HIPPOCAMPUS 24:1120–1128 (2014)
A Longitudinal Study of Stress-Induced Hippocampal Volume Changes in Mice that are Susceptible or Resilient to Chronic Social Defeat Yiu Chung Tse,1 Ixchel Montoya,1 Alice S. Wong,1 Axel Mathieu,1 Jennifer Lissemore,1 Diane C. Lagace,2 and Tak Pan Wong1,3,4*
ABSTRACT: Hippocampal shrinkage is a commonly found neuroanatomical change in stress-related mood disorders such as depression and posttraumatic stress disorders (PTSD). Since the onset and severity of these disorders have been found to be closely related to stressful life events, and as stress alone has been shown to reduce hippocampal volume in animal studies, vulnerability to mood disorders may be related to a susceptibility to stress-induced hippocampal shrinkage. However, a smaller hippocampal volume before stress exposure has also been suggested to confer vulnerability of stressed individuals to PTSD or depression. In this study, we examined the contribution of either innate hippocampal volume differences or hippocampal susceptibility to stress-induced shrinkage to the formation of stress-related psychopathology using longitudinal MRI measurements of hippocampal volume in inbred C57 mice before and after chronic social defeat stress. We found that only half of the stressed C57 mice were susceptible to stress and developed psychopathological behaviors such as social avoidance. The other half was resilient to stress and exhibited no social avoidance. Before exposure to stress, we observed a positive correlation between hippocampal volume and social avoidance. After chronic social defeat stress, we found significant increases in left hippocampal volume in resilient and nonstressed control mice. Intriguingly, this increase in hippocampal volume was not found in susceptible mice, suggesting an arrestment of hippocampal growth in these mice. Our findings suggest that both a susceptibility to stress-induced hippocampal volume changes and a larger hippocampus before stress exposure confer vulnerability to psychoC 2014 Wiley Periodicals, Inc. pathology after chronic stress. V KEY WORDS: chronic stress; individual differences; neuroimaging; resilience; social behavior
INTRODUCTION Certain anatomical characteristics of the hippocampus can be affected by stress (McEwen, 1999; Joels, 2009). Chronic exposure to stressors
1
Douglas Mental Health University Institute, McGill University, 6875 LaSalle Boulevard, Montreal, Quebec, Canada; 2 Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario, Canada; 3 Department of Psychiatry, McGill University, 6875 LaSalle Boulevard, Montreal, Quebec, Canada; 4 Department of Pharmacology and Therapeutics, McGill University, 6875 LaSalle Boulevard, Montreal, Quebec, Canada Grant sponsor: CIHR; Grant number: MOP-106649; Grant sponsor: Douglas CIC’s Pilot Grant (Douglas Institute). Y.C.T. I.M., and A.S.W. contributed equally to this work. *Correspondence to: Tak Pan Wong, Ph.D., Douglas Mental Health University Institute, 6875 LaSalle Blvd., Montreal, QC, Canada H4H 1R3. E-mail: [email protected] Accepted for publication 18 April 2014. DOI 10.1002/hipo.22296 Published online 20 April 2014 in Wiley Online Library (wileyonlinelibrary.com). C 2014 WILEY PERIODICALS, INC. V
can reduce the length and branching of hippocampal dendrites in rodents and primates (Uno et al., 1989; Magarinos and McEwen, 1995). The effects of stress on the hippocampus are likely mediated by glucocorticoids, since chronic glucocorticoid treatment produces similar dendritic shrinkage (Sapolsky et al., 1985; Sapolsky et al., 1990), and blockade of glucocorticoids synthesis abolishes the atrophic effect of stress on hippocampal dendrites (Magarinos and McEwen, 1995). Notably, stressful life events and trauma are closely associated with the onset and symptom severity of depression and post-traumatic stress disorder (PTSD, Breslau et al., 1991; Kendler et al., 1999; Lorenzetti et al., 2009; Muscatell et al., 2009), mood disorders that are frequently accompanied by hippocampal atrophy (McEwen and Sapolsky, 1995; Videbech and Ravnkilde, 2004; Smith, 2005). These findings support a hypothesis that susceptibility to stress-induced hippocampal atrophy underlies certain psychiatric symptoms of mood disorders. For instance, by comparing hippocampal volumes of combat-exposed veterans with or without PTSD, Gurvits et al. observed hippocampal atrophy only in veterans with PTSD (Gurvits et al., 1996). These findings suggest that traumatizing combat experiences may selectively reduce hippocampal volume in individuals who are vulnerable to PTSD. Vulnerability to stress-related mood disorders, however, could be related to innate differences in hippocampal volume that are present before stress exposure. For instance, while stressful rearing conditions exerted no impact on the hippocampal volume of primate offspring (Lyons et al., 2001), offspring with initially small hippocampal volumes exhibited stronger stress responses to the removal of their mothers, suggesting that a small hippocampus predisposed the primates to a higher stress response. Similarly, a small hippocampus may be a pre-trauma vulnerability trait for mood disorders in humans. In a twin study, hippocampal volumes of combat-exposed twins who developed PTSD and their stress-naive twins were discovered to be smaller than combat-exposed veterans who did not suffer from PTSD (Gilbertson et al., 2002). Teenage girls at high risk of developing depression because of their mothers’ medical history of recurrent depression have also been shown to have smaller hippocampal volumes than daughters of mothers with no history of
HIPPOCAMPAL VOLUMES IN MICE WITH DIFFERENT STRESS SUSCEPTIBILITY psychopathology (Chen et al., 2010). Finally, since most studies that revealed hippocampal atrophy in mood disorders are cross-sectional and based on findings obtained after the onset of psychiatric symptoms, the observed hippocampal volume changes could occur before the onset of these symptoms. Nonetheless, without longitudinal studies to reveal smaller hippocampal volumes before the onset of stress-related psychiatric symptoms in mood disorder patients, whether a small hippocampus increases the risk of mood disorders remains an untested hypothesis. Animal studies allow longitudinal measurements of hippocampal volume before and after the formation of stressinduced behavior changes. Using inbred C57 mice that display individual differences in stress susceptibility, we examined whether pre- or post-stress hippocampal volume alterations are related to the development of social avoidance in mice that are susceptible to chronic social defeat stress (Krishnan et al., 2007). Our findings suggest that the vulnerability to developing social avoidance is related to both prestress differences and post-stress changes in hippocampal volume.
MATERIALS AND METHODS Animals Male adult C57BL/6 mice (C57; 8-weeks old) and male CD1 retired breeders were ordered from Charles River Breeding Laboratories. Mice were housed at the Douglas Neurophenotyping Centre for at least a week before the first MRI scan. All care and use of animals was in accordance with the guidelines and policies of the Canadian Council on Animal Care to ameliorate suffering of animals in all work. This study (animal use protocol number no. 5935) was approved by the Facility Animal Care Committee of Douglas Institute, McGill University.
Structural MRI Procedures Mice were scanned by MRI on day 6 and day 22 (Fig. 1A). To facilitate detection of neuroanatomical structures, all C57 mice were injected with a MRI contrast agent, manganese chloride [MnCl2 at 30 mg/kg (Sinkkonen et al., 2004)]), on days 1, 3, 5, 17, 19, and 21 (i.e., three injections before each MRI scan). Mice displayed similar weight gain after each set of MnCl2 injections (control: 0.22 6 0.20 g, susceptible: 0.48 6 0.20 g, resilient: 0.61 6 0.43 g; F(2,15) 5 0.524, P 5 0.602). Twelve C57 mice were stressed by chronic social defeat between day 7 and 16, with half of them belonging to the susceptible group and half to the resilient group. Control mice followed the same schedule for the contrast agent injections and MRI scans without being exposed to the chronic social defeat protocol.
1121
MRI procedures Experimental mice were scanned with a 7 T Bruker Biospec 70/30 USR MRI scanner fitted with the Mini-Imaging Kit (BGA-12S gradient and shim system) and a dedicated mouse brain quadrature 1H surface coil before and after social defeat stress. Briefly, mice were anesthetized by isoflurane at an induction dosage of 5% and maintained at 1–2% during scanning using an integrated nose cone with a tooth bar. Position of the mice was secured using thin foam pads. Body temperature of the mice was maintained stable using a warm air flow (29 C). Respiration rate and depth were monitored throughout the scanning.
MRI parameters Whole brain images were acquired with the standard Bruker TrueFISP pulse sequence with eight different phase advance (DU) parameters resulting in eight images. The final image is the root mean square (RMS) of these eight images for a total scan time under 30 min (Ribot et al., 2013). The RMS composite image was generated using the MINC2 toolbox (http:// www.bic.mni.mcgill.ca/ServicesSoftware/MINC). Parameters: field of view (FOV): 1.8 cm 3 1.8 cm 3 0.9 cm; matrix: 128 3 128 3 64 points; resolution: 140 mm 3 140 mm 3 140 mm; echo time (TE)/repetition time (TR): 2.6/5.2 ms; flip angle 30 ; bandwidth: 50 kHz; phase advance array: 180 , 0 , 90 , 270 , 45 , 225 , 135 , 315 ; averages: 2; acquisition time: 3.5 min for each phase advance.
MRI data analysis Using MIPAV (http://mipav.cit.nih.gov/ (McAuliffe et al., 2001)), the whole brain and hippocampus were manually traced on MRI images by experimenters who were blind to group assignments. Tracing of the hippocampus and whole brain was confirmed by the same experimenter according to the mouse atlas (Franklin and Paxinos, 1997) and an online mouse MRI atlas (http://brainatlas.mbi.ufl.edu). About 28 to 31 horizontal sections that contain the hippocampus were traced (e.g., see Fig. 2). We did not find differences in the number of hippocampus-containing sections between mice (control: 29.7 6 0.37 sections, susceptible: 29.8 6 0.52 sections, resilient: 29.3 6 0.23 sections; F(2,15) 5 0.507, P 5 0.612). Brain tissue in sections that contain the hippocampus was traced for estimation of whole brain volume. In these measurements, structures like the olfactory bulbs, forebrain, and cerebellum were included. Hippocampal volume was normalized by whole brain volume in all experiments.
Chronic Social Defeat Stress Chronic social defeat was performed as previously described, with a test C57 mouse being placed into the space territorialized by a larger and aggressive CD1 retired breeder (Krishnan et al., 2007). CD1 mice were pre-trained to attack intruders with a latency of 1).
Statistics All data are presented as mean 6 SEM. For multiple-group comparisons of behavioral performance and normalized hippo-
campal volume, ANOVA followed by a post hoc Dunnett test (2-sided) were used. We used Pearson correlation coefficients to examine the relationship between initial hippocampal volume/ hippocampal volume changes and social behaviors. All statistical analyses were performed using SPSS statistics or SYSTAT.
RESULTS Susceptible, resilient, and nonstressed control mice spent similar amount of time in the interaction zone of the open field when the perforated enclosure was empty (Fig. 1C; F(2,15) 5 1.044, P 5 0.376). In comparison, when the enclosure contained a CD1 mouse, the susceptible mice spent significantly less time than control and resilient mice in the Hippocampus
1124
TSE ET AL.
interaction zone (F(2,15) 5 8.925, P 5 0.003; post hoc comparison: susceptible vs. control, P 5 0.010; susceptible vs. resilient, P 5 0.002). In addition, the social interaction ratio of susceptible mice was significantly lower than control and resilient mice (F(2,15) 5 10.58, P 5 0.001; post hoc comparison: susceptible vs. control, P 5 0.014; susceptible vs. resilient, P < 0.001). Comparing the corner ratios of these mice also revealed a significance between group difference (Fig. 1D; F(2,15) 5 5.466, P 5 0.016; post hoc comparison: susceptible vs. control, P 5 0.055; susceptible vs. resilient, P 5 0.012). Note that control and resilient mice displayed similar interaction and corner ratios. Finally, all mice displayed similar distances moved in the open field during social interaction screenings (Fig. 1E; repeated-measures ANOVA, interaction between the effect of animal groups and distances moved in the two screenings: F(2,15) 5 1.655, P 5 0.224), suggesting that the susceptibility to social avoidance was likely not related to changes in locomotory activities. Prior to exposure to stress, we measured hippocampal volume using MRI. We separately compared volume changes in left and right hippocampi, since previous studies have revealed left–right asymmetry of stress-induced anatomical changes of this structure (Liu et al., 2011). We did not find differences in the volume of the left (F(2,16) 5 2.43, P 5 0.122) and right hippocampus (F(2,16) 5 1.40, P 5 0.278) between the three animal groups. Interestingly, we found a significant correlation between pre-stress left hippocampal volume and the social interaction ratio (Fig. 3A; r18 5 20.577, P 5 0.012), suggesting an inverse relationship between pre-stress left hippocampal volume and post-stress social interaction. A similar and significant relationship was found between pre-stress right hippocampal volume and the social interaction ratio (Fig. 3B; r18 5 20.653, P 5 0.003). After chronic social defeat, we again did not find differences in the volume of the left (F(2,16) 5 0.10, P 5 0.910) and right hippocampus (F(2,16) 5 0.01, P 5 0.987) between the three animal groups. In addition, no significant relationship between post-stress hippocampal volume and the social interaction ratio was found (left hippocampus: r18 5 20.117, P 5 0.644, Fig. 3C; right hippocampus: r18 5 20.041, P 5 0.871, Fig. 3D). To determine whether the susceptibility to social defeat is related to stress-induced hippocampal volume changes, we computed percent changes in hippocampal volume between the two scans of each mouse (Fig. 4A). We observed an increase in left hippocampal volume between the two MRI scans in control (4.11 6 1.31%) and resilient groups (4.48 6 2.29%), but not in the susceptible group (21.82 6 1.79%; ANOVA comparison of hippocampal volume changes between the three groups: F(2,15) 5 4.414, P 5 0.031; post hoc comparison: susceptible vs. controls, P 5 0.045; susceptible vs. resilient, P 5 0.033; Fig. 4B). We also found a significant correlation between left hippocampal volume changes and the social interaction ratio (Fig. 4C; r18 5 0.588, P 5 0.010), suggesting a direct relationship between left hippocampal volume changes and post-stress social interaction. Similar relationships between right hippocampal volume and social interaction were also Hippocampus
FIGURE 3. A: Scatter plots show the relationship between the volume of the left hippocampus of control (white circles; n 5 6), susceptible (black circles; n 5 6), and resilient mice (gray circles; n 5 6) that were measured in the first MRI scan (pre-stress) and the post-stress social interaction ratio. *P < 0.05, Pearson correlation. B: Scatter plots show the relationship between the volume of the right hippocampus measured in the first MRI scan (pre-stress) and the post-stress social interaction ratio. *P < 0.05, Pearson correlation. C: Scatter plots show the relationship between the volume of the left hippocampus measured in the second MRI scan (post-stress) and the post-stress social interaction ratio. D: Scatter plots show the relationship between the volume of the right hippocampus measured in the second MRI scan (post-stress) and the post-stress social interaction ratio.
observed (Fig. 4D). Although comparing percent changes in right hippocampal volume between the three groups revealed no significant changes (Fig. 4E; F(2,15) 5 1.876, P 5 0.187), there was a significant correlation between right hippocampal volume changes and the social interaction ratio (Fig. 4F; r18 5 0.701, P 5 0.001). The lack of increase in the hippocampal volume of susceptible mice between the two MRI scans could be due to a reported stress-induced weight loss in these mice (Krishnan et al., 2007). However, comparing the weights of control, susceptible, and resilient mice that were measured before the 1st (Fig. 5A; F(2,15) 5 1.878, P 5 0.187) and 2nd MRI scans (F(2,15) 5 1.355, P 5 0.288) revealed no between-group differences. In addition, mice from these three groups displayed similar weight gains between the two MRI scans (Fig. 5B; F(2,15) 5 0.120, P 5 0.888).
HIPPOCAMPAL VOLUMES IN MICE WITH DIFFERENT STRESS SUSCEPTIBILITY
FIGURE 4. A: Scatter plots show changes in normalized left hippocampal volumes of control (white circles; n 5 6), susceptible (black circles; n 5 6), and resilient mice (gray circles; n 5 6) between the two MRI scans that were performed 15 days apart. B: Histograms summarize the percent change in left hippocampal volume of control, susceptible, and resilient mice between the two MRI scans. Only susceptible and resilient mice were stressed by chronic social defeat between these scans. *P < 0.05, post hoc Dunnett test after ANOVA. C: Scatter plots show the relationship between the percent change in left hippocampal volume and the
DISCUSSION This longitudinal MRI study revealed significant differences in stress-induced changes in hippocampal volume between mice that are susceptible and mice that are resilient to chronic social defeat stress. We found an inverse relationship between hippocampal volume measured prior to stress and the social interaction ratio observed after chronic social defeat, suggesting that a larger hippocampus confers higher risk to developing social avoidance after chronic stress. Resilient mice were similar to controls, demonstrating both social interaction behavior and an increase in the volume of the left hippocampus following social defeat stress. Notably, the increase in hippocampal volume before and after the stress was not found in susceptible mice that had a “defeated” phenotype, characterized by social avoidance after chronic stress. Using a within-subject design, to our knowledge, our findings are the first to provide evidence that both pre-stress and post-stress hippocampal volume differences contribute to the development of stress-induced social avoidance.
1125
post-stress social interaction ratio in control (white circles), susceptible (black circles) and resilient mice (gray circles). *P < 0.05, Pearson correlation. D: Scatter plots show changes in normalized right hippocampal volume of control, susceptible, and resilient mice between the two MRI scans. E: Histograms show the percent change in right hippocampal volume of control, susceptible, and resilient mice between the two MRI scans. F: Scatter plots show the relationship between the percent change in right hippocampal volume and the post-stress social interaction ratio. *P < 0.05, Pearson correlation.
We found an inverse relationship between hippocampal volume that was measured prior to stress exposure and the social interaction ratio, suggesting that the risk of developing social avoidance after chronic social defeat is higher in mice with a large hippocampus before stress exposure. A small hippocampus has been suggested to confer vulnerability to PTSD and depression (Gilbertson et al., 2002; Chen et al., 2010). The hippocampus is important in the negative feedback control of the hypothalamic–pituitary–adrenal axis (Herman et al., 2005), and thus a smaller hippocampus could result in an impairment of such feedback and an overproduction of corticosteroids and high stress susceptibility. Nevertheless, consistent with our findings, evidence exists for a link between bigger hippocampal volume and higher stress susceptibility. In healthy young individuals, Pruessner et al. have shown an association between higher social stress-induced cortisol release and larger hippocampal volumes (Pruessner et al., 2007). In older adults, the number of stressful life events experienced has been shown to be associated not only with the risk of developing depression, but also with larger hippocampal volumes (Zannas et al., Hippocampus
1126
TSE ET AL.
FIGURE 5. A: Histograms summarize the weight of control (n 5 6), susceptible (n 5 6), and resilient mice (n 5 6) right before the first (a) and the second MRI scans (b). B: Histograms summarize the weight gain of control, susceptible, and resilient mice between two MRI scans.
2013). Findings such as these question the prevalent assumption that a congenitally small hippocampus increases an individual’s risk for stress-induced neuropathologies. Since a reduction in hippocampal volume associates with cognitive impairment in different brain diseases (Sapolsky, 2000), larger hippocampal volumes in susceptible mice may lead to higher hippocampus-dependent cognitive function. In a human study, subjects with large hippocampal volumes outperformed those with small hippocampal volumes in contextual fear conditioning (Pohlack et al., 2012). The potential increase in hippocampal function in susceptible mice may enhance their fear learning, which could result in stronger fear responses during social interaction than resilient and control mice. Future studies could compare hippocampus-dependent contextual fear conditioning in susceptible and resilient mice to find out if a congenitally bigger hippocampus enhances fear learning in susceptible mice only. We found that between the two MRI scans taken before or after the stress, both nonstressed control and stressed resilient mice displayed increases in left hippocampal volumes. Developmental increases of whole brain and hippocampal volume in rodents has been previously reported. Lau et al. found that between 2.5 and 9 months of age, mice show significant increases in brain volume (Lau et al., 2008). Brain volumes of rodents continuously increase up to almost one year after birth (Maheswaran et al., 2009), perhaps due to delayed cessation of cranial bone growth in rodents when compared with other mammalian species (Kilborn et al., 2002). However, it is important to note that not all longitudinal studies of brain volume in rodents have revealed hippocampal growth. Lee et al. showed that while chronic restraint stress (6 h per day for 3 weeks) decreased hippocampal volume in adult SpragueDawley (SD) rats, longitudinal MRI scanning of nonstressed control rats before and after the same 3-week period of time revealed no changes in hippocampal volume (Lee et al., 2009). The discrepancy of findings between Lee et al. and the current study could be related to differences in the species, the age of the animals, and the type of stressor used. Notably, Lee et al. Hippocampus
used only seven sections to estimate hippocampal volume. The use of 30 sections for the same purpose in the present study may have allowed us to achieve a more precise estimation of hippocampal volume. Although both susceptible and resilient mice were stressed by identical social defeat procedures, only resilient mice exhibited an increase or growth in hippocampal volume after chronic social defeat. Since similar increases in hippocampal volumes were found in nonstressed control mice, our findings suggest that normal hippocampal growth was arrested by chronic stress in susceptible mice only. In parallel to our findings, Isgor et al. performed morphological analyses of fixed brain tissue and revealed hippocampal growth in SD rats within the 3rd postnatal month (Isgor et al., 2004). Interestingly, they showed that hippocampal growth was arrested by chronic variable stress. The lack of hippocampal growth in susceptible mice was not due to stress-related weight loss or reduced food intake, since mice from all three groups displayed similar weight gain between the two MRI scans. Notably, although we found no differences in hippocampal volumes between the three animal groups before stress exposure, we observed a direct correlation between pre-stress hippocampal volume and the social interaction ratio (Fig. 3). We therefore cannot rule out an alternative explanation that hippocampi of susceptible mice have reached their maximum sizes, so no further growth is possible. Future experiments should examine longitudinal changes in hippocampal volumes of susceptible and resilient mice to establish whether susceptible mice reach the maximum hippocampal volume at the age tested in this study (2 months) and find out whether differences in stressinduced hippocampal volume can be maintained between older susceptible and resilient mice (9–12 months) when hippocampal growth subsides. Our findings suggest that there are individual differences in the susceptibility to stress-induced hippocampal volume changes. Individual differences in hippocampal volume changes have similarly been observed in aging studies. The hippocampus has been shown to shrink faster with aging in hypertensive individuals than in normotensive individuals in one longitudinal study (Raz et al., 2005). Serial measurements of hippocampal volume during aging in individuals with different degrees of lifetime mental activity also suggested a protective effect of complex mental activity on the rate of aging-related hippocampal shrinkage (Valenzuela et al., 2008). Individual differences in hippocampal susceptibility to stress-induced volume changes could underlie the formation of psychopathologies in susceptible animals (e.g., social avoidance) and in human patients with mood disorders. Although factors underlying individual differences in stress-induced hippocampal volume changes remain unclear, genetic (Krishnan et al., 2007) and environmental factors (Meaney, 2001) that have been associated with individual differences in stress susceptibility could be responsible. Given the high variability of hippocampal volume in healthy humans (Lupien et al., 2007), individual differences in innate hippocampal volume may contribute to the variability in people’s susceptibility to stress and risk for mood disorders.
HIPPOCAMPAL VOLUMES IN MICE WITH DIFFERENT STRESS SUSCEPTIBILITY We showed that the impact of chronic stress on hippocampal volume in susceptible mice may be asymmetric, with the left hippocampus being more vulnerable to stress. Stress has been shown to induce a larger reduction in volume of the left hippocampus than the right hippocampus in adult rats (Liu et al., 2011). Lateralization of stress susceptibility in the hippocampus could be related to asymmetric expression of corticosteroid receptors in the left and right hippocampus (Hou et al., 2013). Accordingly, the volume of the dentate gyrus of the left hippocampus, but not of the right hippocampus, negatively correlated with saliva cortisol levels in stressed pigs (van der Beek et al., 2004), suggesting an asymmetry in the impact of cortisol on hippocampal volume. However, not all crosssectional studies have revealed specific corticosteroid-induced shrinkage of the left hippocampus. For instance, chronic corticosterone treatment induced a delayed (two months after cessation of corticosterone treatment) shrinkage of the right but not the left hippocampus of adult Long Evans rats in one study (Zach et al., 2010). Additionally, Carrion et al. found that cortisol levels and PTSD symptoms correlate with shrinkage of the right instead of the left hippocampus in children that had experienced trauma (Carrion et al., 2007). Apart from corticosteroid receptor expression, lateralization of stress susceptibility may be related to anatomical (Diamond et al., 1982) or neurochemical asymmetry (Valdes et al., 1981) between the left and right hippocampus in rodents. In conclusion, our findings suggest that the vulnerability to psychopathological outcome after experiencing chronic social defeat stress in rodents is related to both hippocampal susceptibility to stress-induced changes in volume and differences in hippocampal volume before stress exposure. An understanding of the biological factors that underlie susceptibility to stressinduced hippocampal volume changes could be therapeutically useful for increasing resistance to stress-related insults.
Acknowledgments The authors report no biomedical financial interests or potential conflicts of interest. Authors would like to thank Ms. Juliana Hutter and Ms. Jasmine Domagala for their technical assistance.
REFERENCES Breslau N, Davis GC, Andreski P, Peterson E. 1991. Traumatic events and posttraumatic stress disorder in an urban population of young adults. Arch Gen Psychiatry 48:216–222. Carrion VG, Weems CF, Reiss AL. 2007. Stress predicts brain changes in children: A pilot longitudinal study on youth stress, posttraumatic stress disorder, and the hippocampus. Pediatrics 119:509– 516. Chen MC, Hamilton JP, Gotlib IH. 2010. Decreased hippocampal volume in healthy girls at risk of depression. Arch Gen Psychiatry 67:270–276.
1127
Diamond MC, Murphy GM Jr, Akiyama K, Johnson RE. 1982. Morphologic hippocampal asymmetry in male and female rats. Exp Neurol 76:553–565. Franklin KBJ, Paxinos G. 1997. The Mouse Brain in Stereotaxic Coordinates. San Diego: Academic Press. Gilbertson MW, Shenton ME, Ciszewski A, Kasai K, Lasko NB, Orr SP, Pitman RK. 2002. Smaller hippocampal volume predicts pathologic vulnerability to psychological trauma. Nat Neurosci 5:1242–1247. Gurvits TV, Shenton ME, Hokama H, Ohta H, Lasko NB, Gilbertson MW, Orr SP, Kikinis R, Jolesz FA, McCarley RW, Pitman RK. 1996. Magnetic resonance imaging study of hippocampal volume in chronic, combat-related posttraumatic stress disorder. Biol Psychiatry 40:1091–1099. Herman JP, Ostrander MM, Mueller NK, Figueiredo H. 2005. Limbic system mechanisms of stress regulation: Hypothalamopituitary-adrenocortical axis. Prog Neuropsychopharmacol Biol Psychiatry 29:1201–1213. Hou G, Yang X, Yuan TF. 2013. Hippocampal asymmetry: Differences in structures and functions. Neurochem Res 38:453–460. Isgor C, Kabbaj M, Akil H, Watson SJ. 2004. Delayed effects of chronic variable stress during peripubertal-juvenile period on hippocampal morphology and on cognitive and stress axis functions in rats. Hippocampus 14:636–648. Joels M. 2009. Stress, the hippocampus, and epilepsy. Epilepsia 50: 586–597. Kendler KS, Karkowski LM, Prescott CA. 1999. Causal relationship between stressful life events and the onset of major depression. Am J Psychiatry 156:837–841. Kilborn SH, Trudel G, Uhthoff H. 2002. Review of growth plate closure compared with age at sexual maturity and lifespan in laboratory animals. Contemp Top Lab Anim Sci 41:21–26. Krishnan V, Han MH, Graham DL, Berton O, Renthal W, Russo SJ, Laplant Q, Graham A, Lutter M, Lagace DC, Ghose S, Reister R, Tannous P, Green TA, Neve RL, Chakravarty S, Kumar A, Eisch AJ, Self DW, Lee FS, Tamminga CA, Cooper DC, Gershenfeld HK, Nestler EJ. 2007. Molecular adaptations underlying susceptibility and resistance to social defeat in brain reward regions. Cell 131:391–404. Lau JC, Lerch JP, Sled JG, Henkelman RM, Evans AC, Bedell BJ. 2008. Longitudinal neuroanatomical changes determined by deformation-based morphometry in a mouse model of Alzheimer’s disease. Neuroimage 42:19–27. Lee T, Jarome T, Li SJ, Kim JJ, Helmstetter FJ. 2009. Chronic stress selectively reduces hippocampal volume in rats: A longitudinal magnetic resonance imaging study. Neuroreport 20:1554–1558. Liu W, Shu XJ, Chen FY, Zhu C, Sun XH, Liu LJ, Ai YX, Li YG, Zhao H. 2011. Tianeptine reverses stress-induced asymmetrical hippocampal volume and N-acetylaspartate loss in rats: An in vivo study. Psychiatry Res 194:385–392. Lorenzetti V, Allen NB, Fornito A, Yucel M. 2009. Structural brain abnormalities in major depressive disorder: A selective review of recent MRI studies. J Affect Disord 117:1–17. Lupien SJ, Evans A, Lord C, Miles J, Pruessner M, Pike B, Pruessner JC. 2007. Hippocampal volume is as variable in young as in older adults: Implications for the notion of hippocampal atrophy in humans. Neuroimage 34:479–485. Lyons DM, Yang C, Sawyer-Glover AM, Moseley ME, Schatzberg AF. 2001. Early life stress and inherited variation in monkey hippocampal volumes. Arch Gen Psychiatry 58:1145–1151. Magarinos AM, McEwen BS. 1995. Stress-induced atrophy of apical dendrites of hippocampal CA3c neurons: Involvement of glucocorticoid secretion and excitatory amino acid receptors. Neuroscience 69:89–98. Maheswaran S, Barjat H, Rueckert D, Bate ST, Howlett DR, Tilling L, Smart SC, Pohlmann A, Richardson JC, Hartkens T, Hill DL, Upton N, Hajnal JV, James MF. 2009. Longitudinal regional brain Hippocampus
1128
TSE ET AL.
volume changes quantified in normal aging and Alzheimer’s APP x PS1 mice using MRI. Brain Res 1270:19–32. McAuliffe M, Lalonde F, McGarry D, Gandler W, Csaky K, Trus B. Medical image processing, analysis and visualization in clinical research. 2001. In Proceedings of the 14th IEEE Symposium on Computer-Based Medical Systems (CBMS 2001), Bethesda, MD. pp 381–386. McEwen BS. 1999. Stress and hippocampal plasticity. Annu Rev Neurosci 22:105–122. McEwen BS, Sapolsky RM. 1995. Stress and cognitive function. Curr Opin Neurobiol 5:205–216. Meaney MJ. 2001. Maternal care, gene expression, and the transmission of individual differences in stress reactivity across generations. Annu Rev Neurosci 24:1161–1192. Muscatell KA, Slavich GM, Monroe SM, Gotlib IH. 2009. Stressful life events, chronic difficulties, and the symptoms of clinical depression. J Nerv Ment Dis 197:154–160. Pohlack ST, Nees F, Liebscher C, Cacciaglia R, Diener SJ, Ridder S, Woermann FG, Flor H. 2012. Hippocampal but not amygdalar volume affects contextual fear conditioning in humans. Hum Brain Mapp 33:478–488. Pruessner M, Pruessner JC, Hellhammer DH, Bruce PG, Lupien SJ. 2007. The associations among hippocampal volume, cortisol reactivity, and memory performance in healthy young men. Psychiatry Res 155:1–10. Raz N, Lindenberger U, Rodrigue KM, Kennedy KM, Head D, Williamson A, Dahle C, Gerstorf D, Acker JD. 2005. Regional brain changes in aging healthy adults: General trends, individual differences and modifiers. Cereb Cortex 15:1676–1689. Ribot EJ, Pourtau L, Massot P, Voisin P, Thiaudiere E, Franconi JM, Miraux S. 2013. 3D TrueFISP MRI provides accurate longitudinal measurements of glioma volumes in mice. In: Chen CC, editor. Advances in the Biology, Imaging and Therapies for Glioblastoma. Rijeka: InTech. pp 125–144. Sapolsky RM. 2000. Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders. Arch Gen Psychiatry 57:925–935.
Hippocampus
Sapolsky RM, Krey LC, McEwen BS. 1985. Prolonged glucocorticoid exposure reduces hippocampal neuron number: implications for aging. J Neurosci 5:1222–1227. Sapolsky RM, Uno H, Rebert CS, Finch CE. 1990. Hippocampal damage associated with prolonged glucocorticoid exposure in primates. J Neurosci 10:2897–2902. Sinkkonen ST, Vekovischeva OY, Moykkynen T, Ogris W, Sieghart W, Wisden W, Korpi ER. 2004. Behavioural correlates of an altered balance between synaptic and extrasynaptic GABAAergic inhibition in a mouse model. Eur J Neurosci 20:2168–2178. Smith ME. 2005. Bilateral hippocampal volume reduction in adults with post-traumatic stress disorder: A meta-analysis of structural MRI studies. Hippocampus 15:798–807. Uno H, Tarara R, Else JG, Suleman MA, Sapolsky RM. 1989. Hippocampal damage associated with prolonged and fatal stress in primates. J Neurosci 9:1705–1711. Valdes JJ, Mactutus CF, Cory RN, Cameron WR. 1981. Lateralization of norepinephrine, serotonin and choline uptake into hippocampal synaptosomes of sinistral rats. Physiol Behav 27:381–383. Valenzuela MJ, Sachdev P, Wen W, Chen X, Brodaty H. 2008. Lifespan mental activity predicts diminished rate of hippocampal atrophy. PLoS One 3:e2598. van der Beek EM, Wiegant VM, Schouten WG, van Eerdenburg FJ, Loijens LW, van der Plas C, Benning MA, de VH, de Kloet ER, Lucassen PJ. 2004. Neuronal number, volume, and apoptosis of the left dentate gyrus of chronically stressed pigs correlate negatively with basal saliva cortisol levels. Hippocampus 14:688–700. Videbech P, Ravnkilde B. 2004. Hippocampal volume and depression: a meta-analysis of MRI studies. Am J Psychiatry 161:1957–1966. Zach P, Mrzilkova J, Rezacova L, Stuchlik A, Vales K. 2010. Delayed effects of elevated corticosterone level on volume of hippocampal formation in laboratory rat. Physiol Res 59:985–996. Zannas AS, McQuoid DR, Payne ME, Steffens DC, MacFall JR, Ashley-Koch A, Taylor WD. 2013. Negative life stress and longitudinal hippocampal volume changes in older adults with and without depression. J Psychiatr Res 47:829–834.
A Longitudinal Study of Stress-Induced Hippocampal Volume Changes in Mice that are Susceptible or Resilient to Chronic Social Defeat Yiu Chung Tse,1 Ixchel Montoya,1 Alice S. Wong,1 Axel Mathieu,1 Jennifer Lissemore,1 Diane C. Lagace,2 and Tak Pan Wong1,3,4*
ABSTRACT: Hippocampal shrinkage is a commonly found neuroanatomical change in stress-related mood disorders such as depression and posttraumatic stress disorders (PTSD). Since the onset and severity of these disorders have been found to be closely related to stressful life events, and as stress alone has been shown to reduce hippocampal volume in animal studies, vulnerability to mood disorders may be related to a susceptibility to stress-induced hippocampal shrinkage. However, a smaller hippocampal volume before stress exposure has also been suggested to confer vulnerability of stressed individuals to PTSD or depression. In this study, we examined the contribution of either innate hippocampal volume differences or hippocampal susceptibility to stress-induced shrinkage to the formation of stress-related psychopathology using longitudinal MRI measurements of hippocampal volume in inbred C57 mice before and after chronic social defeat stress. We found that only half of the stressed C57 mice were susceptible to stress and developed psychopathological behaviors such as social avoidance. The other half was resilient to stress and exhibited no social avoidance. Before exposure to stress, we observed a positive correlation between hippocampal volume and social avoidance. After chronic social defeat stress, we found significant increases in left hippocampal volume in resilient and nonstressed control mice. Intriguingly, this increase in hippocampal volume was not found in susceptible mice, suggesting an arrestment of hippocampal growth in these mice. Our findings suggest that both a susceptibility to stress-induced hippocampal volume changes and a larger hippocampus before stress exposure confer vulnerability to psychoC 2014 Wiley Periodicals, Inc. pathology after chronic stress. V KEY WORDS: chronic stress; individual differences; neuroimaging; resilience; social behavior
INTRODUCTION Certain anatomical characteristics of the hippocampus can be affected by stress (McEwen, 1999; Joels, 2009). Chronic exposure to stressors
1
Douglas Mental Health University Institute, McGill University, 6875 LaSalle Boulevard, Montreal, Quebec, Canada; 2 Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario, Canada; 3 Department of Psychiatry, McGill University, 6875 LaSalle Boulevard, Montreal, Quebec, Canada; 4 Department of Pharmacology and Therapeutics, McGill University, 6875 LaSalle Boulevard, Montreal, Quebec, Canada Grant sponsor: CIHR; Grant number: MOP-106649; Grant sponsor: Douglas CIC’s Pilot Grant (Douglas Institute). Y.C.T. I.M., and A.S.W. contributed equally to this work. *Correspondence to: Tak Pan Wong, Ph.D., Douglas Mental Health University Institute, 6875 LaSalle Blvd., Montreal, QC, Canada H4H 1R3. E-mail: [email protected] Accepted for publication 18 April 2014. DOI 10.1002/hipo.22296 Published online 20 April 2014 in Wiley Online Library (wileyonlinelibrary.com). C 2014 WILEY PERIODICALS, INC. V
can reduce the length and branching of hippocampal dendrites in rodents and primates (Uno et al., 1989; Magarinos and McEwen, 1995). The effects of stress on the hippocampus are likely mediated by glucocorticoids, since chronic glucocorticoid treatment produces similar dendritic shrinkage (Sapolsky et al., 1985; Sapolsky et al., 1990), and blockade of glucocorticoids synthesis abolishes the atrophic effect of stress on hippocampal dendrites (Magarinos and McEwen, 1995). Notably, stressful life events and trauma are closely associated with the onset and symptom severity of depression and post-traumatic stress disorder (PTSD, Breslau et al., 1991; Kendler et al., 1999; Lorenzetti et al., 2009; Muscatell et al., 2009), mood disorders that are frequently accompanied by hippocampal atrophy (McEwen and Sapolsky, 1995; Videbech and Ravnkilde, 2004; Smith, 2005). These findings support a hypothesis that susceptibility to stress-induced hippocampal atrophy underlies certain psychiatric symptoms of mood disorders. For instance, by comparing hippocampal volumes of combat-exposed veterans with or without PTSD, Gurvits et al. observed hippocampal atrophy only in veterans with PTSD (Gurvits et al., 1996). These findings suggest that traumatizing combat experiences may selectively reduce hippocampal volume in individuals who are vulnerable to PTSD. Vulnerability to stress-related mood disorders, however, could be related to innate differences in hippocampal volume that are present before stress exposure. For instance, while stressful rearing conditions exerted no impact on the hippocampal volume of primate offspring (Lyons et al., 2001), offspring with initially small hippocampal volumes exhibited stronger stress responses to the removal of their mothers, suggesting that a small hippocampus predisposed the primates to a higher stress response. Similarly, a small hippocampus may be a pre-trauma vulnerability trait for mood disorders in humans. In a twin study, hippocampal volumes of combat-exposed twins who developed PTSD and their stress-naive twins were discovered to be smaller than combat-exposed veterans who did not suffer from PTSD (Gilbertson et al., 2002). Teenage girls at high risk of developing depression because of their mothers’ medical history of recurrent depression have also been shown to have smaller hippocampal volumes than daughters of mothers with no history of
HIPPOCAMPAL VOLUMES IN MICE WITH DIFFERENT STRESS SUSCEPTIBILITY psychopathology (Chen et al., 2010). Finally, since most studies that revealed hippocampal atrophy in mood disorders are cross-sectional and based on findings obtained after the onset of psychiatric symptoms, the observed hippocampal volume changes could occur before the onset of these symptoms. Nonetheless, without longitudinal studies to reveal smaller hippocampal volumes before the onset of stress-related psychiatric symptoms in mood disorder patients, whether a small hippocampus increases the risk of mood disorders remains an untested hypothesis. Animal studies allow longitudinal measurements of hippocampal volume before and after the formation of stressinduced behavior changes. Using inbred C57 mice that display individual differences in stress susceptibility, we examined whether pre- or post-stress hippocampal volume alterations are related to the development of social avoidance in mice that are susceptible to chronic social defeat stress (Krishnan et al., 2007). Our findings suggest that the vulnerability to developing social avoidance is related to both prestress differences and post-stress changes in hippocampal volume.
MATERIALS AND METHODS Animals Male adult C57BL/6 mice (C57; 8-weeks old) and male CD1 retired breeders were ordered from Charles River Breeding Laboratories. Mice were housed at the Douglas Neurophenotyping Centre for at least a week before the first MRI scan. All care and use of animals was in accordance with the guidelines and policies of the Canadian Council on Animal Care to ameliorate suffering of animals in all work. This study (animal use protocol number no. 5935) was approved by the Facility Animal Care Committee of Douglas Institute, McGill University.
Structural MRI Procedures Mice were scanned by MRI on day 6 and day 22 (Fig. 1A). To facilitate detection of neuroanatomical structures, all C57 mice were injected with a MRI contrast agent, manganese chloride [MnCl2 at 30 mg/kg (Sinkkonen et al., 2004)]), on days 1, 3, 5, 17, 19, and 21 (i.e., three injections before each MRI scan). Mice displayed similar weight gain after each set of MnCl2 injections (control: 0.22 6 0.20 g, susceptible: 0.48 6 0.20 g, resilient: 0.61 6 0.43 g; F(2,15) 5 0.524, P 5 0.602). Twelve C57 mice were stressed by chronic social defeat between day 7 and 16, with half of them belonging to the susceptible group and half to the resilient group. Control mice followed the same schedule for the contrast agent injections and MRI scans without being exposed to the chronic social defeat protocol.
1121
MRI procedures Experimental mice were scanned with a 7 T Bruker Biospec 70/30 USR MRI scanner fitted with the Mini-Imaging Kit (BGA-12S gradient and shim system) and a dedicated mouse brain quadrature 1H surface coil before and after social defeat stress. Briefly, mice were anesthetized by isoflurane at an induction dosage of 5% and maintained at 1–2% during scanning using an integrated nose cone with a tooth bar. Position of the mice was secured using thin foam pads. Body temperature of the mice was maintained stable using a warm air flow (29 C). Respiration rate and depth were monitored throughout the scanning.
MRI parameters Whole brain images were acquired with the standard Bruker TrueFISP pulse sequence with eight different phase advance (DU) parameters resulting in eight images. The final image is the root mean square (RMS) of these eight images for a total scan time under 30 min (Ribot et al., 2013). The RMS composite image was generated using the MINC2 toolbox (http:// www.bic.mni.mcgill.ca/ServicesSoftware/MINC). Parameters: field of view (FOV): 1.8 cm 3 1.8 cm 3 0.9 cm; matrix: 128 3 128 3 64 points; resolution: 140 mm 3 140 mm 3 140 mm; echo time (TE)/repetition time (TR): 2.6/5.2 ms; flip angle 30 ; bandwidth: 50 kHz; phase advance array: 180 , 0 , 90 , 270 , 45 , 225 , 135 , 315 ; averages: 2; acquisition time: 3.5 min for each phase advance.
MRI data analysis Using MIPAV (http://mipav.cit.nih.gov/ (McAuliffe et al., 2001)), the whole brain and hippocampus were manually traced on MRI images by experimenters who were blind to group assignments. Tracing of the hippocampus and whole brain was confirmed by the same experimenter according to the mouse atlas (Franklin and Paxinos, 1997) and an online mouse MRI atlas (http://brainatlas.mbi.ufl.edu). About 28 to 31 horizontal sections that contain the hippocampus were traced (e.g., see Fig. 2). We did not find differences in the number of hippocampus-containing sections between mice (control: 29.7 6 0.37 sections, susceptible: 29.8 6 0.52 sections, resilient: 29.3 6 0.23 sections; F(2,15) 5 0.507, P 5 0.612). Brain tissue in sections that contain the hippocampus was traced for estimation of whole brain volume. In these measurements, structures like the olfactory bulbs, forebrain, and cerebellum were included. Hippocampal volume was normalized by whole brain volume in all experiments.
Chronic Social Defeat Stress Chronic social defeat was performed as previously described, with a test C57 mouse being placed into the space territorialized by a larger and aggressive CD1 retired breeder (Krishnan et al., 2007). CD1 mice were pre-trained to attack intruders with a latency of 1).
Statistics All data are presented as mean 6 SEM. For multiple-group comparisons of behavioral performance and normalized hippo-
campal volume, ANOVA followed by a post hoc Dunnett test (2-sided) were used. We used Pearson correlation coefficients to examine the relationship between initial hippocampal volume/ hippocampal volume changes and social behaviors. All statistical analyses were performed using SPSS statistics or SYSTAT.
RESULTS Susceptible, resilient, and nonstressed control mice spent similar amount of time in the interaction zone of the open field when the perforated enclosure was empty (Fig. 1C; F(2,15) 5 1.044, P 5 0.376). In comparison, when the enclosure contained a CD1 mouse, the susceptible mice spent significantly less time than control and resilient mice in the Hippocampus
1124
TSE ET AL.
interaction zone (F(2,15) 5 8.925, P 5 0.003; post hoc comparison: susceptible vs. control, P 5 0.010; susceptible vs. resilient, P 5 0.002). In addition, the social interaction ratio of susceptible mice was significantly lower than control and resilient mice (F(2,15) 5 10.58, P 5 0.001; post hoc comparison: susceptible vs. control, P 5 0.014; susceptible vs. resilient, P < 0.001). Comparing the corner ratios of these mice also revealed a significance between group difference (Fig. 1D; F(2,15) 5 5.466, P 5 0.016; post hoc comparison: susceptible vs. control, P 5 0.055; susceptible vs. resilient, P 5 0.012). Note that control and resilient mice displayed similar interaction and corner ratios. Finally, all mice displayed similar distances moved in the open field during social interaction screenings (Fig. 1E; repeated-measures ANOVA, interaction between the effect of animal groups and distances moved in the two screenings: F(2,15) 5 1.655, P 5 0.224), suggesting that the susceptibility to social avoidance was likely not related to changes in locomotory activities. Prior to exposure to stress, we measured hippocampal volume using MRI. We separately compared volume changes in left and right hippocampi, since previous studies have revealed left–right asymmetry of stress-induced anatomical changes of this structure (Liu et al., 2011). We did not find differences in the volume of the left (F(2,16) 5 2.43, P 5 0.122) and right hippocampus (F(2,16) 5 1.40, P 5 0.278) between the three animal groups. Interestingly, we found a significant correlation between pre-stress left hippocampal volume and the social interaction ratio (Fig. 3A; r18 5 20.577, P 5 0.012), suggesting an inverse relationship between pre-stress left hippocampal volume and post-stress social interaction. A similar and significant relationship was found between pre-stress right hippocampal volume and the social interaction ratio (Fig. 3B; r18 5 20.653, P 5 0.003). After chronic social defeat, we again did not find differences in the volume of the left (F(2,16) 5 0.10, P 5 0.910) and right hippocampus (F(2,16) 5 0.01, P 5 0.987) between the three animal groups. In addition, no significant relationship between post-stress hippocampal volume and the social interaction ratio was found (left hippocampus: r18 5 20.117, P 5 0.644, Fig. 3C; right hippocampus: r18 5 20.041, P 5 0.871, Fig. 3D). To determine whether the susceptibility to social defeat is related to stress-induced hippocampal volume changes, we computed percent changes in hippocampal volume between the two scans of each mouse (Fig. 4A). We observed an increase in left hippocampal volume between the two MRI scans in control (4.11 6 1.31%) and resilient groups (4.48 6 2.29%), but not in the susceptible group (21.82 6 1.79%; ANOVA comparison of hippocampal volume changes between the three groups: F(2,15) 5 4.414, P 5 0.031; post hoc comparison: susceptible vs. controls, P 5 0.045; susceptible vs. resilient, P 5 0.033; Fig. 4B). We also found a significant correlation between left hippocampal volume changes and the social interaction ratio (Fig. 4C; r18 5 0.588, P 5 0.010), suggesting a direct relationship between left hippocampal volume changes and post-stress social interaction. Similar relationships between right hippocampal volume and social interaction were also Hippocampus
FIGURE 3. A: Scatter plots show the relationship between the volume of the left hippocampus of control (white circles; n 5 6), susceptible (black circles; n 5 6), and resilient mice (gray circles; n 5 6) that were measured in the first MRI scan (pre-stress) and the post-stress social interaction ratio. *P < 0.05, Pearson correlation. B: Scatter plots show the relationship between the volume of the right hippocampus measured in the first MRI scan (pre-stress) and the post-stress social interaction ratio. *P < 0.05, Pearson correlation. C: Scatter plots show the relationship between the volume of the left hippocampus measured in the second MRI scan (post-stress) and the post-stress social interaction ratio. D: Scatter plots show the relationship between the volume of the right hippocampus measured in the second MRI scan (post-stress) and the post-stress social interaction ratio.
observed (Fig. 4D). Although comparing percent changes in right hippocampal volume between the three groups revealed no significant changes (Fig. 4E; F(2,15) 5 1.876, P 5 0.187), there was a significant correlation between right hippocampal volume changes and the social interaction ratio (Fig. 4F; r18 5 0.701, P 5 0.001). The lack of increase in the hippocampal volume of susceptible mice between the two MRI scans could be due to a reported stress-induced weight loss in these mice (Krishnan et al., 2007). However, comparing the weights of control, susceptible, and resilient mice that were measured before the 1st (Fig. 5A; F(2,15) 5 1.878, P 5 0.187) and 2nd MRI scans (F(2,15) 5 1.355, P 5 0.288) revealed no between-group differences. In addition, mice from these three groups displayed similar weight gains between the two MRI scans (Fig. 5B; F(2,15) 5 0.120, P 5 0.888).
HIPPOCAMPAL VOLUMES IN MICE WITH DIFFERENT STRESS SUSCEPTIBILITY
FIGURE 4. A: Scatter plots show changes in normalized left hippocampal volumes of control (white circles; n 5 6), susceptible (black circles; n 5 6), and resilient mice (gray circles; n 5 6) between the two MRI scans that were performed 15 days apart. B: Histograms summarize the percent change in left hippocampal volume of control, susceptible, and resilient mice between the two MRI scans. Only susceptible and resilient mice were stressed by chronic social defeat between these scans. *P < 0.05, post hoc Dunnett test after ANOVA. C: Scatter plots show the relationship between the percent change in left hippocampal volume and the
DISCUSSION This longitudinal MRI study revealed significant differences in stress-induced changes in hippocampal volume between mice that are susceptible and mice that are resilient to chronic social defeat stress. We found an inverse relationship between hippocampal volume measured prior to stress and the social interaction ratio observed after chronic social defeat, suggesting that a larger hippocampus confers higher risk to developing social avoidance after chronic stress. Resilient mice were similar to controls, demonstrating both social interaction behavior and an increase in the volume of the left hippocampus following social defeat stress. Notably, the increase in hippocampal volume before and after the stress was not found in susceptible mice that had a “defeated” phenotype, characterized by social avoidance after chronic stress. Using a within-subject design, to our knowledge, our findings are the first to provide evidence that both pre-stress and post-stress hippocampal volume differences contribute to the development of stress-induced social avoidance.
1125
post-stress social interaction ratio in control (white circles), susceptible (black circles) and resilient mice (gray circles). *P < 0.05, Pearson correlation. D: Scatter plots show changes in normalized right hippocampal volume of control, susceptible, and resilient mice between the two MRI scans. E: Histograms show the percent change in right hippocampal volume of control, susceptible, and resilient mice between the two MRI scans. F: Scatter plots show the relationship between the percent change in right hippocampal volume and the post-stress social interaction ratio. *P < 0.05, Pearson correlation.
We found an inverse relationship between hippocampal volume that was measured prior to stress exposure and the social interaction ratio, suggesting that the risk of developing social avoidance after chronic social defeat is higher in mice with a large hippocampus before stress exposure. A small hippocampus has been suggested to confer vulnerability to PTSD and depression (Gilbertson et al., 2002; Chen et al., 2010). The hippocampus is important in the negative feedback control of the hypothalamic–pituitary–adrenal axis (Herman et al., 2005), and thus a smaller hippocampus could result in an impairment of such feedback and an overproduction of corticosteroids and high stress susceptibility. Nevertheless, consistent with our findings, evidence exists for a link between bigger hippocampal volume and higher stress susceptibility. In healthy young individuals, Pruessner et al. have shown an association between higher social stress-induced cortisol release and larger hippocampal volumes (Pruessner et al., 2007). In older adults, the number of stressful life events experienced has been shown to be associated not only with the risk of developing depression, but also with larger hippocampal volumes (Zannas et al., Hippocampus
1126
TSE ET AL.
FIGURE 5. A: Histograms summarize the weight of control (n 5 6), susceptible (n 5 6), and resilient mice (n 5 6) right before the first (a) and the second MRI scans (b). B: Histograms summarize the weight gain of control, susceptible, and resilient mice between two MRI scans.
2013). Findings such as these question the prevalent assumption that a congenitally small hippocampus increases an individual’s risk for stress-induced neuropathologies. Since a reduction in hippocampal volume associates with cognitive impairment in different brain diseases (Sapolsky, 2000), larger hippocampal volumes in susceptible mice may lead to higher hippocampus-dependent cognitive function. In a human study, subjects with large hippocampal volumes outperformed those with small hippocampal volumes in contextual fear conditioning (Pohlack et al., 2012). The potential increase in hippocampal function in susceptible mice may enhance their fear learning, which could result in stronger fear responses during social interaction than resilient and control mice. Future studies could compare hippocampus-dependent contextual fear conditioning in susceptible and resilient mice to find out if a congenitally bigger hippocampus enhances fear learning in susceptible mice only. We found that between the two MRI scans taken before or after the stress, both nonstressed control and stressed resilient mice displayed increases in left hippocampal volumes. Developmental increases of whole brain and hippocampal volume in rodents has been previously reported. Lau et al. found that between 2.5 and 9 months of age, mice show significant increases in brain volume (Lau et al., 2008). Brain volumes of rodents continuously increase up to almost one year after birth (Maheswaran et al., 2009), perhaps due to delayed cessation of cranial bone growth in rodents when compared with other mammalian species (Kilborn et al., 2002). However, it is important to note that not all longitudinal studies of brain volume in rodents have revealed hippocampal growth. Lee et al. showed that while chronic restraint stress (6 h per day for 3 weeks) decreased hippocampal volume in adult SpragueDawley (SD) rats, longitudinal MRI scanning of nonstressed control rats before and after the same 3-week period of time revealed no changes in hippocampal volume (Lee et al., 2009). The discrepancy of findings between Lee et al. and the current study could be related to differences in the species, the age of the animals, and the type of stressor used. Notably, Lee et al. Hippocampus
used only seven sections to estimate hippocampal volume. The use of 30 sections for the same purpose in the present study may have allowed us to achieve a more precise estimation of hippocampal volume. Although both susceptible and resilient mice were stressed by identical social defeat procedures, only resilient mice exhibited an increase or growth in hippocampal volume after chronic social defeat. Since similar increases in hippocampal volumes were found in nonstressed control mice, our findings suggest that normal hippocampal growth was arrested by chronic stress in susceptible mice only. In parallel to our findings, Isgor et al. performed morphological analyses of fixed brain tissue and revealed hippocampal growth in SD rats within the 3rd postnatal month (Isgor et al., 2004). Interestingly, they showed that hippocampal growth was arrested by chronic variable stress. The lack of hippocampal growth in susceptible mice was not due to stress-related weight loss or reduced food intake, since mice from all three groups displayed similar weight gain between the two MRI scans. Notably, although we found no differences in hippocampal volumes between the three animal groups before stress exposure, we observed a direct correlation between pre-stress hippocampal volume and the social interaction ratio (Fig. 3). We therefore cannot rule out an alternative explanation that hippocampi of susceptible mice have reached their maximum sizes, so no further growth is possible. Future experiments should examine longitudinal changes in hippocampal volumes of susceptible and resilient mice to establish whether susceptible mice reach the maximum hippocampal volume at the age tested in this study (2 months) and find out whether differences in stressinduced hippocampal volume can be maintained between older susceptible and resilient mice (9–12 months) when hippocampal growth subsides. Our findings suggest that there are individual differences in the susceptibility to stress-induced hippocampal volume changes. Individual differences in hippocampal volume changes have similarly been observed in aging studies. The hippocampus has been shown to shrink faster with aging in hypertensive individuals than in normotensive individuals in one longitudinal study (Raz et al., 2005). Serial measurements of hippocampal volume during aging in individuals with different degrees of lifetime mental activity also suggested a protective effect of complex mental activity on the rate of aging-related hippocampal shrinkage (Valenzuela et al., 2008). Individual differences in hippocampal susceptibility to stress-induced volume changes could underlie the formation of psychopathologies in susceptible animals (e.g., social avoidance) and in human patients with mood disorders. Although factors underlying individual differences in stress-induced hippocampal volume changes remain unclear, genetic (Krishnan et al., 2007) and environmental factors (Meaney, 2001) that have been associated with individual differences in stress susceptibility could be responsible. Given the high variability of hippocampal volume in healthy humans (Lupien et al., 2007), individual differences in innate hippocampal volume may contribute to the variability in people’s susceptibility to stress and risk for mood disorders.
HIPPOCAMPAL VOLUMES IN MICE WITH DIFFERENT STRESS SUSCEPTIBILITY We showed that the impact of chronic stress on hippocampal volume in susceptible mice may be asymmetric, with the left hippocampus being more vulnerable to stress. Stress has been shown to induce a larger reduction in volume of the left hippocampus than the right hippocampus in adult rats (Liu et al., 2011). Lateralization of stress susceptibility in the hippocampus could be related to asymmetric expression of corticosteroid receptors in the left and right hippocampus (Hou et al., 2013). Accordingly, the volume of the dentate gyrus of the left hippocampus, but not of the right hippocampus, negatively correlated with saliva cortisol levels in stressed pigs (van der Beek et al., 2004), suggesting an asymmetry in the impact of cortisol on hippocampal volume. However, not all crosssectional studies have revealed specific corticosteroid-induced shrinkage of the left hippocampus. For instance, chronic corticosterone treatment induced a delayed (two months after cessation of corticosterone treatment) shrinkage of the right but not the left hippocampus of adult Long Evans rats in one study (Zach et al., 2010). Additionally, Carrion et al. found that cortisol levels and PTSD symptoms correlate with shrinkage of the right instead of the left hippocampus in children that had experienced trauma (Carrion et al., 2007). Apart from corticosteroid receptor expression, lateralization of stress susceptibility may be related to anatomical (Diamond et al., 1982) or neurochemical asymmetry (Valdes et al., 1981) between the left and right hippocampus in rodents. In conclusion, our findings suggest that the vulnerability to psychopathological outcome after experiencing chronic social defeat stress in rodents is related to both hippocampal susceptibility to stress-induced changes in volume and differences in hippocampal volume before stress exposure. An understanding of the biological factors that underlie susceptibility to stressinduced hippocampal volume changes could be therapeutically useful for increasing resistance to stress-related insults.
Acknowledgments The authors report no biomedical financial interests or potential conflicts of interest. Authors would like to thank Ms. Juliana Hutter and Ms. Jasmine Domagala for their technical assistance.
REFERENCES Breslau N, Davis GC, Andreski P, Peterson E. 1991. Traumatic events and posttraumatic stress disorder in an urban population of young adults. Arch Gen Psychiatry 48:216–222. Carrion VG, Weems CF, Reiss AL. 2007. Stress predicts brain changes in children: A pilot longitudinal study on youth stress, posttraumatic stress disorder, and the hippocampus. Pediatrics 119:509– 516. Chen MC, Hamilton JP, Gotlib IH. 2010. Decreased hippocampal volume in healthy girls at risk of depression. Arch Gen Psychiatry 67:270–276.
1127
Diamond MC, Murphy GM Jr, Akiyama K, Johnson RE. 1982. Morphologic hippocampal asymmetry in male and female rats. Exp Neurol 76:553–565. Franklin KBJ, Paxinos G. 1997. The Mouse Brain in Stereotaxic Coordinates. San Diego: Academic Press. Gilbertson MW, Shenton ME, Ciszewski A, Kasai K, Lasko NB, Orr SP, Pitman RK. 2002. Smaller hippocampal volume predicts pathologic vulnerability to psychological trauma. Nat Neurosci 5:1242–1247. Gurvits TV, Shenton ME, Hokama H, Ohta H, Lasko NB, Gilbertson MW, Orr SP, Kikinis R, Jolesz FA, McCarley RW, Pitman RK. 1996. Magnetic resonance imaging study of hippocampal volume in chronic, combat-related posttraumatic stress disorder. Biol Psychiatry 40:1091–1099. Herman JP, Ostrander MM, Mueller NK, Figueiredo H. 2005. Limbic system mechanisms of stress regulation: Hypothalamopituitary-adrenocortical axis. Prog Neuropsychopharmacol Biol Psychiatry 29:1201–1213. Hou G, Yang X, Yuan TF. 2013. Hippocampal asymmetry: Differences in structures and functions. Neurochem Res 38:453–460. Isgor C, Kabbaj M, Akil H, Watson SJ. 2004. Delayed effects of chronic variable stress during peripubertal-juvenile period on hippocampal morphology and on cognitive and stress axis functions in rats. Hippocampus 14:636–648. Joels M. 2009. Stress, the hippocampus, and epilepsy. Epilepsia 50: 586–597. Kendler KS, Karkowski LM, Prescott CA. 1999. Causal relationship between stressful life events and the onset of major depression. Am J Psychiatry 156:837–841. Kilborn SH, Trudel G, Uhthoff H. 2002. Review of growth plate closure compared with age at sexual maturity and lifespan in laboratory animals. Contemp Top Lab Anim Sci 41:21–26. Krishnan V, Han MH, Graham DL, Berton O, Renthal W, Russo SJ, Laplant Q, Graham A, Lutter M, Lagace DC, Ghose S, Reister R, Tannous P, Green TA, Neve RL, Chakravarty S, Kumar A, Eisch AJ, Self DW, Lee FS, Tamminga CA, Cooper DC, Gershenfeld HK, Nestler EJ. 2007. Molecular adaptations underlying susceptibility and resistance to social defeat in brain reward regions. Cell 131:391–404. Lau JC, Lerch JP, Sled JG, Henkelman RM, Evans AC, Bedell BJ. 2008. Longitudinal neuroanatomical changes determined by deformation-based morphometry in a mouse model of Alzheimer’s disease. Neuroimage 42:19–27. Lee T, Jarome T, Li SJ, Kim JJ, Helmstetter FJ. 2009. Chronic stress selectively reduces hippocampal volume in rats: A longitudinal magnetic resonance imaging study. Neuroreport 20:1554–1558. Liu W, Shu XJ, Chen FY, Zhu C, Sun XH, Liu LJ, Ai YX, Li YG, Zhao H. 2011. Tianeptine reverses stress-induced asymmetrical hippocampal volume and N-acetylaspartate loss in rats: An in vivo study. Psychiatry Res 194:385–392. Lorenzetti V, Allen NB, Fornito A, Yucel M. 2009. Structural brain abnormalities in major depressive disorder: A selective review of recent MRI studies. J Affect Disord 117:1–17. Lupien SJ, Evans A, Lord C, Miles J, Pruessner M, Pike B, Pruessner JC. 2007. Hippocampal volume is as variable in young as in older adults: Implications for the notion of hippocampal atrophy in humans. Neuroimage 34:479–485. Lyons DM, Yang C, Sawyer-Glover AM, Moseley ME, Schatzberg AF. 2001. Early life stress and inherited variation in monkey hippocampal volumes. Arch Gen Psychiatry 58:1145–1151. Magarinos AM, McEwen BS. 1995. Stress-induced atrophy of apical dendrites of hippocampal CA3c neurons: Involvement of glucocorticoid secretion and excitatory amino acid receptors. Neuroscience 69:89–98. Maheswaran S, Barjat H, Rueckert D, Bate ST, Howlett DR, Tilling L, Smart SC, Pohlmann A, Richardson JC, Hartkens T, Hill DL, Upton N, Hajnal JV, James MF. 2009. Longitudinal regional brain Hippocampus
1128
TSE ET AL.
volume changes quantified in normal aging and Alzheimer’s APP x PS1 mice using MRI. Brain Res 1270:19–32. McAuliffe M, Lalonde F, McGarry D, Gandler W, Csaky K, Trus B. Medical image processing, analysis and visualization in clinical research. 2001. In Proceedings of the 14th IEEE Symposium on Computer-Based Medical Systems (CBMS 2001), Bethesda, MD. pp 381–386. McEwen BS. 1999. Stress and hippocampal plasticity. Annu Rev Neurosci 22:105–122. McEwen BS, Sapolsky RM. 1995. Stress and cognitive function. Curr Opin Neurobiol 5:205–216. Meaney MJ. 2001. Maternal care, gene expression, and the transmission of individual differences in stress reactivity across generations. Annu Rev Neurosci 24:1161–1192. Muscatell KA, Slavich GM, Monroe SM, Gotlib IH. 2009. Stressful life events, chronic difficulties, and the symptoms of clinical depression. J Nerv Ment Dis 197:154–160. Pohlack ST, Nees F, Liebscher C, Cacciaglia R, Diener SJ, Ridder S, Woermann FG, Flor H. 2012. Hippocampal but not amygdalar volume affects contextual fear conditioning in humans. Hum Brain Mapp 33:478–488. Pruessner M, Pruessner JC, Hellhammer DH, Bruce PG, Lupien SJ. 2007. The associations among hippocampal volume, cortisol reactivity, and memory performance in healthy young men. Psychiatry Res 155:1–10. Raz N, Lindenberger U, Rodrigue KM, Kennedy KM, Head D, Williamson A, Dahle C, Gerstorf D, Acker JD. 2005. Regional brain changes in aging healthy adults: General trends, individual differences and modifiers. Cereb Cortex 15:1676–1689. Ribot EJ, Pourtau L, Massot P, Voisin P, Thiaudiere E, Franconi JM, Miraux S. 2013. 3D TrueFISP MRI provides accurate longitudinal measurements of glioma volumes in mice. In: Chen CC, editor. Advances in the Biology, Imaging and Therapies for Glioblastoma. Rijeka: InTech. pp 125–144. Sapolsky RM. 2000. Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders. Arch Gen Psychiatry 57:925–935.
Hippocampus
Sapolsky RM, Krey LC, McEwen BS. 1985. Prolonged glucocorticoid exposure reduces hippocampal neuron number: implications for aging. J Neurosci 5:1222–1227. Sapolsky RM, Uno H, Rebert CS, Finch CE. 1990. Hippocampal damage associated with prolonged glucocorticoid exposure in primates. J Neurosci 10:2897–2902. Sinkkonen ST, Vekovischeva OY, Moykkynen T, Ogris W, Sieghart W, Wisden W, Korpi ER. 2004. Behavioural correlates of an altered balance between synaptic and extrasynaptic GABAAergic inhibition in a mouse model. Eur J Neurosci 20:2168–2178. Smith ME. 2005. Bilateral hippocampal volume reduction in adults with post-traumatic stress disorder: A meta-analysis of structural MRI studies. Hippocampus 15:798–807. Uno H, Tarara R, Else JG, Suleman MA, Sapolsky RM. 1989. Hippocampal damage associated with prolonged and fatal stress in primates. J Neurosci 9:1705–1711. Valdes JJ, Mactutus CF, Cory RN, Cameron WR. 1981. Lateralization of norepinephrine, serotonin and choline uptake into hippocampal synaptosomes of sinistral rats. Physiol Behav 27:381–383. Valenzuela MJ, Sachdev P, Wen W, Chen X, Brodaty H. 2008. Lifespan mental activity predicts diminished rate of hippocampal atrophy. PLoS One 3:e2598. van der Beek EM, Wiegant VM, Schouten WG, van Eerdenburg FJ, Loijens LW, van der Plas C, Benning MA, de VH, de Kloet ER, Lucassen PJ. 2004. Neuronal number, volume, and apoptosis of the left dentate gyrus of chronically stressed pigs correlate negatively with basal saliva cortisol levels. Hippocampus 14:688–700. Videbech P, Ravnkilde B. 2004. Hippocampal volume and depression: a meta-analysis of MRI studies. Am J Psychiatry 161:1957–1966. Zach P, Mrzilkova J, Rezacova L, Stuchlik A, Vales K. 2010. Delayed effects of elevated corticosterone level on volume of hippocampal formation in laboratory rat. Physiol Res 59:985–996. Zannas AS, McQuoid DR, Payne ME, Steffens DC, MacFall JR, Ashley-Koch A, Taylor WD. 2013. Negative life stress and longitudinal hippocampal volume changes in older adults with and without depression. J Psychiatr Res 47:829–834.
