Original Paper Neuroendocrinology 1992 ;56:719-723

John R. Glowa “ Mark A. Geyerh Phil W. Gold" Esther M. Sternberg"

Differential Startle Amplitude and Corticosterone Response in Rats

Clinical Neuroendocrinology Branch, NIMH, Bethesda, Md., Department of Psychiatry, University of California at San Diego, Calif., USA

Abstract

Different rat strains exhibit large differences in hypothalamic pituitary-adre­ nal activity that have been used to determine the role of the neuroendocrine system in susceptibility to autoimmune disease. To further characterize poten­ tial behavioral correlates of these differences, the amplitude of the acoustic (ASR) and tactile (TSR) startle response and the corticosterone response to acoustic startle stimuli were compared between two histocompatible strains, Lewis (LEW/N) and Fischer (F344/N) rats, as well as outbred Harlan Sprague-Dawley (SD) rats. Startle stimuli elicited larger ASR and TSR in LEW/N rats than in F344/N rats, with SD rats exhibiting an intermediate response. The ASR habituated at a similar rate in LEW/N and F344 rats, while the ASR did not habituate in SD rats. After handling and placement in the startle chambers, the three strains did not differ in control levels of cortico­ sterone. In contrast, exposure to acoustic startle stimuli increased corticoste­ rone 5-fold in F344/N rats and 2-fold in SD rats, but had no effect on cortico­ sterone in LEW/N rats. These findings suggested an inverse relationship between the amplitude of the ASR and hypothalamic-pituitary-adrenal activa­ tion across strains. This relationship was further supported by a high negative correlation between corticosterone level and ASR amplitude within the F344/N group.

Lewis (LEW/N) rats serve as an experimental model for susceptibility to inflammatory disease [ 1]. In response to a single injection of streptococcal cell wall peptidoglycan polysaccharide (SCW), LEW/N rats develop inflam­ matory arthritis which mimics human rheumatoid arth­ ritis. In contrast, the histocompatible Fischer (F344/N) strain is resistant to SCW-induced inflammation [2], This

Received: August 26. 1991 Accepted after revision: April 3. 1992

susceptibility to inflammatory disease in LEW/N rats and resistance in F344/N rats has been related to differ­ ences in hypothalamic-pituitary-adrenal (HPA) axis re­ sponsiveness in these two strains [3]. Compared to F344/N rats, LEW/N rats exhibit profoundly depressed corticotropin-releasing hormone (CRH), adrenocorticotropin (ACTH) and corticosterone responses to SCW as

John R. Glowa. PhD CNF./NIM H Building I4D. Room 3 11 Bethesda, MD 20892 (USA)

Downloaded by: University Toronto Libr. 142.150.190.39 - 10/11/2017 4:25:58 PM

Key Words Startle Stress Strain-dependent differences

Methods Procedure

Sixty female rats. 20 each of the F344/N. LF.W/N and Harlan Sprague-Dawley (SD) strains, weighing approximately 100 g, were individually housed, and maintained under a reversed 12-hour lightidark cycle (lights off at 07:00 h) with ad libitum access to food and water. One week after arrival rats were tested in groups of 4, with the strains equally distributed over testing runs. Testing oc­ curred between 09:00 and 13:00 h. As detailed elsewhere [10], rats were placed individually in a startle chamber with a 70-dB [A] back­ ground noise for a 5-min acclimatization period. The chambers (SRLAB, San Diego Instruments, Inc., San Diego, Calif., USA) consisted of a Plexiglas cylinder mounted on a Plexiglas base and placed within a sound-attenuating chamber. A piezoelectric accele­ rometer attached to the Plexiglas base transduced the startle re­ sponse. Stabilimeter readings were rectified, digitized on a 0-4095 scale, and recorded by the SRLAB system, with 100 1-ms readings collected starting at stimulus onset. The average of these 100 read­ ings was used as the measure of startle amplitude. Acoustic noise bursts (115 dB [A], 30 ms) were generated by the SRLAB and presented by a loudspeaker mounted 24 cm above the rat. Tactile stimuli were 30-ms air puffs controlled by the microcomputer. Air was maintained at 25 lb/in2 and delivered through a 8-mm tube directed at the rat’s back. On selected trials, 80 dB [A], 30 ms pre­ pulse stimuli were presented 100 ms before the acoustic startle stimuli.

720

On the first day of testing, 60 rats were tested in a session con­ sisting of three successive components: 7 acoustic stimuli; 7 tactile stimuli, and 14-trial component in which the acoustic stimuli were presented with and without the prepulse stimuli, to assess prepulse inhibition [cf. 10]. Individual trials were separated by a variable in­ terval averaging 15 s; components were separated by a 1-min pause. On the second day of testing (2 day's after the first session), 30 of the same 60 animals were exposed to a session in which the 5-min accli­ matization period was followed by 6 blocks of 10 acoustic trials ( 15 min), to assess potential differences in the habituation of acous­ tic startle. The other 30 rats were exposed to the startle chambers and the background noise for 20 min, but without startle stimuli (sham control). Following each of these sessions, the animals were removed from the chambers, moved to an adjacent room, and rap­ idly decapitated. Trunk blood was collected on ice, centrifuged at 1500 rpm, at 4 °C, and the serum frozen on dry ice for subsequent analysis. Biochemical Analysis

Serum levels of corticosterone were assessed by the use of a stan­ dard radioimmunoassay kit (ICN Biomedicals, Inc., Costa Mesa, Calif., USA). Detection limits were approximately 25 ng/ml. Interand intra-assay control variabilities were 2.4 and 3.7%, respectively. Statistical Analysis

The startle data for each component of the first session was ana­ lyzed using analysis of variance (ANOVA, BMDP) with a betweengroups design [5], Prepulse inhibition was assessed as the percent reduction in startle amplitude produced by the prepulse, as detailed elsewhere [10], Reactivity and habituation differences in the second session were assessed using a mixed-design ANOVA, with trial blocks as a repeated measure. Corticosterone levels were also ana­ lyzed using analysis of variance (ANOVA, BMDP) with a betweengroups design [II], Post-hoc comparisons were done with Tukey’s studentized range method. Spearman’s correlation coefficient was used to assess the relationship between corticosterone level and startle amplitude.

Results

Figure I shows the mean amplitude of the startle re­ sponses elicited by acoustic or tactile stimuli in the three strains of rats. The ANOVA revealed a significant group effect for both acoustic (F[2,56| = 14.55, p < 0.0001) and tactile stimuli (F[2,56] = 10.22, p < 0.0002). With both stimulus modalities, LEW/N rats responded with consistently greater startle amplitudes than did the F344/N rats. The SD rats responded at an intermediate level that was significantly above the F344/N and not sig­ nificantly different from the LEW/N rats. All three groups exhibited greater than 50% prepulse inhibition (data not shown) and did not differ on this measure (F12,56] = 0.28, n.s.). The second experiment assessed the degree and com­ parability of habituation to ASR across strains. Figure 2

Glowa/Geyer/Gold/Sternberg

Strain Differences in Startle Responses

Downloaded by: University Toronto Libr. 142.150.190.39 - 10/11/2017 4:25:58 PM

well as to a variety of inflammatory, neurotransmitter and other environmentally relevant stimuli [3,4]. As glucocor­ ticoids have potent anti-inflammatory and immunosup­ pressive effects, the differential inflammatoiy response in these strains may be a direct consequence of strain differ­ ences in hypothalamic CRH responsiveness [2]. In addition to their differences in inflammatory and HPA axis response, behavioral differences have also been observed between LEW/N and F344/N strains. Expo­ sure to an open field or to swim stress resulted in changes in behavior which were consistent with a relative hypo- or hyperresponsiveness of the HPA axis to these stressful situations [3, 5]. The startle response is an easily quanti­ fied reflex contraction of the skeletal musculature in re­ sponse to intense acoustic (ASR) or tactile (TSR) stimula­ tion [6] that is known to be altered in stressful situations [7], Previous studies have shown that the central adminis­ tration of CRH can potentiate the ASR in rats [8, 9], sup­ porting the possibility that differences in endogenous lev­ els of CRH or sensitivity to the consequences of its release may predict differences in the magnitude of the ASR. The present study was designed to assess potential differences in the ASR in LEW/N and F344/N rats. Since SpragueDawley rats are intermediate in their response to arthritic challenge [3], their response to startle was also studied.

Fig. 3. Mean corticosterone (± SEM) in F 344/N, SIX and LEW/N strains for sham control and acoustic startle-exposed rats.

Fig. 2. The mean magnitude of the startle response in F344/N ( • ) , SD (A ) and LEW/N (A ) strains as a function of repeated exposures to acoustic startle stimuli. Abscissa 6 blocks of 10 trials each. Groups consisted of 9 or 10 animals each.

illustrates the group means in response to 6 successive blocks of 10 acoustic startle stimuli. The strain-related dif­ ferences in this experiment confirmed the difference seen earlier between the LEW/N and F344/N rats. The mixed-design ANOVA revealed a significant group effect

(F[2,28] = 9.50 p < 0.0005). Subsequent Tukey’s com­ parisons confirmed that the LEW/N rats exhibited larger responses than the F344/N rats during each and every block of trials. Responses by the SD rats were signifi­ cantly lower than the LEW rats during the first 2 blocks and significantly higher than the F344/N rats during the last 4 blocks. The ANOVA also revealed a significant group-by-trials interaction (F[ 10,140] = 43. 3, p < 0.0001), suggesting that the strains exhibited different rates of startle habituation. Subsequent repeatedmeasures ANOVAs on each strain confirmed that signifi­ cant habituation occurred in the LEW/N rats (F[5,50] = 6.05, p < 0.002) and the F344/N rats (F[5,45] = 10.46, p < 0.0002), but not in the SD rats (F[5,45] = 2.12, n.s.). As shown in figure 2, the magnitude of the ASR de­ creased across trial blocks in both the LEW/N and F344/N rats. In contrast, the ASR did not habituate in the SD rats. Thus, while levels of response in the SD rat were near those of the F344/N rat at the beginning of the session, they were nearly identical to those of the LEW/N rats at the end of the session. Corticosterone levels at the end of these sessions were elevated in comparison to sham-exposed animals (fig. 3). The ANOVA revealed a significant main effect of the startle experience on corticosterone (F]l,53] = 27.67, p < 0.0001), a significant strain effect (F[2,53] = 17.82), and a significant exposure-by-strain interaction (F[2,53] = 16.66, p < 0.0001). Tukey’s comparisons indicated that virtually all these effects were attributable to the se­ lective increase in corticosterone induced by the startle experience only in the F344/N animals. That is, F344/N rats exhibited an approximately 5-fold increase in levels of corticosterone as a function of exposure to the acoustic startle sessions, even though they habituated to the stimu­ lus in terms of startle amplitude.

721

Downloaded by: University Toronto Libr. 142.150.190.39 - 10/11/2017 4:25:58 PM

Fig. 1. The mean magnitude of the startle response F344/N, SD, and LEW/N strains as a function of type of startle. The mean (± SEM) responses elicited by the 7 acoustic stimuli (left bars) or the 7 tactile stimuli (right bars) during the first test session are shown. Each group consisted of 19 od 20 animals.

Discussion

The present study showed that LEW/N and F344/N rats differ in their response to acoustic as well as tactile startle stimuli. These differences were replicated and ex­ tended by showing that the initial differences between these strains were maintained throughout the habituation to startling stimuli. By contrast, SD rats differed from the other two strains in startle habituation. Furthermore, while corticosterone increased 5-fold in F344/N rats in response to startle stimuli, it only doubled in SD rats and did not change in LEW/N rats. Thus, the present results are consistent with previous studies demonstrating differ­ ences in HPA activity across these strains [2-4, 12]. Previous studies have shown that LEW/N rats exhibit a blunted HPA response, the SD rat an intermediate re­ sponse, and the F344/N an exaggerated response, to a variety of inflammatory [1, 2], pharmacological [4, 12], and environmental challenges [3]. The present study con­ firms and extends these findings to another category of events, acoustic and tactile startle stimuli, and shows that, in addition to differential corticosterone responses to these stimuli, these strains also exhibit different behav­ ioral responses to environmental challenges. The differ­ ence in behavioral response suggests that central mecha­ nisms may be involved in these differences in enroendocrine response. In support of this possibility, a recent study showed that intracerebroventricular administration of CRH produces a significant behavioral effect in F344/N rats, but not in LEW/N rats, in the open field [5], Since CRH has been shown previously to potentiate the ASR, it is paradoxical that a strain reported to have diminished hypothalamic CRH activity (LEW/N) ap­ pears more responsive to startle stimuli than a strain re­ ported to have enhanced CRH activity (F344/N). How­ ever, the correlation between levels of corticosterone and ASR amplitude found for F344/N rats further supported this observation. One possibility for this discrepancy is that corticoste­ rone levels themselves may influence the startle response. Corticosterone has previously been shown to have behav­

722

ioral effects analogous to anxiolytic drugs [ 13], and classi­ cal anxiolytics attenuate the startle response [8], Similar behavioral effects have been seen with another neuroste­ roid, tetrahydrodeoxycorticosterone [14], This steroid is increased to a greater extent in F344/N rats that in LEW/N rats in response to stressors [Purdy, pers. commun.]. Previous studies have also shown that central lim­ bic structures thought to play an important role in startle [15], have an abundance of corticosterone receptors [16]. However, the difference in startle response was apparent from the beginning of the session, when corticosterone levels would not yet be elevated. These strain-dependent differences in startle response are, therefore, more likely to be associated with events occurring before corticoste­ rone release, further supporting a potential role for CRH. The present results are also consistent with a study that compared startle amplitudes in SD and F344 rats of dif­ ferent ages [17]. F344 rats consistently exhibited startle amplitudes lower than those of SD rats. On the other hand, SD females which were intended to serve as the ‘normal' controls, did not habituate in the present study. Since this earlier study also observed habituation in male SD rats, current findings may reflect a gender-dependent effect in habituation. The same study [17] however failed to observe overall differences in habituation in startle be­ tween male and female F344 rats. Previous studies have noted other strain-dependent differences in startle response. For example, apomorphine blocked the prepulse inhibition of the ASR without affecting startle amplitude in male Wistar rats while apomorphine had no effect on prepulse inhibition but in­ creased startle amplitude in SD rats [18]. In mice, adre­ nalectomy differentially increased the magnitude of the ASR in response to nicotine in a strain-dependent man­ ner [19], These differences in startle response to CNS-active drugs further support the central mediation of straindependent differences in startle response. However, the increased response in adrenalectomized mice suggests that this mediation may be a result of adrenal feedback mechanisms. The current study provides novel data on HPA func­ tion during the assessment of the startle response. Al­ though the startle is a reflexive response to intense stimuli, the present results suggest that these stimuli need not be stressors. While corticosterone levels were clearly elevated in F344/N rats exposed to startle stimuli, they were only doubled in SD rats and not increased at all in LEW/N rats. Since the increase in corticosterone levels in F344/N rats is a likely consequence of the unique suscep­ tibility of that strain to a wide range of environmental

Glowa/Geyer/Gold/Sternberg

Strain Differences in Startle Responses

Downloaded by: University Toronto Libr. 142.150.190.39 - 10/11/2017 4:25:58 PM

Within the group of F344/N rats, 1 animal was an ex­ treme outlier and obscured an apparent negative correla­ tion between level of corticosterone and ASR amplitude. With this animal excluded, the rank-order correlation be­ tween level of corticosterone and ASR amplitude was -0.700 (d.f. = 8, p < 0.05). No significant correlations were evident in the other two groups.

stimuli, the magnitude of the corticosterone response in the other strains suggests that the startle stimuli were minimally stressful. This observation suggests that the ASR may represent a benign, well-defined paradigm suit­ able for the study of genetic and central components of the neuroendocrine response to behaviorally relevant events.

Acknowledgments This work was supported in part by a grant front the National Institute on Drug Abuse (DA02925). M.A. Geyer was supported by a Research Scientist Development Award from the National In­ stitute of Mental Health (MHOOI88). The authors sincerely thank Ms. Valli Keith and Ms. Virginia Masten for their expert assistance in the conduct of these studies.

References 6 Davis M, Gendelman DS, Tischler MD, Gendelman PM: A primary acoustic startle circuit: Lesion and stimulation studies. J Neurosci 1982;6:791-805. 7 Brown J. Kalish H. Farber I: Conditioned fear as revealed by magnitude of startle response to an auditory stimulus. J Exp Psychol 1951; 41:317-327. 8 Swerdlow GR, Geyer MA, Vale WW, Koob GF: Corticotropin-releasing factor poten­ tiates acoustic startle in rats: Blockade by chloridiazepoxidc. Psychopharmacology 1986; 88:147-152. 9 Swerdlow NR. Britton KT, Koob G F: Poten­ tiation of acoustic startle by corticotropin-releasing factor (CRF) and by fear are both re­ versed by a helical CRF (9-41). Neuropsycho­ pharmacology 1989;2:285-292. 10 Geyer MA, Swerdlow NR. Mansbach RS, BralT DL: Startle response models of sensori­ motor gating and habituation deficits in schi­ zophrenia. Brain Res Bull 1990;25:485-498. 11 Dixon WJ (ed): BMDP Statistical Software Manual. Berkeley. University of California Press, 1988. 12 Sternberg EM, Young WS III, Bernardini R, Calogero AE, Chrousos GP, Gold PW, Wilder RL: A central nervous system defect in biosynthesis of corticotropin-releasing hor­ mone is associated with susceptibility to strep­ tococcal cell wall arthritis in Lewis rats. Proc Natl Acad Sci USA 1989;86:4771-4775.

13 File SE, Vellucci SA, Wendland S: Corticoste­ rone - an anxiogenic or an anxiolytic agent? J Pharm Pharmacol 1979;31:300-305. 14 Crawley JN, Glowa JR. Paul SM, Majcwska MD: Anxiolytic activity of an endogenous adrenal steroid. Brain Res 1986:398:382385. 15 Davis M: Pharmacological and anatomical analysis of fear conditioning using the fearpotentiated startle paradigm. Behav Neurosci 1986;100:814-824. 16 Gerlach JL, McEwen BS: Rat brain binds adrenal steroid: radioautography of hippo­ campus with corticosterone. Science 1972; 175:1133-1136. 17 Rinald PC, Thompson RF: Age, sex and strain comparison of habituation of the startle response in the rat. Physiol Behav 1985; 35:9-13. 18 Rigdon GC: Differential effects of apomorphine on prepulse inhibition of acoustic startle reflex in two rat strains. Psychopharma­ cology 1990;102:419-421. 19 Pauly JR, Ullman EA, Collins AC: Strain dif­ ferences in adrenalectomy-induced alterations in nicotine sensitivity in the mouse. Pharma­ col Bioehem Behav 1990;35:171-179.

723

Downloaded by: University Toronto Libr. 142.150.190.39 - 10/11/2017 4:25:58 PM

1 Lehman TJA, Allen JB, Plotz PH, Wilder R: Lactobacillus casei cell wall-induced arthritis in rats: Cell wall fragment distribution and persistence in chronic arthritis-susceptible LEW/N and -resistant F344/N rats. Arthritis Rheum 1984;27:939-942. 2 Sternberg EM, Hill JM, Chrousos GP, Kamilaris T, Listwak SJ. Gold PW, Wilder RL: In­ flammatory mediator-induced hypothalamicpituitary-adrenal axis activation is defective in streptococcal cell wall arthritis susceptible Lewis rats. Proc Natl Acad Sci USA 1989; 86:2374-2378. 3 Sternberg EM, Glowa JR, Smith M. Calogero AE, Listwak SJ, Aksentijevich S, Chrousos GP. Wilder RL, Gold PW: Susceptibility to in­ flammatory arthritis and abnormal behavioral responses to stress in rats are linked to a com­ mon defect in neuronal responsiveness. Brain Res 1992:570:54-60. 4 Sternberg EM, Wilder RL, Chrousos GP, Gold PW: Stress responses and the pathogen­ esis of arthritis; in McCubbin JA, Kaufmann PG, Nemeroff CB, Peters S (eds): The Role of Neuropeptides in Stress Pathogenesis and Systemic Disease New York, Academic Press. 1991. pp 287-300. 5 Glowa JR, Gold PW, Sternberg EM: Differ­ ential behavioral response in LEW/N and F344/N rats: Effects of corticotropin releas­ ing hormone. Prog Neuropsychopharmacol Biol Psychiatry 1992;16:549-560.

Differential startle amplitude and corticosterone response in rats.

Different rat strains exhibit large differences in hypothalamic pituitary-adrenal activity that have been used to determine the role of the neuroendoc...
920KB Sizes 0 Downloads 0 Views