Introduction
Autism is a developmental disorder, which manifests itself during early childhood. In the autistic child, communications and social interactions are severely impaired. Unable to learn from the natural environment as most children do, the child with autism generally shows little interest in the world or people around him. Although some children with autism develop normally and even acquire advanced skills, most exhibit a wide range of behavioral problems. In reality, autism affects the way a person comprehends, communicates and relates to others. Autism was originally thought to be primarily a psychiatric condition. However, further investigation showed that genetic and environmental factors are implicated in the pathogenesis of autism. The effects of environmental factors such as infections and toxic chemicals on gene expression result in biochemical, immunological and neurological disorders found in children with autism.

Involvements of different systems in autism
Similar to many complex diseases, genetic and environmental factors including infections, xenobiotics, dietary proteins and pep-tides, play a critical role in the development of autism. The effects of environmental factors on genetic makeup result in immune, gastrointestinal, neurological, biochemical and neuroimmunological abnormalities. Based on extensive research, which led to publications of three different manuscripts and two review articles, we postulated that autism is induced by infectious agent antigens, toxic chemicals and dietary proteins. This process begins in the gastrointestinal tract but manifests itself in the brain. These factors will be explained in detail in the following sections.

The role of infectious agents in autism
Many infectious agents, including Streptococcus, measles, Rubella, Cytomegalo virus, Varicella zoster, Herpes type-6 and others have long been suspected as etiologic factors in autism. Maternal or post-maternal exposure to these infectious agents may result in neurological disorders including autism. Using the observation that maternal infection increases the risk of schizophrenia and autism in offspring, recently it has been shown that respiratory infection of pregnant mice (both BALB/c and C57BL/6 strains) with the human influenza virus resulted in offspring that displayed highly abnormal behavioral responses as adults. As in schizophrenia and autism, these offspring displayed deficits in prepulse inhibition (PPI) in the acoustic startle response. Compared with control mice, the infected mice also showed striking responses to the acute administration of antipsychotic and psychomimetic drugs. Moreover, these mice were deficient in exploratory behavior in both open-field and novel-object tests, and they were deficient in social interaction. At least some of these behavioral changes were likely attributable to the maternal immune response itself. They concluded that abnormal levels of cytokine production, which interfere with neuroimmuno-communications, are responsible for abnormal development of the brain.

Until recently, there has been little direct evidence readily available in support of the molecular mimicry hypothesis and to clearly de-lineate the role of infectious agents as a cause for neurological disorders. For example, studies in mice have shown that infection with Theiler’s virus elicits an inflammatory response in the CNS that progresses to chronic experimental autoimmune encephalomyelitis. Epitopes of Streptococcal M proteins have also been shown to evoke antibodies that cross-react with human brain neuronal cell basal ganglia, which are potentially involved in the pathogenesis of Sydenham’s chorea (associated with acute rheumatic fever).

To support the proposed role of infectious agents in autism in several research projects which resulted in publication in a respected scientific journal, we showed that bacterial toxins and heat shock proteins could promote development of peptidase antibodies in children with autism and patients with autoimmune disease. In these studies, by searching for a mechanism underlying autoimmunity in autism, we postulated that gliadin peptides, heat shock protein (HSP-60) and streptokinase (SK) bind to different peptidases. Binding results in autoimmunity. We assessed this hypothesis inpatients with autism and in those with mixed connective tissue diseases. Concomitant with the appearance of anti-gliadin and anti-HSP antibodies, children with autism and patients with autoimmune disease developed anti-DPPI, anti-DPP IV and anti-CD13 autoantibodies. These antibodies may be synthesized as a result of gliadin and HSP-60 binding to different peptidases since a significant percentage of autoimmune and autistic sera were associated with elevated IgG, IgM or IgA antibodies against all three peptidases, gliadin and HSP-60. In these studies we propose that superantigens (e.g., SK, HSP-60), dietary proteins (i.e., gliadin peptides) in individuals with predisposing HLA molecules bind to aminopeptidases and induce autoantibodies against peptides and tissue antigens. From our results we conclude that binding of bacterial superantigens to DPP IV, DPP I or CD13 is responsible for autoantibody production in children with autism and in patients with autoimmune diseases.

The role of heavy metals and other toxic chemicals in autism.
Xenobiotics have been suspected to contribute to the induction of autoimmunity. Many environmental chemicals or drugs are toxic to hosts, and their detoxification is achieved primarily in the liver. During their metabolism, they may form reactive metabolites, which can then modify cellular proteins to form neo antigens. The precise mechanisms that lead to modification of self-proteins and the molecular requirements for this modified self to induce tolerance breakdown remain to be established. However, it is important to note that the direct toxic effect of xenobiotics is usually dose dependent and may be evident in the majority of individuals shortly after drug in take; hence, they are relatively easy to identify. In contrast, the immune-mediated effects that follow the intake of drugs or xenobiotics may take a prolonged period of time to be clinically manifest, making the identification of the causative agents a formidable task.

For a chemical compound to lead to an autoimmune response, it is generally thought that the compound must first become covalently bound to a carrier protein. Immune reactions to drugs or their metabolites can develop when a hapten carrier complex inter-acts with gut-associated lymphoid tissues (GALT) that constitute the largest lymphoid organ. If covalent adducts of drugs or other chemical compounds are formed in GALT, it seems reasonable that they may lead to immune responses and chemically-induced Type I- Type IV allergic reactions. 

Among many toxicants, thimerosal or ethyl mercury in vaccines has been associated with immune injuries described in children with autism. Contrary to many haptens that bind covalently to a single amino acid, such as lysine, metal complexes consist of a central metal ion composed of four different amino acids, and hence they possess increased complex stability. To demonstrate possible binding of ethyl mercury to DPP IV and CD69, in a very recent study, we postulated that in addition to infectious agent antigens such as Streptokinase, ethyl mercury (xenobiotic) binds to different lymphocyte receptors and tissue antigens. We, therefore, propose that bacterial antigens and thimerosal (ethyl mercury) in individuals with pre-disposing HLA molecules, bind to CD26 or CD69 and induce antibodies against these molecules as well as to lymphocyte receptors and tissue antigens. In conclusion, this study is apparently the first to demonstrate that bacterial toxins and xenobiotics bind to lymphocyte receptors and/or tissue enzymes (see Fig. 24), resulting in autoimmune reaction in children with autism. Results of this study will be published in the International Journal of Immunopathology and Pharmacology by the end of the year 2003, upon which reports will become available.

Binding of dietary peptides to different tissue enzymes may promote development of peptidase antibodies in children with autism.
Opioid peptides are available from a variety of food sources. These dietary proteins and peptides, including casein, casomorphins, gluten (GLU) and gluteomorphins, can stimulate T-cells, induce peptide-specific T-cell responses, and abnormal levels of cytokine production, which may result in inflammation, autoimmune reactions and disruption of neuroimmune communications. In celiac disease (CD), a majority of patients who express HLA-dQ2 and/or DQ8 react to a 33-mer peptide and 15 other T-cell stimulatory peptides. This peptide binding to HLA-DQ2and HLA-DQ8 molecules is most efficient when negatively charged amino acids are present at anchor positions in the peptide. Yet GLU contains very few negatively charged amino acids, which makes GLU-derived peptides low affinity ligands for HLA-DQ2 and–DQ8. This paradox has been solved by finding that enzyme tissue transglutaminase, target of endomysium-specific antibodies in CD patients, can modify GLU peptides by conversion of glutamine residues into glutamic acid, which introduces negative charges favored for binding.

A majority of children with autism can not tolerate wheat and milk proteins or peptides and hence elimination of these peptides from the diets significantly improves their conditions. This clinical finding correlates with laboratory results reported earlier by our group in children with autism and by different investigators in MS-like syndromes. They found that an encephalitogenic T-cell response to myelin oligodendrocyte glycoprotein (MOG) could be either induced or alternatively suppressed as a consequence of immunological cross-reactivity or "molecular mimicry" with the extra-cellular IV-like domain of milk protein called but yrophilin (BTN). We detected IgG, IgM and IgA antibodies against nine specific neuron-specific antigens in the sera of children with autism. These antibodies were found to bind with different encephalitogenic molecules that have sequence homologies to a milk protein. 

Indeed, when we tested IgG, IgM and IgA antibodies against milk peptides, we found that every single serum with ELISA values higher than 0.3 O.D. against neurological antigens also exhibited high levels of antibodies against neurological antigens and antibodies against milk peptides in a higher percentage of experimental sera. Similar to milk peptides, antibodies against different gliadin peptides have also been described in celiac disease, gluten ataxia and recently in children with autism.

Test results are for physician use only.