There is growing evidence that immunological mechanisms contribute to the pathophysiology of TS, at least in a subgroup of TS patients. We hypothesized that TS has a common underlying immune mechanism that possibly causes a disturbed clearing of infectious agents. As innate immunity plays an important role in the first-line immune response, in our study we focused on monocytes/macrophages and associated parameters.
Our results are in line with previous findings: Leckman et al. [19
] reported signs of an increased activation of the cellular type-1 immune response in TS. At baseline, the authors found higher concentrations in TS patients than in controls of both the T cell (type 1) cytokine IL-12 and monocyte-derived TNF-alpha; the latter finding was in contrast to our findings. Even higher concentrations of these parameters were found in patients in a state of symptom exacerbation. Interestingly, these observations were more frequent in non-PANDAS than in PANDAS patients. Leckman et al. found no differences in the levels of other cytokines such as IL-1, IL-4, IL-6, IL-10, and IFN-γ [19
]. In our sample, TNF-alpha serum levels were lower in TS patients than in healthy controls. Besides methodological differences, several factors could explain the differences between our findings. For example, our sample had a higher number of unmedicated patients. Our medicated patients received mainly dopamine antagonists, whereas patients in the study by Leckman et al. received mainly alpha-agonists. Certain antipsychotics are known to increase TNF-alpha levels [36
], so our sample may have actually had even lower TNF-alpha levels if they had been in an unmedicated state. Although our sample size was limited, and the subgroups were particularly small, we did not find a significant influence of medication on any parameters, including TNF-alpha. Another contributing factor could be age [37
] and disease stage. In our sample one limitation was the age range of study participants. This is why subanalyses were performed: healthy children/adolescents showed the highest serum TNF-alpha levels, followed by healthy adults, TS children/adolescents, and TS adults. When the subgroup of TS children/adolescents and TS adults were compared with the respective healthy control age groups, differences were seen between both age groups in almost all parameters. Given the high co-morbidity of OCD and TS [23
], our results in regard to TNF-alpha levels are in accordance with reports of decreased TNF-alpha levels in OCD [20
]. In our sample, the subgroup analysis of patients with and those without co-morbid OCD symptoms revealed no significant difference in the levels of any of the determined parameters, including TNF-alpha. This result indicates that in our sample the occurrence of co-morbid OCD seems to have no relevant influence on TNF-alpha levels. The same is true for other parameters, although the small size of the samples limits the interpretation of the findings. However, the trend of stronger effects in the comparison of healthy controls with TS patients with co-morbid ODC is noteworthy and in parallel with the findings of Gabbay et al. concerning IL-12 [24
]. It has to be mentioned that in our study the subgroups of TS patients and TS patients with OCD could eventually suffer from a co-morbid ADHD. The structured clinical interviews did not show hints for ADHD in all study participants, however, in order to exclude ADHD entirely, a specific ADHD rating scale should be applied in future studies. Another limitation of this study is that the used rating scales (MOCI and TSGS) are reliable, but are less frequently used then for example the Yale-Brown Obsessive Compulsive Scale and do not represent the ‘gold standard’. In the present sample the mean value for OCD in the TS group was 13 out of a maximum of 30. Therefore these patients suffered only from a relatively mild OCD. More homogeneous patient samples and further studies are required to answer the question whether TS, TS with comorbid OCD, and OCD have to be seen as spectrum disorders with common causes or different pathophysiological mechanisms.
Since cytokines have to be seen as a network [38
], TNF-alpha levels have to be discussed within the context of our other findings. Also, the decreased levels of IL1-ra suggest a decreased release of monocyte-derived cytokines and point to a decreased production of IL-1. Since IL-1 is released in a paracrine manner, the levels of circulating IL-1 in the blood often are below the limit of detection, are unstable and show a short half-life; therefore, the validity of estimating IL-1 serum levels is critically discussed. IL1-ra is a more stable and reliable laboratory parameter. It is produced and released from activated cells of the monocyte lineage [39
]. No differences in IL-6 levels were detected between TS patients and healthy controls. IL-6 is assigned pro- and anti-inflammatory characteristics and plays a key role in the acute phase response, especially in direct resolutions of acute infections and septic shock. It is likely to play a detrimental role in chronic disease [40
The most prominent finding—the lower levels of sCD14 in TS—could also reflect a decreased monocyte activation state and a possible susceptibility to infectious agents. The reduced sCD14 levels in TS patients may point to a blunted innate immunity and could indicate incomplete elimination of infectious agents and a chronic latent infectious state in TS. The most common and sensitive marker of inflammation is CRP. Although CRP levels were within the normal range in our sample of TS patients, the distinct but significant increase compared to controls could point to a latent underlying inflammatory process, similar to the role of high sensitivity CRP in the pathogenesis of atherosclerosis [41
]. Neopterin, an unspecific marker of inflammation found in various inflammatory conditions such as infection and autoimmune syndromes [42
], has been used as a measure of cell-mediated immunity [43
]. The higher levels of neopterin in the TS group in our study reflects the activation of a Th-1 or type-1 immune response, since the type-1 cytokine Interferon-γ (INF-γ) induces neopterin production [44
]. The findings of increased neopterin values in TS patients in our study are in line with previous findings. Hoekstra et al. reported significantly elevated neopterin levels in TS patients, indicating an immune activation [45
In summary, our results point to changes in innate immunity in TS patients compared with healthy controls: sCD14 and the products of activated monocytes, TNF-alpha, and IL1-ra showed lower levels. The number of circulating monocytes, however, was increased. Whether this increase in monocytes might be a compensatory mechanism cannot be determined on the basis of the present data and needs to be investigated in future studies. These changes suggest a dysregulated immune system in TS, which could have implications for susceptibility to infections or autoimmunity.
In our study, not just one but several different parameters showed results pointing to the under-activation of components of the monocyte lineage associated with a possible subclinical inflammatory reaction. Moreover, the findings result from different laboratory methods, i.e. an immunoassay technique, the Luminex bead technology, and automated analyzer methods. As a caveat, it has to be noted that TNF-alpha, IL-6, IL1-ra, CRP, and neopterin serum levels were within the normal ranges and often close to the detection limits. Therefore, the increases or decreases in relation to the control group can only be interpreted as clues for potential trends. No functional tests were performed, and the data reflect only a single time point measurement of the peripheral immune system. Nevertheless, little is known about immunopathological mechanisms in TS, so that these preliminary data contribute to the understanding of monocyte regulation and innate immunity in TS.
To build a possible bridge between the peripheral immune system and the nervous system, one interesting fact is that microglia cells derive from the monocyte/macrophage lineage. Furthermore, cerebral hyperintensities in magnetic resonance images, which can be a sign of neuroinflammatory processes, seem to be associated with TS [46
]. However, it remains unclear whether tic symptoms are partly due to the damage caused directly by infections or by the generated autoantibodies, or both, or due to changes in the immune balance. In both cases, neuronal signaling would be disturbed and could lead to tics. The central dopamine systems are most likely to play an important role in TS and can be modulated by immune function. In their review, Martino et al. [4
] propose a pathophysiological model for immune-mediated dopamine synthesis via CaM kinase II activation. In this model, infections might induce cross-reactive anti-neural antibodies. First-line inflammatory responses might be promoted by infections and may potentiate dopamine release by autonomic fibers and modulate peripheral immune cells.
The immune constellation in our sample could be in accordance with an impaired innate immune response, associated with an imbalance of the specific (type-1 and type-2) immune responses, the type-2 responses possibly leading to infections, autoimmune phenomena, and inflammatory reactions. Nevertheless, one must take into account that TS is a heterogeneous syndrome that may have different underlying pathological mechanisms. The limited size of our sample—in particular with respect to interfering variables such as age, medication and co-morbidity—means that further intense research regarding the role of inflammation, infection, and the immune system in TS is required. Particularly functional tests and the analysis of genetic polymorphisms that encode for inflammatory cytokines or sCD14 would be of interest.