Inflammation and the genes, molecules, and biological pathways that lead to inflammatory processes influence many important and disparate biological processes and disease states that are quite often not generally considered classical inflammatory or autoimmune disorders. These include development, reproduction, aging, tumor development and tumor rejection, cardiovascular pathologies, metabolic disorders, as well as neurological and psychiatric disorders. This paper compares parallel aspects of autism and inflammatory disorders with an emphasis on asthma. These comparisons include epidemiological, morphometric, molecular, and genetic aspects of both disease types, contributing to a hypothesis of autism in the context of the immune based hygiene hypothesis. This hypothesis is meant to address the apparent rise in the prevalence of autism in the population.

One of the challenges in the early study of the molecular basis of classical autoimmune disorders was the attempt to establish the relevance of highly variable and fluctuating immune serum proteins and cell populations to disease etiology. That is, are fluctuations in any set of cytokines, immune mediators, or T cell populations, causative or are they epiphenomena due to peripheral effects of target tissue destruction, transient common infections, or more importantly, are they echoes of long ago infections. There is an ever present “which came first, the chicken or the egg” nature in the study of highly variable immune mediators. Are oligoclonal antibody bands found in the CSF of multiple sclerosis patients 25 related to the etiology of the disease or are they end stage phenomena? Do alterations in cytokines from a patient with systemic lupus erythematosus have a role in disease etiology or are they late stage responses to tissue destruction brought on by other mechanisms? Similarly, are immune aberrations in autism disease causing or are they epiphenomena?

Numerous reports have described imbalances in immune and inflammatory processes in autistic patients, including aberrations in antibody levels, cytokines, and cellular subsets 7 , 8 , 9 , 10 , 11 , 12 . Additionally, recent reports have described an increased frequency of HLA-A2 13 and HLA-DR4 14 antigens in autism. Interestingly, epidemiological studies have provided evidence for the association of asthma and allergies 15 or autoimmune disorders in families with autistic children 16 , 17 , 18 , 19 . The exact significance of immune abnormalities and the relationship of infections, immunizations, allergies, inflammation, or other aspects of immune response to disease etiology are unclear and controversial. Alterations of immune and inflammatory processes in autism have recently been reviewed 3 , 20 , 21 , 22 - 24 .

Many theories regarding the biological basis of autism have been suggested, including neurodevelopmental, exposure to environmental toxins, particularly to mercury 2 , and immune 3 hypotheses. More recently, theories of hyper-systemizing and assortative mating 4 , 5 and hyper-dopamine 6 have been proposed. At this time there is little definitive evidence to support any single theory of the fundamental biological nature of autism.

Autism is an enigmatic childhood disorder of unknown origin. It is characterized by developmental, language, and social deficits, ranging in severity from patients with profound deficits to individuals that are high functioning. Although the underlying etiological basis of autism has eluded researchers, the genetic heritability of autism is quite strong 1 . Specifically what genes are involved and how they contribute to the disease phenotype is unclear.

Other comparisons of autism to asthma and autoimmune/inflammatory disorders

In addition to imbalances in immune molecular mediators, there are other seemingly unrelated parallels in the study of immune and inflammatory disorders as compared to autism that, when viewed collectively, may provide additional support for shared aspects of disease etiology between immune and inflammatory disorders and autism. These include; sex bias, birth order, age-of-onset, neonatal head circumference, increasing prevalence in the population, rural versus urban disease comparisons, and shared molecular and genetic markers.

Disease onset and sex bias In asthma and in autism presentation is in early childhood. Both disease types have an age of onset in early childhood; 2-4 years for children with autistic disorder 26 and 3 to 6 for wheeze and asthma 27. In addition, both autism and asthma display a skewed sex bias toward boys. This bias is approximately 4:1 boys to girls in autism 1 and approximately 2:1 in asthma 28,29. It is well known that in most adult autoimmune and inflammatory disorders, including asthma, there is a predominance of adult women with the diagnosis. However, less well known is that prior to puberty this skewing is toward boys 30. This male bias prior to puberty may be true in other immune mediated disorders as well such as multiple sclerosis 31, Type 1 diabetes, and thyroiditis.

Birth order Some studies have shown birth order to be relevant in atopic disorders as well as autism. In both cases, being first born may carry a greater risk for disease than later births. In a large study of 11, 371 Italian young men those with no siblings had the highest level of serum IgE sensitization. An inverse association was observed between number of siblings at time of testing and prevalence of high atopy p < 0.0001 32. Similar findings have been shown in for atopic disease in Crete 33, asthma, eczema-urticaria and hay fever in Scotland 34, asthma with allergic rhinitis in Denmark 35 and asthma, allergy, and eczema in the Netherlands 36. These observations are thought to be related to increased transmission of childhood infections due to a growing family size in the context of the hygiene hypothesis (see below). Similarly, the risk of autism has been shown in some cases to be related to birth order in the same direction as asthma and atopic disorders, with risk decreasing with a greater number of older siblings in the United States 37,38,39, Western Australia 40 and England 41.

Increased neonatal head circumference Increased neonatal head circumference has been found in both autism and asthma. Increases in neonatal head circumference have been associated with asthma and atopy. In particular, head circumference has been associated with elevated serum IgE levels and hay fever disorders 42-45. Increased neonatal head circumference or macrocephaly is a robust finding in autism with the largest effect between the ages of 2-5 46,47,48,49,50,51. This brain size difference is largely back to normal by adolescence. The biological basis for this increase is unknown although genetic, infectious, and inflammatory mechanisms have been proposed 50 (see PTEN below).

Increase in prevalence in the population: parallel “epidemics” Both autism and asthma have had reports of apparent increases in the population over the last 30 years. Numerous studies show general increases in prevalence in both asthma 52,53 and autism 54-58, at roughly similar rates over the last 30 years. In both disease types this has been often referred to as an “epidemic” 54,59. In both disease types this apparent increase is controversial. Changes in diagnostic classifications and access to health care resources have confounded the interpretation of prevalence estimates in the study of asthma and autism. Significant increases in disease prevalence over a short time in evolutionary terms suggest that purely genetic mechanisms may not be solely responsible 60. Given the strong heritability of autism, changing environmental modifiers in the context of the background genetics of autism may be important over the past 30 years. There have been similar increases in the prevalence in classical autoimmune diseases over the same time span as well, including Type 1 diabetes 61,60,62.

Rural vs Urban disease distribution Both autism and asthma appear to show uneven geographical distributions in disease prevalence. Differential susceptibility or resistance to asthma and allergies is found in urban environments versus rural or farm environments 63,64,65. Although the exact mechanistic basis of the difference is not known, this distribution pattern of disease is thought to have an inverse relationship to infection and is central to the hygiene hypothesis (see below). The geographical distribution of autism is less clear although there is evidence that there may be an urban versus rural distribution. This has been found in epidemiological studies from multiple countries including Denmark 66, the United States 67, England 55,68 and Japan 69. Interestingly, in studies of autism that analyzed numerous familial risk factors, a major risk factor for autism was increasing degree of urbanization 55,68. In a study from the US, the urban versus rural distribution was attributable to mercury exposure in the environment, however this may reflect an industrialized versus rural pattern as well 67. The Inuit of northern Canada may provide an interesting population case study. This isolated rural population exists in crowded living conditions, with high levels of mercury and other environmental toxins in the diet 70. However, autism is essentially non-existent in the Inuit. In a recent report Fombonne, et al., state; “No case of autism has ever been reported in an Inuit child in the past 15 years 71. In parallel, asthma and atopic disorders are uncommon in Inuit children, even with very high rates of lower respiratory infections prior to age 2 and particularly high rates of childhood smoking (31.9%) 72.

Molecular and genetic markers shared with inflammatory/ autoimmune diseases Like many common human disorders, autism, asthma, and autoimmune disorders have been studied using genetic linkage and genetic association approaches. The chromosomal regions identified in linkage studies and the specific variants of genes identified in genetic association studies are quite often not unique to any one disorder. Many, if not most, genes in the human genome have broad based effects influencing different cells and tissues at different times of development under the influence of different environmental modifiers. In the context of common human disease, important regulatory genes may effect disease susceptibility differently when found in combination with different disease associated alleles 73.

ADRB2, beta(2)-Adrenergic receptor The gene for the beta(2)-Adrenergic receptor encodes a member of the G protein-coupled receptor superfamily and is expressed on epithelial and endothelial cells of the lung, mast cells, as well as airway smooth muscle cells. ADRB2 activation is thought to work through increased intracellular cAMP levels 74. Polymorphisms in ADRB2, including the Glu27 allele, have been studied in multiple disease states including hypertension 75, atopic dermatitis 76, Graves disease 77, rheumatoid arthritis 78,79, obesity 80 and in particular, asthma 81. ADRB2 is of major interest in asthma as it may be involved in lung function as well as response to β 2 -Adrenergic agonists 82,83. ADRB2 polymorphisms may not influence asthma incidence or prevalence but may influence persistence of asthmatic symptoms 84. Importantly, the Glu27 allele of ADRB2 has recently been associated with autism in twins 85 as well as in the AGRE autism cohort 86.

PTEN-Phosphatase and tensin homolog PTEN, phosphatase and tensin homolog, is central to phosphoinositide metabolism as an important regulatory checkpoint in the PI3K/ATK signaling pathway, effecting multiple downstream processes including immune function, cell growth, cell survival, and differentiation 87-89. PTEN has been shown to play a role in lymphocyte proliferation, systemic autoimmunity, and autoimmune disease 87,88, as well as in benign tumors of the gastrointestinal tract. In relation to disease, PTEN has been implicated in bronchial asthma and allergic inflammation 90. Interestingly, PTEN has been implicated in macrocephaly (OMIM # #153480) and Cowden disease (OMIM #158350). PTEN has been implicated in autism as well, in particular, within a subset of autistic individuals with macrocephaly 91,92. A recent report described a patient with a PTEN mutation having autistic features, macrocephaly as well as nodular lymphoid hyperplasia of the small and large intestinal mucosa 93. Moreover, a mouse model with specific deletions of PTEN in selected neuronal cell types resulted in macrocephaly, changes in social interactions, and increased responses to sensory stimuli, suggesting a model for autistic spectrum disorder 94.