Introduction

Post-traumatic stress disorder (PTSD) is a chronic and disabling anxiety disorder that may develop in survivors of a traumatic event.¹ PTSD is currently defined by the coexistence of three clusters of symptoms: re-experiencing, avoidance and hyperarousal, persisting for at least 1 month.²

Animal models, although limited in their ability to comprehend human complexities, are an invaluable tool in the research of markers for psychiatric disorders in general, and PTSD in particular. A good PTSD animal model of clinical conditions, should fulfill four criteria:^{3, 4} etiological, face, construct and predictive validity. Unlike many other mental disorders, the diagnostic criteria for PTSD specify an etiological factor, which is an exposure to a life-threatening, traumatic event.⁵

Animal models differ from one another in their face, construct and predictive validities. A number of animal models have been developed, mimicking many of the behavioral and physiological changes seen in PTSD-like behavior. These models use different stress paradigm and show a wide range of behavioral and physiological manifestation in addition to different therapeutic effects of drugs. The application of direct physical stressors, such as electric shock,^{6, 7, 8, 9} underwater trauma¹⁰ and restraint stress,^{11, 12} are the most widely used methods of applying a stressor to laboratory animals. Nevertheless, the use of exogenous stimuli that closely mimic those seen in nature, such as exposure to a live predator,^{13, 14, 15} a predatory cue^{16, 17, 18, 19} or psychological stress,^{20, 21} might have greater etiological relevance, thereby leading to improved modeling and analysis of fear and anxiety states.

Stressed rats tend to show PTSD-like behavior such as increased immobility, decreased grooming and rearing,¹⁸ decreased exploratory behavior and decreased food consumption.²² The 'freezing' response has been used as a behavioral measure of anxiety or fear.²³ Among all the unconditioned stressors, the predator stress seems to be the most potent stressor, since its effects on fear/anxiety potentiation can last for 3 weeks.¹³

In addition to behavioral models which utilize different components of environmental stimuli, many researchers use genetic models such as inbred strains, selected lines, linkage studies and gene targeting techniques (reviewed in Clement et al.⁴).

Most studies using PTSD animal models refer to the entire exposed PTSD group as a uniform population.^{7, 9, 24} A recent study showed that animals respond to stress heterogeneously and that it is essential to divide them into distinct groups according to magnitude of response and to study them accordingly.¹⁵ This study emphasizes the immediate need for a new approach, which categorizes PTSD-like behavior individually. We took this model and further analyzed the animals for up to 28 days by treating each animal data individually according to three clusters of symptoms: manifestation of freezing behavior when (a) situated alone in an open field paradigm, (b) situated with a habituated companion and (c) after exposure to a hyperarousal event. Furthermore, we assessed predictive validity by testing the effect of citalopram (a serotonin selective reuptake inhibitor (SSRI)) on these three different symptoms.

Based on this setup, we looked for a new putative biological marker. Biological alterations may underline the onset and persistence of PTSD symptoms.^{4, 25, 26} Such alterations are likely to be associated with differential gene transcription and may last long term after the exposure to the triggering event. The long-term alterations are likely to be associated with the PTSD state, while alterations that fade with time are more likely to represent the response to the acute stressful event.²⁷

The WFS1 gene is located in chromosome 4. This gene undergoes alternative splicing and in the general population has polymorphism. The product of this gene (Wolframin) is a glycoprotein located in the endoplasmic reticulum. This glycoprotein is highly distributed in the pancreas, heart and several brain areas. The function of Wolframin is completely unknown. Analysis of the structural features provides experimental evidence that Wolframin contains nine transmembrane segments and is embedded in the membrane into higher molecular weight complexes. In the rat brain, it is predominantly present in neurons of the hippocampus CA1, amygdaloid area, olfactory tubercle and the superficial layer of the allocortex.

Wolfram syndrome is caused by mutations in WFS1. These mutations are rare, and its recessive disorder is characterized by early-onset, non-autoimmune diabetes mellitus, optic atrophy, deafness and further neurological and endocrinological abnormalities.

Swift et al.²⁸ reported that 60% of Wolfram syndrome patients, when homozygous for WFS1 alleles, showed psychiatric symptoms in patients with high prevalence of depression.²⁹ Heterozygous carriers are at a 26-fold risk of psychiatric hospitalization, predisposed to psychiatric illness. Moreover, 1% of the population are carriers of a WFS1 mutation, which may contribute to susceptibility for psychiatric diseases.

Exposure of animals to predators or predator odors activates the hypothalamic-pituitary-adrenal axis plasma and significantly increases corticosterone in male laboratory rats.¹⁵ Even very brief (for 5 min) exposure to predator odor activates the WSF1 gene in the amygdala and in other brain structures.³⁰

In the present study we (i) modified the existing animal model of PTSD (ii) re-test its face and predictive validities and (iii) suggest a putative candidate gene, WSF1, as a biological marker for PTSD.

Methods

Animals and treatments

Adult male Sprague–Dawley rats (250–300 g; Harlan, Rehovot, Israel) were used and housed under conditions of constant temperature (22 °C) and 50% humidity, with a 12-h light–12-h dark cycle. The animals were housed three per cage, two experimental rats together with a third companion rat. The same three rats remained together until the end of the study. Food and water were provided ad libitum. Animals were allowed to habituate in the animal house of Bar-Ilan University for 14 days.

All animal procedures were approved by the Bar-Ilan University Animal Care Committee and were carried out in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

Behavioral measurements

The following behaviors conditions were measured. Freezing time was monitored (videotaped) using an observer apparatus (Noldus, the Netherlands). The rat was tested alone (5 min), with its habituated companion (5 min), and after exposure to a loud noise (hyperarousal event) during 'post-startle response' (5 min). The loud voice was a 36.50.3 dB pick/scale A over baseline noise of 550.5 dB measured by Quest instrument, Quest Technologies, model 2900, calibrated by QC-10 calibrator 114 dB- 1000 Hz, which was recorded and was played back to the tested rats in each 'hyperarousal' condition. After the exposure to the litter, the rat was moved to the 'open field' in its home cage, and during the test, the tested-rat was not removed from the open field nor touched. The three experimental conditions took place in the same box consecutively. The freezing time was calibrated to the percentage (%) of total time (see Results). 'Freezing' was defined when the rat stayed still without any movement for at least 2 s.

In the preliminary study, freezing, grooming, sniffing, climbing over, staying in the corner, crawling under and genital investigation were measured. After an initial analysis of the above parameters, we concluded that the freezing parameter was the only parameter that was significantly different between exposed PTSD-like and exposed non-PTSD-like rats and acute exposed rats.

Behavioral procedure

The model procedure spreads over 10 weeks as illustrated in Figure 1 and contained several stages.

Figure 1.

Flow chart of the behavioral model procedure.

Full figure and legend (58K)

Habituation

Rats were initially habituated in the 'open field' with their cage companion for 14 days, 5 min per day. The 'open field' was a 90 90 30 cm plastiglass box, which was placed under the camera. All experiments were performed between 0800 and 1400 in daylight. The video and computer equipment were situated in a separate room where all video and observation analyses were performed.

Baseline

Baseline for each of the conditions (exploration, with a companion rat and after a hyperarousal event) was measured on day 15 of habituation.

Initial exposure

One week later (day 22), experimental rats were exposed to litter with a cat scent (trauma). The trauma was performed in a separate testing area. The temperature and humidity in the exposed area were identical to the housing and the testing area conditions. Individual rats were placed in a clean plastic cylinder containing 125 ml of cat litter (litter which the cat used during the 24 h before the experiment) for 30 min, after which they were transferred to the 'open field', where the behavioral parameters were measured as described above. The cylinder was cleaned between each exposed rat and the same cat was used through the whole set of experiment. The group of rats, which were not exposed to the trauma, was housed in separate cages in a separate housing area during all experiments. The control group was exposed to the exact same conditions but to litter without a cat scent. Controls (that is, not exposed to trauma) were always tested 1 day before the exposed rats. The 'open field' where the rats were tested was cleaned between each tested rat.

Re-exposure (Re₁)

One week following the initial exposure (day 29), rats were re-exposed to litter with the same texture (exposures were in the same conditions of 30 min to the litter without the cat scent) and tested for behavior under three separate conditions.

Re-re-exposure (Re₂)

Four weeks after the initial exposure (day 50), rats were again exposed to litter with the same texture, but without the cat scent. Immediately after this exposure, they were tested for behavior under three separate conditions and then defined as exposed PTSD-like (PTSD-like) or exposed non-PTSD-like rats based on their behavioral results.

Behavioral data analysis

The range of each measured behavioral parameter was determined for all animals after habituation (week 3). We analyzed the data collected at baseline and applied the explore procedure using SPSS 11 to define the range of the population. The upper and lower levels of this range were treated as 'normal baseline' and alterations from this range were used to define PTSD-like behavior. Animals were subsequently divided into two groups: 'exposed PTSD-like' (exhibiting behavior above 'normal baseline' in all three conditions) and 'exposed non-PTSD-like' (non-PTSD-like animals exhibiting at least one behavior in the 'normal baseline' range).

Treatments

After identifying exposed PTSD-like animals, these animals were randomly separated into two groups. Disposable micro-osmotic pump (Model 2ML2, ALZET, DURECT Corporodion, Cupertino, CA, USA) filled with saline containing citalopram HBr (Lundbeck A/S, Copenhagen-DK; 20 mg/kg/day) or saline as a vehicle control was implanted subcutaneously. The pump continuously released solutions for 14 days (5 l h^-1). At the completion of the 14 days of citalopram or saline treatment, the rats were re-tested in the open field for behavioral performance under the three conditions. The control group (which was exposed to litter without the cat scent on the 'initial exposure'; see Figure 1) also received citalopram or saline treatment. The observer and experimenter were blinded as to the treatment.

Molecular methods

cDNA synthesis

Rats were decapitated 24 h after last behavior analysis and their brains were removed rapidly. Serial 0.5 mm sections were cut and tissue punches of the basolateral amygdala, CA1 of the Hippocampus, as previous described,³¹ were frozen at -70 °C until extraction.

Total RNA was isolated separately from both the basolateral amygdala and the CA1 sections from the control-saline, PTSD-saline and PTSD-citalopram-treated rats by single-step method (TriReagent, Sigma, Rehovot, Israel) using the manufacturers recommended procedure.

First-strand cDNA synthesis was carried out in a final reaction of 20 l in the buffer provided by MmuLV-reverse-transcriptase (Roche, Mannheim, Germany). Three micrograms of total RNA, 20 pg oligo-dT primer (Roche) and 1 mM of dNTP mix (Roche) were added to the buffer. After heating to 65 °C for 2 min and cooling to 4 °C for 1 min, 50 U MmuLV-reverse-transcriptase and 20 U of RNase inhibitor (Roche) were added, and the reverse transcription was allowed to proceed at 37 °C for 60 min tubes without MmuLV-reverse-transcriptase served as negative controls.

Quantitative RT-PCR

The resulting first-strand cDNA (2 ml) was added to the PCR mixture containing 0.2 mM of dNTP mix, 1 mM of each oligonucleotide primer and 2.5 U Taq DNA polymerase in the buffer supplied by the enzyme manufacturer (Roche) for a final reaction volume of 50 l. The primers used in the study are as follows:

WFS1 up 5'-GACCTCAGCCTGACTCCAGT-3';

primer down 5'-CTTTCCGGTCAGCTAGGCAC-3';

-actin up 5'-GGTATGGGTCAGAAGGACTCC-3';

primer down 5'-TCAGGATCTTCATGAGGTAGTC-3'.

Reactions were initially denatured at 94 °C for 2 min. PCR was then preformed using a thermal cycler (MJ Research, Watertown, MA, USA). For WFS1, the thermal cycler was programmed for 10 cycles, denaturation at 94 °C for 1 min, primer annealing at 55 °C for 1 min and primer extinction at 72 °C for 1 min, then 30 cycles, consisting of denaturation at 94 °C for 1 min, annealing at 60 °C for 1 min and primer extinction at 72 °C for 1 min using the optimal conditions for quantitative analysis (40 cycles for WFS1), the levels of the WFS1 mRNA were compared to -actin levels. The PCR products were analyzed on 1% agarose gels containing athidium bromide, and visualized under ultraviolet light. The bands were scanned by a ultraviolet scanner, and the ratio between the target gene and -actin gene band densities was used for quantitative evaluation. Image densitometric analysis was preformed using the NIH ImajeJ software by David Chow, Division of Computer Research and Technology, NIH 1998 edition.

Statistical analysis

An analysis of variance with a Bonferroni or Tamhane post hoc tests was used to compare the freezing time (as % of upper limit) between groups (non-exposed control, exposed stressed, exposed non-stressed). A paired two-sample Student's t-test was used to compare freezing behavior of treated animals. A Pearson's ²-test was used to examine whether there is a statistical dependence between treatment and percent of freezing time. A probability value of P<0.05 was considered significant. Data are expressed as the means.e.m.

Results

Behavioral results

A total of 176 rats were tested in the model. Preliminary results indicated the incidence of PTSD-like animals using our procedure to be about 16%. Therefore, 131 rats were exposed to the traumatic event (litter with cat odor) resulting in 21 PTSD-like animals and 45 were used as control animals (exposed to litter without the odor). Analyzing all baseline samples of behavior (explore procedure) found the upper level for excluding outliers (95% confidence) in the 'exploration' and 'hyperarousal' to be 1.5 interquartile range and for the 'companion condition' 2 interquartile range. According to the descriptive statistics, four rats were excluded from the study as outliers.

Analysis of the behavior before pharmacological treatment

The mean value of freezing time is presented in Table 1. Following the second re-exposure (Re₂, 35 days post trauma), rats were divided into two groups: exposed PTSD-like ('PTSD'; n=21, 16.04%) and exposed non-PTSD-like ('non-PTSD'; n=110) rats. Identification of the animals as PTSD and non-PTSD was depicted only after the second re-exposure and only if all three behavior clusters were above the defined baseline. Retrospective scattered behavior of all the animals in each of the three groups during the three checkpoints is compared in Figure 2. Inconsistency in alterations of behavior can be recognized after the traumatic event (litter with cat scent) and first reminder 'exposure' (litter without cat scent). The data indicate that, in our setup, only after Re₂ PTSD-like rats are distinguishable.

Figure 2.

Freezing behavior is presented in alone, with companion and hyper-arousal conditions separately (a, b and c respectively), for each testing time post baseline (exposure, Re₁ and Re₂). Black circles represent 'control' group, open circles represent 'non-PTSD' group and gray circles represent 'PTSD' group. The gray rectangle represents the range of 'normal baseline' and the numbers represent the percent of animals that exhibit freezing behavior above this normal range. Significantly different values were detected with one-way analysis of variance followed by post hoc bonferroni or tamhane. ^*P<0.05 control vs non-PTSD, ^**P<0.05 control vs PTSD, ^***P<0.05 PTSD vs non-PTSD. Note that the division to 'PTSD' and 'non-PTSD' groups was performed after Re₂, therefore the figures were constructed in relation to retrospect data.

Full figure and legend (96K)

Table 1 - Mean value of freezing time as percent of upper limit of 'normal baseline' for three animal groups.

Full table

Analysis of the behavior after pharmacological treatment

PTSD-defined rats (total 21) were divided randomly into two groups. One group was treated with saline (n=11) and the other with citalopram (n=10). The magnitude of changes in freezing (normalized values) was determined by dividing the percentage of total time freezing during the 'test' (after 14 days of treatment) by the percentage of total time freezing in the 'Re₂' (pretreatment). As seen from the results (Figure 3a), the average 'companion condition' was significantly improved (40%). Inter-analysis of this improved group showed that 60% of the rats were above 30% improvement and 20% remitted (values were restored to within the range of baseline). 'Exploration' and 'hyperarousal' were not significantly affected by citalopram compared to saline, when analyzed for each rat individually (data not shown) or averaged (Figure 3a).

Figure 3.

Effect of citalopram treatment on exposed PTSD-like and control rats. Changes in freezing (normalized values) in exposed PTSD-like rats (a) treated with saline (white column, n=11) or citalopram (black column, n=10) or control rats (b) treated with saline (n=10) or citalopram (n=10), for 14 days in the three conditions separately. The magnitude of changes in freezing (normalized values) was determined by dividing the percent of total time freezing during the test (after 14 days of treatment) by the percent of total time freezing in the Re₂ (before treatment). Decreased values indicate improvements in freezing behavior. The mean of normalized freezing valuess.e.m. are presented (Paired t-test, P<0.05).

Full figure and legend (16K)

Normalized values for control-treated rats (Figure 3b) showed no differences between the two treatments under all three conditions. Unexpectedly, the percentage of freezing time was increased in both treatments, but eventually remained within the 'normal baseline' range of the population.

Wolframin expression in brain regions

Transcript levels of Wolframin gene were elevated in the CA1 hippocampus and basolateral amygdala of PTSD rats vs controls treated with saline. Transcript levels in animals exposed to the same traumatic procedure without developing enhanced freezing (non-PTSD-like rats) were identical to controls. Treatment of PTSD animals with citalopram for 14 days normalized this elevation in both regions (Figure 4).

Figure 4.

Quantitative analysis of WFS1 mRNA expression. WFS1 mRNA expression levels in the CA1 and the Amygdala (a and b) following saline or citalopram treatment. Different values (meanss.e.m.) significantly detected (a, F_(3,11)=8.64, P<0.05; b, F_(4,10)=32.83, P<0.01, accordingly) using one-way analysis of variance followed by post hoc bonferroni. ^*P<0.01, PTSD-saline vs PTSD-citalopram, ^**P<0.05, non-PTSD vs control-saline and control-saline vs PTSD-saline, ^***P<0.05 control-saline vs PTSD saline.

Full figure and legend (68K)

Discussion

The primary aim of the present study was to suggest a putative biological marker for PTSD using an animal model that is based upon individual behavioral analysis. This model enables us to identify and separate the animals, which in the long term, after exposure to the trauma, will develop symptoms that mimic the PTSD-like disorder (hereafter 'exposed PTSD-like animals').

Most animal models known today refer to the animal group exposed to a stressful event as if it were a homogenous population. We admit that this approach overlooks the nature of the human pattern response to a stressful stimulus, which resembles heterogeneity as not all individuals cope with a stressful event in a similar manner. Although animal models do not presume to reflect human disorders in an accurate way, they try to approximate them. Hence, referring to all exposed animals as a uniform 'exposed PTSD-like population' without relating individually to the behavior of each exposed animal is literally the same as referring to all persons involved in a traumatic event as exhibiting a kind of PTSD-like behavior.

In human psychiatric practice, people suffering from PTSD are diagnosed, using structured questionnaires, such as CAPS (clinician-administered PTSD scale), according to three clusters criteria ('re-experiencing', 'avoidance' and 'hyperarousal'). We picked similar criteria for our analysis of maladapted animals and measured freezing time to a familiar open area when alone, with a nonspecific companion, and after a standardized loud voice (named them 'companion condition' and 'hyperarousal', accordingly). With caution, we suggest that these parameters may parallel the clinical manifestations of social withdrawal (isolation) and hyperarousal. Re-experiencing is modeled by given the trauma reminder aspect of second exposure to unscented cat litter and named 'exploration' after second exposure.

Moreover, we analyzed each animal individually, longitudinally and compared it to its own line and to the average of the normal naive population. Using this approach, our results show that 16.04% of the rats that were exposed and then re-exposed to the stressor developed PTSD-like symptoms. This result correlates well with statistics reported in human studies.^{32, 33} Accordingly, following exposure to a traumatic event, 15–25% of the exposed population developed PTSD, whereas the majority showed spontaneous recovery within a year of the traumatic event.¹ Estimates of the lifetime prevalence of PTSD from surveys of the general adult population have ranged from 1 to 12.3%.⁵ Although a higher lifetime prevalence of around 30% has been reported for Vietnam veterans and female victims of rape,³⁴ a recently published reanalysis of the NVVRS data in Vietnam veterans reported lower PTSD rates.^{35, 36}

Our model is based on the model developed by Cohen and the second author (JZ)¹⁵ and this study supports further the face validity of this model in mimicking clinical PTSD. However, the model presented in this paper stresses the importance of the reminder 'component' (that is, the re-experiencing phenomenon.). As can be seen in Figure 2, a single exposure was not sufficient to distinguish between those animals that will eventually display PTSD-like symptoms. Moreover our model also stresses the need for long-term evaluation of the symptoms following the exposure to the trauma.

In order to test predictive validity and allocate a possible biological marker, we then proceeded to test the effect of citalopram, an SSRI used in PTSD patients, in our model. We found a significant reduction as a result of citalopram treatment, in 'companion condition' (50%), with an average though nonsignificant reduction in the 'avoidance' (50%) and 'hyperarousal' (30–34%) clusters. A case report³⁷ of two gulf war veterans that were treated with citalopram was also in line with these results, reporting an improvement in the social isolation while the flashbacks and intrusive thoughts continued. However, other studies reported that treatment of PTSD subjects with citalopram (20–40 mg per day, for 8–10 weeks) resulted in a significant reduction of total symptoms of PTSD and in all three above-mentioned symptom clusters separately.^{38, 39, 40, 41} A comparative study of the efficacy of citalopram treatment in children/adolescents and adults found that the mean reduction in total PTSD symptoms ranged from 39 to 54%, whereas only around 65% of the subjects were classified as responders at end point.⁴⁰ It is possible that longer treatment with citalopram could reveal more significant behavioral results in our 'exposed PTSD-like' rats, since human studies show an optimal efficacy of the drug only after 4–6 weeks of treatment.^{38, 39} Elaboration of animal studies using citalopram as compared to other drugs is required to enable more accurate comparison to clinical results.

Testing the average level of improvement in percentage of total freezing time, as exhibited by saline- and citalopram-treated control animals, is important for excluding decreased physical performance as an effect of surgery or injections. Results show an increase in freezing time after treatment of controls that might be related to the stressful procedure of injection. Moreover, if citalopram does actually increase the freezing time of control animals and decreases the freezing time in exposed PTSD-like animals, then the results of the 'exposed PTSD-like' animals are underestimated and even more significant.

Most studies on PTSD concentrate on long-term alterations in sympathetic and HPA axis reactivity.^{42, 43} Some studies describe alterations in the immune modulations.^{42, 44, 45, 46, 47, 48, 49} One pivotal study suggested that changes in peripheral blood mononuclear genes may serve as a fingerprint.²⁶ However, it was unable to conclude that these changes are merely informative or bear relevance for the pathogenesis of PTSD. Nonetheless, while these alterations were suggested as expressing a modulator role in PTSD, these markers are peripheral and not necessarily reflect the alteration in the site of the disease. We tried to locate a central (brain) marker since we believe that such marker might be more indicative of the post-traumatic syndrome pathology that lead to the disease.

Human studies indicating that WFS1 gene may play a role in psychiatric disorders were first suggested by Swift et al.⁵⁰ They reported that heterozygous carriers of WFS1 mutations are 26-fold more likely to require psychiatric hospitalization than non-carriers. On the basis of these observations, WFS1 was investigated in several studies as a candidate for psychiatric disorders.^{12, 51, 52, 53} The relevance of these variants is difficult to interpret since a control population was not included. Nevertheless, haplotype analysis revealed a common GTA haplotype, formed by single-nucleotide polymorphisms 684C/G, 1185C/T and 1832G/A, conferring risk for affective disorders and anxiety.²⁹

As suggested by others, early changes in gene expression have little relevance to the implantation of a chronic illness if it does not persist on the long term.⁵⁴ We postulate that changes in WFS1 in the brain, specifically in the CA1 and amygdala, may serve as a 'reporter gene' for PTSD since alteration in its expression is associated with the development of long-term PTSD. Moreover, WFS1 remained unchanged in animals that were exposed to the same traumatic procedure but did not develop PTSD behavior (exposed non-PTSD-like rats). Chronic treatment with citalopram normalized long-term changes of WFS1 in PTSD-like animals parallel to improvement in their behavioral manifestation (Figure 4). At this point we cannot indicate when after the traumatic event the expression of this gene is increased, but we can say that its alteration is long term. In addition, at this time we cannot define the WFS1 mechanism in PTSD as there is not yet available knowledge regarding wolfram function.

Although in this study we did not test the distribution of WFS1 mutation as a possible cause of differences in the response to the trauma, it is tempting to speculate that individuals carrying a WFS1 mutant gene will be more vulnerable to develop PTSD. Hofmann⁵⁵ has described molecular abnormalities in the WFS1 gene and has shown that specific mutations in the WFS1 gene affect protein stability resulting in lower wolfram (protein) levels. Preliminary data from our laboratory have shown that in PTSD-like rats, both WFS1 expression and translation levels are lowered (not published observation).

Identification of markers in PTSD, such as Wolframin, may enable future screening for and identification of individuals vulnerable to the development of PTSD after exposure to a traumatic event. Providing these individuals with prophylactic treatment may prevent or decrease the prevalence of development of the disorder.

References

1. Davidson JR, Stein DJ, Shalev AY, Yehuda R. Posttraumatic stress disorder: acquisition, recognition, course, and treatment. J Neuropsychiatry Clin Neurosci 2004; 16: 135–147. | PubMed | ISI |
2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM IV). American Psychiatric Press: Washington, DC, 1994.
3. Bloom FE, Kupfer DJ. Psychopharmacology: The Fourth Generation of Progress. Raven Press: New York, 1995.
4. Clement Y, Calatayud F, Belzung C. Genetic basis of anxiety-like behaviour: a critical review. Brain Res Bull 2002; 57: 57–71. | Article | PubMed | ISI |
5. Nutt D, Davidson JRT. Post-Traumatic Stress Disorder Diagnosis, Management and Treatment. Taylor & Francis: London, 2000.
6. Servatius RJ, Ottenweller JE, Natelson BH. Delayed startle sensitization distinguishes rats exposed to one or three stress sessions: further evidence toward an animal model of PTSD. Biol Psychiatry 1995; 38: 539–546. | Article | PubMed | ChemPort |
7. Pynoos RS, Ritzmann RF, Steinberg AM, Goenjian A, Prisecaru I. A behavioral animal model of posttraumatic stress disorder featuring repeated exposure to situational reminders. Biol Psychiatry 1996; 39: 129–134. | Article | PubMed | ChemPort |
8. Garrick T, Morrow N, Shalev AY, Eth S. Stress-induced enhancement of auditory startle: an animal model of posttraumatic stress disorder. Psychiatry 2001; 64: 346–354. | PubMed | ChemPort |
9. Koba T, Kodama Y, Shimizu K, Nomura S, Sugawara M, Kobayashi Y et al. Persistent behavioural changes in rats following inescapable shock stress: a potential model of posttraumatic stress disorder. World J Biol Psychiatry 2001; 2: 34–37. | PubMed | ChemPort |
10. Richter-Levin G. Acute and long-term behavioral correlates of underwater trauma—potential relevance to stress and post-stress syndromes. Psychiatry Res 1998; 79: 73–83. | Article | PubMed | ChemPort |
11. Liberzon I, Krstov M, Young EA. Stress-restress: effects on ACTH and fast feedback. Psychoneuroendocrinology 1997; 22: 443–453. | Article | PubMed | ISI | ChemPort |
12. Torres IL, Gamaro GD, Vasconcellos AP, Silveira R, Dalmaz C. Effects of chronic restraint stress on feeding behavior and on monoamine levels in different brain structures in rats. Neurochem Res 2002; 27: 519–525. | Article | PubMed | ChemPort |
13. Adamec RE, Shallow T. Lasting effects on rodent anxiety of a single exposure to a cat. Physiol Behav 1993; 54: 101–109. | Article | PubMed | ChemPort |
14. Adamec RE, Burton P, Shallow T, Budgell J. NMDA receptors mediate lasting increases in anxiety-like behavior produced by the stress of predator exposure-implications for anxiety associated with posttraumatic stress disorder. Physiol Behav 1999; 65: 723–737. | Article | PubMed | ChemPort |
15. Cohen H, Zohar J, Matar M. The relevance of differential response to trauma in an animal model of posttraumatic stress disorder. Biol Psychiatry 2003; 53: 463–473. | Article | PubMed |
16. Cohen H, Kaplan Z, Kotler M. CCK-antagonists in a rat exposed to acute stress: implication for anxiety associated with post-traumatic stress disorder. Depress Anxiety 1999; 10: 8–17. | Article | PubMed | ChemPort |
17. Cohen H, Benjamin J, Kaplan Z, Kotler M. Administration of high-dose ketoconazole, an inhibitor of steroid synthesis, prevents posttraumatic anxiety in an animal model. Eur Neuropsychopharmacol 2000; 10: 429–435. | Article | PubMed | ChemPort |
18. Morrow BA, Redmond AJ, Roth RH, Elsworth JD. The predator odor, TMT, displays a unique, stress-like pattern of dopaminergic and endocrinological activation in the rat. Brain Res 2000; 864: 146–151. | Article | PubMed | ChemPort |
19. Dielenberg RA, Hunt GE, McGregor IS. 'When a rat smells a cat': the distribution of Fos immunoreactivity in rat brain following exposure to a predatory odor. Neuroscience 2001; 104: 1085–1097. | Article | PubMed | ChemPort |
20. Makino S, Shibasaki T, Yamauchi N, Nishioka T, Mimoto T, Wakabayashi I et al. Psychological stress increased corticotropin-releasing hormone mRNA and content in the central nucleus of the amygdala but not in the hypothalamic paraventricular nucleus in the rat. Brain Res 1999; 850: 136–143. | Article | PubMed | ChemPort |
21. Funada M, Hara C. Differential effects of psychological stress on activation of the 5-hydroxytryptamine- and dopamine-containing neurons in the brain of freely moving rats. Brain Res 2001; 901: 247–251. | Article | PubMed | ChemPort |
22. Ottenweller JE, Natelson BH, Pitman DL, Drastal SD. Adrenocortical and behavioral responses to repeated stressors: toward an animal model of chronic stress and stress-related mental illness. Biol Psychiatry 1989; 26: 829–841. | Article | PubMed | ChemPort |
23. Feldman RS, Meyer JS, Quenzer LF. Affective disorders. In: Feldman RS, Meyer JS, Quenzer LF (eds). Principles of Neuropharmacology. Sinauer: Sunderland, 1997, pp 819–861.
24. Mechiel KS, De Boer SF. A robust animal model of state anxiety: fear-potentiated behaviour in the elevated plus-maze. Eur J Pharmacol 2003; 463: 163–175. | Article | PubMed | ChemPort |
25. Ozer EJ, Best SR, Lipsey TL, Weiss DS. Predictors of posttraumatic stress disorder and symptoms in adults: a meta-analysis. Psychol Bull 2003; 129: 52–73. | Article | PubMed | ISI |
26. Segman RH, Shefi N, Goltser-Dubner T, Friedman N, Kaminski N, Shalev AY. Peripheral blood mononuclear cell gene expression profiles identify emergent post-traumatic stress disorder among trauma survivors. Mol Psychiatry 2005; 10: 500–513, 425. | Article | PubMed | ISI | ChemPort |
27. Hope BT, Crombag HS, Jedynak JP, Wise RA. Neuroadaptations of total levels of adenylate cyclase, protein kinase A, tyrosine hydroxylase, cdk5 and neurofilaments in the nucleus accumbens and ventral tegmental area do not correlate with expression of sensitized or tolerant locomotor responses to cocaine. J Neurochem 2005; 92: 536–545. | Article | PubMed | ChemPort |
28. Swift M, Swift RG. Wolframin mutations and hospitalization for psychiatric illness. Mol Psychiatry 2005; 10: 799–803. | Article | PubMed | ChemPort |
29. Koido K, Koks S, Nikopensius T, Maron E, Altmae S, Heinaste E et al. Polymorphisms in Wolframin (WFS1) gene are possibly related to increased risk for mood disorders. Int J Neuropsychopharmacol 2005; 8: 235–244. | Article | PubMed | ISI | ChemPort |
30. Koks S, Planken A, Luuk H, Vasar E. Cat odour exposure increases the expression of Wolframin gene in the amygdaloid area of rat. Neurosci Lett 2002; 322: 116–120. | Article | PubMed | ChemPort |
31. Zangen A, Overstreet DH, Yadid G. High serotonin and 5-hydroxyindoleacetic acid levels in limbic brain regions in a rat model of depression: normalization by chronic antidepressant treatment. J Neurochem 1997; 69: 2477–2483. | PubMed | ChemPort |
32. Breslau N, Davis GC, Andreski P, Peterson E. Traumatic events and posttraumatic stress disorder in an urban population of young adults. Arch Gen Psychiatry 1991; 48: 216–222. | PubMed | ChemPort |
33. Breslau N, Kessler RC, Chilcoat HD, Schultz LR, Davis GC, Andreski P. Trauma and posttraumatic stress disorder in the community: the 1996 Detroit Area Survey of Trauma. Arch Gen Psychiatry 1998; 55: 626–632. | Article | PubMed | ChemPort |
34. Jordan BK, Schlenger WE, Hough R, Kulka RA, Weiss D, Fairbank JA et al. Lifetime and current prevalence of specific psychiatric disorders among Vietnam veterans and controls. Arch Gen Psychiatry 1991; 48: 207–215. | PubMed | ChemPort |
35. Dohrenwend BP, Turner JB, Turse NA, Adams BG, Koenen KC, Marshall R. The psychological risks of Vietnam for U.S. veterans: a revisit with new data and methods. Science 2006; 313: 979–982. | Article | PubMed | ChemPort |
36. Resnick HS, Kilpatrick DG, Dansky BS, Saunders BE, Best CL. Prevalence of civilian trauma and posttraumatic stress disorder in a representative national sample of women. J Consult Clin Psychol 1993; 61: 984–991. | Article | PubMed | ChemPort |
37. Khouzam HR, el Gabalawi F, Donnelly NJ. The clinical experience of citalopram in the treatment of post-traumatic stress disorder: a report of two Persian Gulf War veterans. Mil Med 2001; 166: 921–923. | PubMed | ChemPort |
38. Seedat S, Lockhat R, Kaminer D, Zungu-Dirwayi N, Stein DJ. An open trial of citalopram in adolescents with post-traumatic stress disorder. Int Clin Psychopharmacol 2001; 16: 21–25. | Article | PubMed | ChemPort |
39. Seedat S, Stein DJ, Emsley RA. Open trial of citalopram in adults with post-traumatic stress disorder. Int J Neuropsychopharmacol 2000; 3: 135–140. | Article | PubMed | ChemPort |
40. Seedat S, Stein DJ, Ziervogel C, Middleton T, Kaminer D, Emsley RA et al. Comparison of response to a selective serotonin reuptake inhibitor in children, adolescents, and adults with posttraumatic stress disorder. J Child Adolesc Psychopharmacol 2002; 12: 37–46. | Article | PubMed |
41. Tucker P, Potter-Kimball R, Wyatt DB, Parker DE, Burgin C, Jones DE et al. Can physiologic assessment and side effects tease out differences in PTSD trials? A double-blind comparison of citalopram, sertraline, and placebo. Psychopharmacol Bull 2003; 37: 135–149. | PubMed |
42. Southwick SM, Bremner JD, Rasmusson A, Morgan III CA, Arnsten A, Charney DS. Role of norepinephrine in the pathophysiology and treatment of posttraumatic stress disorder. Biol Psychiatry 1999; 46: 1192–1204. | Article | PubMed | ChemPort |
43. Yehuda R. Current status of cortisol findings in post-traumatic stress disorder. Psychiatr Clin North Am 2002; 25: 341–368, vii. | Article | PubMed |
44. McEwen BS, Magarinos AM. Stress effects on morphology and function of the hippocampus. Ann N Y Acad Sci 1997; 821: 271–284. | Article | PubMed | ChemPort |
45. Baker DG, Ekhator NN, Kasckow JW, Hill KK, Zoumakis E, Dashevsky BA et al. Plasma and cerebrospinal fluid interleukin-6 concentrations in posttraumatic stress disorder. Neuroimmunomodulation 2001; 9: 209–217. | Article | PubMed | ChemPort |
46. Spivak B, Shohat B, Mester R, Avraham S, Gil-Ad I, Bleich A et al. Elevated levels of serum interleukin-1 beta in combat-related posttraumatic stress disorder. Biol Psychiatry 1997; 42: 345–348. | Article | PubMed | ChemPort |
47. Maes M, Lin AH, Delmeire L, Van Gastel A, Kenis G, De Jongh R et al. Elevated serum interleukin-6 (IL-6) and IL-6 receptor concentrations in posttraumatic stress disorder following accidental man-made traumatic events. Biol Psychiatry 1999; 45: 833–839. | Article | PubMed | ISI | ChemPort |
48. Kawamura N, Kim Y, Asukai N. Suppression of cellular immunity in men with a past history of posttraumatic stress disorder. Am J Psychiatry 2001; 158: 484–486. | Article | PubMed | ChemPort |
49. Miller RJ, Sutherland AG, Hutchison JD, Alexander DA. C-reactive protein and interleukin 6 receptor in post-traumatic stress disorder: a pilot study. Cytokine 2001; 13: 253–255. | Article | PubMed | ChemPort |
50. Swift RG, Polymeropoulos MH, Torres R, Swift M. Predisposition of Wolfram syndrome heterozygotes to psychiatric illness. Mol Psychiatry 1998; 3: 86–91. | Article | PubMed | ChemPort |
51. Crawford J, Zielinski MA, Fisher LJ, Sutherland GR, Goldney RD. Is there a relationship between Wolfram syndrome carrier status and suicide? Am J Med Genet 2002; 114: 343–346. | Article | PubMed | ISI |
52. Martorell L, Zaera MG, Valero J, Serrano D, Figuera L, Joven J et al. The WFS1 (Wolfram syndrome 1) is not a major susceptibility gene for the development of psychiatric disorders. Psychiatr Genet 2003; 13: 29–32. | Article | PubMed | ISI |
53. Sequeira A, Kim C, Seguin M, Lesage A, Chawky N, Desautels A et al. Wolfram syndrome and suicide: evidence for a role of WFS1 in suicidal and impulsive behavior. Am J Med Genet B Neuropsychiatr Genet 2003; 119: 108–113. | Article | PubMed |
54. Khanim F, Kirk J, Latif F, Barrett TG. WFS1/Wolframin mutations, wolfram syndrome, and associated diseases. Hum Mutat 2001; 17: 357–367. | Article | PubMed | ISI | ChemPort |
55. Hofmann S, Philbrook C, Gerbitz KD, Bauer MF. Wolfram syndrome: structural and functional analyses of mutant and wild-type Wolframin, the WFS1 gene product. Hum Mol Genet 2003; 12: 2003–2012. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Lundbeck A/S, Copenhagen-DK (Adam Keeney) for the generous support establishing this study and providing citalopram HBr. The results of this paper are patent pending (Patent Number 10/549.596).