- Top of page
- Materials and methods
- Supporting Information
Post-traumatic stress disorder (PTSD) is an anxiety disorder that develops in predisposed individuals following a terrifying event. Studies on isogenic animal populations might explain susceptibility to PTSD by revealing associations between the molecular and behavioural consequences of traumatic stress. Our study employed four inbred mouse strains to search for differences in post-stress response to a 1.5-mA electric foot shock. One day to 6 weeks after the foot shock anxiety, depression and addiction-like phenotypes were assessed. In addition, expression levels of selected stress-related genes were analysed in hippocampus and amygdala. C57BL/6J mice exhibited up-regulation in the expression of Tsc22d3, Nfkbia, Plat and Crhr1 genes in both brain regions. These alterations were associated with an increase of sensitized fear and depressive-like behaviour over time. Traumatic stress induced expression of Tsc22d3, Nfkbia, Plat and Fkbp5 genes and developed social withdrawal in DBA/2J mice. In 129P3/J strain, exposure to stress produced the up-regulation of Tsc22d3 and Nfkbia genes and enhanced sensitivity to the rewarding properties of morphine. Whereas, SWR/J mice displayed increase only in Pdyn expression in the amygdala and had the lowest conditioned fear. Our results reveal a complex genetic background of phenotypic variation in response to stress and indicate the SWR/J strain as a valuable model of stress resistance. We found potential links between the alterations in expression of Tsc22d3, Nfkbia and Pdyn, and different aspects of susceptibility to stress.
Post-traumatic stress disorder (PTSD) is a debilitating anxiety disorder that occurs in certain individuals in response to an extreme stress (Nemeroff et al. 2006; Olff et al. 2005; Yehuda & LeDoux 2007). A key question is, why do some people develop PTSD while others appear to experience only few negative effects following a potentially traumatic event.
Clinical studies provide essential findings, but research on human patients is constrained. Therefore, a number of animal models of PTSD has been proposed (Adamec et al. 2008; Cohen et al. 2011; Mikics et al. 2008; Philbert et al. 2011). We have used an inescapable shock paradigm as a mouse model of PTSD (Siegmund & Wotjak 2007b) due to its ability to produce long-lasting effects and employed four inbred mouse strains (C57BL/6J, DBA/2J, SWR/J and 129P3/J). Their variable behavioural phenotypes (Korostynski et al. 2006; Solecki et al. 2009), like dissimilarities in activity level (Paulus et al. 1999; Tang et al. 2002), impulsivity (Gubner et al. 2010) or anxiety-like behaviour (Milner & Crabbe 2008), may indicate their different susceptibility to trauma-related disorders.
There is evidence that stress can alter molecular mechanisms that lead to long-lasting plastic changes in brain regions, such as amygdala and hippocampus (Fenoglio et al. 2006; McEwen et al. 2011; Pittenger & Duman 2008; Radley et al. 2011). Interactions between genetic susceptibility and environment can determine risk of anxiety disorders, but little is known about specific factors underlying such disorders. Several genes have been previously linked to trauma-induced pathologies. Fkbp5 gene is a regulator of glucocorticoid receptor sensitivity and hypothalamic-pituitary-adrenal (HPA) axis reactivity that may influence coping strategies after a stressor (Binder 2009; Touma et al. 2011). Corticotrophin-releasing hormone type 1 receptor gene (Crhr1) has been reported to play a role in moderating the effect of early-life trauma on anxiety responses in adulthood (Amstadter et al. 2011; Wang et al. 2012) and in modulating stress-induced dendritic remodelling (Wang et al. 2011). Prodynorphin (Pdyn)-derived peptides (dynorphins, alpha-neo-endorphin and beta-neo-endorphin) were reported to modulate anxiety-like behaviours (Wittmann et al. 2009) and dynorphins are supposed to play a role in the formation and extinction of fear memories (Bilkei-Gorzo et al. 2012). Tsc22d3 gene is another potential molecular factor for the stress-related disorders due to its dependence on HPA axis activation and putative involvement in dendritic spines alterations (Ayroldi & Riccardi 2009; Piechota et al. 2010; Yang et al. 2008). Other genes were chosen on the basis of their hypothetical role in triggering plastic changes in the brain. Nf-κB signalling is involved in processes such as memory and neuronal remodelling (Mattson et al. 2000; Romano et al. 2006; Yeh et al. 2004), while the tissue plasminogen activator (Plat) gene is supposed to play a role in hippocampus-dependent learning and stress-induced anxiety (Pang et al. 2004; Pawlak et al. 2003).
By combining phenotypic, genetic and molecular factors, we aimed to identify the stress-susceptible/stress-resilient mouse strains and gene expression patterns potentially characteristic for a specific stress response.
- Top of page
- Materials and methods
- Supporting Information
Psychiatric diseases are affected by complex interactions between genes and diverse environmental factors that modify expression of these genes. One of such factors is stress. Inter-strain dissimilarity in PTSD-like symptoms examined in this study might reflect the individual differences in susceptibility to PTSD in humans.
Our results of CF and SF are consistent with other studies showing fear responses in the C57BL/6J strain (Radulovic et al. 1998; Schimanski et al. 2007; Siegmund & Wotjak 2007b). In accordance with previous papers (Belzung et al. 2001; Cohen et al. 2008), we suggest that a strong stressor is able to produce the same level of fear in C57BL/6J and DBA/2J mice. A constant level of fear in the 129P3/J mice can be explained by the fact that this strain exhibits poor fear extinction (Camp et al. 2009) and no habituation in anxiety-related behaviour (Salomons et al. 2010). The most intriguing observation from our experiments is the presence of an irregular pattern of associative and non-associative fear responses in SWR/J mice. There are explanations for such results. First, hyperarousal does not depend on trauma-related contextual memory, and therefore may lead to a different level of freezing (Siegmund & Wotjak 2007a). Moreover, SWR/J mice may exhibit a more ‘active’ response to stress by escaping or attacking, rather than freezing to trauma-related cue. We also conclude that low level of CF within SWR/J strain may not result from the perception of pain (i.e. the electric foot shock) as its foot shock sensitivity was comparable to those displayed by C57BL/6J and DBA/2J.
According to manifestations of numbing of general responsiveness after trauma, DBA/2J was the only strain that markedly increased avoidance. These results are comparable to previous data showing avoiding interaction in DBA/2J (Moy et al. 2008; Uchiumi et al. 2008). Low avoidance and high social interaction observed in SWR/J mice confirm their strong activity and may correspond to sociability and tendency to attacking described previously (Moy et al. 2008). Increased floating in the FST in the C57BL/6J strain suggests that these mice are more susceptible to post-shock depressive-like behaviour. However, in SWR/J mice increased floating may not be related to depression due to the co-occurring high level of struggling. We suggest that SWR/J use both passive and active coping strategies with stress.
Finally, we explored the comorbidity of PTSD and addiction, which is often observed in humans (Chilcoat & Breslau 1998; Jacobsen et al. 2001). Recent evidence suggests that uncontrollable stress (inescapable shock) increases the effect of morphine-induced CPP (Rozeske et al. 2009). Low morphine-induced CPP before and after stress in SWR/J mice confirmed our hypothesis that this strain might be resilient to the long-term behavioural consequences of trauma. As it has been previously reported, the 129P3/J strain has low morphine tolerance (Kest et al. 2002) and low physical dependence (Kest et al. 2002). Strong morphine preference after foot shock suggests that this strain can be used in models of susceptibility to drug addiction after stress.
By using various behavioural tests, we were able to obtain a model that corresponds to Diagnostic and Statistical Manual of Mental Disorders, fourth edition, text revision (DSM-IV-TR) criteria and reveals a wide profile of PTSD-like symptoms. Our study may provide information about molecular mechanisms that initiate long-term behavioural symptoms by revealing alterations in gene expression, discussed below.
We analysed genes potentially involved in different stress-related mechanisms. In our model we observed moderate increases in the Fkbp5 and Crhr1 mRNA levels only in individual mouse strains – Fkbp5 in DBA/2J and Crhr1 in C57BL/6J strain. Alterations in expression of those two genes can correspond to the specific symptoms of PTSD, i.e. up-regulation of Fkbp5 may be related to the increase in post-traumatic avoidance typical for the DBA/2J strain. Accordingly, the link between genetic variation in Fkbp5 gene and harm avoidance in humans was previously found (Shibuya et al. 2010). We also suggest that the increase of Crhr1 in C57BL/6J may correspond to accumulation of SF. It has been recently shown that Crhr1-deficient mice have reduced SF (Thoeringer et al. 2012).
We found comparable increases in the expression of Tsc22d3 and Nfkbia in C57BL/6J, DBA/2J and 129P3/J strains. Tsc22d3 is a marker of glucocorticoid action and has already been considered in the context of its neuronal function (Ayroldi & Riccardi 2009; Korostynski et al. 2007; Yachi et al. 2007). Our recent studies suggest that Tsc22d3 knockdown provokes changes in spine morphology (Piechota et al. 2010). Therefore, based on alterations in expression ofTsc22d3 we hypothesize there to be an involvement of this gene in vulnerability to long-lasting consequences of traumatic stress. It was also shown that the NF-κBpathway mediates cortical neurons remodelling and dendrite branching through interaction with Notch signalling (Bonini et al. 2011). Our results, in combination with current literature, suggest that the NF-κB pathway may act as a central integrator of the stress response with cell plasticity (Memet 2006). A rapid increase in the expression of the NF-κB protein inhibitor, probably in response to NF-κB activation, may further support this hypothesis. Plat gene, up-regulated in C57BL/6J and DBA/2J mice, has been previously associated with experience-induced synaptic plasticity (Shiosaka 2004; Skrzypiec et al. 2008). The stress-induced up-regulation of Tsc22d3, Nfkbia and Plat genes in the amygdala and hippocampus may indicate activation of the three independent molecular pathways. Interestingly, prodynorphin, previously described as a marker of stress (Knoll & Carlezon 2010), was up-regulated in amygdala only in the SWR/J strain. However, unaffected behaviour and transcription of the remaining genes suggest that the SWR/J strain may be resilient to traumatic stress. We hypothesized that in case of SWR/J mice Pdyn expression is a marker of stress resistance. Reduction of Pdyn expression in amygdala has been previously found in patients with major depression (Hurd 2002). Mice lacking dynorphin had an enhanced cue-dependent fear conditioning and delayed extinction (Bilkei-Gorzo et al. 2012). Thus, it is possible that Pdyn up-regulation in response to a traumatic event may protect against development of stress-related psychiatric disorders.
We are convinced that three strains (C57BL/6J, DBA/2J and 129P3/J) are more vulnerable to traumatic stress than the SWR/J strain is. The data previously published (Cohen et al. 2008) indicated that C57BL/6J displayed a greater corticosterone response to stress than other inbred strains studied and had greater persistence of hippocampal long-term potentiation (Matsuyama et al. 1997). Both C57BL/6J and DBA/2J revealed the most pronounced changes in shock-induced gene expression and they appear to provide an adequate model of PTSD behavioural symptoms – particularly, C57BL/6J for modelling of hyperarousal and depressive-like behaviour, and DBA/2J for modelling of social withdrawal.
To the best of our knowledge, we are the first group to investigate SWR/J mice in an animal model of PTSD and reveal their lack of susceptibility to trauma, especially in two clusters of symptoms: re-experiencing (weak fear responses to trauma-related cues) and avoidance/numbing of general responsiveness (lack of social withdrawal and flexibility in coping strategies with stress). Although poor retention of CF may be considered in the context of stress vulnerability (Acheson et al. 2011), fear response to trauma-related internal or external cues is one of the criteria required for PTSD diagnosis (Nemeroff et al. 2006; Siegmund & Wotjak 2006; Yehuda & LeDoux 2007). The absence of PTSD-like behavioural symptoms and the lack of changes in expression of HPA axis activity-dependent molecular markers make the SWR/J strain potentially ‘stress-proof’. Considering that the pre-existing factors are able to influence not only vulnerability, but also resilience-related traits, we identified the SWR/J strain as a valuable model of stress resistance.
To summarize, our data supports the idea that genetic factors are important for phenotypic variation in response to stress. We propose candidate genes (Tsc22d3, Nfkbia and Plat) that may play a dynamic role in the development of long-lasting consequences of traumatic event. Conversely, we hypothesize that up-regulation of the Pdyn transcript in response to stress may provide resistance to anxiety disorders. In order to generate new strategies for the diagnosis and efficient therapy of PTSD, future research should be focused on objective parameters of the disease, such as changes in transcription of genes and interactions between diverse signalling pathways.