We describe a rat model of post traumatic stress disorder (PTSD) that reveals the persistent alterations in neuroendocrine function and the delayed long-term, exaggerated fear response, characteristic of PTSD patients. The animal model and methods described here are useful for correlating biomarkers in brain nuclei, which are mechanistic but cannot be measured in patients, with biomarkers in peripheral white blood cells, which can.
Identification of biomarkers representing the evolution of the pathophysiology of Post Traumatic Stress Disorder (PTSD) is vitally important, not only for objective diagnosis but also for the evaluation of therapeutic efficacy and resilience to trauma. Ongoing research is directed at identifying molecular biomarkers for PTSD, including traumatic stress induced proteins, transcriptomes, genomic variances and genetic modulators, using biologic samples from subjects’ blood, saliva, urine, and postmortem brain tissues. However, the correlation of these biomarker molecules in peripheral or postmortem samples to altered brain functions associated with psychiatric symptoms in PTSD remains unresolved. Here, we present an animal model of PTSD in which both peripheral blood and central brain biomarkers, as well as behavioral phenotype, can be collected and measured, thus providing the needed correlation of the central biomarkers of PTSD, which are mechanistic and pathognomonic but cannot be collected from people, with the peripheral biomarkers and behavioral phenotypes, which can.
Our animal model of PTSD employs restraint and tail shocks repeated for three continuous days – the inescapable tail-shock model (ITS) in rats. This ITS model mimics the pathophysiology of PTSD 17, 7, 4, 10. We and others have verified that the ITS model induces behavioral and neurobiological alterations similar to those found in PTSD subjects 17, 7, 10, 9. Specifically, these stressed rats exhibit (1) a delayed and exaggerated startle response appearing several days after stressor cessation, which given the compressed time scale of the rat’s life compared to a humans, corresponds to the one to three months delay of symptoms in PTSD patients (DSM-IV-TR PTSD Criterian D/E 13), (2) enhanced plasma corticosterone (CORT) for several days, indicating compromise of the hypothalamopituitary axis (HPA), and (3) retarded body weight gain after stressor cessation, indicating dysfunction of metabolic regulation.
The experimental paradigms employed for this model are: (1) a learned helplessness paradigm in the rat assayed by measurement of acoustic startle response (ASR) and a charting of body mass; (2) microdissection of the rat brain into regions and nuclei; (3) enzyme-linked immunosorbent assay (ELISA) for blood levels of CORT; (4) a gene expression microarray plus related bioinformatics tools 18. This microarray, dubbed rMNChip, focuses on mitochondrial and mitochondria-related nuclear genes in the rat so as to specifically address the neuronal bioenergetics hypothesized to be involved in PTSD.
1. Animal Behavioral Model of PTSD
2. Brain Dissection
3. Gene Microarray of Mitochondrial & Mitochondria-related Nuclear Genes
To study rat mitochondrial functions in brain tissues, we have recently developed the rat mitochondrion-neuron focused microarray (rMNChip) and bioinformatics tools for rapid identification of differential pathways in brain tissues 18. rMNChip contains 1,500 genes involved in mitochondrial functions, stress response, circadian rhythms and signal transduction. The bioinformatics tool includes an algorithm for computing of differentially expressed genes, and a database for straightforward and intuitive interpretation for microarray results.
Washing Solution | Volume | 20 X SSC | 10%SDS | ddH2O |
0.5 X SSC/0.01 SDS | 500 ml | 12.5 ml | 0.5 ml | up to 500 ml |
0.5 X SSC | 500 ml | 12.5 ml | up to 500 ml | |
0.1 X SSC | 500 ml | 2.5 ml | up to 500 ml | |
0.01 X SSC | 500 ml | 0.25 ml | up to 500 ml |
4. Blood Sample Collection and Plasma CORT Concentration Measurement
5. Representative Results
Figure 1. Rats are restrained and exposed to tail shock. Subsequent body weight, plasma corticosterone concentration, and acoustic startle response are measured.A: Stress Exposure: Animals are restrained by being immobilized in a ventilated plexiglass tube. Forty electric shocks (2 mA, 3 sec duration;) are delivered to their tails at semi-random intervals of 150 to 210 sec. B and C: Stress retards gain in body weight during growth: Body weight and food and water consumption are measured immediately prior to stress (Day -3), on the day of the three days of stress and then every other day there after up to Day 14. The lack of gain of body weight during stress is never compensated. D: Stress increases plasma corticosterone concentration. E: Acoustic Startle: Animals are tested one day before stress (day-1) as a baseline reading and 12 days following the final day of the consecutive 3 days of the stress. Data for each group – Stress and Control – are expressed as percent of acoustic startle on day 12 relative to day -1. Stress markedly increases the acoustic startle reflex. Click here to view larger figure.
Figure 2. Dissection of the amygdala, hippocampus, and hypothalamus from the brain slice. A: Rat brain, ventral view: Arrows point to the middle cerebral arteries. B: The brain block, ventral view, ready to be transported to the Vibratome. C: The brain block glued to the vibratome tray, caudal side up, cortex facing the blade. The caudal brain has already been cut away exposing the caudal hippocampus. The block is now ready for the 2,500 μm slice containing the major part of the hippocampus to be taken. D: The 2,500 μm thick slice containing the caudal hippocampus. E: The isocortex (ISO) is peeled from the hippocampus (HC) and the hippocampus peeled from the midbrain. F: The 2,500 μm thick slice containing the amygdala and the rostral hippocampus. G: The isocortex has been excised and the amygdala (Amyg) resected. H: The hypothalamus (HT) is excised and displaced. Click here to view larger figure.
Figure 3. Expression levels for mRNA of the glucocorticoid receptor (GR) and minerocorticoid receptor (MR) in amygdala, hippocampus, hypothalamus, and frontal cortex before (C, control) and after (Str) tail-shock stress. Units are proportion of control; bars are S.E.M. A: Amygdala: Stress decreases minerocorticoid receptor mRNA expression. B: Hippocampus: Stress increases minerocorticoid receptor mRNA expression. C: Hypothalamus: Stress increases minerocorticoid receptor mRNA expression. D: Frontal Cortex: Stress decreases glucocorticoid receptor mRNA expression as well as minerocorticoid receptor mRNA expression.
The (121bp) PCR primers for rat GR are: | 1f.CCACTGCAGGAGTCTCACAA 1rAACACCTCGGGTTCAATCAC |
The (99 bp) PCR primers for rat MR are: | 1f.GCCTTCAGCTATGCCACTTC 1rAACGTCGTGAGCACCTTTCT |
Figure 4. Cluster and heatmap of RNA differentially expressed from 64 genes derived from 5 rat brain tissues, including cerebellum (CL), cerebrum (CR), frontal cortex (FC), hypothalamus (HT), and hippocampus (HC). Color map indicates fold changes in down- (green) and up- (red) expressed genes. (A) Cluster and heatmap of the normalized signal intensities of 9 measurements for each of 64 genes derived from 15 microarray experiments for these five brain tissues. The expression of each gene was measured by technical triplicates and experimental triplicates. (B) Cluster and heatmap of the mean RNA levels of these 9 measurements of each of the 64 genes. These results show clear differences in mitochondrial gene expression and therefore related functions, which support our hypothesis that different brain regions have different energy demands. Click here to view larger figure.
Our application of the rMNChip and bioinformatic tools led to identification of a cluster and heatmap of 64 genes with differentially expressed RNA derived from 5 rat brain tissues including cerebellum (CL), cerebrum (CR), frontal cortex (FC), hypothalamus (HT), and hippocampus (HC) (Figure 4). These data demonstrate the clear differences in mitochondrial gene expression and therefore related functions. The results demonstrate that the different brain regions demand different amount of energy in order to carrying out corresponding brain functions.
The diagnosis of PTSD is based on self reported psychiatric symptoms (DSM IV) by potential subjects. No well defined biomarker is currently available to access the pathophysiological status of potential PTSD patients. PTSD is a disorder provoked by life threatening traumatic events and the main psychiatric symptoms remain present in the survivor’s life for months and even years after the initial events. The most prominent and persistent symptoms revealed in patients with PTSD are hypervigilance, delayed exaggerated startle response 14, 15, 16 and an apparent compromise of the HPA axis. In humans these symptoms remain, or appear with a delay of three months, after cessation of the traumatic stressor 24. Our current model for biomarker studies of PTSD employs restraint and tail shocks repeated for three continuous days (2-hr sessions of 40, 2 mA tailshocks) – the inescapable tail-shock model (ITS) in rats weighing 150 gram. This ITS model has been shown to mimic to a substantial extent the pathophysiology of PTSD 17, 7, 4, 10. Our lab and other labs have verified that the ITS model of stress in rats induces behavioral and neurobiological alterations that are similar to those found in PTSD subjects 17, 7, 10, 9. Specifically, these stressed rats exhibit (1) a delayed and exaggerated startle response appearing several days after stressor cessation, which given the compressed time scale of the rat’s life compared to a humans, corresponds to the one to three months delay of symptoms in PTSD patients(DSM-IV-TR PTSD Criterian D/E 13), (2) enhanced plasma corticosterone (CORT) for several (10) days, indicating compromise of the hypothalamopituitary axis (HPA), and (3) retarded body weight gain after stressor cessation, corresponding to the dysfunction of metabolic regulation of PTSD. There is no evidence in the literature that fox odor, predator exposure, or fear potentiated startle response, exhibit these persistent behavioral and neuroendocrinologic phenotypes associated with PTSD.
Rats exposed to a single stress session (1DS) have exhibited transient, but not the persistent abnormalities displayed by 3DS rats 17 The present experiment compared the startle response of 3DS and 1DS rats 4, 7, and 10 days after stressor cessation. Consistent with previous work, stressed rats exhibited elevated basal plasma CORT levels the first day post stress 17. These CORT levels were sensitive to the number of stressor exposures with higher CORT levels in 3DS rats than in 1DS rats. As for startle response, the 1DS rats exhibit an exaggerated startle response 7 days post stressor, whereas startle sensitization only becomes apparent 10 days post stressor in 3DS rats. Thus, the appearance of an exaggerated startle response after stressor cessation appears to be related to the number of stress session exposures. The 3DS stressed model appears to be useful to gain insight into the altered expression of biomarkers associated with the symptoms of PTSD and the key measurable behavioral phenotypes associated with the timing following the cessation of stress. Genomic results presented provide proof of principle for applying rMNChip and bioinformatics tools to identify differential pathways, and gene and protein biomarkers, which will greatly facilitate systems-biological study and understanding of molecular mechanisms underlying complex and multifactorial neurologic disorders, including PTSD.
While our paradigm does not delve into the cognitive and more complex behavioral aspects of PTSD, we note that altered sleep patterns in the ITS model 1 correspond to the difficulty falling and staying asleep and the nightmares of PTSD patients 11 (DSM-IV-TR PTSD Criteria D 13), and the deficiencies in escape/avoidance learning and learning of an appetitive task in the ITS model 12 corresponds to the poor concentrations and memory deficits of PTSD 5 (DSM-IV-TR PTSD Criteria C 13). The current model correlates well with the key symptoms characteristic of PTSD and provides a good model for correlating peripheral biomarkers of PTSD, which can be collected from patients, with central, mechanistic biomarkers, which cannot 6.
The authors have nothing to disclose.
This work was supported by the CDMRP, USUHS Grants G188LE, G188MG, and G188QC (to HL), and the USUHS Center for the Study of Traumatic Stress.