March 2011
Summary on Round Table Discussion on Immunology and Psychiatry

This section was chaired by Prof. Brian Leonard and Prof. Volker Arolt. The questions and remarks were lead mainly by the invited discussants.

Question 1: Is the effect of Selective Serotonin Reuptake Inhibitor (SSRI) on rheumatoid arthritis anti-inflammatory reaction or the outcome through brain function? (F Fallarino)
Discussions: It is possible that be both anti-inflammatory effect and effect through brain function are involved.

The possible mechanisms of direct anti-inflammatory action were discussed as follows:
1. Map kinase mechanism since the extracellular signal-regulated kinase pathway is known to mediate neurotrophic actions and synaptic plasticity (R Straub)
2. Fluoxetine is potent inhibitor of interferon-γ, tumour necrosis factor-α and cyclooxygenase-2 in T cells (TJ O’Connor)
3. The effect antidepressants on Ca++ transport mechanism and other non-antidepressant effects should be considered to be involved in anti-inflammatory effects (D van Calker)

Remark 1: It is necessary to consider the effects of different antidepressants and different anti-psychotics. Lumping all together and considering as antidepressant effects or antipsychotic effects should be avoided. (M. Rappaport)

Discussions: The following points were discussed based on the above remarks:
1. It is very important to consider the different antidepressants or antipsychotics differently. Most of the pharmacogenetic studies failed and most possible reason would be due to lumping of all different drugs in one category (V Arolt)    
2. Regarding the genetic studies, intermedeophenotypes should be considered. The direct link between solid biochemical or morphological data and DNA data should be considered as well (HA Drexhage)
3. Proper psychopathological rating is also very essential to subtype under major disorder. Regarding the effect of different medication or pharmaceutical compounds on subtype of depression, for example, would be more meaningful (V Arolt)

Remark 2: Previously, it was mainly considered on Th1 and Th2 type of immune responses but it is now extended to Th17 and IL9 responses. (R Straub)

Discussions: The following discussions were made based on the above remark.
1. Instead of Th1 and Th2 type, pro-inflammatory or anti-inflammatory may be more appropriate (MJ Schwarz)
2. Th17 should be considered as marker for chronic inflammation (R Straub)
3. Both Th17 and Treg signatures are increased in schizophrenia. Regarding Th1 and Th2, it is not clear cut Th1 or Th2 increased or decreased signature expression between depression and schizophrenia though clearly different signature was observed (HA Drexhage)
4. The reason for the above Th1 and Th2 finding could be due to the increased Th1 type of immune reaction occurred in the beginning and is followed by Th2 type of reaction (F Fallarino)

Remarks 3: Cytokines are released from the subset of cells in the bone marrow or lymphoid organ and most are acting as transcription factors. Therefore, it would be difficult to conclude that increased circulating cytokines would induce some effects on the brain (R Jack).

Discussions: The following points were discussed based on the above remarks:
1. The circulating cytokines should have some effect on the brain since treatment with interferon-α, interferon-β and IL-2 could induce depressive or psychotic symptoms in the patients. Moreover, those cytokines could induce microglia activation and activated microglia also produce those cytokines. (HA Drexhage, M Rappaport, V Arolt, UCMeier)
2. Another example is the sickness behaviour in which the hypothalamo-hypophysial-adrenal axis is activated in case of systemic endotoxin treatment (R Straub)

Question 2: Can we say the cytokine balance in the brain by measuring the blood cytokines? (R Jack)

Discussions: The following answers and discussions were made:
1. There are several mechanisms that peripheral cytokines can cross blood brain barrier (R Straub)
2. Not all the cytokines could cross blood brain barrier but some cytokines are synthesized in the brain as well. Moreover, some cytokines work through different mechanisms in different way. One example is IL-6 which can be neurotoxic but can also be protective. (D van Calker)
3. There is definite connection between microglia or brain immune cells and the peripheral immune cells. An example is crossing blood brain barrier (B Bogerts)
4. In multiple sclerosis, it is clear that the cytotoxic cells can cross blood brain barrier (UC Meier)
5. Not only the blood brain barrier, but also the communication through the sensory pathway and IL-1 is the example (R Straub, HA Drexhage)
6. Another example is increased in peripheral IL-6 could induce change in brain IL-6 and could disturb cognitive function (M Rappaport)
7. Those connections are right for particular cytokines but we cannot generalize those pathways for all cytokines (R Jack)

Question 3: Is that true that low grade inflammation is treated by antidepressants or antipsychotics? Why ketamine has antidepressant action without acting on inflammation? Should inflammatory parameters be the target for treatment of psychiatric disorders? (BE Leonard)

Discussions: The following points were discussed:
1. The reduction of cytokine levels after treatment with antidepressant can also be due to the fact that the depressive symptoms which are stressful to the body are resolved in stead of the direct inhibitory action of antidepressants (V Arolt)
2. If we could treat with a therapy which is specific to one cytokine, we could answer those questions (R Straub)
3. Since there is heterogeneity in those psychiatric disorders, the cytokines may be useful as markers for treatment response in subgroup of patients (MJ Schwarz)
4. That could be possible. For example even rheumatoid arthritis cannot be considered just one type because there are B-cell dependent and T-cell dependent types (R Straub)

Question 4: How sure are the neuroprotective and neurotoxic effects of the kynurenines? (M Rappaport)

Discussions: The following points were discussed based on above questions:
1. It was proven in in-vitro experiments as well as in animal experiments and  human studies also showed that the kynurenic acid which is more of protective is reduced in the blood of patients with major psychiatric disorders (AM Myint)
2. Biomarkers are not only for to indicate the disease but also could indicate the organ changes. Is there any finding that kynurenines can indicate the organ changes? (B Sperner-Unterweger)
3. There was one study showing the association between kynurenines and fMRI changes in adolescent depression (AM Myint)
4. It would be more useful to combine the immunological markers with other markers such as genetic to indicate early diagnosis or subtype of the disease (AM Myint)
5. Physiological biomarkers should also be considered. Sympathetic and parasympathetic markers should be considered as well (BE Leonard)
6. It would be necessary to do study endophenotypes. The symptom centred approached should be used (B Bogerts)
7. As long as biomarkers are concerned, the reproducibility of the findings are important (M Rappaport)
8. Therefore, the standardization of the methods used to analyse the markers is important (AM Myint)
9. The standardization of patient recruitment in the studies are also important (MJ Schwarz)
10. May be pathogenetically related diagnostic system should be used (N Müller)
11. There are some tolls for that such as Newcastle Index (V Arolt)
12. Direct association with the symptoms should be studied (HA Drexhage)
13. The association between biomarkers and treatment effect is also important (R Straub)
14. How sure is that kynurenic acid is protective (M Rappaport)
15. Kynurenic acid can prevent the toxic effect of glutamate. Kynurenic acid is mainly secreted by astrocytes and prevents astrocytes apoptosis. Astrocytes produce neurotrophic factors (A Harkin)
16. It would be necessary to measure the concentration in the precise brain area in association with morphological changes (BE Leonard)

Remarks 4: General remarks were made by several discussants on how to translate the results from animal models. When we establish a model, the strain difference should be considered. For example, when we study the immune changes we use Black 6 mice, which is already T-helper type 1 dominant, the result would answer the response of Th1 dominant type of person. Therefore it is not possible to generalize and translate one to one literally from animal model to human. Especially in psychiatry or psychology, since interview is also important, it would not be possible to translate the animal model results to human. For example, the cognitive deficit finding in the animal model could be because of lack of attention due to sickness behaviour. (BE Leonard, A Harkin, TJ O’Connor, AM Myint, R Straub)

Moreover the use of in-vitro model as a preliminary experiment was also discussed. For example, the action of  IL-6 which can be neuroprotective as well as neurotoxic depended on the presence of another transcription factors such as SOCS3 in case of IDO activity. Such kind of finding from in-vitro model could be a basis for some animal models such as genetically transformed models such as knock-out or overexpression. (BE Leonard, D van Calker, F Fallarino, R Straub)

Remarks 5: A general remarks was made on the morphological studies. Since morphological studies in human brain could be done only post-mortem, the effect of medications could not be excluded.

Discussions: The following questions and discussions were made:
1. In this case what would be the best parameter to study in post mortem brain tissues? (B Bogerts)
2. Genetic markers would be possible (HA Drexhage)
3. paraffin fixed tissue can be successfully used for cytokine studies. We will have to carefully subgroup post mortem material. Cases with short disease duration (if possible) might give us better insight into pathophysiology than chronic cases (UC Meier)
4. To get brain tissue from epilepsy study if the patient is also depressed but there is a high risk of getting pathological tissue around the focus (V Arolt and B Bogerts)
5. Possibility of using stem cells is raised (R Yolken, V Arolt)

Remarks 6: The importance of environmental factor in pathophysiology of schizophrenia was discussed. Even though genetic factors are involved in schizophrenia, since there is gene environment interaction, the environmental factors should be considered. One important environmental factor is prenatal or early life infection. (HA Drexhage, D van Calker, MJ Schwarz, R Yolken)

Discussions: Since prenatal or early life infection is discussed as a risk factor for schizophrenia, a question was raised why incidence of schizophrenia is not much higher in the third world countries where childhood infections are common. The actual association between infections and schizophrenia should be studied. It is also dependent on how the body system react to which type of infection. The hygiene hypothesis should also be considered in abnormal or hypersensitive immune response. The role of immune response to symbiotic bacteria or normal flora in the pathophysiology of psychiatric diseases should also be studied. The environmental pollution such as dioxin can also effect on Th1/Th2 immune response. (V Arolt, R Yolken, R Jack, AM Myint, HA Drexhage, F Fallarino)    

The discussion was closed by the chairs with the concluding remark that further studies should be developed based on the points discussed and based on the new contacts established from this meeting.    

« Back
Send this to a colleague