- Email: [email protected]
Lack of effect of an early stressful life event on sensorimotor gating in adult rats
Schizophrenia Research 41 (2000) 365–371 www.elsevier.com/locate/schres
Lack of effect of an early stressful life event on sensorimotor gating in adult rats Julia Lehmann, Christopher R. Pryce, Joram Feldon * Behavioural Biology Laboratory, Swiss Federal Institute of Technology Zu¨rich, Schorenstrasse 16, CH-8603 Schwerzenbach, Switzerland Received 1 January 1999; accepted for publication 27 April 1999
Abstract Hypotheses of the etiology of schizophrenia emphasize the important role of perinatal insults in predisposing individuals to the development of the disease, so that an animal model in which a discrete postnatal manipulation of the infant social environment yields schizophrenia-like behavior in adulthood would be valuable in terms of the study of the neural substrate and treatment of schizophrenia. Schizophrenics demonstrate a deficit in sensorimotor gating (prepulse inhibition), and a similar phenomenon has been described in adult rats following the administration of direct and indirect dopamine agonists. Recently it has been reported that a 24 h separation of rat pups from the mother results in a disruption of prepulse inhibition at adulthood. Here we report a study which investigated the same phenomenon but which, in contrast to the previous study, utilized unrelated subjects all derived from different dams. Maternal separation was conducted for 24 h with pups aged 4, 9 or 18 days and these subjects, together with non-separated controls, were tested at age 3 months in terms of their prepulse inhibition in the acoustic startle response paradigm. Maternal separation did not disrupt prepulse inhibition. Comparison of males and females (with a maximum of one opposite-sex sibling) demonstrated that acoustic startle response and prepulse inhibition of this response was enhanced in males relative to females. This study indicates that 24 h maternal separation does not provide a robust model for studying the effects of early environmental insults on the long-term abnormal development of sensorimotor gating. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Litter effects; Long-term development; Maternal separation; Rat model; Schizophrenia; Sensorimotor gating
1. Introduction It has been consistently reported that schizophrenic patients demonstrate a disruption of prepulse inhibition (PPI ) of the acoustic startle response (ASR) (Braff et al., 1978; Braff et al., 1992; American Psychiatric Association, 1994). PPI is an operational measure of sensorimotor gating. The ASR is a reflex response observed * Corresponding author. Tel.: +41-1-825-7416; fax: +41-1-825-7417. E-mail address: [email protected] (J. Feldon)
following presentation of a sudden pulse of loud noise, and PPI is a reduction in the magnitude of the ASR that typically occurs when the acoustic startle pulse is preceded by a weak acoustic prepulse. The deficit in PPI exhibited by schizophrenics has been interpreted as evidence that impaired sensorimotor gating is an important symptom of schizophrenia and that it might underlie the cognitive symptoms of this disease. PPI of the ASR has also been developed as an index of sensorimotor gating in the rat, and is measured using virtually identical techniques and parameters to those used in human subjects/
0920-9964/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S0 9 2 0 -9 9 6 4 ( 9 9 ) 0 0 08 0 - 8
366
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
patients. In the rat, several studies indicate that the brain dopamine (DA) systems are critical substrates for the mediation of ASR sensorimotor gating (see Swerdlow et al., 1992). Rat PPI is impaired by the DA agonist apomorphine and by the indirect agonist amphetamine. As these effects are obtained in the rat using essentially the same ASR–PPI paradigm as that used in humans, and as neuroleptic drugs (both typical and atypical ) antagonize these DA-induced disruptions, the rat ASR–PPI paradigm has been proposed as an appropriate model of the sensorimotor gating symptoms of schizophrenia, with construct, face and predictive validity (Swerdlow et al., 1994). Animal model studies of PPI and other schizophrenia-like symptoms have attempted to move away from pharmacological manipulations, and these studies have been influenced by the human evidence that disturbances in early normal brain development may induce schizophrenia, leading to the neurodevelopmental hypothesis of schizophrenia ( Weinberger, 1987). One important animal model to have been informed by the neurodevelopmental hypothesis is that of neonatal hippocampal lesions, as developed by Lipska et al. (1993). This is based on excitotoxic lesions within the hippocampal area in 7 day old rat pups, which result in hyper-responsiveness to DA agonists, hypo-responsiveness to DA antagonists and impaired PPI; as in schizophrenia, it has been reported that the behavioral symptoms of early hippocampal lesion do not emerge until after puberty (Lipska et al., 1995). Currently there is increasing interest in the validation of animal models derived via perinatal environmental manipulation. Here the rationale is that manipulation of the rat during a sensitive phase of neurobehavioral development, involving deprivation of salient features of the normal environment, will induce important effects on development of an otherwise non-manipulated central nervous system and its functioning ( Ellenbroek and Cools, 1998). If animal models based on environmental manipulations do yield robust and consistent effects on sensorimotor gating, then we are presented with the very attractive scenario of a non-pharmacological/lesioned neurodevelopmental animal model for an important aspect of schizophrenia, a
disease for which environmentally-induced neurodevelopmental disturbance is one of the major hypotheses of underlying etiology (for a review see Feldon and Weiner, 1992). Two such manipulations are maternal separation performed for a single 24 h period during pup–dam dependency, and social isolation performed chronically immediately post-weaning. It has been reported that social isolation disrupts PPI in the rat (Geyer et al., 1993; Wilkinson et al., 1994; Varty and Higgins, 1995), although two recent studies conducted in our laboratory have called into question the robustness of this model, in Wistar rats at least (Domeney and Feldon, 1998; Weiss et al., 1999). It has recently been reported that 24 h maternal separation at postnatal day 3, 6 or 9 induces disruption of PPI in adult Wistar rats, and in males and females to the same extent ( Ellenbroek et al., 1998). This is potentially a very important finding, for the reasons outlined above. The study by Ellenbroek et al. used a methodology in which the 8–10 pups in each control or treatment group per experiment were provided by a small number of dams and, in the case of the major experiment, by two dams only: that is, the average coefficient of relatedness of each subject to one-half of the other subjects in its experimental group was 0.5. Treating subjects from the same litter as independent samples will decrease within-treatment variance and thereby increase the likelihood of obtaining treatment effects (Denenberg, 1977; Spear and File, 1996). Such a ‘litter effect’ has been reported for the great majority of behavioral paradigms in which it has been tested for to-date, including auditory startle habituation (Buelke-Sam et al., 1985). For example, it has been demonstrated that treating just two pups per litter as independent observations nearly triples the likelihood of obtaining statistically significant effects with prenatal treatments (Holson and Pearce, 1992). Against this background, the aim of the study described here was to analyze the effects of 24 h maternal separation on PPI in subjects that were unrelated. If we observed a decrease in PPI of the ASR as a result of 24 h maternal separation under conditions that used independent samples, then this would add considerably to the proposal that this represents a robust environmental animal
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
model of PPI disruption in schizophrenia. However, if 24 h maternal separation did not yield a decrease in PPI, this would strongly suggest that this environmental manipulation is inappropriate for the modelling of disrupted PPI.
2. Materials and methods 2.1. Animals The experiment was carried out with male and female Wistar rats, all bred in-house (Zur:WIST[ HanIbm], Animal Services, Swiss Federal Institute of Technology Zu¨rich, Schwerzenbach) and under constant husbandry conditions of reversed cycle lighting ( lights on: 1900–0700 h) in a temperature (21±1°C ) and humidity (55±5%) controlled animal facility. Subjects were derived from 48 different litters produced by 48 different dams; that is, each of 48 dams contributed one male (total n=48) and one female (n=48) adult offspring to the experiment. Dams were checked twice daily for litters and the day of birth was assigned as postnatal day (PND) 0. Within 24 h of birth all litters from which subjects were derived were culled to the same litter size and composition of four males and four females. In order to minimize disturbance to litters that might confound effects of the maternal separation procedure (see below), cage cleaning was carried out once only, at postnatal day 11 for all litters, between birth and weaning on day 21. At weaning subjects were placed in group cages (Perspex Macrolon type IV, 59.0×38.5×20.0 cm), four same-sex animals per cage, with each group of four animals derived from different litters and belonging to the same treatment group. Subjects remained undisturbed until testing at age 3 months, with food (Nafag 9431, Eberle Nafag AG, Gossau, CH ) and water available ad libitum. 2.2. Maternal separation Three experimental groups and one control group were studied. Maternally separated subjects were separated from their mothers for 24 h, 1800– 1800 h, at either postnatal day 4 (MS4; n=12
367
males, 12 females), or day 9 (MS9; n=12, 12), or day 18 (MS18; n=12, 12). To carry out maternal separation the dam was removed to another cage while the pups remained in their home cage. The home cage was then transferred to a separate room and placed on a heat pad set at 33°C. After 24 h the pups were returned to the colony room and the mother returned to the home cage. Control animals (CON; n=12, 12), apart from culling and the single cage cleaning, remained undisturbed with their mothers from birth until weaning. 2.3. Acoustic startle response and prepulse inhibition There were 12 subjects per treatment/sex group. In each group, four or eight subjects were tested in an open field paradigm one week prior to inclusion in this study. ANOVA analysis indicated that this experience did not have an effect on ASR ( p>0.37) or PPI ( p>0.27). Rats were run in balanced squads of four. The acoustic startle apparatus consisted of four startle chambers (model SR-LAB, San Diego Instruments, San Diego, CA) each containing a transparent Plexiglas tube (diameter 8.2 cm, length 20 cm) mounted on a Plexiglas frame within a ventilated enclosure. Acoustic pulses and prepulses were presented via a speaker mounted 24 cm above the tube. A piezoelectric accelerometer mounted below the frame detected and transduced movement within the tube. Subjects were placed inside the tube and during a 5 min acclimatization period were exposed to a background white noise of 70 dB[A] that continued throughout the session. Then subjects were exposed to four startle pulses of 120 dB[A] white noise, each of 30 ms duration, to determine basal acoustic startle response (ASR). Following this, six blocks of 11 trials were presented to test for PPI. Each block comprised four different trial types: startle pulse only (two trials), prepulse followed 100 ms after prepulse onset by startle pulse (one trial for each prepulse intensity), prepulse only (one trial for each prepulse intensity), and no stimulus (one trial ). Prepulses each took the form of 20 ms of white noise, and had an intensity of either 72, 76, 80, or 84 B[A]. The four trial types were presented in a pseudorandom order within
368
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
each block, with an inter-trial interval of 10–20 s. Startle amplitude (in arbitrary units) was measured by the accelerometer during 100 ms after startle pulse onset and these readings were rectified and amplified and the average startle amplitude during 100 1 ms readings was calculated by computer.
p<0.005), and a significant sex×prepulse intensity interaction (F =4.85; p<0.005) (Fig. 2). Males 3,264 and females demonstrated similar PPIs following prepulses of 72 dB[A] and males demonstrated enhanced PPI relative to females following prepulse intensities of 76, 80, and 84 dB[A].
2.4. Data analysis 4. Discussion The effect of maternal separation on subjects’ ASR was determined by ANOVA with main factors of sex and treatment and repeated measures for the different startle pulses (16 in total ). The percentage PPI induced by each prepulse intensity was calculated as: [100−(100×startle amplitude at prepulse trial )/(startle amplitude at startle pulse alone trial )]. The effect of maternal separation on subjects’ PPI was determined by ANOVA with main factors of sex and treatment and repeated measures for the four different prepulse intensities.
3. Results The 2×4×16 ANOVA did not yield either a main effect or an interaction involving the factor of maternal separation on ASR ( p>0.73). There was a significant main effect of sex (F =81.79; 1,88 p<0.001) and of the repeated measurement factor of startle pulse (F =14.1; p<0.0001), and 15,1320 a significant sex×startle pulse interaction (F =3.4; p<0.0001). As presented in Fig. 1, 15,1320 these analyses reflect the absolute sex difference in ASR and the sex-dependent habituation of ASR across trials. Females displayed relatively low ASR and a shallow ASR habituation curve across trials 1–16, whereas males displayed initially very high ASR and strong habituation after four to five trials. Turning to PPI, the 2×4×4 ANOVA did not yield either a main effect or an interaction involving the factor of maternal separation ( p>0.27). There was a significant main effect of the repeated measurement factor of prepulse intensity (F =91.9; p<0.0001), reflecting, as shown in 3,264 Fig. 2, the gradual increase in PPI as a function of the intensity of the prepulse stimulus. There was a significant sex difference in PPI (F =8.6; 1,88
In this experiment, designed to study the effects on prepulse inhibition of a single 24 h period of maternal separation, we have demonstrated that sensorimotor gating is not influenced by this environmental manipulation. Unrelated adult rats, male and female, were unaffected in terms of the development of PPI relative to controls, following 24 h of maternal deprivation at day 4, 9 or 18. This negative finding contradicts a recent report that a 24 h maternal separation does yield a PPI deficit in adulthood ( Ellenbroek et al., 1998). The major difference between these two studies in terms of experimental design is that, whereas in the Ellenbroek et al. study subjects in each condition were derived from a small number of dams only, in the present study all subjects were derived from different dams. Given the conclusive evidence that the likelihood of obtaining significant statistical effects is increased markedly by including related subjects (siblings) in the same treatment group (Denenberg, 1977; Spear and File, 1996), then it is very plausible that the different findings of these two otherwise similar studies are attributable to the differences in between-subject relatedness. Therefore, the present study contradicts the notion that the chronic effects of a single 24 h period of maternal separation on sensorimotor gating, as measured by acoustic PPI, represents a robust animal model for the deficits in this function presented by schizophrenics. Although the litter effect is by far the most important difference between this and the Ellenbroek et al. study, it is of course possible that even when two laboratories work with the same environmental model, and both correctly use unrelated subjects, they will still obtain different results. Indeed, interlaboratory reproducibility represents a very important validation step for animal models.
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
369
Fig. 1. Amplitude of acoustic startle response to 16 startle pulses of 120 dB[A] white noise in male and female rats. Values in the line graph are means±SEMs for 48 males and 48 females (12 per MS4, MS9, MS18, CON group) and values in the inset histogram are overall means±SEMs.
Line differences within a rat strain may be important (both our study and that of Ellenbroek et al. were performed with outbred strains of Wistar rats but there could be line differences in relevant phenotypes). Methodological differences could also be important: for example, Zimmerberg and Shartrand (1992) have demonstrated that ambient temperature during maternal separation affects development of the dopaminergic system (our study was conducted with infants placed on a heat pad whereas there is no mention of such in Ellenbroek et al., 1998), so that behavioral paradigms which are sensitive to maternal separation should control ambient temperature. However, the MS procedure employed by Zimmerberg and
Shartrand (1992) consisted of a repeated MS between PND 2–15 for 6 h per day, and any comparisons with the present MS design must be tentative. While we did not find any evidence for a consistent effect of maternal deprivation on the development of PPI in males or females in this study, we have demonstrated that males — both with and without a prior experience of 24 h maternal separation — exhibit a markedly higher ASR than females and that males exhibit enhanced PPI at prepulse intensities of 76, 80 and 84 dB[A]. To our knowledge, there are no previous reports of a sex difference in prepulse inhibition in the rat, and we report on the consistent evidence for sex differ-
370
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
Fig. 2. Prepulse inhibition (calculated according to the formula described in Materials and methods) as a function of prepulse intensity in male and female rats. Values are means±SEMs for 12 subjects per treatment/sex group. Overall means±SEMs are represented in the inset histogram.
ences in ASR and PPI obtained in our laboratory in detail elsewhere (Lehmann et al., in press). Furthermore, our studies are in accord with the human evidence that men demonstrate enhanced PPI relative to women (Swerdlow et al., 1997). That we obtained a finding contrary to previous reports of no sex difference (e.g. Ellenbroek et al., 1998) could well, again, be attributable to designing our study to maximize unrelatedness between subjects. Studies based on a small number of litters which compare several male with several female siblings are less likely to find a sex difference than a study such as ours in which each male had a maximum of one female sibling in the group under comparison. Given that males demonstrated both enhanced ASR and enhanced PPI, we need to be cautious in our interpretation of the latter as a ‘real’ sex difference when it might be a consequence of the high absolute ASR. However, other studies conducted recently in our laboratory provide indirect support for the conclusion that enhanced PPI in males is independent of their high
startle responses (see Lehmann et al., in press). For example, in a between-strain study of Lewis and Fischer rats we found markedly elevated ASR in Lewis rats but no differences in PPI, and in a within-strain study of Lewis rats we found no sex difference in ASR and enhanced PPI in males (unpublished data). In summary, the present study has demonstrated that a 24 h period of maternal separation conducted during early, mid or late maternal dependency does not lead to a deficit in sensorimotor gating in male or female Wistar rats, and that sensorimotor gating is enhanced in male Wistar rats compared with females. These findings are based on experiments conducted with unrelated subjects and are therefore free from the very confounding effects which high coefficients of relatedness can introduce. This study demonstrates that a single maternal deprivation, which in principle represents an attractive and advantageous neurodevelopmental model, is unlikely to provide a robust model for studying the effects of early
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
environmental insults on the long-term abnormal development of sensorimotor gating.
Acknowledgements This work was supported by a grant from the Swiss Federal Institute of Technology Zu¨rich. We thank Animal Services for the maintenance and care of the animals.
References American Psychiatric Association, 4th edn. 1994. Diagnostic and Statistical Manual of Mental Disorders, DSM-IV. American Psychiatric Association, Washington, DC. Braff, D., Stone, C., Callaway, E., Glick, E., Bali, L., 1978. Prestimulus effects on human startle reflex in normals and schizophrenics. Psychophysiology 15, 339–343. Braff, D.L., Grillon, C., Geyer, M.A., 1992. Gating and habituation of the startle reflex in schizophrenic patients. Arch. Gen. Psychiatry 49, 206–215. Buelke-Sam, J., Kimmel, C.A., Adams, J., Nelson, C.J., Vorhees, C.V., Wright, D.C., Omer, V.S., Korol, B.A., Butcher, R.E., Geyer, M.A., Holson, J.F., Kutsch, C.L., Wayner, M.J., 1985. Collaborative behavioral teratology study: results. Neurobehav. Toxicol. Teratol. 7, 591–624. Denenberg, V.H., 1977. Assessing the effects of early experience. In: Myers, R.D. ( Ed.), Methods in Psychobiology Vol. 3. Academic Press, New York, pp. 127–147. Domeney, A.M., Feldon, J., 1998. The disruption of prepulse inhibition by social isolation in the Wistar rat: How robust is the effect? Pharmacol. Biochem. Behav. 59, 883–890. Ellenbroek, B.A., Cools, A.R., 1998. The neurodevelopment hypothesis of schizophrenia: clinical evidence and animal models. Neurosci. Res. Comm. 22, 127–136. Ellenbroek, B.A., van den Kroonenberg, P.T.J.M., Cools, A.R., 1998. The effects of an early stressful life event on sensorimotor gating in adult rats. Schizophrenia Res. 30, 251–260. Feldon, J., Weiner, I., 1992. From an animal model of an attentional deficit towards new insights into the pathophysiology of schizophrenia. J. Psychiatric Res. 26, 345–366. Geyer, M.A., Wilkinson, L.S., Humby, T., Robbins, T.W., 1993. Isolation rearing of rat produces a deficit in prepulse inhibition of acoustic startle similar to that in schizophrenia. Biol. Psychiatry 34, 361–372.
371
Holson, R.R., Pearce, B., 1992. Principles and pitfalls in the analysis of prenatal treatment effects in multiparous species. Neurotoxicol. Teratol. 14, 221–228. Lehmann, J., Pryce, C.R., Feldon, J., in press. Sex differences in the acoustic startle response and prepulse inhibition in Wistar rats. Behav. Brain Res. Lipska, B.K., Jaskiw, G.E., Weinberger, D.R., 1993. Postpubertal emergence of hyper-responsiveness to stress and to amphetamine after neonatal excitotoxic hippocampal damage: a potential animal model of schizophrenia. Neuropsychopharmacology 9, 67–75. Lipska, B.K., Swerdlow, N.R., Geyer, M.A., Jaskiw, G.E., Braff, D.L., Weinberger, D.R., 1995. Neonatal excitotoxic hippocampal damage in rats causes post-pubertal changes in prepulse inhibition of startle and its disruption by apomorphine. Psychopharmacology 122, 34–43. Spear, L.P., File, S.E., 1996. Methodological considerations in neurobehavioral teratology. Pharmacol. Biochem. Behav. 55, 455–457. Swerdlow, N.R., Caine, S.B., Braff, D.L., Geyer, M.A., 1992. The neural substrates of sensorimotor gating of the startle reflex: a review of recent findings and their implications. J. Psychopharmacol. 6, 176–190. Swerdlow, N.R., Braff, D.L., Taaid, N., Geyer, M.A., 1994. Assessing the validity of an animal model of deficient sensorimotor gating in schizophrenic patients. Arch. Gen. Psychiatry 51, 139–154. Swerdlow, N.R., Hartman, P.L., Auerbach, P.P., 1997. Changes in sensorimotor inhibition across the menstrual cycle: Implications for neuropsychiatric disorders. Biol. Psychiatry 41, 452–460. Varty, G.B., Higgins, G.A., 1995. Examination of drug-induced and isolation-induced disruptions of prepulse inhibition as models to screen antipsychotic drugs. Psychopharmacology 122, 15–26. Weinberger, D.R., 1987. Implications of normal brain development for the pathogenesis of schizophrenia. Arch. Gen. Psychiatry 44, 660–669. Weiss, I.C., Feldon, J., Domeney, A.M., 1999. Isolation rearinginduced disruption of prepulse inhibition: further evidence for fragility of the response. Behav. Pharmacol. 10, 139–149. Wilkinson, L.S., Killcross, S.S., Humby, T., Hall, F.S., Geyer, M.A., Robbins, T.W., 1994. Social isolation in the rat produces developmentally specific deficits in prepulse inhibition of the acoustic startle response without disrupting latent inhibition. Neuropsychopharmacology 10, 61–72. Zimmerberg, B., Shartrand, A.M., 1992. Temperature-dependent effects of maternal separation on growth, activity, and amphetamine sensitivity in the rat. Dev. Psychobiol. 25, 213–226.
Lack of effect of an early stressful life event on sensorimotor gating in adult rats Julia Lehmann, Christopher R. Pryce, Joram Feldon * Behavioural Biology Laboratory, Swiss Federal Institute of Technology Zu¨rich, Schorenstrasse 16, CH-8603 Schwerzenbach, Switzerland Received 1 January 1999; accepted for publication 27 April 1999
Abstract Hypotheses of the etiology of schizophrenia emphasize the important role of perinatal insults in predisposing individuals to the development of the disease, so that an animal model in which a discrete postnatal manipulation of the infant social environment yields schizophrenia-like behavior in adulthood would be valuable in terms of the study of the neural substrate and treatment of schizophrenia. Schizophrenics demonstrate a deficit in sensorimotor gating (prepulse inhibition), and a similar phenomenon has been described in adult rats following the administration of direct and indirect dopamine agonists. Recently it has been reported that a 24 h separation of rat pups from the mother results in a disruption of prepulse inhibition at adulthood. Here we report a study which investigated the same phenomenon but which, in contrast to the previous study, utilized unrelated subjects all derived from different dams. Maternal separation was conducted for 24 h with pups aged 4, 9 or 18 days and these subjects, together with non-separated controls, were tested at age 3 months in terms of their prepulse inhibition in the acoustic startle response paradigm. Maternal separation did not disrupt prepulse inhibition. Comparison of males and females (with a maximum of one opposite-sex sibling) demonstrated that acoustic startle response and prepulse inhibition of this response was enhanced in males relative to females. This study indicates that 24 h maternal separation does not provide a robust model for studying the effects of early environmental insults on the long-term abnormal development of sensorimotor gating. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Litter effects; Long-term development; Maternal separation; Rat model; Schizophrenia; Sensorimotor gating
1. Introduction It has been consistently reported that schizophrenic patients demonstrate a disruption of prepulse inhibition (PPI ) of the acoustic startle response (ASR) (Braff et al., 1978; Braff et al., 1992; American Psychiatric Association, 1994). PPI is an operational measure of sensorimotor gating. The ASR is a reflex response observed * Corresponding author. Tel.: +41-1-825-7416; fax: +41-1-825-7417. E-mail address: [email protected] (J. Feldon)
following presentation of a sudden pulse of loud noise, and PPI is a reduction in the magnitude of the ASR that typically occurs when the acoustic startle pulse is preceded by a weak acoustic prepulse. The deficit in PPI exhibited by schizophrenics has been interpreted as evidence that impaired sensorimotor gating is an important symptom of schizophrenia and that it might underlie the cognitive symptoms of this disease. PPI of the ASR has also been developed as an index of sensorimotor gating in the rat, and is measured using virtually identical techniques and parameters to those used in human subjects/
0920-9964/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S0 9 2 0 -9 9 6 4 ( 9 9 ) 0 0 08 0 - 8
366
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
patients. In the rat, several studies indicate that the brain dopamine (DA) systems are critical substrates for the mediation of ASR sensorimotor gating (see Swerdlow et al., 1992). Rat PPI is impaired by the DA agonist apomorphine and by the indirect agonist amphetamine. As these effects are obtained in the rat using essentially the same ASR–PPI paradigm as that used in humans, and as neuroleptic drugs (both typical and atypical ) antagonize these DA-induced disruptions, the rat ASR–PPI paradigm has been proposed as an appropriate model of the sensorimotor gating symptoms of schizophrenia, with construct, face and predictive validity (Swerdlow et al., 1994). Animal model studies of PPI and other schizophrenia-like symptoms have attempted to move away from pharmacological manipulations, and these studies have been influenced by the human evidence that disturbances in early normal brain development may induce schizophrenia, leading to the neurodevelopmental hypothesis of schizophrenia ( Weinberger, 1987). One important animal model to have been informed by the neurodevelopmental hypothesis is that of neonatal hippocampal lesions, as developed by Lipska et al. (1993). This is based on excitotoxic lesions within the hippocampal area in 7 day old rat pups, which result in hyper-responsiveness to DA agonists, hypo-responsiveness to DA antagonists and impaired PPI; as in schizophrenia, it has been reported that the behavioral symptoms of early hippocampal lesion do not emerge until after puberty (Lipska et al., 1995). Currently there is increasing interest in the validation of animal models derived via perinatal environmental manipulation. Here the rationale is that manipulation of the rat during a sensitive phase of neurobehavioral development, involving deprivation of salient features of the normal environment, will induce important effects on development of an otherwise non-manipulated central nervous system and its functioning ( Ellenbroek and Cools, 1998). If animal models based on environmental manipulations do yield robust and consistent effects on sensorimotor gating, then we are presented with the very attractive scenario of a non-pharmacological/lesioned neurodevelopmental animal model for an important aspect of schizophrenia, a
disease for which environmentally-induced neurodevelopmental disturbance is one of the major hypotheses of underlying etiology (for a review see Feldon and Weiner, 1992). Two such manipulations are maternal separation performed for a single 24 h period during pup–dam dependency, and social isolation performed chronically immediately post-weaning. It has been reported that social isolation disrupts PPI in the rat (Geyer et al., 1993; Wilkinson et al., 1994; Varty and Higgins, 1995), although two recent studies conducted in our laboratory have called into question the robustness of this model, in Wistar rats at least (Domeney and Feldon, 1998; Weiss et al., 1999). It has recently been reported that 24 h maternal separation at postnatal day 3, 6 or 9 induces disruption of PPI in adult Wistar rats, and in males and females to the same extent ( Ellenbroek et al., 1998). This is potentially a very important finding, for the reasons outlined above. The study by Ellenbroek et al. used a methodology in which the 8–10 pups in each control or treatment group per experiment were provided by a small number of dams and, in the case of the major experiment, by two dams only: that is, the average coefficient of relatedness of each subject to one-half of the other subjects in its experimental group was 0.5. Treating subjects from the same litter as independent samples will decrease within-treatment variance and thereby increase the likelihood of obtaining treatment effects (Denenberg, 1977; Spear and File, 1996). Such a ‘litter effect’ has been reported for the great majority of behavioral paradigms in which it has been tested for to-date, including auditory startle habituation (Buelke-Sam et al., 1985). For example, it has been demonstrated that treating just two pups per litter as independent observations nearly triples the likelihood of obtaining statistically significant effects with prenatal treatments (Holson and Pearce, 1992). Against this background, the aim of the study described here was to analyze the effects of 24 h maternal separation on PPI in subjects that were unrelated. If we observed a decrease in PPI of the ASR as a result of 24 h maternal separation under conditions that used independent samples, then this would add considerably to the proposal that this represents a robust environmental animal
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
model of PPI disruption in schizophrenia. However, if 24 h maternal separation did not yield a decrease in PPI, this would strongly suggest that this environmental manipulation is inappropriate for the modelling of disrupted PPI.
2. Materials and methods 2.1. Animals The experiment was carried out with male and female Wistar rats, all bred in-house (Zur:WIST[ HanIbm], Animal Services, Swiss Federal Institute of Technology Zu¨rich, Schwerzenbach) and under constant husbandry conditions of reversed cycle lighting ( lights on: 1900–0700 h) in a temperature (21±1°C ) and humidity (55±5%) controlled animal facility. Subjects were derived from 48 different litters produced by 48 different dams; that is, each of 48 dams contributed one male (total n=48) and one female (n=48) adult offspring to the experiment. Dams were checked twice daily for litters and the day of birth was assigned as postnatal day (PND) 0. Within 24 h of birth all litters from which subjects were derived were culled to the same litter size and composition of four males and four females. In order to minimize disturbance to litters that might confound effects of the maternal separation procedure (see below), cage cleaning was carried out once only, at postnatal day 11 for all litters, between birth and weaning on day 21. At weaning subjects were placed in group cages (Perspex Macrolon type IV, 59.0×38.5×20.0 cm), four same-sex animals per cage, with each group of four animals derived from different litters and belonging to the same treatment group. Subjects remained undisturbed until testing at age 3 months, with food (Nafag 9431, Eberle Nafag AG, Gossau, CH ) and water available ad libitum. 2.2. Maternal separation Three experimental groups and one control group were studied. Maternally separated subjects were separated from their mothers for 24 h, 1800– 1800 h, at either postnatal day 4 (MS4; n=12
367
males, 12 females), or day 9 (MS9; n=12, 12), or day 18 (MS18; n=12, 12). To carry out maternal separation the dam was removed to another cage while the pups remained in their home cage. The home cage was then transferred to a separate room and placed on a heat pad set at 33°C. After 24 h the pups were returned to the colony room and the mother returned to the home cage. Control animals (CON; n=12, 12), apart from culling and the single cage cleaning, remained undisturbed with their mothers from birth until weaning. 2.3. Acoustic startle response and prepulse inhibition There were 12 subjects per treatment/sex group. In each group, four or eight subjects were tested in an open field paradigm one week prior to inclusion in this study. ANOVA analysis indicated that this experience did not have an effect on ASR ( p>0.37) or PPI ( p>0.27). Rats were run in balanced squads of four. The acoustic startle apparatus consisted of four startle chambers (model SR-LAB, San Diego Instruments, San Diego, CA) each containing a transparent Plexiglas tube (diameter 8.2 cm, length 20 cm) mounted on a Plexiglas frame within a ventilated enclosure. Acoustic pulses and prepulses were presented via a speaker mounted 24 cm above the tube. A piezoelectric accelerometer mounted below the frame detected and transduced movement within the tube. Subjects were placed inside the tube and during a 5 min acclimatization period were exposed to a background white noise of 70 dB[A] that continued throughout the session. Then subjects were exposed to four startle pulses of 120 dB[A] white noise, each of 30 ms duration, to determine basal acoustic startle response (ASR). Following this, six blocks of 11 trials were presented to test for PPI. Each block comprised four different trial types: startle pulse only (two trials), prepulse followed 100 ms after prepulse onset by startle pulse (one trial for each prepulse intensity), prepulse only (one trial for each prepulse intensity), and no stimulus (one trial ). Prepulses each took the form of 20 ms of white noise, and had an intensity of either 72, 76, 80, or 84 B[A]. The four trial types were presented in a pseudorandom order within
368
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
each block, with an inter-trial interval of 10–20 s. Startle amplitude (in arbitrary units) was measured by the accelerometer during 100 ms after startle pulse onset and these readings were rectified and amplified and the average startle amplitude during 100 1 ms readings was calculated by computer.
p<0.005), and a significant sex×prepulse intensity interaction (F =4.85; p<0.005) (Fig. 2). Males 3,264 and females demonstrated similar PPIs following prepulses of 72 dB[A] and males demonstrated enhanced PPI relative to females following prepulse intensities of 76, 80, and 84 dB[A].
2.4. Data analysis 4. Discussion The effect of maternal separation on subjects’ ASR was determined by ANOVA with main factors of sex and treatment and repeated measures for the different startle pulses (16 in total ). The percentage PPI induced by each prepulse intensity was calculated as: [100−(100×startle amplitude at prepulse trial )/(startle amplitude at startle pulse alone trial )]. The effect of maternal separation on subjects’ PPI was determined by ANOVA with main factors of sex and treatment and repeated measures for the four different prepulse intensities.
3. Results The 2×4×16 ANOVA did not yield either a main effect or an interaction involving the factor of maternal separation on ASR ( p>0.73). There was a significant main effect of sex (F =81.79; 1,88 p<0.001) and of the repeated measurement factor of startle pulse (F =14.1; p<0.0001), and 15,1320 a significant sex×startle pulse interaction (F =3.4; p<0.0001). As presented in Fig. 1, 15,1320 these analyses reflect the absolute sex difference in ASR and the sex-dependent habituation of ASR across trials. Females displayed relatively low ASR and a shallow ASR habituation curve across trials 1–16, whereas males displayed initially very high ASR and strong habituation after four to five trials. Turning to PPI, the 2×4×4 ANOVA did not yield either a main effect or an interaction involving the factor of maternal separation ( p>0.27). There was a significant main effect of the repeated measurement factor of prepulse intensity (F =91.9; p<0.0001), reflecting, as shown in 3,264 Fig. 2, the gradual increase in PPI as a function of the intensity of the prepulse stimulus. There was a significant sex difference in PPI (F =8.6; 1,88
In this experiment, designed to study the effects on prepulse inhibition of a single 24 h period of maternal separation, we have demonstrated that sensorimotor gating is not influenced by this environmental manipulation. Unrelated adult rats, male and female, were unaffected in terms of the development of PPI relative to controls, following 24 h of maternal deprivation at day 4, 9 or 18. This negative finding contradicts a recent report that a 24 h maternal separation does yield a PPI deficit in adulthood ( Ellenbroek et al., 1998). The major difference between these two studies in terms of experimental design is that, whereas in the Ellenbroek et al. study subjects in each condition were derived from a small number of dams only, in the present study all subjects were derived from different dams. Given the conclusive evidence that the likelihood of obtaining significant statistical effects is increased markedly by including related subjects (siblings) in the same treatment group (Denenberg, 1977; Spear and File, 1996), then it is very plausible that the different findings of these two otherwise similar studies are attributable to the differences in between-subject relatedness. Therefore, the present study contradicts the notion that the chronic effects of a single 24 h period of maternal separation on sensorimotor gating, as measured by acoustic PPI, represents a robust animal model for the deficits in this function presented by schizophrenics. Although the litter effect is by far the most important difference between this and the Ellenbroek et al. study, it is of course possible that even when two laboratories work with the same environmental model, and both correctly use unrelated subjects, they will still obtain different results. Indeed, interlaboratory reproducibility represents a very important validation step for animal models.
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
369
Fig. 1. Amplitude of acoustic startle response to 16 startle pulses of 120 dB[A] white noise in male and female rats. Values in the line graph are means±SEMs for 48 males and 48 females (12 per MS4, MS9, MS18, CON group) and values in the inset histogram are overall means±SEMs.
Line differences within a rat strain may be important (both our study and that of Ellenbroek et al. were performed with outbred strains of Wistar rats but there could be line differences in relevant phenotypes). Methodological differences could also be important: for example, Zimmerberg and Shartrand (1992) have demonstrated that ambient temperature during maternal separation affects development of the dopaminergic system (our study was conducted with infants placed on a heat pad whereas there is no mention of such in Ellenbroek et al., 1998), so that behavioral paradigms which are sensitive to maternal separation should control ambient temperature. However, the MS procedure employed by Zimmerberg and
Shartrand (1992) consisted of a repeated MS between PND 2–15 for 6 h per day, and any comparisons with the present MS design must be tentative. While we did not find any evidence for a consistent effect of maternal deprivation on the development of PPI in males or females in this study, we have demonstrated that males — both with and without a prior experience of 24 h maternal separation — exhibit a markedly higher ASR than females and that males exhibit enhanced PPI at prepulse intensities of 76, 80 and 84 dB[A]. To our knowledge, there are no previous reports of a sex difference in prepulse inhibition in the rat, and we report on the consistent evidence for sex differ-
370
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
Fig. 2. Prepulse inhibition (calculated according to the formula described in Materials and methods) as a function of prepulse intensity in male and female rats. Values are means±SEMs for 12 subjects per treatment/sex group. Overall means±SEMs are represented in the inset histogram.
ences in ASR and PPI obtained in our laboratory in detail elsewhere (Lehmann et al., in press). Furthermore, our studies are in accord with the human evidence that men demonstrate enhanced PPI relative to women (Swerdlow et al., 1997). That we obtained a finding contrary to previous reports of no sex difference (e.g. Ellenbroek et al., 1998) could well, again, be attributable to designing our study to maximize unrelatedness between subjects. Studies based on a small number of litters which compare several male with several female siblings are less likely to find a sex difference than a study such as ours in which each male had a maximum of one female sibling in the group under comparison. Given that males demonstrated both enhanced ASR and enhanced PPI, we need to be cautious in our interpretation of the latter as a ‘real’ sex difference when it might be a consequence of the high absolute ASR. However, other studies conducted recently in our laboratory provide indirect support for the conclusion that enhanced PPI in males is independent of their high
startle responses (see Lehmann et al., in press). For example, in a between-strain study of Lewis and Fischer rats we found markedly elevated ASR in Lewis rats but no differences in PPI, and in a within-strain study of Lewis rats we found no sex difference in ASR and enhanced PPI in males (unpublished data). In summary, the present study has demonstrated that a 24 h period of maternal separation conducted during early, mid or late maternal dependency does not lead to a deficit in sensorimotor gating in male or female Wistar rats, and that sensorimotor gating is enhanced in male Wistar rats compared with females. These findings are based on experiments conducted with unrelated subjects and are therefore free from the very confounding effects which high coefficients of relatedness can introduce. This study demonstrates that a single maternal deprivation, which in principle represents an attractive and advantageous neurodevelopmental model, is unlikely to provide a robust model for studying the effects of early
J. Lehmann et al. / Schizophrenia Research 41 (2000) 365–371
environmental insults on the long-term abnormal development of sensorimotor gating.
Acknowledgements This work was supported by a grant from the Swiss Federal Institute of Technology Zu¨rich. We thank Animal Services for the maintenance and care of the animals.
References American Psychiatric Association, 4th edn. 1994. Diagnostic and Statistical Manual of Mental Disorders, DSM-IV. American Psychiatric Association, Washington, DC. Braff, D., Stone, C., Callaway, E., Glick, E., Bali, L., 1978. Prestimulus effects on human startle reflex in normals and schizophrenics. Psychophysiology 15, 339–343. Braff, D.L., Grillon, C., Geyer, M.A., 1992. Gating and habituation of the startle reflex in schizophrenic patients. Arch. Gen. Psychiatry 49, 206–215. Buelke-Sam, J., Kimmel, C.A., Adams, J., Nelson, C.J., Vorhees, C.V., Wright, D.C., Omer, V.S., Korol, B.A., Butcher, R.E., Geyer, M.A., Holson, J.F., Kutsch, C.L., Wayner, M.J., 1985. Collaborative behavioral teratology study: results. Neurobehav. Toxicol. Teratol. 7, 591–624. Denenberg, V.H., 1977. Assessing the effects of early experience. In: Myers, R.D. ( Ed.), Methods in Psychobiology Vol. 3. Academic Press, New York, pp. 127–147. Domeney, A.M., Feldon, J., 1998. The disruption of prepulse inhibition by social isolation in the Wistar rat: How robust is the effect? Pharmacol. Biochem. Behav. 59, 883–890. Ellenbroek, B.A., Cools, A.R., 1998. The neurodevelopment hypothesis of schizophrenia: clinical evidence and animal models. Neurosci. Res. Comm. 22, 127–136. Ellenbroek, B.A., van den Kroonenberg, P.T.J.M., Cools, A.R., 1998. The effects of an early stressful life event on sensorimotor gating in adult rats. Schizophrenia Res. 30, 251–260. Feldon, J., Weiner, I., 1992. From an animal model of an attentional deficit towards new insights into the pathophysiology of schizophrenia. J. Psychiatric Res. 26, 345–366. Geyer, M.A., Wilkinson, L.S., Humby, T., Robbins, T.W., 1993. Isolation rearing of rat produces a deficit in prepulse inhibition of acoustic startle similar to that in schizophrenia. Biol. Psychiatry 34, 361–372.
371
Holson, R.R., Pearce, B., 1992. Principles and pitfalls in the analysis of prenatal treatment effects in multiparous species. Neurotoxicol. Teratol. 14, 221–228. Lehmann, J., Pryce, C.R., Feldon, J., in press. Sex differences in the acoustic startle response and prepulse inhibition in Wistar rats. Behav. Brain Res. Lipska, B.K., Jaskiw, G.E., Weinberger, D.R., 1993. Postpubertal emergence of hyper-responsiveness to stress and to amphetamine after neonatal excitotoxic hippocampal damage: a potential animal model of schizophrenia. Neuropsychopharmacology 9, 67–75. Lipska, B.K., Swerdlow, N.R., Geyer, M.A., Jaskiw, G.E., Braff, D.L., Weinberger, D.R., 1995. Neonatal excitotoxic hippocampal damage in rats causes post-pubertal changes in prepulse inhibition of startle and its disruption by apomorphine. Psychopharmacology 122, 34–43. Spear, L.P., File, S.E., 1996. Methodological considerations in neurobehavioral teratology. Pharmacol. Biochem. Behav. 55, 455–457. Swerdlow, N.R., Caine, S.B., Braff, D.L., Geyer, M.A., 1992. The neural substrates of sensorimotor gating of the startle reflex: a review of recent findings and their implications. J. Psychopharmacol. 6, 176–190. Swerdlow, N.R., Braff, D.L., Taaid, N., Geyer, M.A., 1994. Assessing the validity of an animal model of deficient sensorimotor gating in schizophrenic patients. Arch. Gen. Psychiatry 51, 139–154. Swerdlow, N.R., Hartman, P.L., Auerbach, P.P., 1997. Changes in sensorimotor inhibition across the menstrual cycle: Implications for neuropsychiatric disorders. Biol. Psychiatry 41, 452–460. Varty, G.B., Higgins, G.A., 1995. Examination of drug-induced and isolation-induced disruptions of prepulse inhibition as models to screen antipsychotic drugs. Psychopharmacology 122, 15–26. Weinberger, D.R., 1987. Implications of normal brain development for the pathogenesis of schizophrenia. Arch. Gen. Psychiatry 44, 660–669. Weiss, I.C., Feldon, J., Domeney, A.M., 1999. Isolation rearinginduced disruption of prepulse inhibition: further evidence for fragility of the response. Behav. Pharmacol. 10, 139–149. Wilkinson, L.S., Killcross, S.S., Humby, T., Hall, F.S., Geyer, M.A., Robbins, T.W., 1994. Social isolation in the rat produces developmentally specific deficits in prepulse inhibition of the acoustic startle response without disrupting latent inhibition. Neuropsychopharmacology 10, 61–72. Zimmerberg, B., Shartrand, A.M., 1992. Temperature-dependent effects of maternal separation on growth, activity, and amphetamine sensitivity in the rat. Dev. Psychobiol. 25, 213–226.