Post-Traumatic Stress Disorder (PTSD) is a complex mental disorder that develops in a subset of people exposed to a traumatic event and results in long-lasting emotional and functional impairments (American Psychiatric Association [APA], 2000). Individual susceptibility to developing PTSD has long been acknowledged (Yehuda et al., 2000; Gilbertson et al., 2002; Childress et al., 2013; Logue et al., 2018), yet it is poorly understood. To facilitate investigations into individual susceptibility, we have previously postulated a unifying framework of risk, susceptibility and sequelae factors (Alexander et al., 2020)

We define susceptibility factors as those aspects that are present and can be identified before trauma and manipulated such that they affect the onset and progression of PTSD after trauma (Alexander et al., 2020). To identify and study susceptibility factors, we developed an animal model, RISP, (Revealing Individual Susceptibility to PTSD-like phenotype). Using RISP, one can classify male rats before trauma as Resilient or Susceptible to developing a PTSD-like phenotype based on a priori set of criteria of acoustic startle response and anxiety-like behaviors 4 days after exposure to a mild stressor which does not induce fear conditioning. These are temporary changes in behavior in response to the mild stressor that return to baseline if the animals do not subsequently experience a traumatic event. If however, they do experience a traumatic even (strong foot shock), Susceptible, but not Resilient, rats display a PTSD-like phenotype with impaired fear extinction and sustained increase in acoustic startle response (Nalloor et al., 2011).

Using the RISP model in conjunction with a sensitive cellular imaging technique called Arc/Homer1a catFISH (cellular compartment analysis of temporal activity by fluorescent in situ hybridization using Arc and Homer 1a) we previously reported a putative susceptibility factor- impaired hippocampal learning-induced plasticity before trauma (Nalloor et al., 2014). During spontaneous spatial learning Susceptible, compared to Resilient, rats have impaired expression of immediate-early genes (IEGs) in the ventral/temporal hippocampus. Susceptible rats also showed altered pattern of IEG expression in the dorsal/septal hippocampus during the same low stress event. These differences were detected before the animals experienced a PTSD-inducing traumatic event suggesting that altered hippocampal function is a putative susceptibility factor. Such findings are consistent with reports in humans where hippocampal structural and functional deficits before trauma have been associated with increased severity of PTSD (Gilbertson et al., 2002; Gilbertson et al., 2006; Hayes et al., 2011; Xie et al., 2018).

Other brain regions functionally related to the hippocampus such as the medial prefrontal cortex (mPFC) and amygdala have also been shown to have structural and functional alterations in people with PTSD (Milad et al., 2009; Pitman et al., 2012; Harnett et al., 2020; Xiao et al., 2022). It is unclear, however, if these alterations exist before trauma and thus contribute to susceptibility to PTSD or develop because of the trauma. We hypothesize that these alterations exist before trauma and thus, may be putative susceptibility factors based on the evidence that susceptible male rats have increased inflammation in the mPFC and that inflammation disrupts neuronal function (Shanazz et al., 2023; Tastan and Heneka, 2024). If true, this will suggest that susceptibility emerges in a network of brain regions functionally related to the hippocampus rather than just the hippocampus.

We tested the hypothesis that mPFC has altered function in Susceptible compared to Resilient rats before PTSD-inducing trauma by examining plasticity in the Prelimbic (PL) and Infralimbic (IL) regions of the mPFC induced by a non-traumatic event using Arc/Homer 1a catFISH (Vazdarjanova et al., 2002; Vazdarjanova et al., 2006). For rigor, we used tissue from the same animals in which we had demonstrated alterations in Arc/Homer 1a expression in the hippocampus, but not the basolateral amygdala, during spatial exploration. This task was chosen to assess brain function under relatively non-stressful conditions to eliminate the possibility that the examined IEG expression is due to processing of a potentially traumatic event that can lead to PTSD-like behavior. Evidence in support of this hypothesis will satisfy the first step in identifying mPFC deficit as a susceptibility factor, i.e., it is present and identifiable before trauma.

Arc (activity-regulated cytoskeleton-associated protein) and Homer1a (H1a) are effector IEGs that are very low at baseline and are coordinately expressed during learning (Steward and Worley, 2001; Bottai et al., 2002; Vazdarjanova et al., 2002; Bramham et al., 2008; Bramham et al., 2010). They are reliable markers of plasticity, not solely neuronal activity (Guzowski et al., 2006; Carpenter-Hyland et al., 2010) and are necessary for memory consolidation (Fagni et al., 2002; Ramírez-Amaya et al., 2005; Szumlinski et al., 2006; Bramham et al., 2010). Furthermore, Arc expression reliably predicts the pattern of plasticity assessed electrophysiologically a day after learning (Guzowski et al., 2006; Carpenter-Hyland et al., 2010). Additionally, Arc and H1a mRNA have different lengths such that, using a full-length Arc probe and a similar sized probe targeting the 3’ UTR of H1a, it is possible to dissociate neuronal ensembles that have initiated plasticity during two events experienced 20–30 min apart. Neurons with H1a foci of transcription mark those responding to the first event and the ones with Arc foci mark neurons responding to the second event. Neurons with both H1a and Arc foci have responded to both events (Vazdarjanova et al., 2002; Chawla et al., 2004; Nalloor et al., 2012; Nalloor et al., 2014; Figure 1B). One advantage of Arc/H1a catFISH is that the dissociation of the neuronal ensembles responding to these events can be done across multiple brain regions in the same animals. Combined with the RISP model, Arc/H1a catFISH allowed us to examine whether relative plasticity across different brain regions is altered in Susceptible compared to Resilient rats.

We also hypothesized that any differences in plasticity between Resilient and Susceptible rats will also be detected functionally with appropriate cognitive tests. Attentional Set-Shifting is a task that is commonly used to test cognitive flexibility in animals (Heisler et al., 2015; Britten et al., 2023). Impairments in this task are associated with structural changes and functional deficits in the mPFC (Birrell and Brown, 2000; Bissonette et al., 2013; Chen et al., 2023). Therefore, we tested whether the Resilient and Susceptible rats differ on their performance on the Attentional Set-Shifting task.

To begin evaluating the translational relevance of these hypotheses, we compared pre-deployment performance on a battery of cognitive tests of military personnel who developed post-deployment PTSD and those who did not. We hypothesized that susceptibility factors also exist in humans and can be functionally detected before a PTSD-inducing trauma.

The design of Experiment 1 is summarized in Figure 1A. The behavior of these animals during classification was previously reported in Nalloor et al. (2014). Rats classified as Resilient (n = 8) or Susceptible (n = 6) were allowed to explore a novel box (50 cm × 10 cm × 19 cm) twice for 5 min, 20 min apart (Event 1 and Event 2, respectively). The rats were in their home cages during the 20 min between the Events. Fear behavior and locomotion were assessed. Freezing, a measure of fear behavior, was defined as lack of all movement except for respiration. Number of Crossings, a measure of locomotor activity, was defined as all four paws and the base of the tail crossing over the midline of the box. Immediately after Event 2, rats were quickly anesthetized (< 1.5 min) and decapitated. Caged control (n = 3) rats were removed from the cages either before the first or second behavioral event and were immediately anesthetized for brain harvesting and assessing of baseline Arc/H1a expression.

The experimental design is summarized in Figure 3A. Similar to Experiment 1, rats were classified using the RISP model. Resilient (n = 11) and Susceptible (n = 9) rats were food restricted to maintain their body weight at ∼85% of pre-restriction weight.

Attentional Set-Shifting task: Details of the order of discrimination tasks are provided in Figure 3B.

The timepoints of collection of the deidentified human data are provided in Figure 5A. The pre-trauma cognitive assessment was performed using ANAM (Automated Neuropsychological Assessment Metrics, Vista Life Sciences, CO), a battery of computerized cognitive tests that also evaluates simple reaction time at the first and second half of the battery. The ANAM data was collected before combat exposure. At this time, none of the subjects met the criteria for PTSD diagnosis. Subsequently, all subjects encountered severe combat related traumatic events that resulted in TBI. Following this trauma, some developed PTSD (PTSD; n = 3) and some did not (NoPTSD; n = 6). The PTSD evaluation and diagnosis were performed by qualified clinicians at Eisenhower Army Medical Center, Augusta, GA.

Both NoPTSD and PTSD groups were compared on the following outputs of their pre-deployment ANAM: Simple Reaction Times, Accuracy, Throughput and Reaction Time for each cognitive test. In order to normalize for possible differences in individual reaction time, we calculated Cognitive Processing Time:

Cognitive Processing Time = Reaction Time for test – (Simple Reaction Time 1 + Simple Reaction Time 2)/2

Two-way repeated measures ANOVA with Group and Event (Figure 2) or Task (Figure 4) as independent variables was used to compare IEG expressing ensembles or number of trails to criterion, respectively. Statistically significant main effects were followed up by Fisher’s LSD post hoc for between group comparisons and one-way ANOVA for within group comparisons. The difference between groups in the ratio of ensemble sizes (Event 2/Event1) was evaluated using one group t-test with a hypothesized difference of 1. In the human dataset, one-way ANOVAs were used to compare performance of No PTSD and PTSD groups in the different cognitive tests (Figures 5B–D) (GraphPad 10). Differences were considered statistically significant at p < 0.05.

The set of studies reported here aimed to test the hypothesis that mPFC deficits are a putative susceptibility factor for PTSD that can be detected before trauma in both rats and humans. The results provided evidence in support of this hypothesis at both the cellular and behavioral levels in male rats and at the behavioral level in humans. At the cellular ensemble level, male rats pre-classified according to the RISP model as Susceptible to developing a PTSD-like phenotype had a different pattern of plasticity-induced IEG expression in the mPFC than Resilient rats. On the behavioral level, Susceptible rats showed deficits under higher cognitive demand before trauma. Similarly, male military personnel who later developed combat-related PTSD showed longer cognitive processing time in a PFC- and hippocampus-dependent task prior to deployment.

At the cellular level, Susceptible rats differed from Resilient rats in the relative contribution of PL and IL in spontaneous spatial learning. All rats engaged PL and IL regions of the mPFC during spatial exploration as evidenced by the presence of above-baseline number of neurons expressing plasticity-associated immediate- early genes (Arc and H1a) in both groups. However, there was a difference in the pattern of Arc/H1a expression between Resilient and Susceptible rats such that Resilient rats consistently engaged a larger ensemble in PL than IL while this PL dominance was not seen in the Susceptible rats. Additionally, there was a decrease in the percentage of double labeled cells in the PL of the Susceptible compared to Resilient rats. Combined with evidence from the literature where activation of PL ensembles have been linked to continuous behavioral microstates (Zhang et al., 2022) this suggests that flexible control of behavior may be diminished in Susceptible rats. The fact that these differences are present before trauma suggests that these brain regions may respond differently to a traumatic event.

The PL and IL have distinct roles in cognition and emotional regulation (Giustino and Maren, 2015; Capuzzo and Floresco, 2020). However, reciprocal connections between these brain regions allow information exchange that leads to flexible and adaptive behavior (van Aerde et al., 2008; Marek et al., 2018). Both PL and IL also receive unidirectional projections from the hippocampus (Hoover and Vertes, 2007; Sánchez-Bellot et al., 2022). and are implicated in fear learning and extinction (Quirk and Mueller, 2007; Maren et al., 2013; Giustino and Maren, 2015). Our previous data show that in these Susceptible animals the dorsal and ventral hippocampus, but not the basolateral amygdala, has altered function compared to Resilient animals before trauma (Nalloor et al., 2014). Combined with the current findings, these data suggest that susceptibility may be the product of alterations in the brain network that included the hippocampus and the mPFC. Furthermore, the hippocampus and the PFC are in brain regions that are well documented to have impaired functioning in people with established PTSD (Pitman et al., 2001; Bremner, 2007; Harnett et al., 2020). This raises the possibility that susceptible individuals develop PTSD in part due to experiencing a traumatic event differently than resilient individuals (Alexander et al., 2020).

The differences on the cellular level that are reported here (Experiment 1) were detectable because of the unique strengths offered by the Arc/H1a catFISH. Due to the restricted time of learning-induced Arc/H1a expression (< 10 min) as detected by intranuclear foci of transcription, the temporal separation of the two events (30 min), and the very low level of baseline Arc and H1a expression (< 1%), the activated neuronal ensembles are highly specific to each event. This specificity allows a more precise evaluation of changes in patterns of activation across multiple brain regions. The prominent differences in the PL and IL were not in the size of the ensembles induced by the two events but in the relative activation across the two brain regions. In addition, these findings are based on a large number of neurons per brain region and thus are representative of each region.

The differences at the cellular ensemble level were consistent with findings at the behavioral level in a hippocampal/mPFC dependent task (Attentional Set Shifting) (Birrell and Brown, 2000; Barense et al., 2002; Cernotova et al., 2021). While there were no differences in the initial learning of discrimination and reversal tasks, Susceptible rats showed impaired performance compared to Resilient rats when cognitive load increased. As the training progressed, Resilient rats improved their performance during reversals, while Susceptible rats did not. Notably, Susceptible rats started off slightly better than Resilient in the Complex Discrimination and initial Intradimensional Reversal tasks. However, they were slightly impaired in the second Intradimensional Reversal and then significantly impaired at the subsequent Extradimensional Reversal compared to Resilient rats. These findings suggest reduced baseline cognitive flexibility in male Susceptible rats. Such deficits may contribute to PTSD susceptibility by changing the experience of the trauma and anchoring it to the most aversive aspect.

The behavioral findings demonstrated in male rats paralleled those we observed in the human data set. Male military personnel who developed PTSD post-trauma had cognitive deficits that were detectable pre-trauma. Similar to rats, these differences were not universal to all tasks but were specific to a hippocampus/mPFC-dependent task (spatial Match-To-Sample) (Monk et al., 2002; Habeck et al., 2005; Schon et al., 2008; Daniel et al., 2016). The differences in cognitive processing time were not based on motivation or general inability to understand or perform the tasks, as accuracy was not affected. They were also not due to individual differences in simple reaction time, as the latter was subtracted from the individual task reaction time for each participant to derive the Cognitive Processing Time. These considerations demonstrate the specificity of the reported findings. These findings demonstrate that pre-trauma functional differences exist and can be identified in humans as well.

In addition to the strengths discussed, the study has several limitations. First, although it would have been ideal to examine alterations in patterns of plasticity during the Attentional Set-Shifting task, this was not possible due to the temporal specificity of the Arc/H1a catFISH method: performing the task took much longer than the time window that is optimal for inducing and detecting Arc/H1a intranuclear foci of transcription (3–7 min). Second, the presented data is based on studies that included only male subjects. PTSD is a sexually dimorphic disorder, so the conclusions drawn may only be applicable to males. Additional studies in female subjects are required before generalizing findings to both sexes. Third, the human data presented here is preliminary as it is based on a small sample size. Furthermore, all military personnel in this data set experienced TBI which could have influenced the PTSD diagnosis. However, the human data findings are consistent with the animal data and thus warrant larger scale study in humans to examine cognitive determinants of susceptibility to PTSD with or without physical trauma.

Despite the stated limitations, the findings suggest that mPFC deficits are a putative susceptibility factor for PTSD. In order to test whether these deficits satisfy all criteria of being a susceptibility factor, it is necessary to test if enhancing mPFC function in susceptible individuals will confer resilience to developing PTSD. This hypothesis will be addressed in future studies. Building resilience is crucial for minimizing the number of people suffering from PTSD, given that it is difficult to treat, and treatments are resource intensive and benefit only a subpopulation of people suffering from PTSD.