;168: Lie YS, Petropoulos CJ. Advances in quantitative PCR technology: 5 nuclease assays. Curr Opin Biotechnol 1998;9: Gruber AR, Lorenz R, Bernhart SH, Neuboock R, Hofacker IL. The Vienna RNA websuite. Nucl Acid Res 2008;36:W SantaLucia J Jr. Physical principles and visual- OMP software for optimal PCR design. Meth Mol Biol 2007;402: Marras SAE, Kramer FR, Tyagi S. Efficiencies of fluorescence resonance energy transfer and contact-mediated quenching in oligonucleotide probes. Nucl Acids Res 2002;30:e Smieja M, Mahony JB, Goldsmith CH, Chong S, Petrich A, Chernesky M. Replicate PCR testing and probit analysis for detection and quantitation of Chlamydia pneumoniae in clinical specimens. J Clin Microbiol 2001;39: Clinical Chemistry 58:11 (2012)

12 Faltin B, Walde S, Roth G, Zengerle R, von Stetten F. Mediator Probe PCR: A novel approach for detection of real-time PCR based on label-free primary probes and standardized secondary universal fluorogenic reporters Efficiency of fluorescence quenching The reaction mixture contains 0.02 U / µl DNase I (Fermentas GmbH, St. Leon-Rot, Germany) in reaction buffer, 200 nm fluorogenic oligonucleotide and dih 2 O to adjust the volume to 50 µl. For negative controls DNase I was replaced by dih 2 O. The mixture was incubated at 37 C for 10 min and distributed in five aliquots. The fluorescence signal was acquired every 15 s at 37 C (repeated 60 times) using the Corbett Rotor-Gene 6000 thermocycler (Corbett Research, Pty, now acquired by Qiagen GmbH, Germany). For each oligonucleotide the values from cycle 20 to 30 were averaged and normalized to the corresponding untreated control. Supplemental Data Figure 1: Efficiency of fluorescence quenching. Specific hydrolysis probes (left panel) and universal reporters (right panel) used in this study. FAM labeled probes are depicted in grey, Cy5 labeled probes in black.

13 Faltin B, Walde S, Roth G, Zengerle R, von Stetten F. Mediator Probe PCR: A novel approach for detection of real-time PCR based on label-free primary probes and standardized secondary universal fluorogenic reporters Influence of elongation time on quantification In different PCR experiments elongation time was reduced from 50 s to 6 s (Supplemental Data Figure 2). It has to be stated that the nominal times include data acquisition which takes 5 s per read-out for all reaction tubes. Therefore, shorter elongations times are not applicable due to technical constraints. As expected cycle of quantification values (C q ) increase with shorter elongation time for all DNA concentrations tested with both techniques (Supplemental Data Figure 2A & B). The reaction efficiencies were 79 % (mediator probe PCR) and 83 % (hydrolysis probe PCR) for 50 s elongation, 87 % and 92 % for 35 s, 84 % and 88 % for 20 s, 90 % and 86 % for 10 s, and both 90 % for 6 s. (Supplemental Data Figure 2C & D). The precision for back-calculated copy numbers amplified with different elongation times was 5.7 % (mediator probe PCR) and 9.0 % (hydrolysis probe PCR) for 10 5 copies per reaction, 8.3 % and 11.8 % for 10 4 copies per reaction, and 6.2 % and 27.6 % for 10 3 copies per reaction. For subsequent experiments 45 s was chosen as elongation time and applied within a universal thermocycling protocol for different targets.

14 A Mediator probe PCR B Hydrolysis probe PCR C D CV = 5.7 % CV = 9.0 % CV = 8.3 % CV = 11.8 % CV = 6.2 % CV = 27.6 % Supplemental Data Figure 2: Influence of elongation time on quantification. A serial dilution of HPV18 DNA (10 3 to 10 5 copies per reaction) was amplified with different elongation time in each experiment. Elongation time was reduced from 50 s to 6 s in individual experiments. Plots of input copy number (abscissa) vs C q value (ordinate) for mediator probe PCR (A) and hydrolysis probe PCR (B). Plots of input copy number (abscissa) vs back calculated copy number (ordinate) for mediator probe PCR (C) and hydrolysis probe PCR (D). The inter-assay imprecision is indicated next to the dots (C and D).

15 Faltin B, Walde S, Roth G, Zengerle R, von Stetten F. Mediator Probe PCR: A novel approach for detection of real-time PCR based on label-free primary probes and standardized secondary universal fluorogenic reporters Alignment of Mediator Probe, Universal Reporter and Hydrolysis Probe sequences Sequence alignments were performed using the Basic Local Alignment Sequence Tool (BLAST ) suite (BLASTN ). The mediator, universal reporter, and hydrolysis probe sequences (query) were checked for hits on the given targets (subject), respectively. The E value was set to 0.01 as significance threshold. Each hydrolysis probe was aligned against its dedicated target (Supplemental Data Table 1). The respective mediator probes gave comparable results as their 3 region consisted of the hydrolysis probe sequence. In contrast, no significant hit was found for the universal reporter sequences. Also alignment of the mediator sequences revealed no significant similarity. Supplemental Data Table 1: Overview of sequence alignment results. Score and E value are given for the sequence alignment of each universal reporter, mediator, mediator probe and hydrolysis probe against all targets. Oligonucleotide Target HPV18 S. aureus H. sapiens E. coli E value Score E value Score E value Score E value Score Universal reporter 01 * a * a * a * a Universal reporter 02 * a * a * a * a Mediator 01 * a * a * a * a Mediator 02 * a * a * a * a HPV18 mediator probe 5.00E * a * a * a HPV18 hydrolysis probe 1.00E * a * a * a S. aureus mediator probe * a 1.00E * a * a S. aureus hydrolysis probe * a 1.00E * a * a H. sapiens mediator probe 01 * a * a 5.00E * a H. sapiens mediator probe 02 * a * a 1.00E * a H. sapiens hydrolysis probe * a * a 1.00E * a E. coli mediator probe * a * a * a 1.00E E. coli hydrolysis probe * a * a * a 4.00E a No significant similarity found (E value threshold: 0.01)

16 Faltin B, Walde S, Roth G, Zengerle R, von Stetten F. Mediator Probe PCR: A novel approach for detection of real-time PCR based on label-free primary probes and standardized secondary universal fluorogenic reporters Supplemental Data Table 2: Intra-assay imprecision. Calculated copy numbers (# output), standard deviation (SD), precision (% CV), and accuracy (%) for different initial HPV18 copy numbers amplified with mediator probe PCR (left panel) and hydrolysis probe PCR (right panel), respectively. Input copy number Mediator probe PCR Hydrolysis probe PCR # # output SD Precision Accuracy # output SD Precision Accuracy (% CV) (%) (% CV) (%)

17 Faltin B, Walde S, Roth G, Zengerle R, von Stetten F. Mediator Probe PCR: A novel approach for detection of real-time PCR based on label-free primary probes and standardized secondary universal fluorogenic reporters Supplemental Data Table 3: Inter-assay imprecision. Calculated copy numbers (# output), standard deviation (SD), precision (% CV), and accuracy (%) for different initial HPV18 copy numbers amplified with mediator probe PCR (left panel) and hydrolysis probe PCR (right panel), respectively. Input copy number Mediator probe PCR Hydrolysis probe PCR # # output SD Precision Accuracy # output SD Precision Accuracy (% CV) (%) (% CV) (%)