Cytokine fluctuation during acute stress is correlated to life trauma

Our research found that there are biomarkers (cytokines) significantly increased during an acute stress response, that these cytokines are directly correlated to the level of trauma one experiences in life, and that some are also significantly related to prior PTSD diagnosis.

Abstract

BACKGROUND

Multiple studies have demonstrated that human neurobiology and behavior are inextricably linked to the activity of our immune systems. Trauma is associated with a multitude of immune system changes; reflecting this, posttraumatic stress disorder (PTSD) is often comorbid with immune-related conditions such as autoimmune disorders. To further investigate this phenomenon, we tested our hypothesis that cytokine fluctuations during and after an acute stress response correlates with experienced life trauma.

METHODS

Using a prospective observational approach, this cohort study measured biomarker profiles in firefighter participants (n = 63), with 9 participants having prior PTSD diagnoses and 54 without prior PTSD diagnoses. In addition, life trauma scores were determined from all participants using the Life Events Checklist 5 (LEC-5) survey. Baseline salivary biomarker concentrations were determined, along with levels immediately before, immediately after, and 1 hour following a standardized stressful training event. Biomarkers measured using these salivary samples included 42 cytokines and 6 steroid and thyroid hormones. The concentrations of these markers were then correlated, using Pearson correlation coefficients, with the participants' LEC-5 scores. t Tests were also performed to compare cytokine values between the populations with and without prior PTSD diagnosis.

RESULTS

Included in the cytokine panel were interleukin (IL)-8, IL-10, IL-1B, GCSF, IL1-Ra, Groα, IFNa2, PDGFAA, and VEGF, all of which demonstrated positive correlation at various time points in individuals with increased severity of LEC-5 scores (and thus increased experienced life trauma). Concentrations of Groα, PDGFAA, IL1-Ra, IL-1a, Mip1a, IL-1a, IL-6, Mip1b, TNFα, and TGFα were also found to be significantly altered at various time points in participants with prior PTSD diagnoses, demonstrating some overlap with the LEC-5 Pearson correlations.

CONCLUSION

The results support our hypothesis and demonstrate that LEC-5 scores are indeed significantly correlated to cytokine concentrations and fluctuations surrounding a stress test.

LEVEL OF EVIDENCE

Diagnostic Tests or Criteria; Level IV.

Multiple studies have suggested a complex relationship between the human immune system and our neurobiology. As evidence of this, lymphocytes and macrophages have been shown to have membrane surface receptors for neurotransmitters including noradrenaline, acetylcholine, dopamine, serotonin, γ-aminobutyric acid, endorphins, and hormones such as corticotropin, thus allowing them to respond to signals from the nervous system in a variety of scenarios. This includes the stress response, which can be activated in traumatic situations.¹ Furthermore, reflecting the bidirectional relationship that exists, cytokines released by immune cells can themselves induce the release of corticotrophin-releasing hormone, arginine vasopressin, adrenocorticotropic hormone, and glucocorticoids from the hypothalamic pituitary adrenal axis with its associated tissues and can interfere with neuroplasticity and neurogenesis in the central nervous system (CNS).^2,3 Suffice it to say, neither system exists completely independent of the other.

In an age where psychological conditions such as depression and anxiety demand further attention, understanding this relationship is of immediate relevance. In a review by Johnson et al.,⁴ they explored the effects of acute and chronic stress on the release of inflammatory cytokines within the brain. They found trends demonstrating a significant increase in the release of inflammatory cytokines interleukin (IL)-1β, TNFα, and IL-6, especially in the hippocampus (which is thought to play a central role in anxiety disorders).^4,5 These increases were especially pronounced in subjects experiencing chronic stress. Other studies have found similar trends. In relation to psychiatric disorders such as posttraumatic stress disorder (PTSD), Hoge et al.⁶ found that, of subjects suffering from PTSD or panic disorders, 85% had detectable levels of at least six common proinflammatory cytokines, compared with only 25% of control meeting the same criteria. This suggests that patients suffering from PTSD and similar disorders are likely to present with a generalized proinflammatory state.⁶ These relationships being established, to date, and to our knowledge, there does not exist an “immune profile” detailing the correlations between cytokine levels/dysregulation and an individual's experienced trauma and PTSD.

Posttraumatic stress disorder is a trauma/stress-associated psychiatric disorder that can develop in an individual after being involved in or witnessing an event that the individual perceives as traumatizing or extreme.⁷ It can present with a variety of symptoms including flashbacks, severe anxiety, dissociative episodes, and fleeing or combative behaviors.⁷ Interestingly however, not everyone who experiences or witnesses the same event will go on to develop PTSD. This suggests an underlying risk profile, perhaps immune modulated, toward the development of PTSD that has yet to be fully elucidated.

Although PTSD is primarily recognized as a psychiatric disorder, it also manifests with physiological dysfunctions, especially with regard to the immune system.⁸ In 2015, there was a large-scale retrospective study conducted on 666,269 Iraq war veterans. This study discovered a twofold increase in the risk of autoimmune diseases in individuals with PTSD compared with those without any psychiatric illness, and a 51% increased risk when compared with individuals with other psychiatric illnesses.⁹ This strongly suggests a considerable connection between trauma, whether from a chronic or acute stressor, and immune system dysregulation. To quantitatively describe this relationship between stress, inflammatory cytokines/biomarkers, and the immune system, we used the Life Events Checklist 5 (LEC-5) survey.

The LEC-5 is a survey that can be used to determine the number and severity of life trauma events individuals have experienced, allowing for the establishment of their exposure to a “PTSD Criterion A event”.¹⁰ Utilization of the LEC-5 as a measure of trauma exposure has validity demonstrated in numerous studies.^10,11 The survey includes experiences such as natural disasters, sexual assault, combat exposure, transportation accidents, and other similarly potentially traumatizing events. This survey allowed for the assignment of a “severity score” to each participant based on their potentially traumatic life experiences and a correlation analysis of that score to individualized biomarker panel results. Using this score, participants’ prior PTSD diagnoses, and participants’ measured biomarker concentrations, we aimed to further describe the link between trauma, acute stress, and immune modulators. Analysis was conducted in order to test our hypothesis that cytokine fluctuations in the time surrounding and during an acute stress response correlate with experienced life trauma.

PATIENTS AND METHODS

This is a prospective observational study investigating levels of salivary cytokines, steroid, and thyroid hormones of participants at baseline, immediately before, immediately after, and 1 hour after a psychophysical challenge containing both emotionally and physically stressful components. Levels of salivary cytokines and hormones were collected throughout the stress response for the purpose of comparing them to LEC-5 results via Pearson correlations and to evaluate by t test any significant differences in cytokine levels between the populations with and without prior PTSD diagnoses.

Participants

Our cohort consisted of 63 self-consented participants, and our study was conducted under institutional review board approval (approval number, 2019-0092). All participants were Fire Service recruits enrolled in a local fire academy program. This was a necessary factor because the psychophysical stress test (see procedure) we used was also a requirement to graduate from the fire academy and be employed. This study was performed over the course of 1 year during 2020 (April and June), using two full recruit classes at the academy.

The demographics self-reported by our participants were as follows: the mean age of participants was 30.6 years. Of the first responders, there were 58 males and 5 females (both gender and biological sex). Among them, 77.63% were White; 6.9%, Black; 6.9%, Hispanic; 5.17%, multiracial; 1.7%, Chinese; and 1.7%, Indian. These data were collected for completeness and to highlight instances where our data may or may not be completely representative (i.e., the sample is White predominated). Seventy-six percent of participants had prior work experience as emergency medical technicians, and 14% had experience in the military. Within our sample of firefighter recruits were nine individuals with a prior PTSD diagnosis (15.5%). Four of our participants were lost following consenting and initial sampling. One was infected with COVD-19, two others sustained injuries requiring removal from the study, and one participant's data were excluded because of being unavailable for a significant portion of the testing period. All participants indicated that they were not taking any anti-inflammatory medications or steroids at the time of the study.

After consenting, all participants were administered the LEC-5 and Hardiness Resilience Gauge surveys. For the purposes of this paper, we will focus on the LEC-5. The average LEC-5 score was 26.17 (possible range, 0–40). To establish a biomarker baseline, saliva samples were also collected from all participants, following the same procedure that would be used to take samples at the three points surrounding their stress test. The salivary sample was evaluated in a 42-plex cytokine/chemokine panel investigating levels of EGF, eotaxin-1, FGF-2, Flt-3 L, Fractalkine, G-CSF, GM-CSF, GROα, IFN-α2, IFN-γ, IL-1α, IL-1β, IL-1ra, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12 (p40), IL-12 (p70), IL-13, IL-15, IL-17A, IL-18, IP-10, MCP-1, MCP-3, MDC, MIP-1, MIP-1β, PDGF-AA, PDGF-AB/BB, RANTES, sCD40L, TGFα, TNFα, TNFβ, and VEGF-A, along with a 6-plex steroid-thyroid hormone panel investigating levels of cortisol, progesterone, estradiol, T3, T4, and testosterone.

Procedure

LEC-5 Scoring

As described previously, each participant was administered the LEC-5 at the onset of the project, and each participant was assigned a “severity score” based thereon. Points were assigned using the following guidelines: 3 points for each traumatic event on the checklist experienced, 2 points for each witnessed, and 1 point for each learned about. These scores were then correlated to cytokine salivary concentrations at each time point surrounding the stress test, along with Δ's of the biomarkers (i.e., prestress vs. stress).

Stress Test

The stress test administered to the participants was one already built into, as well as required, to pass and receive employment following their Fire Service academy training. It included both physical and psychological stressors, requiring them to navigate an overheating house while blindfolded, with the main priority being not removing their mask.

Saliva Sampling

As previously mentioned, saliva samples were collected from participants at baseline and then again at three periods surrounding the stress test. For the purposes of this paper, we will focus only on the three samples taken surrounding the stress test. The prestress sample was taken immediately before the event, another sample was taken directly after the event (stress), and the final sample was taken an hour following the stress event (recovery). For all participants, the stress test occurred within 2 hours of the prestress saliva sample, which was collected around 8:00 on the morning of the event. During this time, participants did not eat; however, drinking was not restricted. Saliva was collected using the whole stimulated saliva method. Using this method, each participant chewed sugar free gum for 5 minutes to stimulate salivation before 1 mL of saliva was collected, stored on ice with a protease inhibitor, and then shipped on dry ice to have biomarker concentrations determined using Eve Technologies' (Alberta, Canada) cytokine and steroid/thyroid assays.¹²

Data Analysis

To evaluate our data and assess associations between LEC-5 severity scores and biomarker levels at each collection period, Pearson correlation coefficients (r) were generated. Specifically, r was computed between LEC-5 severity scores and biomarker levels for prestress (saliva sample immediately prior to stress test), stress (immediately following stress test), and recovery (1 hour after stress test). To evaluate the association of PTSD with variations in biomarker levels, the biomarker panels of the 9 individuals with previously diagnosed PTSD were then compared by t test with those of the 51 controls without any prior diagnosis of PTSD.

RESULTS

LEC-5 and Cytokines

As shown in Table 1 (described hereinafter), we found that, at various time points, the following cytokines were significantly (p < 0.05) and positively correlated to LEC-5 severity scores regardless of prior PTSD diagnoses: IL-10, IL-1B, GCSF, IL1-Ra, Groα, IFNa2, PDGFAA, and VEGF (full summary statistics are available in Supplemental Digital Content, Supplementary Data 2, http://links.lww.com/TA/D17). Interleukin-10 and IL-8 were found to be significantly elevated in multiple time point Δ's. Specifically, IL-10 was elevated during the stress collection point when compared with recovery, prestress versus stress, and when directly correlated to LEC-5 scores. Interleukin-8 had the highest r value of any of the significantly correlated fluctuations, with a value of 0.40659 at the stress versus recovery Δ. It was also found to be significantly increased in the prestress versus recovery Δ.

TABLE 1.

Cytokines Correlate Significantly With LEC-5 Severity Scores

Cytokine	Time Point	r Value	p	n
IL-8	Stress vs. recovery	0.40659	0.0014	59
IL-10	Stress vs. recovery	0.31042	0.0167	59
IL-1B	Stress vs. recovery	0.30142	0.0215	58
IL-10	Prestress	0.40582	0.0017	57
GCSF	Prestress	0.33194	0.0117	57
IL1-Ra	Prestress	0.30433	0.0214	57
IL-8	Prestress vs. recovery	0.33001	0.0122	57
Groα	Prestress vs. stress	0.3371	0.0103	57
IFNa2	Prestress vs. stress	0.31328	0.0177	57
PDGFAA	Prestress vs. stress	0.3015	0.0227	57
VEGF	Prestress vs. stress	0.30004	0.0234	57
IL-10	Stress	0.30981	0.0169	57

Time points throughout stress response and Δ's between time points are indicated. Those not compared with other time points have been correlated to LEC-5 scores directly.

PTSD and Cytokines

Reporting only significant results (p < 0.05), at the prestress time point (Fig. 1), GROα, PDGFAA, IL-1Ra, IL-1a, and Mip1a were found to be increased in participants with a prior PTSD diagnosis, while TNFb was found to be decreased. At the stress time point (Fig. 2), IL-1a, IL-6, IL-8, Mip1b, TNFα, and TGFα were found to be elevated in participants with a prior PTSD diagnosis, while progesterone and T4 were found to be reduced. At the recovery time point (Fig. 3), both IL-6 and TNFα were found to be increased with no markers decreased. Worth noting from these PTSD results, GROα, PDGFAA, IL-8, and IL1-Ra were all also found to be positively correlated to LEC-5 severity (full summary statistics are available in Supplemental Digital Content, Supplementary Data 3, http://links.lww.com/TA/D18).

Figure 1.

Prestress. Altered cytokine levels at the prestress time point in participants with a prior PTSD diagnosis compared with those without. It includes those found to be increased (Groα, PDGFAA, IL-1Ra, IL-1a, Mip1a) and decreased (TNFb).

Figure 2.

Stress. Altered cytokine levels at the stress time point in participants with a prior PTSD diagnosis compared with those without. It includes markers found to be increased (IL-1a, IL-6, IL-8, Mip1b, TNFα, and TGFα) and decreased (progesterone and T4).

Figure 3.

Recovery. Increases in cytokines at the recovery time point in participants with a prior PTSD diagnosis compared with those without. Those markers are IL-6 and TNFα.

Some cytokines were found to have altered Δ's when compared with participants without a PTSD diagnosis as well. For the prestress versus stress Δ, TGFα, IL-6, IL-8, FLt3L, and MDC were increased and eotaxin was decreased. For the stress versus recovery Δ, progesterone, PDGFAA, and VEGF-A were increased, and none were decreased. For the prestress versus recovery Δ, IL-1Ra, IP10, and VEGF-A were increased. These can be visualized in Supplemental Digital Content (Supplementary Data 4, http://links.lww.com/TA/D19, and Supplemental Digital Content, Supplementary Data 5, http://links.lww.com/TA/D20).

Missing/Excluded Data

Worth noting, two of the cytokines tested for, IL-17a and IL-3, were out of range/below the minimum value determined by the standard curve for all study participants and were thus ultimately excluded from the study as a whole. All other analyte values were included in the data analysis at all time points where concentrations were detectable within the sample. There were instances in which n was brought below 59 because of individual analytes not being detected in samples at specific time points. Within Table 1 for example, this brought some sample sizes to a low of 57. The lowest sample size in the study as a whole was 56 for IFNy at the prestress time point. For a complete list of instances where n was below 59, see Supplemental Digital Content (Supplementary Data 2, http://links.lww.com/TA/D17, and Supplementary Data 3, http://links.lww.com/TA/D18).

DISCUSSION

The aim of our research was to quantifiably demonstrate a direct link between the amount of trauma experienced by individuals and biological response to acute stress. Our data show that there are cytokines that, when correlated to LEC-5 severity scores, demonstrably change during an acute stress response based on an individual's prior exposure to traumatic events, specifically, IL-8, IL-10, IL-1B, GCSF, IL1-Ra, Groα, IFNa2, PDGFAA, and VEGF. These changes support the hypothesis and suggest that increasing levels of psychological trauma have a quantifiable effect on the immune system during acute stress. These cytokines are also depicted in Table 1. Furthermore, some of these same cytokines were found to have similarly and significantly altered concentrations in individuals with prior PTSD diagnoses. These common cytokines include IL-8, IL1Ra, GROα, and PDGFAA.

To better understand the potential significance of these elevations, we will briefly describe their established functions in the body. Interleukin-8 and GROα (also known as CXCL1) both play a well-established role in neutrophil activation and chemotaxis.^13–15 This process is foundational in the promotion of cell-mediated inflammation. High expression of IL-8 has been implicated in the pathogenesis of COVID-19 and its progression to acute respiratory distress syndrome.¹³ Interestingly, IL-1Ra is an anti-inflammatory cytokine that competitively inhibits the receptor IL-1R1. Interleukin-1R1, typically activated by IL-1 (or IL-1a, which we also found to be elevated), is implicated in the induction of a proinflammatory state and the release of IL-6. Interleukin-6 was found to also be elevated in participants with prior PTSD diagnoses, along with a broad spectrum of other proinflammatory cytokines.¹⁶ This action by IL-1Ra has been shown in animal models to play an important role in protection against cytokine-mediated toxicities and as a potentially viable mechanism for the treatment of rheumatoid arthritis in humans.^13,17 In our study, this increased secretion of IL-1Ra could perhaps be suggestive of an attempt by the body to diminish the hypersecretion of IL-6 and other proinflammatory mediators witnessed in the stress and recovery period. PDGFAA is a cytokine that holds its primary function in wound healing and cell senescence.^18,19

While the exact clinical significance of these cytokine fluctuations cannot be specifically elucidated because of the limited scope of this study (i.e., whether it is protective, harmful, maladaptive, etc.), the discovery of significant and consistent changes validates the importance of further study regarding the subject and, furthermore, the development of biomarker-based “immune profiles.” Such profiles could potentially be used in the development of susceptibility screenings (i.e., susceptibility to the development of PTSD), personalized treatments, and preventative medicine for those with profiles demonstrating risk for developing psychological dysregulations as a result of trauma.

Considering the findings mentioned previously, future research could investigate how these cytokine changes affect an individual's interpretation of and physiological reaction to acute stress. Considering that not all individuals who experience the same traumatic event develop PTSD, expounding upon this interaction holds potential value in furthering our understanding of the connection between experienced trauma and the development of psychiatric maladaptive disorders such as PTSD and depression. Worth noting, this question highlights one of the limitations regarding our study—the ability to extrapolate precise cause and effect from the data. Because of the short timeline for data collection, it cannot be confidently determined whether the increasing or decreasing levels of cytokines are the result of protective or maladaptive responses. A longitudinal observational study among populations more organically exposed to potentially stressful/traumatic experiences could be more suited to investigate the precise effect of these fluctuations.

In previous studies based off of a general hypothesis that a proinflammatory state (without investigating specific cytokines or inflammatory markers) may contribute to psychological depression, it was discovered that the pairing of selective serotonin reuptake inhibitors with the administration of COX-2–inhibiting nonsteroidal anti-inflammatory drugs significantly increased the rate of depression remission.²⁰ In addition, it has been demonstrated that treatment of multiple sclerosis with interferons (proinflammatory cytokines), is associated with a high risk of developing psychiatric disorders including treatment emergent depression.²¹ These studies once again demonstrate the correlation between proinflammatory states and the development of maladaptive psychiatric illness.

The mechanisms behind how psychiatric illness develops is beyond the scope of this report; however, Perez et al.²² offer many potential connections and explanations. One topic they describe is the effect of increased concentrations of IL-6 (which we found to be increased in participants with prior PTSD diagnoses) on neurogenesis. Hippocampal cells are some of the only neuronal cells that continue to be generated beyond the developmental stages of life and are known to be critical in the regulation of stress, memory, and of the hypothalamic pituitary adrenal axis.²³ Astoundingly, it was discovered that incubation of hippocampal progenitor cells with IL-6 decreased neurogenesis by half.²² They described IL-1B as having similar effects—another cytokine we found to be positively correlated to increasing LEC-5 scores. In summary, there appears to be a strong correlation between proinflammatory cytokines and CNS cellular proliferation and function.

Many of the cytokines found to be increased in our study fall into the “proinflammatory” category. An exception to this generalization is IL-10, which we found to be increased in correlation to LEC-5 scores. Interleukin-10 is a notable anti-inflammatory interleukin. While increased concentrations of IL-10 have been noted by others investigating effects of acute stress on salivary inflammatory markers,²⁴ this once again demonstrates the limitations of this study in determining conclusively whether the changes in inflammatory marker concentrations are detrimental or protective. Does this increase in IL-10 represent a dysregulation of its expression with potential unknown adverse effects? Conversely, is the rise of IL-10 simply a systemic response to compensate for an already developed inflammatory state?

This question and our study may provide further insight into how the acute stress response relates to individualized allostatic load. Allostatic load refers to the wear and tear on the body that accumulates as a response to repeated stress.²⁵ The psychoneuroimmunological axis and associated pathways are critical for regulating these responses, and the body can help to mediate the effects of the allostatic load by developing resilience to stress. Future research efforts should focus on resilience strategies, such as stress inoculation training, to improve health outcomes for individuals faced with harmful allostatic loads.

Having described many of our findings, it is important to further delineate the limitations and biases of this study. Some of these include the limited sample size; the lack of ethnic, gender, age, and occupational diversity within the sample; the potential biases introduced by the use of induced salivary sampling rather than blood draws; and the simulated nature of the stress test. Considering the lack of diversity in the demographics of our patients (i.e., primarily male, White, and generally physically active), the precise applicability of our findings to the general population and thus the reproducibility of our results may vary as a result. Along the same vein, another concern is the relatively small sample size. This, combined with the lack of subject diversity, limits the strength of our statistical analyses and inhibits its generalizability. It is also worth noting that, while our research has demonstrated positive correlation between the stated biomarkers and increasing LEC-5 scores (and/or prior PTSD diagnoses), these findings likely do not independently define a causal relationship. For example, these associations could potentially coexist as a result of events untested for in our methods (e.g., allostatic load). We also question whether the data gathered from our cohort with prior PTSD diagnoses stem from the effects of a direct PTSD maladaptive response, a compensatory response aimed at healing or protecting the individual, or a combination of the two (PTSD status was not reevaluated at the onset of the study).

In conclusion, the CNS and immune system are intricately connected in ways that have yet to be fully described in the literature. This study has shown that there is indeed a quantifiable and consistent response between the CNS and a multitude of biomarkers, which interact with the immune system, in the setting of the acute stress response. The cytokines found to be positively correlated to increasing LEC-5 scores included IL-8, IL-10, IL-1B, GCSF, IL-1Ra, Groα, IFNa2, PDGFAA, and VEGF. Those significantly associated with prior PTSD diagnoses included Groα, PDGFAA, IL-1Ra, IL-8, IL-1a, Mip1a, IL-6, Mip1b, TNFα, and TGFα. In an effort to further the treatment and understanding of psychological dysregulation as a result of acute stress and trauma, continued research to further elucidate the value and meaning of these detected patterns, as well as others yet to be discovered, is warranted.