the science

It is well documented that there is an association between autism and immune dysfunction although the exact nature of this relationship is not fully understood (DelGuidice, 2003). Studies in peripheral immune functions have had conflicting findings while cerebral spinal fluid and brain autopsy studies have been more productive. Both microglia and astrocytes play important roles in neuronal function and homeostasis. Changes in astroglia and microglia can produce marked neuronal and synaptic changes that are likely to contribute to CNS dysfunction observed in autism. Neuronal dysfunction and abnormalities in cortical organization may also be responsible for pathophysiological responses that may lead to neuroglial activation, reactions that may subsequently increase the magnitude of neuronal dysfunction. Post-mortem brain tissues from 11 autistic patients have demonstrated the presence of an active and ongoing neuroinflammatory process in the cerebral cortex and white matter, and notably in the cerebellum. The microglia and astroglia had marked activation compared to controls. In the brains of autistic patients immunostaining of the cerebellum showed a marked reactivity of Bergmann’s astroglia in areas of neuronal loss within the Purkinje cell layer as well as a marked astroglial reaction in the granular cell layer and cerebral white matter. Cytokines were also increased including IL-6 which is a proinflammatory cytokine. CSF findings from patients with autism following developmental regression were found to have a proinflammatory profile of cytokines in particular interferon-gamma and MCP-1 a chemoattractant and protein for monocytes and macrophages compared to a control group (Pardo et al., 2005; Zimmerman, 2006).

Immune aberrations consistent with a dysregulated immune response that have been reported in autistic children include abnormal or skewed T helper cell type 1 (TH1/TH2 cytokine profiles, decreased lymphocyte numbers, decreased T cell mitogen response, and the imbalance of serum immunoglobulin levels. Autism has also been linked with autoimmunity and in association with immune-based genes including human leukocyte antigen (HLA0-DRB1 and complement C4 alleles (Ashwood et al., 2006). One of the most consistent biological findings in autism is elevated whole blood serotonin (5HT) levels found in about 1/3 of cases. 5HT is an immunomodulator (Burgess et al., 2006). It has been shown that there is a significantly increased occurrence of autoimmune disorders in the families of autistic individuals (Sweeten et al., 2003). Further, a particular major histocompatiblity complex haplotype B44-Sc 30-DR4 has been shown to occur much more frequently in individuals with autism or their mothers versus an unrelated control group (Daniels et al., 1995). In addition, the frequency of human leukocyte antigen A2 alleles has been found to be significantly increased in autistic subjects compared with normal allelic frequencies form the National Marrow Donors Program (Torres et al., 2006).

An underlying mechanism of immune response is cytokine production and/or inhibition, including proinflammatory cytokines. Increased levels of proinflammatory cytokines have been found in individuals with autoimmune diseases (Simopoulos, 2002). A study by Jyonouchi et al. (2001) found excessive innate immune responses, most evident in TNF-alpha production, in a 47.9 to 60% of ASD patients depending on the stimulant. Another study found significantly elevated levels of tumor necrosis factor receptor II in 35 subjects with autism compared with controls (Zimmerman et al., 2005). Furthermore, after stimulation with cow’s milk protein and its major components in peripheral mononuclear cells of persons with ASD and gastrointestinal symptoms there was elevated TNF alpha/IL-12 compared to control subjects. Peripheral mononuclear cells from subjects with ASD without a history of gastrointestinal symptoms had produced more TNF alpha/IL-12 with cow’s milk protein but not to individual components (Jyonouchi et al., 2005a). Another study by the same group showed that when peripheral mononuclear cells from children with ASD were stimulated with the endotoxin lipopolysaccaride, only the cells of the children with gastrointestinal symptoms showed a positive association between TNF alpha levels produced by the endotoxin and TNF levels produced by cow’s milk protein and its major components (Jyonouchi et al., 2005b). The cells also produced lower levels of IL-10 a counter regulatory cytokine compared to controls. However, even children with ASD without gastrointestinal symptoms produced elevated TNF-alpha levels in their peripheral mononuclear cells. Also, significantly increased production of IFN-Gamma and IL-1RA and a trend towards significantly increased production of IL-6 and TNF-alpha were found in the whole blood of autistic children (Croonenberghs et al., 2002a). Vargas et al. (2005) developed cytokine profiles from brain tissue of eleven autistic individuals obtained at autopsy and from cerebrospinal fluid of seven living patients. Active neuroinflammatory processes were found in the brain tissue samples and proinflammatory cytokines, including markedly elevated MCP-1 (macrophage chemoattractant protein) in the CSF of living subjects. In one study, IgG2 and IgG4 serum concentrations were also found to be elevated in autistic patients (Croonenberghs et al., 2002b). Another study showed immune activation in peripheral blood and intestinal mucosa of autistic patients with gastrointestinal symptoms with levels on CD3+, TNFalpha+ and CD3+IFNgamma+ similar to patients with Crohn’s disease (Ashwood and Wakefield, 2006). One problem with peripheral studies is that it is not clear when measurements are taken at single time points and among subjects of different ages if there is true immune dysfunction or dysmaturation that changes with age (Pardo et al., 2005).

Also, there is evidence to show that altered cytokine levels are associated with behavioral changes in animals and humans. Direct administration of IL-1, IL-2 and IL-6 in mice results in profound behavioral changes, including digging, exploration, rearing and grooming (Zalcman et al., 1998). A study of transgenic mice with a null mutation for interleukin-6 showed lowered ambulation and exploration behavior, reflecting fearfulness (Armario et al., 1998). In mice bred for differences in social behavior, nonaggressive mice had significantly lower gamma-interferon and IL-2 production and significantly lower NK activity than aggressive mice (Petitto et al., 1994).

Two of the current pharmacological treatments for aggression and agitation in autism have been moderately successful. In a double blind, placebo controlled study, the atypical anti-psychotic risperidone was shown to improve symptoms of autism, including repetitive behavior, aggression, anxiety or nervousness, depression, irritability and the overall behavioral symptoms of autism (McDougle et al., 1998). Anecdotal evidence and reports of clinicians also appear to support the efficacy of risperidone, although there have been reports of its benefits diminishing over time. Lithium has also been used with some success in a variety of psychiatric disorders, including autism (McDougle et al., 2003). Both risperidone and lithium have been shown to have significant immunoregulatory effects. Cazzullo et al. (2002) examined cytokine profiles in 15 schizophrenic patients before and after a three month treatment with risperidone. Increased levels of IL-10 (a suppressor of proinflammatory cytokines), and decreased levels of interferon-gamma (a proinflammatory cytokine) were found in the patients after treatment with risperidone. Studies of lithium have also shown it to have profound immunoregulatory effects in animals and humans. In a study of the effect of lithium on ex vivo cytokine production, Rapaport and Manji (2001) found a decrease in the levels of proinflammatory cytokines IL-2 and interferon-gamma and an increase in the anti-inflammatory cytokines IL-4 and IL-10. In a study of healthy human subjects, lithium had major immunoregulatory effects, including increases in both proinflammatory and negative immunoregulatory cytokines or proteins when stimulated by lipopolysaccaride and phytohemaglutinin (Maes et al., 1999).

Helminth worms, in particular the porcine whipworm (Trichuris Suis), have been studied as a potential treatment for Crohn’s disease and ulcerative colitis. This avenue of research is an outgrowth of the “hygiene hypothesis,” which states that the increase in autoimmune disorders is a result of improved hygiene in the developed world. Recently, a study using porcine whipworm as a treatment for Crohn’s disease and ulcerative colitis showed extremely positive results, with 3 of 4 Crohn’s disease and all three patients with ulcerative colitis going into remission (Summers et al., 2003). A larger follow up study using the same treatment on a group of 29 patients with moderately severe Crohn’s disease resulted in 72.4% of the patients going into remission (Summers et al., 2005a). Another trial which was placebo controlled involving patients with ulcerative colitis showed improvement at 12 weeks in 13 of 30 patients on active treatment compared to 4 of 24 patients showing improvement with placebo (Summers et al., 2005b). These results lend credibility to the hygiene hypothesis and suggest that the return of helminths to the human gut may allow positive interaction with the immune system and lead to a decrease in or elimination of autoimmune response. This immunoregulatory effect could also potentially decrease or eliminate symptoms of aggression and agitation in autism.

In summary, there is evidence to suggest a possible causal relationship between increased levels of proinflammatory cytokines and symptoms of aggression and agitation in autism. New immunomodulatory treatments for other autoimmune disorders should be investigated as possible treatments for these symptoms. The recent extremely promising results from the use of TSO in the treatment of Crohn’s disease and ulcerative colitis warrant examination as a potential treatment for symptoms of agitation and aggression in individuals with autism.