Post-traumatic stress disorder (PTSD) may develop as a failure of the body to reverse the acute stress response. Because of the sustained stress of the sympathetic hyperarousal state in PTSD, also immune functioning has early been hypothesized to be affected. Therefore, in this study the notion has been favored that genes involved in stress and immune responses are differentially transcribed in PTSD patients.
Two approaches have been pursued simultaneously: (1) re-evaluation of metabolic parameters tentatively associated with PTSD, and (2) cDNA microarray investigations using ‘stress/immune chips’. Recently, similar experiments have been carried out in patients suffering from acute traumatic events, using Affymetrix chips.1 In our study, considerable efforts have been made to accurately standardize any condition possibly interfering with measurements. Whole blood was used and, very importantly, the group of PTSD patients was homogeneous in that all of them experienced the same traumatic event (Ramstein air show catastrophe, 1989). Despite the long-time interval between the traumatic event and the time of this study, in these individuals' features characteristic for PTSD in the questionnaire interviews and typical psychosomatic symptoms still persist. However, on the metabolic level no differences between the patients and healthy controls could be detected with cortisol, adrenaline, noradrenaline, vanillylmandelic acid, homovanillic acid, and the cytokines interleukin (IL)-6, IL-1β and tumor necrosis factor-α (Supplementary Information 1).
The genes represented on our home-made ‘stress/immune chips’ had been selected based on up-to-date knowledge about their role in inflammation, apoptosis, stress response and related pathways. We included eight patients and eight controls (Supplementary Information 2). The chip experiments were carried out according to Zieker et al.2 After normalization of the data, we applied three statistical methods, the paired t-test, the Wilcoxon's test and the rank product method3 (Supplementary Information 3). Four up- and 14 downregulated genes (altogether ∼5% of total valid transcripts compared) have been obtained that showed statistical significance for at least one of the three statistical methods (Table 1).
In the present investigation, we have been focusing on the downregulated genes. Seven of those genes and one additional gene, the xc(−) glutamate–cystine–antiporter were selected for confirmations by quantitative (real-time) polymerase chain reaction (PCR), because they appeared to be physiologically connected. Real-time PCR results confirmed regulation of all of them except 3-phosphoglycerate dehydrogenase (Supplementary Information 4).
Most downregulated transcripts are associated with immune functions or with reactive oxygen species (ROS). The first set of transcripts encompasses thioredoxin reductase (TXR), superoxide dismutase (SOD) and the xc(−) transporter. Selenocysteine is the essential component for the catalytic activity of selenoenzymes,4 like TXR. There might be a connection between the availability of selenium in PTSD patients and the amount of TXR in their blood. It has been shown that SOD reduces the O2− radical to H2O2. In this context, we assume that H2O2 production is decreased owing to the downregulation of SOD. Therefore, temporarily reduced availability of antioxidative enzymes like SOD and TXR may lead to repetitive bursts of OH-radical overload,5 which cannot be efficiently compensated for. Hence, ROS accumulation may be one reason for the higher risk of PTSD patients to develop autoimmune diseases like insulin-dependent diabetes6 or cardiovascular diseases.7
Glutathione (GSH) is the major antioxidative intracellular redox buffer and cysteine (Cys) is a rate-limiting precursor for GSH synthesis. The primary uptake mechanism of cystine (Cys2) is mediated by the chloride-dependent xc(−) cystine–glutamate exchanger.8 Its downregulation as reported here may influence GSH replenishment through reduced Cys2 uptake. In this way, xc(−) transporter connects the oxidative pathway with SOD as a key enzyme and the GSH system (Figure 1). This exchanger also plays a crucial role in the maintenance of extracellular glutamate in brain and has been implicated in relapse to drugs.9 Interestingly, relapse can be attenuated by treatment of animals with N-acetyl cysteine (NAC), an xc(−) exchanger agonist.10 We, therefore, hypothesize that treatment of PTSD patients with NAC may alleviate stress burden by increasing the efficiency of Cys2/glutamate exchange.
The major transcripts of the second group of genes revealed by the microarray studies are IL-18, IL-16 and endothelial differentiation sphingolipid G-protein-coupled receptor 1 (EDG1) mRNAs. It has been shown that IL-18 is regulated by different mechanisms, one of which is its induction by H2O2.11 As suggested above, H2O2 might be decreased in PTSD patients owing to downregulated SOD. Further, IL-18 expression in keratinocytes reportedly is downregulated by corticotropin-releasing hormone (CRH).12 This downregulation may also occur in PTSD patients who typically express hyperactive central CRH.13 We believe that our data showing downregulation of IL-18 mRNA may result from decreased levels of H2O2 and increased levels of CRH. Pro-IL16 is cleaved to its mature bioactive form by caspase-3,14 which interacts with caspase-2. These enzymes are regulated by sphingosine 1-phosphate via EDG1. Downregulation of EDG1 leads to activation of caspase-3, which may be part of a mechanism to convert more pro-IL16 into its mature form.
In summary, using cDNA microarrays, we clearly see changes on the mRNA level in PTSD patients even 16 years after the traumatic event. The transcripts identified here have not been associated with PTSD yet. The majority of the molecular targets can be grouped around ROS-related metabolism. Even cytokines, like IL-18, can be viewed in this context. Additional transcripts on the gene lists may also be important in PTSD, but in a context distinct from ROS-related events. Whether or not the gene products are more suitable targets for drug therapy can only be tested on the protein level. Nonetheless, the transcripts identified in this investigation are promising candidates and may support the notion of temporary perturbances in the oxidative (stress) system and in specific immune parameters of PTSD patients.