Verified by 
It's probably a symptom of our innate tendency to hyper-mentalize that we think of our brains as our principal organ of adaptation to our environment and prime means of survival. But a moment's reflection reveals that our immune system is in many ways very similar--and is certainly just as critical to our survival and reproductive success. Both nervous and immune system exemplify what genes do to meet the challenge of changing environments for which they cannot legislate in advance: they build expert systems to deal with the problems for them in real time.
I have heard it claimed that the immune system processes as much information as the brain, for example in producing anti-bodies, but whatever the truth, both systems process data on a prodigious scale. And that's just the start of it. Both nervous and immune system
• are constitutionally sensitive to environmental influences;
• learn, can be conditioned, and are prone to placebos;
• have innate and acquired responses;
• have both short and long term memories;
• show high individual variability and critical self/non-self discrimination;
• can react to harmless threats (allergy, phobia);
• can occasion internal conflicts (auto-immune disorders, psychosis);
• and are products of evolutionary arms races--indisputably so in the case of the immune system and arguably so in that of the nervous system according to the imprinted brain theory.
We have known since the 1964 rubella epidemic in the USA that infection during pregnancy is a major risk factor for autism, and recent research suggests that this is the tip of an iceberg of such prenatal effects of infection and immune system responses on neuro-development resulting not only in autistic spectrum disorders (ASDs) but psychotic spectrum disorders (PSDs) such as schizophrenia. Indeed, as a recent paper by Meyer, Feldon and Dammann points out, such an association with perinatal infection appears not to be pathogen-specific in the case of either ASD or PSD, but related to the immunological response itself.
These authors suggest that abnormal levels of pro-inflammatory cytokines and other mediators of inflammation during critical periods of early brain development may contribute to an increased susceptibility to schizophrenia or autism. They cite epidemiological evidence for a significant association between high maternal levels of pro-inflammatory cytokines during pregnancy and elevated risk of PSD in the offspring. And as I noted in a previous post, there appear to be links between ASD and asthma.
However, Meyer, Feldon and Dammann also point out that autism but not schizophrenia seems to be characterized by relatively severe chronic inflammation. In the case of schizophrenia, they suggest that the immune response works the opposite way, with suppression of fetal inflammation instead of persistence of it as in autism, and with latent rather than chronic inflammation being the outcome in schizophrenia.
As the diagram above illustrating their model suggests, this immunological approach to ASD and PSD bears a striking similarity to the diametric model proposed by the imprinted brain theory (below). Indeed, Meyer, Feldon and Dammann cite the diametric model explicitly and implicitly invoke it when they add the genetic factors which surely contribute to the outcome. As they put it, "The genetic background may critically influence the transition of acute fetal inflammation into either persistence or suppression of inflammation. ... The genetic background further contributes to the emergence of unique brain dysfunctions independently of or in interaction with the outlined inflammatory pathways."
As yet, little is known about how parent-specific imprinted genes and X or Y sex chromosome ones interact with the immune system. However, genome-wide scans have revealed that genes most strongly associated with schizophrenia/bipolar risk are to be found in the mother's--not the sufferer's--major histocompatibility complex (containing the key genes for the immune system). There is already good evidence linking homosexuality in males to the mother's immune response via the H-Y male-specific histocompatibility antigen, and a comparable link with schizophrenia is credible. Indeed, you might see both as instances of maternal/female genes asserting themselves against paternal/male ones. And of course, if this is so, it makes sense that, as the new immunological model proposes, the mother's genes should suppress fetal inflammation. The father's genetic self-interest, on the other hand, is on the side of his genes invested in the fetus, so he should favor full fetal immunological reactions. And at the very least, the fetus is a problem for the mother's immune system, not the father's--he isn't gestating it!
As the diagram above suggests, the imprinted brain theory proposes exactly how critical genes affect brain development and cognition, but given the prime importance of the immune system and its many parallel features with the brain, it is by no means far-fetched to suspect that something similar is going on there. At the very least, this intriguing new model proposes one possible way in which the nervous and immune systems may interact, revealing yet another remarkable equivalence between them--not to mention adding a new, immunological dimension to the diametric model of mental illness.
(With thanks and acknowledgement to Graham Rook for kindly bringing this paper to my attention.)
