4.0 Discussion:
Overall, the current results suggest that despite their common actions in the inflammatory response, the pro-inflammatory cytokines are differentially associated with psychological symptoms related to MDD, with some interesting sex-specific effects. The pro-inflammatory cytokines IL-1α and IL-6 were significantly higher in those with MDD than healthy controls; however, a group by sex interaction for IL-6 and post-hoc analysis revealed that IL-6 was significantly higher in MDD versus control only for females. Both IL-1α and IL-6 were associated with all psychological symptoms measured. In contrast, TNF-α did not differ between those with versus without MDD and was correlated with anxiety and hostility only. Despite none of the inflammatory cytokines being different between sexes, there was a sex difference in their associations with psychological symptoms; these were associated with IL-6 and TNF-α, but not IL-1α in females, and with IL-1α but neither IL-6 nor TNF-1α in males. In contrast to hypotheses, the pro-inflammatory cytokines were not significantly associated with adiposity or cardiac health measures.
The current finding of significantly higher plasma IL-1α with MDD is consistent with a previous study (Simon et al. 2008) and add to the limited reports on examination of IL-1α in MDD. Interestingly, in the current study, although IL-6 was similarly higher in the MDD group, this was with a sex interaction and IL-6 was only higher in females with MDD versus without MDD with no difference between MDD males and control males. While previous studies also reported that individuals with MDD had higher IL-6 than controls, they did not examine any sex differences (Al-Hakeim et al. 2015; Munjiza et al. 2018). This gives the assumption of a difference between groups for both males and females, which may be incorrect. The current results highlight how important it is to test for the sex effect. Given females are more prone to MDD (Kessler et al. 2005), show more pronounced inflammatory responses (Wegner et al. 2017) and have higher CVD-related mortality rate than males (Gao et al. 2019), and the female-specific association of IL-6 with depression here, it is worth further investigating a potential role for IL-6 in mediating these relationships.
Both IL1-α and IL-6 were significantly correlated with the GSI score as well as scores of depression, anxiety, hostility and stress, and accounted for unique variance in each of these psychometric measures, when included alongside TNF-α and sex in multiple linear regression analyses. Interestingly, they both were independent predictors of depression and hostility, while only IL1-α was predictive of stress and only IL-6 was predictive of anxiety. Similarly, previous studies in non-depressed cohorts reported associations of IL-1α with depressive symptom severity (Suarez et al. 2003) and with hostility (Suarez et al. 2004), and associations of IL-6 with depressive symptoms, anxiety (Reichenberg et al. 2001) and hostility (Suarez et al. 2003; Suarez et al. 2004). The positive associations of IL-1α and IL-6 with stress in the current study is expected, since stress stimulates the release of glucocorticoids and catecholamines which themselves promote inflammation, and the pro-inflammatory cytokines are also involved in the adaptation to stressors (McEwen, 2008). The current results are important since they extend beyond the previous studies in non-depressed individuals to report on the association of pro-inflammatory cytokines with specific psychopathology symptoms in participants with MDD.
In the current study, TNF-α concentration did not differ between diagnostic groups or between females and males and was only correlated with the psychological measures of anxiety and hostility, and not with the GSI, depression severity or stress. Although this result agrees with previous research (Cilan et al. 2012; Karlović et al. 2012; Cassano et al. 2017; Obermanns et al. 2021; Bürhan-Çavuşoğlu et al. 2021), others have reported that TNF-α was significantly higher in MDD than healthy controls (Tulgu et al. 2003; Liu et al. 2012; Al-Hakeim et al. 2015; Köhler et al. 2017). However, two of these latter studies had small sample size (N = 43) (Tulgu et al. 2003) and (N = 60) (Al-Hakeim et al. 2015), included depressed patients who were using anti-depressed treatment (Al-Hakeim et al. 2015), and they were not age and sex matched with the control participants (Tulgu et al. 2003); all these factors may have confounded the findings.
TNF-α had a more discriminatory profile than IL-1α and IL-6 in terms of associations with psychological symptoms, being correlated only with anxiety and hostility. This agrees with previous research that reported an association of TNF-α with anxiety (Reichenberg et al. 2001; Postal et al. 2016) and hostility (Suarez et al. 2004; Takahashi et al. 2018), albeit among healthy participants. Of note is that there is a well-established relationship between hostility and CMD risk: hostility is linked with high blood pressure (Spicer and Chamberlain, 1996), elevated levels of low-density lipoproteins (Suarez et al. 1998) and greater risk of coronary heart disease (Wong et al., 2013), all associated with inflammatory processes (Goldbacher and Matthews, 2007; Nabi et al. 2010). Given these pieces of evidence, TNF-α could be associated with the risk of CMD directly by its pro-inflammatory nature and indirectly through its link with hostility. Although TNF-α correlated significantly with hostility and anxiety, when it was entered in regression analyses alongside the other pro-inflammatory cytokines in the current study, it did not significantly account for unique variance in psychological symptoms. This is perhaps due to the positive correlation between IL-6 and TNF-α, and since IL-6 was a stronger predictor of psychological symptoms than TNF-α, IL-6 accounted for unique variance in most types of psychological symptoms.
Investigating the sex effect on the association of pro-inflammatory cytokines with psychological symptoms showed compelling results in the current study. IL-1α was correlated with all psychological symptoms in males but none in females, while IL-6 and TNF-α were correlated with all psychological symptoms, except not TNF-α with depression severity, in females, but none in males. IL-1α may be a risk factor for depression, hostility and stress in males, and IL-6 and TNF-α could be risk factors to being anxious, hostile and stressed in females. These novel findings of sex-specific relationships between pro-inflammatory cytokines and psychopathology symptoms in a clinical cohort with MDD may be related to sex differences in MDD prevalence, symptoms and treatment response (Seney et al. 2021).
In the current study, none of the pro-inflammatory cytokines were correlated with CMD risk indices of waist circumference, BMI, blood pressure or heart rate. Although one study presented an association between pro-inflammatory cytokines (IL-1α) and obesity (Um et al. 2011), this was in non-depressed participants. A previous study among people with MDD reported that IL-6 and TNF-α were both correlated with BMI (Shelton et al. 2015); however, these individuals were obese or severely obese with BMI greater than 30 Kg/m2. The lower mean BMI of participants in the current study, in the overweight rather than the obese range, may have contributed to the lack of significant association between the inflammatory cytokines and BMI or waist circumference. Similarly, participants in the current study were in the normotensive range on average, and the lack of correlation between any of the pro-inflammatory cytokines and blood pressure likely indicates that pro-inflammatory cytokines are more relevant to high blood pressure.
There were several limitations to the current study. Firstly, the cohort was normotensive, and most participants were non-obese. This likely affected the non-significant association between pro-inflammatory cytokines with blood pressure and BMI. Being a cross-sectional study allowed only for associations to be assessed rather than examination of any causative effects as per a longitudinal study design. Further, since research supports a role of IL-1β in the development of adipose tissue inflammation, insulin resistance and metabolic syndrome (Ballak et al. 2015), it would be important to assess IL-1β alongside IL-1α in a cohort with MDD who have physical and physiological CMD risk indices such as high blood pressure and BMI.
In conclusion, of the plasma pro-inflammatory cytokines, only IL-1α was elevated in MDD versus control, IL-6 was higher in MDD versus control among females, and TNF-α did not differ between groups. The latter finding and the lack of a main effect of sex on IL-1-α and TNF-α were contrary to our hypothesis. The group by sex interaction for IL-6 and sex specific associations between pro-inflammatory cytokines and psychometrics suggests potentially differential roles of these cytokines and mental health between males and females. These results of sex differences in inflammatory processes linked to depressive symptoms are compelling and could be aetiologically important in depression interventions and treatments.