DTIC. . rl-lecte " "'ORT DOCUMENTATION PAGE. AUru I11B NWED QAXK c. ADDRESS (Ot.) State. end ZFCod) 7b. ADDRESS (City. State, end ZIP Code)

Transcription

1 -UNCLASSIFIED SE5URITY CL.AS5IPFCArI O T S PAgE " "'ORT DOCUMENTATION PAGE 1b. RT VI MblIC, " T L. " 2. A D-A DISTRIBUTION/ AVALAI i2b OR Approved for public release; distribution unlimited. 4 PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANIZATION REPORT NuMsER(S) SR89-21 Ga. NAME OF PERFORMING ORGANIZATION 6b. OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION Armed Forces Radiobiology l (f applicable) Research Institute AFRRI 6c. ADDRESS (Ot.) State. end ZFCod) 7b. ADDRESS (City. State, end ZIP Code) Bethesda, MD Ba. NAME OF FUNDING /SPONSORING 6b. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER ORGANIZATIONI (f a/pp1llcae) Defense Nuclear Agency DNA 8c. ADDRESS (City, State. and ZIPCode) 10. SOURCE OF FUNDING NUMBERS Washington, DC PROGRAM IPROJECT TAS WORK UNIT ELEMENT NO. NO.. CESSION NO. NWED QAXK TITLE (Icldude Securft CFAI1catlon) (see reprint) 12. PERSONAL AUTHOR(S) Pellmar, T. C., and Neel, K. L. 13a. TYPE OF REPORT 13ROTME COVERED 14. DATE, OF REPORT (Yom, MWodI Day) Si. PAGE COUNT Reprint I FROM TO0 o9 16. SUPPLEMENTARY NOTATION 17 COSATI CODES 15. SUBJECT TERMS (Contirwe on reverse if necessary and 4demny by block number) FIELD I GROUP SUB-GROUP '9 AaSTRACT (Continue on reverse if necessary and identiff by block ntmber) DTIC. rl-lecte AUru I11B 20- DSTRIBUTION I AVAILABIUTY OF ABSTRACT 21. ABSTRACT SECURITY CLASSIFICATION 0 UNCLASSIFIEDOJNUMITtO C3 SAME AS RPT. OTIC USER 22. NAME OF RESPONSIBLE INDIVIDUAL 22b. TELEPHONE (id Area CE Glora R~gieo I(202) ! SDP FIESMO 2Gloria Ruggiero 7 c O FORM MAR 83 APR edition may be used urmi exhausted. SECUR:TY CLASSIFICATION OF T'WS OAGE All other editions are obsolete. UNCLASSIFIED

2 Fr'' Radi(it Riol,, &, Vol 6., pp '. IS-) l I) 81) s ) Printed in the USA All right, rekscr~d NXI I/ I 'crgji, ii Prc ph A PORC1111 RAODOUMV Original Contribution 11CIENTFIC P&It ra SRl89-21 OXIDATIVE DAMAGE IN THE GUINEA PIG HIPPOCAMPAL SLICE TIRRY C. Pnt.N1.iAR' and KATIlRNN L. Nt-t-t- Physiology Department. Armed Forces Radiobioloey Research Intitutc. Bethesda. %I) UIS.A. (Re(eived 3 Ma 198h. Revi.s'd 29 July I885: At'celted 9 1ugu.t I9881 Abstract-Free radicals and active oxygen compounds are implicated in brain ischernia and head tiauma. Pico tus studies have shown that free radicals, generated by radiation and through the Fenton reaction, produce both synaptic and postsynaptic damage in the hippocampal brain slice.' - ' To evaluate the contribution of oxidation to the observed damage. the actions of the oxidants, chloramine-t and N-chlorosuccinimide (NCS). were studied on electrophysiological responses in the hippocampal slice isolated from the brains of guinea pigs. Electrical stimulation of afferents to neurons of the CA I region of hippocampus evoked a population postsynaptic potential (population PSP) in the dendritic layer and a population spike in the cell body layer. Chloramine-T (25-50(0 /im) and NCS ( pim) decreased the population spike in a dose-dependent manner (ED, 125 pm and I I(X) IM. respectively). Input/output curves revealed that both the population spike and the population PSP were significantlycduced with both oxidants. but. the ability of the population PSP to produce a population spike was not impaired. fhese studies suggest that oxidation reactions can account for the synaptic component of the damage produced by free radicals but can rot account for the postsynaptic effects. Keywords-Chloramine-T. N-chlorosuccinimide. Oxidation. Free radical, Hippocampus INTRODUCTION efficacy (synpatic damage) and impair mechanisms of Active oxygen compounds are generated normally in spike generation (postsynaptic damage). The evidence vivo 4-5 ' but the presence of superoxide dismutase, cat- suggests that the molecular mechanisms underlying alase, peroxidase and a variety of antioxidants limit synaptic and postsynaptic damage are different.' In the the concentrations to non-toxic levels. In the event of present study, the oxidants chloramine-t and N-chloan ischemic attack or head trauma, levels of these ac- rosuccinimide were tested to evaluate the contribution tive oxygen species increase. 7 ' however the source of of an oxidation reaction to free radical damage. The these compounds is unclear. Peroxide and superoxide results indicate that the oxidants can account for imcould be generated from xanthine oxidase in local en- pairment of synaptic function but not for postsynaptic dothelial cells'"'; Beckman et al.' calculated concentrations of superoxide and peroxide as high as 70 p M/ min and 170 pm/min, respectively, following an is- MATERIALS AND METHODS chemic episode. Active oxygen compounds could also Hippocampal slices were prepared from the brains be secreted by the microglia invading a region of in- of euthanized male Hartley guinea pigs as previously Jury. ' Another possible source is the generation of free radicals in neurons during reperfusion when a burst of described.' 2 The slices ( pm thick) were in- cubated in artificial cerebrospinal fluid (ACSF) (coinoxidative metabolism results in the release of incom- position in mm: 124 NaCI, 3 KCI. 2.4 CaCl,, 1.3 pletely reduced oxygen (i.e.. superoxide and MgSO 4, 1.24 KHPO 4,. 10 glucose, 26 NaHCOI equiperoxide).' librated with 95% 0,/5% CO 2 ) at room temperature Previous studies have shown that active oxygen spe- for 1-2 hours to allow recovery from dissection. A cies produce functional damage in neurons.' 2 1 Hydro- slice was then transferred to a submerged slice recordgen peroxide and ionizing radiation decrease synaptic ing chamber and continually perfused (1-2 ml/min) with oxygenated ACSF. All experiments were done at *Author to whom correspondence should be addressed. 30' ± I' C. Solutions of chloramine-t (CT) and N- 467

3 468 T. C. Pi i i sir and K. L. Ni i chlorosuccinimide (NCS) (Sigma Chemical Company) 4 A. were prepared fresh daily. Potentials were recorded with a high gain DC am- > 3 plifier and were digitized, stored, and analyzed on an E LSI minicomputer. A bipolar stainless steel,, 2 stimulating electrode (DKI) was positioned in stratum radiatum to activate the Schaffer collateral pathway as A 0 well as other afferents to the CAI pyramidal cells. a- Constant current stimuli ( ma, 200 its) were provided at 0.20 Hz. Field potentials were recorded using OZ glass microelectrodes filled with 2M NaCI and having a resistance of less than 10 Mil. One recording electrode Volley (m) was placed in the cell body layer of CAI region 0.8 B. to record the somatic response (population spike). The population spike is the extracellularly recorded action >; potential occurring synchronously in a population of E 0.6- CAI pyramidal cells. in some experiments, a second a. 0.4 recording electrode was positioned in the stratum ra- 0 diatum to record the dendritic response (population W postsynaptic potential. population PSP) and the afferent volley. The population PSP is the extracellularly 0 2 recorded synaptic potential activated synchronously in the dendrites of CA I pyramidal cells. The afferent vol- Volley (mv) ley is the potential produced by the activation of fibers 4 C. in the stimulated pathway. Following placement of the electrodes, baseline recordings were obtained for a minimum of 30 min to E ensure stability of the tissue. Stimulus intensity was W set to a value that produced approximately a half-max- L. 2 imal response. If during this period the responses W a. changed substantially, the experiment was discarded. 0 a-. A dose of either chloramine-t or NCS was then applied o PSP Slope (mv/msec) 80.. Fig. 2. Input-output curves averaged for 9 slices exposed to loo1 CT IM chloramine-t. Control curves: solid line: chloramine-t curses: ' 60- dotted line. A: Graph shows that chloramine-t reduces the popu- :: lation spike for a given afferent volley amplitude. B: Chloranuine- 40 T decreases the ability of the afferent volley to produce a snaptic response. C: Chloramine-T has no affect on the ability of a PSP to 20 L NCS evoke an action potential. Chloramne-T causes synaptic but no postsynaptic deficits. continuously for thirty minutes. Electrophysiological 0 2C responses were continuously monitored throughout this Dose,Mj period: every 5 min 8 traces were averaged. Drug was Fig. I. Dose response curves for Chloramine-T (CT) and NCS. then washed off and the tissue was perfused with ACSF Varying doses of chloramine-t and NCS were applied to the hip- for another 30 min. To determine the dose-response pocampal brain slice for 30 min. The percent decrease in the pop- relationships, a minimum of four experiments at each ulation spike at 30 min in comparison with control (i.e., before drug) is plotted vs. the dose of drug used. The number of experiments dose of NCS and chloramine-t were performed. The with Chloramine-T at 25 pm = 6: 50 /im = 7: 100 IN = 6: 250 effectiveness of the drug was expressed as the per- IM = 7; 500 IM = 5:1000 /M = 3. The number of experiments centage decrease in amplitude at the 30 min time point with NCS at 625 PM = 5: 750 IM = 4: 875 pim = 6: 1000 IM = 7; 1250 pm = 1: 1500 #M = S. Squares: as compared to control amplitude. The 30 min time Chloramine-T, Circles: NCS. point was chosen for 2 reasons: I) The drug effects

4 sdani t in hippoc i Ipus 40) frequently appeared to level off within this time period CAI in a dose dependent manner (Fig. I ). Li doses and 2) this protocol allowed comparison with previous of chloramine-t (25 itm) produced little or no effect experiments on hydrogen peroxide using it 30 nin cx- during or after exposure to the drug. Intermediate doses posure. All experiments at a single dose were averaged (0) /pm) noticeably decreased the amplitude of and standard errors (SEM) were calculated. the population spike during the 30 minute exposure to Input-output (I/0) curves were constructed for each choramine-t. The response recovered approximately drug at a dose that produced approximately a 4(K/ to its original size within a 30 min period of wash with decrease in population spike amplitude. Somatic and normal ACSF. When exposed to higher doses of chlordendritic traces (it = 4) were averaged at each stimulus amine-t (250-I jm). the amplitude of the popuintensity ranging from 0. 1 to 1.0 ma. The data from lation spike decreased sharply during exposure and 5 experiments for NCS and for 8 experiments for chlor- remained depressed throughout the wash. The dose amine-t were averaged to obtain composite curves, response curve (Fig. I1 shows that the concentration The 1/0 curves consisted ofthree relationships: I) vo,- for a half maximal effect is 125 lim. Chloramine-T is Icy vs. population spike 2) volley vs. population PSP. maximall cffc,'tive at doses greater than about 500 and 35 population PSP vs. population spike. The volley ItM. vs. population spike curve reflects the ability of the The dose response curve for N-chlorosuccinimide 0ics"Yiaptic activity to elicit an action11 polfltda! :) the (NCS) (Fig. I) shows that NCS icquiicd highci,,- postsynaptic cell. The population spike vs. population centrations than chloramine-t to produce comparable PSP reflects primarily postsynaptic mechanisms (i.e.. damal'e. The onset and reversal of NCS action was the effectiveness of a postsynaptic depolarization to very similar to, although slightly more gradual than. evoke an action potential). The volley vs. population that of chloramine-t The dose of NCS to produce a PSP reflects the synaptic mechanisms (i.e.. the ability half maximal effect was approximately I. I mam. Doses of presynaptic activity to produce a synaptic potential). of 2 mm and greater caused a maximal decrease in the Thus. analysis of these curves provides information on population spike amplitude. the mechanisms of oxidative damage. The input/output Input-output (I/O) experiments were done to deterdata were computer-fitted with the equation for a sic- mine the site of damae of the oxidaints. The dose of moid curve and analyzed as previously described.' Dif- chloramine-t chosen for these experiments ( 100 p M) ferences between control and treated curves were tested was expected from the dose response curve to produce. for significance by comparing the residual sum of on the average, a 4 01/ decrease in the population spike. squares for the individual curves with the residual sum The I/0 curves (a = 8) show that for a given afferent of squares for a curve fit to a composite of control and volley input, both the population spike (Fig. 2A) and test data. Significance was accepted at p < the popultion PSP (Fig. 2B) were significantly reduced by chloramine-t. The ability of the population PSP to REStULTS produce a population spike, however, was not altered (Fig. 2C. The population PSP was equally effective Chloramine-T decreased the amplitude of the pop- in evoking a spike before and during exposure to the ulation spike elicited by orthodromic stimulation of oxidant. Fig. 3 shows sample traces from a typical slice Control hloramine-t Superimposed Somatic A. Dendritic X B. Fig. 3. Somatic ltop traces) and dendritic responses (bottom traces) from slice treated A ith 111 NI chloramitc-t-. (onlrotl (left) and treated (middle) traces are superimposed on the right. A: At the same stimulus sirength. Chloraininc-T causes a decrease in both the population spike and the population PSP. B: When the stimull' strength in chloramine-t "kas increased to produce the same dendritic response as control. the evoked population spike %as the sanle is control. Calibratiot (sqtrc aave on lower left): I mv isomatic). 0.5 mv (dendriticlj 2 nns.

5 470 T. C' P1il RIM5k and K. I_ Nil H Control Chloramine-T Superimposed Fig. 4. The antidrornic spike was evoked by stlntulation ol the alves and recorded in tc cell body laser of CAI. It ssas unaffected by 250 tim chloramine-t (center trace) and can he seen best when the traces are uperimlposed (right traces). Calibration (,square wave on lower left t: I mv, 2 Ins. treated with 100 pm chloramine-t. Both population 3 A. spike and population PSP were reduced. If. while still in chloramine-t, stimulus strength was increased to > produce a population PSP comparable to control, the 2 evoked population spike is comparable in amplitude Q) (Fig. 3B). These data suggest that the observed de- crease in the population spike with exposure to chlor- C amine-t is due completely to the reduction in the a_... synaptic response. Previous reports ' 4 indicate that chloramine T at similar doses removes sodium inactivation, broad- Volley (mv) ening the sodium action potential in squid, crayfish, frog and toad axons. To evaluate the actions of chlor- Z B. amine-t directly on the sodium action potential in this preparation. the effect on the antidromically elicited E 0.6 population spike was evaluated. Axons of the CAI > pyramidal cells were stimulated with a bipolar stainless 0.4..,D steel electrode in the alveus. The resulting antidromic Qo,... spike was recorded from the cell body layer of CA I. (n At a dose that maximally reduced the synaptically- 0.2 activated (orthodromic) population spike (250 ptm), 0. chloramine-t had no effect on the antidromic action potential (Fig. 4) (n = 4). Volley (mv) Input/output experiments were repeated with NCS V m. at a dose of 1.0 mm (n = 5). As with chloramine-t. 3 C. the population spike (Fig. 5A) and the population PSP (Fig. 5B) evoked by equivalent afferent volleys were E both reduced. Yet population PSPs of equivalent size. 2 before and during exposure to NCS. produced equiv- CL alent population spikes (Fig. 5C. NCS reduced the a. population PSP for the same size afferent volley, but 0 the effectiveness of that PSP was unaltered. As with chloramine-t, these results demonstrate that only syn- 0". aptic damage and no postsynaptic deficits were pro duced by NCS. PSP Slope (mv/msec) r DISCUSSION Fig. 5. Input-output curves averaged for 8 slices exposed to I rnm NCS. The results are identical to those obtained for chlorarnine-t. Control curves: solid line: NCS curves: dotted line. A. For a given afferent volley, the population spike is decreased. B: For a given atferent volley, the synaptic response is decreased. C: For a given synaptic response. the evoked population spike is the same both in control and in NCS. NCS causes only synaptic damage and no post- synaptic damage. The oxidants, chloramine-t and N-chlorosuccinimide. were observed to cause a decrease in the orthodromic potentials in the CA I region of the hippocampus in vitro. While the synaptic potentials were reduced

6 Oxidants in hippocampu, 471 in size, they were not impaired in their ability to evoke idants, the contribution of an oxidative reaction to the action potentials in the postsynaptic neurons. The re- observed damage was considered. duced synaptic efficacy occurred at doses that did not Chloramine-T and NCS are oxidizing agents that alter the antidromically-evoked spike in the same neu- appear to be fairly specific for methionine and cysteine ronal population. residues of membrane proteins." These actions are Chloramine-T and NCS have been reported to re- similar to the oxidizing effects of hydrogen peroxide. ', move inactivation of the sodium current in the giant The oxidants in the present study were able to produce axon of squid 4 '" and crayfish" and in the myelinated damage in the hippocampal brain slice, but they could axons of the toad' 2 " and frog. 5 Although many of the not completely mimic the effects of radiation and perstudies used chloramine-t in higher doses (I to 10 oxide. This provides further support for the hypothesis mm), Wang' 2 reported that in toad fibers, a dose as that two separate mechanisms underlie the synaptic and low as 89 pm was effective in broadening the action the postsynaptic damage produced by radiation and by potential 3-fold after only 7 min. Longer exposures peroxide. The present study showed that the oxidants further prolonged the action potential and began to specifically impaired synaptic efficacy. suggesting that reduce the amplitude. In contrast, in the present study an oxidation reaction could account for the peroxidain the hippocampal slice preparation 250 itm chlora- tive and radiation damage at this site. Yet. the oxidants mine-t for 30 min did not alter the antidromic spike had no effect on postsynaptic generation of action powhich is predominantly a sodium-dependent action po- tentials. The limited dose rate dependence of radiation tential. One major difference between the present ex- damage' and the iron potentiation of pi',side periments and those testing chloramine-t in axons is damage' 2 are also more consistent with a lipid perthat in the present study, potassium currents were not oxidation mechanism than with an oxidation reaction. pharmacologically blocked. The presence of the po- In conclusion, the observations reported here sugtassium component of the action potential could oh- gest that oxidative damage may account for the synscure an action of the oxidants to broaden the action aptic component of free radical damage in the nervous potential through removal of soditm inactivation. In system but can not account for the postsynaptic two studies'"'' chloramine-t (1-14 mm) was also deficits. found to decrease squd the aons deresedsodum potassium Nethe current nacivaionor in toad and Atknouh'cdngemen-We iha uk tdr. P. (Guntcr-Smith for hclptul rn squid axons. Neither decreased sodium inactivation or ments on the manuscript. Supported by the Armed Forces Radiodecreased potassium current can account for the de- biology Research Institute. Delnse Nuclear Agency. under work crease in synaptic efficacy seen in the present study. unit Views presented in this paper are those of the authors: If anything. one would predict that both mechanisms 'no should endorsement by the Defense Nuclear Agency has been given or be inferred. Research 'as conducted according to the prinwould increase the duration of the action potential, ciples enunciated in the 'Guide lfr the Care and ULe of Laboratory increase the presynaptic calcium influx and enhance Animals' prepared by the Institute of Laboratory Animal Resources. transmitter release. National Research Council. Previous studies' 23 showed that ionizing radiation REFERENCES and hydrogen peroxide could produce damage in an isolated neuronal system. In an aqueous environment I. Pellmar. T. Electrophysiological correlates of peroxide damage ionizing radiation produces free radicals including.oh in guinea pig tippocampus in vitro. Braii Res. 346: : and O, -. Hydrogen peroxide can react with iron and 2. Pellmar. T. Peroxide alters neuronal excitability in CAI recion through the Fenton reaction produce hydroxyl free rad- of guinea pig hippocampus in vitro. Neuro.wi. 23: icals. Both procedures decreased synaptic responses Tolliver. J. M.: Pellmar. T. C. Effects of ionizing radiation on produced by orthodromic stimulation of stratum radia- hippocampal excitability. Rodiation Res. 112: : tum suggesting either a decrement in presynaptic re- 4. Sinet, P. M.: Heikkela. R. E.: Cohen. G. Hydrogen peroxide lease or impaired function of the postsynaptic receptor/ production by rat brain in %ivo. J.,euroclhm. 34: , ionophore complex (synaptic damage). In addition, the 5. Fridovich. I. Superoxide radical: an endogenous toxicant. Ann. synaptic potentials that were elicited were less capable Rei. Pharnacol. Toii',l. 23: : of evoking an action potential in the hippocampal neu- 6. Halliwell. B.: Gutteridge. J. M. C. Oxygen radicals and the nervous system. Trends it,tosii. 8:22-26: rons suggesting that membrane properties of the soma 7. Dcmopoulas. H. B.: Flamm. E.: Selignian. M.: Pietronigro. D. or axon hillock were altered in some way (postsynaptic D. Oxygen free radicals in central nervous systen ischenia and damage). Two separate mechanisms were postulated trauma. In: Autor. A. P.. ed. I'otho/og\ of oi\ioepi. New, York: Academic Press: 1982: for these two actions since they showed very different S. Kogure. K.: Arai, H.: Abe. K.: Nakano. M, Free radical damage dependencies on the dose rate of radiation.' Since free of the brain following ischemia. Prie. Brain Res. 63: : radicals produced by ionizing radiation (such as the Beckman. J. S.. Liu. T. H.. Hogan, E. L.. Freeman, B. A. and hydroxyl free radical) and hydrogen peroxide are ox- Hsu. C. Y. Oxygen free radicals and xanthine oxidase in cerebral

7 C. Pit i lmk and K... NiII ischemnic injurv in the rat. Soc. Nemro.'.A.sr. 13:1498 ahtr. 14. Wan. (;. K.: Bnrd\% ick. M. S I and hatiin. 1) C Rclmoal of ( sodiurn channel inactivation in squid axoin ilh the omidant chlor- 10. Rr sen. G. M.: Freenan. B. A )etection oji,upcroidc gen- arinc-t. J. Gei. PhxNiol. 86:28) 312: crated by endotheli:i ceclls. Pron. Nad. Acad. S(i USI. 15. Drc%'. 6. Ef'ects,f* cl loramnic-t i.n chare, mo cie.inn anrd 81: : tract,n t npen channels, it tr,i miles of Ran icr. I'Iu n,'nr\ I I. Colton, C. A.: Gilbert. D. 1.. Production of superoxide anion, Arch 409: : by a CNS macrophagc. the microglia. PIEBS 223: Huang. J. M C. Tantu\. J.: Ych. J. Z. Remo\al (it sodium inactivation and hlck of sodiumri channels h chlnramme-t in 12. Wan, G. K. Irreversible mioditication ot sodium channel in- cra) fish and squid axin,. lli, hn.n. J. 52: : 197. activation in toad mvelinated nerve fibres,. h\ the nixidant clilnr- 17. Means. (i. V.: FcnN. R. L. Chermical nmiodification (1 proteins. aminc-t. J. Plrvsiol. 346: : San :rancisco:holdcn-i)a.: Wan'. G. K. Modification (of sodium channel inaclivation ii I. Schechter. Y.: Burlstein. Y.: and Patchomnik. A. A sclcctic single mylinated nerve fibers b methionine-reactive chcmi- oxidation (it net hioilic rcsidue, in protciri. Bio, iin cals. Bio qiin J. 46: : : AeOssior. For NTIS GRA&I DTIC TAB Unannounced Justifleatlon Availability Codes java i and/or Dist 3pecial '', 00II lii III