On the origins of autism: The Quantitative Threshold Exposure hypothesis

Medical Hypotheses 85 (2015) 798–806

Contents lists available at ScienceDirect

Medical Hypotheses journal homepage: www.elsevier.com/locate/mehy

On the origins of autism: The Quantitative Threshold Exposure hypothesis S. Crawford ⇑ Department of Biology, Southern Connecticut State University, 501 Crescent Street, New Haven, CT 06515, United States

a r t i c l e

i n f o

Article history: Received 6 March 2015 Accepted 8 October 2015

a b s t r a c t The Quantitative Threshold Exposure (QTE) hypothesis is a multifactorial threshold model that accounts for the cumulative effects of risk factor exposure in both the causation of autism spectrum disorder (ASD) and its dramatic increase over the past 30 years. The QTE hypothesis proposes that ASD is triggered by the cumulative effects of high-level exposure to endogenous and environmental factors that act as antigens to impair normal immune system (IS) and associated central nervous system (CNS) functions during critical developmental stages. The quantitative threshold parameters that comprise a cumulative risk for the development of ASD are identified by the assessment of documented epidemiological factors that, in sum, determine the likelihood that ASD will occur as a result of their effects on critically integrated IS and CNS pathways active during prenatal, neo-natal and early childhood brain maturation. The model proposes an explanation for the relationship between critical developmental stages of brain/immune system development in conjunction with the quantitative effects of genetic and environmental risk factors that may interface with these critical developmental windows. This model may be useful even when the individual contributions of specific risk factors cannot be quantified, as it proposes that the combined quantitative level of exposure to risk factors for ASD rather than exposure to any one risk factor per se defines threshold occurrence rates. Ó 2015 Elsevier Ltd. All rights reserved.

Introduction Autism represents a broad-spectrum group of neurological disorders associated with language and social impairments that range in severity from the mild dysfunction characterized by Asperger’s Syndrome, to classic autism, that, in its extreme form, causes severe cognitive and developmental deficits classified as pervasive developmental disorder-not otherwise specified (PDD-NOS). Up to 50% of autism cases are also associated with some degree of cognitive impairment. Since first described by Kanner in 1943, both the biological and environmental factors with potential links to autism have received close scrutiny, while diagnostic and behavioral criteria have been elaborated in the form of a spectrum of behaviors that critically involve deficits in brain development affecting socialization and verbal communication [1]. The dramatic upswing in the incidence of autism spectrum disorders (ASDs) over the past several decades has unleashed a flurry of hypotheses to account for this disturbing trend. Two Support from Connecticut State University Research Grant 2014–15. ⇑ Tel./fax: +1 203 392 6215. E-mail address: [email protected] http://dx.doi.org/10.1016/j.mehy.2015.10.006 0306-9877/Ó 2015 Elsevier Ltd. All rights reserved.

decades ago, autism was detected in 1 in 1000 children; as of 2014 an estimated 1 in 88 children has been identified with ASD [2] (see Fig. 1). This increase cannot be fully accounted for by the refinement of detection and diagnostic procedures. Over the past decade, a large body of research has probed the nature of these etiological risk factors [3]. Approximately 10% of children with ASD show genetic, neurologic or metabolic differences that may be etiologically linked to this disorder. As detection and surveillance of autistic disorders in recent years have included neuro-imagery alongside monitoring of age-appropriate behavior, neuro-structural differences in the developing brain in children with ASD have been identified [4]. In this context, inflammation has been identified as a contributing factor responsible for these neuro-structural differences. Additional studies have also identified genetic changes, pre-natal conditions, vaccines and even the environmental milieu as contributory factors in the development of autism [5]. Nevertheless, no specific link has been established between any of these potential risk factors and the development of ASD. Rather, specific genetic profiles and epigenetic factors, such as maternal prenatal immune system dysregulation/activation, have shown a correlated increased risk for the development of autism.

S. Crawford / Medical Hypotheses 85 (2015) 798–806

799

Fig. 1. Rise in autism prevalence 1975–2012.

The Quantitative Threshold Exposure hypothesis The Quantitative Threshold Exposure (QTE) hypothesis is a multifactorial threshold model that accounts for the cumulative effects of risk factor exposure in both the causation of ASD and its dramatic increase over the past 30 years. The QTE hypothesis proposes that ASD is triggered by the cumulative effects of exposure to endogenous and environmental factors that act as antigens that impair normal immune system (IS) and associated central nervous system (CNS) functions during critical developmental phases. The QTE hypothesis for ASD is an elaboration of the classical threshold model of disease involving multifactorial risk factors (see Fig. 2). The threshold model presents a normal distribution of risk factors implicated in disease liability with an associated threshold level of quantitative risk factor exposure that precipitates the onset of a multifactorial disorder. An individual will not express a multifactorial disease trait unless the quantitative effects of risk factors, of genetic and/or environmental composition, reach a critical threshold. The threshold model of disease is of particular utility in assessing multifactorial disorders involving critical developmental/exposure windows such as cleft palate and spinal bifida.

Fig. 2. The threshold model for multifactorial disease.

The quantitative threshold parameters that comprise a cumulative risk for the development of ASD are identified by the assessment of documented epidemiological factors that, in sum, determine the likelihood that ASD will occur as a result of their effects on critically integrated IS and CNS pathways active during prenatal, neo-natal and early childhood brain maturation. The model rejects the concept that a single genetic or environmental agent is the cause of most cases of ASD. Rather, the QTE hypothesis proposes that it is the quantitative exposure level to any number or combination of genetic and environmental risk factors at critical developmental stages that determines whether the threshold exposure level is sufficient to cause ASD (see Fig. 3). The QTE hypothesis suggests that the threshold model of multifactorial disease can also explain the developmental dys-regulation responsible for ASD. The model proposes an explanation for the relationship between critical developmental stages of brain/ immune system development in conjunction with the quantitative effects of genetic and environmental risk factors that may interface with these critical developmental windows. This model may be useful even when the individual contributions of specific risk factors cannot be quantified, as it proposes that the combined quantitative level of exposure to risk factors for ASD rather than exposure to any one risk factor per se defines threshold occurrence rates. The hypothesis predicts that the greater the number of risk factors and the quantitative amounts of each to which a child is exposed prenatally and in early postnatal life, the greater the likelihood of developing ASD. These risk factors are multifaceted in origin, and have been identified in extensive epidemiological studies to include genetic predisposition, maternal fetal exposure to infectious disease, inflammatory and autoimmune phenomena, as well as exposure to antigenic and pro-inflammatory environmental factors. Taken together, this cumulative exposure may cause an individual to cross the threshold boundary to develop ASD. The model further predicts that, once the threshold exposure level sufficient to impair CNS development to cause ASD is attained, additional increases in quantitative risk factor exposure may determine the severity of ASD, thereby accounting for the spectrum or range of neurological impairments identified in children with ASD.

800

S. Crawford / Medical Hypotheses 85 (2015) 798–806

Increasing Risk for ASD

 Epidemiologically defined risk factors for ASD.  Genetic risk factors associated with ASD.  Primary causes and effects of brain dysfunction associated with ASD.  Identified positive correlations between risk factor incidence and prevalence (I/P) rates and I/P rates of ASD.

ASD

THRESHOLD

Increasing levels of Risk Factor Exposure

Each of these areas will be discussed in the following sections as they are implicated in ASD by the QTE hypothesis.

NO ASD

A. The balance of exposure versus effect determines Outcome.

Low Risk for ASD

ASD

Epidemiologically defined risk factors for ASD

ASD

THRESHOLD Low level Exposure to Risk Factors

NO ASD

B. Low level exposure falls below threshold effect on ASD.

High Risk for ASD ASD

High Level Exposure to Risk Factors

THRESHOLD

NO ASD

C. Increasing exposure levels to risk factors, both and antenatal, has increased the likelihood of crossing the threshold of ASD occurrence and in increased overall incidence rates over the past 20+ years. Fig. 3. The QTE hypothesis is based on the threshold model of multifactorial disorders. The hypothesis proposes that the sum of total of genetic and environmental risk factors that may affect integrated IS/CNS associations at critical windows of pre-natal and infant development may tip the balance at a critical threshold to cause ASD. Further elaboration of cause and effect relationships between each risk factor and IS/+CNS development may permit a quantitative assessment of relative risk.

The timing as well as the level of exposure to ASD risk factors may critically impact not only whether the disorder threshold is met, but also the severity of ASD. This combined quantitative/ developmental timing of exposure criteria has previously been applied to risk factors associated with congenital developmental disorders such as spinal bifida, microencephaly and cleft palate/ hare lip syndromes. The severity of each of these disorders is tightly associated with the timing of pre-natal events that interfere with programmed morphogenetic CNS development pathways. In ASD also, the level of exposure necessary to trigger an impairment of CNS development may vary significantly with the timing of exposure to determine threshold impact and the severity of ASD. Earlier prenatal exposure may impact programmed CNS pathways that occur normally during gestational growth, whereas post-natal exposure may affect stages of brain development programmed to occur in the first several years of life. The elaboration of this hypothesis involves the following parameters as they relate to the development of ASD:

Despite important advances in our understanding of the physiological basis for ASD, it is currently impossible to predict which children will develop this disorder; one can only identify critical biological and environmental risk factors that may increase the likelihood of its occurrence in an individual. Among the critical epidemiological factors associated with ASD is age of incidence, newborn to about 3 years; this coincides with the timing of critical stages of postnatal development in both the brain and the immune system. This diagnostic window is an essential parameter for identifying potential genetic and environmental factors that may be responsible for the development of ASD. Some of these factors may originate in pre-natal exposure to agents and/or abnormal gene products that may disrupt the normal process of brain development; however, the fact that both IS and CNS maturation occur in postnatal life suggests that post-natal exposure to exogenous factors and/or endogenous gene products important to these later stages of development may also play a critical role in the genesis of this disorder. Prenatal infection and autoimmune disorders may be risk factors for ASD. Recent studies by Comi et al. [6] showed that the mean number of autoimmune disorders is greater than average in families with autism. The most common autoimmune disorders prevalent in families with autism are type 1 diabetes, rheumatoid arthritis, hypothyroidism and lupus. This link illustrated between autism and autoimmune disorders suggests that in some cases, abnormal immune system function combined with certain environmental factors may increase the probability of autism. Obesity and gestational diabetes, increasingly common complications of pregnancy, may also contribute to fetal neuro-inflammation that may underlie or facilitate the development of ASD. Levels of interleukin-6 (IL-6) are markedly increased in gestational diabetes, with especially high levels in the intrauterine environment and in the cerebellum of the autistic brain where IL-6 levels were reportedly increased by 89% in a study of autistic patients [7]. IL-6 promotes inflammation and leukocyte migration to developing neural tissue of the fetus. Thus, neuro-inflammation during embryonic development may act to disrupt normal synaptic patterning that may be linked to the capacity of IL-6 to cause anomalies of neural cell adhesion, myelination, and over-expression of granule cell excitatory synapses. Recent studies have begun to evaluate the effects on immune system functions and ASD risk of exposure to various environmental factors in perinatal and neonatal stages of development. Several studies have shown increased risk for ASD due to congenital exposure to rubella, cytomegalovirus, and pharmacological exposure to thalidomide and valproic acid [8–12]. Furthermore, recent epidemiologic research has labeled the perinatal window as the most susceptible to environmental factor influences, specifically maternal prenatal medications and infection [13]. Mothers who suffered from prenatal maternal urinary tract, upper respiratory and vaginal infections have a higher risk of having a child with ASD. Several research studies have provided evidence that both pathogenic and commensal microbes can trigger IS dysregulation and

S. Crawford / Medical Hypotheses 85 (2015) 798–806

autoimmune responses linked to neurological impairment. In this regard, several recent clinical studies have reported that the severity of childhood cognitive impairment was associated with specific anti-brain antibody profiles [14]. Moreover, mid-pregnancy blood samples taken from mothers of children with autism have shown higher concentrations of inflammatory cytokines than mothers of children without autism. Additional environmental factors with suggested links to autism include thimerosol in pediatric vaccines [15] the measles– mumps rubella vaccine [16], acetaminophen [17], organophosphate pesticides [18], and excess vitamin exposure in utero [19]. Lead and mercury (in fish) exposure are also neurotoxic. In this regard, high levels of fish consumption and its potential to affect mercury levels in pregnant women may serve as an important risk factor for ASD in countries such as Japan that consume large amounts of fish. Vitamin D (VD) deficits [20] have been linked to ASD. Vitamin D intake levels have received significant attention since vitamin D plays an important role in IS and CNS development. VD is known to play an essential role as a neurotrophic factor in CNS development as it upregulates critical neural growth factors, including nerve growth factor (NGF), glial cell neurotropic factor and others [21]. Grant and Cannelle have shown that autistic individuals have lower than normal levels of circulating VD and that ASD incidence is higher in areas of the world with shorter sun light intervals and in highly pigmented individuals in which UV associated VD uptake is reduced [22]. Moreover, Cannelle has provided evidence that VD supplementation may alleviate some symptoms of ASD [23]. In contrast, Bittner [24] has argued that infants and young children have high intake levels of vitamin D as a result of early weaning to vitamin D fortified milk and other foods. Research by Schultz et al. [25] suggested that ASD incidence is lower in breast-fed babies than formula fed babies and that the longer the duration of breastfeeding the lower the risk of ASD. There are many differences in these two modes of feeding; however, it should be noted that formula and milk are both VD supplemented whereas human milk contains low levels of VD [26]. Related research has shown that children with ASD have low levels of N-acetylgalactosaminidase (Nagalase), an enzyme that blocks the activity of the vitamin D3 binding protein (VDBP) to Gc macrophage activating factor (GcMAF), which is required for macrophage activation [27] The net result is that decreased Nagalase activity results in increased levels of macrophage activation. Interestingly, increased levels of nagalase have also been reported in systemic lupus erythematosus, an autoimmune disorder whose incidence is elevated in families of children with ASD. These observations suggest that the altered metabolic use of VD, independent of intake levels, may be a relevant factor in ASD.

Genetic risk factors associated with ASD Precision genetic screening has failed to identify a specific gene responsible for this disorder, although it is entirely possible that a spectrum of genetic changes may predispose to the development of ASD (see Fig. 4). Genetic analyses of DNA from individuals with ASD have demonstrated the presence of rare inherited copy number variations (CNVs) associated with this disorder. In addition, over one hundred potential candidate genes have been identified; however, none yet has been directly linked to ASD [28,29] (see Fig. 2). Among the most prominent of these are the serotonin transporter 5-hydroxytryptophan (5-HTT) and Raelin (RELN) genes, both on chromosome 7 and neurolignan genes on the X chromosome. Genes associated with ASD located on the X chromosome might explain, in part, the increased prevalence of ASD in males. A large, sophisticated twin study by Hallmayer et al. [30]

801

Fig. 4. Genes and mutations implicated in recent genetic studies of individuals with ASD.

showed a significantly lower concordance rate between genetic factors and ASD risk than suggested by earlier studies, therefore, indirectly placing more emphasis on environmental causes of ASD. This study, interestingly, showed a higher concordance rate for dizygotic twins than for siblings, pointing to the intrauterine environment as a potentially important factor in ASD. Functional studies of gene transcription in individuals with autism have provided evidence of reduced or abnormal patterns of synaptic transmission and immune system dysfunction [31]. These findings correlate with earlier studies showing evidence of neuroinflammation in autopsied brains from autistic individuals [32]. The autistic transcriptome shows a deficit in gene expression patterns in frontal and temporal cortex regions associated with differentiated brain function, a pattern that emerges normally during fetal development. However, it should be noted that synapse formation and connectivity pathways initiated in the last trimester of fetal development continue for several years of post-natal development, thereby extending the sensitivity window to ASD dysfunction to the first several years of life. Kostovisc and Judas published data suggesting that the levels of SCGmRNA levels are lower in the brains of individuals exposed to cannabis during the timeframe when intra-and extra-cortical axon pathways are formed in the developing brain [33]. Primary causes and effects of brain dysfunction associated with ASD Chief among the more important biological assessments is the documented association between aberrations of immune system functions and brain development. The confluence of clinical data suggesting a link between autism and dys-regulated immune function and research in developmental biology that shows the integration of central nervous system (CNS) and immune system (IS) formation early in life presents a coherent model for understanding the biological basis for this disorder. Beyond these critical correlations is the necessity of identifying specific roles for genetic and environmental factors that may have etiological significance as risk factors for the development of ASD. Direct evidence of a link between abnormal brain maturation and dysregulated immune system function can be seen in clinical research studies of the brains of individuals with autism. Several important research studies have identified differences in the immunologic composition of the cerebral spinal fluid, microglia and macrophages in autistic individuals that may impact neo-natal and post-natal brain development [34]. Recent studies suggest that an important biological target for this immune

802

S. Crawford / Medical Hypotheses 85 (2015) 798–806

Fig. 5. The Quantitative Threshold Model (QTE) of ASD implicates the sum total combined effects of exposure to diverse genetic, physiological and environmental risk factors that, taken together, interfere with normal IS/CNS critical prenatal and postnatal IS/CNS development windows. Levels of exposure have increased significantly in the past several decades, which may account for increased incidence of ASD.

dysfunction are the Purkinje neurons located in the cerebellum that play a critical role in elaborating neural connections and sensory input throughout the brain, including the pre-frontal cortex [35]. Elevated levels of microglia in association with decreased Purkinje cell density in the cerebellum provide an anatomic connection illustrating the interface of these two critical organ systems and their association with abnormal brain function. Additional pathological changes in autopsied brains show involvement of the brainstem, cerebellum, hippocampus, amygdala, entorhinal cortex, fusiform gyrus, cingulate cortex as well as increased size of the frontal lobes and number of dendritic spines suggesting an early, in utero origin of the developmental deficit [4]. Longitudinal brain imaging studies have shown that these changes were accompanied by brain overgrowth, further suggesting in utero involvement [36]. Identified positive correlations between risk factor incidence and prevalence (I/P) rates and I/P rates of ASD The ‘‘Quantitative Threshold Exposure” (QTE) hypothesis proposes that over-exposure to a plethora of antigenic stimuli in utero and in the early postnatal years may overwhelm the developing immune system, thereby provoking dysregulated responses that may pose an enhanced risk for ASD as a result of secondary effects

on CNS development. Taken together, the data obtained from these etiological studies of the origins of ASD represent only the beginnings of an explanation for the extraordinary and alarming increase in its incidence over the past fifty years. A critical question involves whether or not it is possible to use the available epidemiological data to identify (a) potential cause(s) of this dramatic upswing in ASD incidence in recent years. Ironically, the most important clue to the cause of ASD risk may be the statistic itself that defines a meteoric rise in its incidence in the past thirty years. Statistical assessments have indicated that the observed increase in ASD incidence is real, and cannot be fully accounted for by increased diagnosis. Moreover, the extraordinary rate of increase in ASD incidence and prevalence (I/P) over the past several decades rules out certain etiological risk factors as the primary cause of ASD. For example, the physiological role of genetic mutations as a cause of ASD, based on known mutational frequency rates, is far too low to account for the observed increase in ASD incidence. This observation essentially rules out genetic mutations as the sole cause of ASD. What we are left with, then, as a potential clue to better understanding the origins of ASD, is its dramatically increased incidence and prevalence (I/P) over the past several decades that strongly suggests that a dramatic change and/or increase in exposure to etiological factors that predispose to ASD in prenatal and early postnatal years may account, at least in

S. Crawford / Medical Hypotheses 85 (2015) 798–806

part, for the statistical increase in ASD incidence rates. Thus, identifying environmental changes that have occurred over the past several decades with the potential to impact brain development in utero and early in life may permit the identification of key causal factors of ASD. Epidemiologists have identified many risk factors for ASD whose levels have changed significantly over the past half century, and it is very likely that, taken together, these may account for a significant fraction of the increased incidence rates for ASD over the past quarter century. Many of these risk factors involve the association of immunological function with brain development (see Fig. 5). The following serves as a partial list of risk factors whose increased incidence/prevalence (I/P) levels may be associated with the increased I/P of autism. Increased incidence of pediatric infectious disease Epidemiological data on pediatric illnesses show not only an elevated incidence of ASD over the past several decades, but also in the incidence of childhood infections, asthma and other autoimmune disorders, whose rate of increase has occurred in parallel to that of ASD [37–39]. This correlation is strengthened by the observation that risk factors linked to asthma have been also implicated in the etiology of ASD. Many of the cited immunological risk factors for asthma, whose early childhood incidence has increased in parallel to ASD over the past several decades, involve similar immunological factors linked to ASD. For example, it has been well documented that immune cytokines play an important role in normal brain development as well as contributing to pathological injury to the developing brain. Epidemiological factors that impact these levels during critical developmental windows may impact the incidence rates of disorders whose underlying pathology is linked to these altered immune system components, including asthma and ASD. Clinical data suggest that the incidence rates of early childhood infections have increased significantly over the past several decades. For example, a study in which 2540 children were followed from birth to age 10 showed that early childhood respiratory infections increased the risk for the development of asthma [40]. Moreover, a clinical study by Wu et al. [41] provided evidence that episodes of lower respiratory tract infection by Respiratory Syncytial Virus (RSV) in early life are associated with an increased risk of developing asthma. Yet another study by Jackson et al. [42] showed that RSV associated infant viral infections were highly predictive of the development of asthma by age 6 in a cohort of high-risk infants. Additional data provided by the COPSAC study have shown that newborn (1 month) airway infections associated with Streptococcus pneumoniae, Haemophilus influenzae, and/or Moraxella catarrhalis, correlated with a higher risk of wheezing symptoms, elevated blood eosinophil numbers, increased total IgE, and a higher rate of development of asthma by school age [43]. Clinical research studies on the patterns of childhood infections have shown that, despite the availability of antibiotics and vaccines responsible for dramatic decreases in infant mortality rates in the 20th century in the developed world, rates of infant and early childhood infectious disease have been affected by changes in early childhood care practices over the past half century. According to Monthly Labor Review, one of the most dramatic changes in the US society in the past 75 years has been the increasing percentage of women, including mothers of infants, that work outside the home, resulting in a substantial increase in the number of infants in daycare [44]. In 2014 the percentage of US children of working mothers in daycare under a year old was 57.1%. Moreover, incidence rates for infections in infancy are higher for babies cared for in multiple settings, particularly daycare. A study by Bell et al. in 1989 of 843 children under 36 months of age showed that children cared for at home had 2.03 infections on average during

803

the six month study period [45]. Adjusted rates of excess infection rates were 0.09 in relatives’ homes, 0.1 in daycare homes, 0.79 in day care centers and 0.66 in multiple settings. Moreover, children in daycare centers were 4.5 times more likely to be hospitalized for infections than children in other care settings. Interestingly, the strongest predictor of early childhood illness was the number of children in a single room in a caregiving facility: the greater the number of children, the greater the infection rate. An additional study by Hurwitz et al. showed a statistically significant increase in respiratory illness rates in infants from 6 weeks to 17 months who attended group daycare [46]. Yet another clinical study by Ethelberg et al. showed that between 1980 and 1996 rates for hospitalization of infants suffering from bronchiolitis increased significantly, particularly among children younger than 6 months, in which the increase was 239% [47]. Moreover, the rate of hospitalization for children with bronchiolitis increased for all age groups during this time period. These infectious disease rate increases correlated with increased rates of early childhood daycare enrollment during this period. Among children of working mothers aged between 1 and 2 years, the enrollment percentage increased from 12% in 1982 to 25% in 1993. For children 0–12 months, this percentage rose from 5% to 20% during this time frame. The authors concluded that the trend of increasing childcare enrollment in group daycare may be associated with increased rates of RSV associated bronchiolitis in very young children. According to a study by Hagerhed-Engman et al., children who attended day care had an increased risk of airways infections, eczema and food allergies [48]. A total of 10,851 children, 1–6 years were assessed in this study. This increased risk was mainly seen in younger children 1–4 years. These studies have demonstrated that there is a parallel between changes in childcare arrangements in early years, increased incidence of early childhood infections and the increased incidence of asthma and ASD. To the extent that excessive or dysregulated immune system stimulation at critical developmental windows may impact brain development, the increased rate of early childhood infection brought about by these demographic changes may, in part, be responsible for increased ASD I/P.

Changes in infant/early childhood feeding practices Modern caregiving trends increasingly have involved shorter duration of exclusive breastfeeding as the primary source of infant nutrition as well as the introduction of a variety of solid foods early in life. Earlier childrearing practices into the 20th century in the US and most parts of the developed world involved breastfeeding (often exclusively) for most of the 0–3 years critical IS/CNS developmental window associated with ASD risk. Several important research studies have shown that specific infant feeding patterns such as early introduction of bottle feeding or solid food in place of longer periods of exclusive breastfeeding may lead to reduced lung and airway growth and increased risk of autoimmune disorders, including childhood asthma [49]. The link between breastfeeding duration and asthma risk may be associated with breast milk components, including IgA, cytokines, glycans and longchain fatty acid effects on innate immune system stability. These stabilizing effects of breast milk on the immune system may also have a protective effect against ASD. Patterns of decreased duration of breastfeeding and/or early introduction of a variety of solid foods may have a contributory role in increasing the antigenic risk factor profile for asthma and ASD. The correlation between infant feeding pattern and ASD risk is particularly compelling given the documented association between gastrointestinal dysfunction and ASD [50].

804

S. Crawford / Medical Hypotheses 85 (2015) 798–806

Increased obesity/diabetes type-2 in parents of child-bearing age Maternal (and possibly paternal) obesity comprises another important risk factor for altered brain development and ASD. Currently, approximately 34% of women of childbearing age in the US are obese; these numbers have increased appreciably over the past several decades. A US study of 1311 mothers and children between 2005 and 2012 showed that severely obese mothers (BMI > 35) are three times more likely to have a child who develops ASD [51]. The increased risk of ASD in children of obese mothers was independent of other risk factors, including pregnancy weight gain, gestational diabetes, duration of breastfeeding, postnatal depression or infant birth weight. Maternal gestational diabetes (GDM) is often associated with obesity and represents an independent risk factor for ASD. A study by Feig et al. conducted in Ontario, Canada showed that the incidence of GDM and pre-GDM in pregnancy doubled between 1996 and 2010 [52]. Increased use of tetrahydrocannabinol (THC) during child-bearing years In utero exposure to drugs that interact with neural pathways has been implicated as an important risk factor for ASD. Cannabis/tetrahydrocannabinol (THC) is the most widely used psychotropic drug; its use has increased substantially over the past 20 years; moreover, more recent formulations of the drug display enhanced potency due to changes in preparation methods [53]. Currently, cannabis use during pregnancy is estimated at 10%. Recent studies by Passey et al. [54] and Shabani et al. [55] have shown that neural deficits can result from in utero cannabis exposure. Siniscalco et al. [27] have suggested that the endocannabinoid (EC) system may play an important role in the integrated IS/CNS developmental pathway that is dysregulated in autism. Their research has shown that the cannabinoid receptor type 2 (CBR2) signal pathway is upregulated in peripheral blood mononuclear cells (PBMCs) from children with ASD. This finding raises the possibility that the endocannabinoid (EC) system may be associated with ASD. In contrast to elevated CB2R levels in PBMCs, the authors found reduced levels of bone marrow differentiated macrophages (BMDCs) suggesting several altered immunologic functions associated with CBR2 levels. Other research suggests that the altered CB2 levels observed in autistic children could be the consequence of infections or differences in the microbial composition of the gut, based on findings in mouse models of ASD [57]. Additional research findings by Muhl et al. [58] on CB2 receptor metabolism in long-term cannabis users suggest a possible relationship to EC2 receptor levels in autistic children. The study analyzed peripheral blood levels of CB-1, CB-2 proteins and mRNA levels of the CB agonist anandamide in individuals who had used cannabis at least 20 times (lifetime use) but not for 6 months prior to tests for CB levels. The authors found that both CB-2 and anandamide levels (but NOT CB-1 levels) were elevated in the study cohort compared to individuals who had used cannabis five times or less (lifetime exposure). These data are remarkably similar to the findings of Sinilcasco et al. [56] showing that only the level of CB-2, but not CB-1, was elevated in peripheral blood samples taken from autistic children. These findings suggest that fetal or early childhood exposure to cannabis may explain the observed increase in CB-2 levels in the peripheral blood mononuclear cells of children with ASD. Moreover, recent studies show that obesity is associated with increased levels of 2-arachidonoylglycerol cannabinoid agonist and CB2 receptor protein levels in adipose tissue and the liver [59–62], suggesting a further link between maternal obesity and cannabis use as independent risk factors for ASD as a result of their metabolic effects on endocannabinoid signal pathways.

Endogenous cannabinoids bind to type-1 cannabinoid receptors in the CNS to guide neural pattern formation and network connectivity in the developing brain. Research by Cutando et al. provides evidence that THC binding of EC-1 receptors as a consequence of subchronic cannabis exposure may affect these signal pathways, at least in part, by activating microglial cells important to neural function [63]. Similar patterns of cerebellar microglial activation have been documented in the brains of autistic children, suggesting similar pathways may be involved. Tortoriello et al. [64] have recently determined that THC affects EC-1 receptor signaling in the developing fetal brain by altering fetal cortical circuitry, further implicating THC as a potential cause of autism. Discussion In contrast to some current hypotheses on the origins of ASD that implicate specific environmental risk factors (such as early childhood vaccines), the QTE hypothesis does not associate a specific individual risk factor with the risk of developing ASD. Rather, with regard to the nature of ASD risk factors, the model proposes the following:  No one specific risk factor, genetic or environmental, is responsible for ASD.  Risk factors for ASD comprise environmental and/or in utero exposure to any infectious, organic or inorganic materials with antigenic properties capable of eliciting immune response activity.  Additional risk factors for ASD involve genetic factors/polymorphisms and/or mutations that may impact IS/CNS associations during development and/or immune reactivity to antigenic stimulation.  The quantitative cumulative level of exposure to risk factors and the developmental window in which this exposure threshold occurs determines ASD incidence and severity. The QTE accounts for the precipitous rise in ASD incidence over the past several decades, in that it proposes that the combined cumulative levels of exposure to risk factors in utero and in the early postnatal years has increased in parallel to the increased incidence of ASD (see Table 1). The QTE hypothesis is supported by evidence showing that the level of pre-natal, neo-natal and early childhood exposure to epidemiological risk factors for ASD has

Table 1 A list of identified risk factors for ASD whose incidence/prevalence rates have increased over the past 30 years, in parallel with the time frame in which I/P rates for ASD have increased significantly. Risk factor for ASD

>Incidence/prevalence over past 30 years

Pre-natal infection Prenatal auto-immune disorders Gestational type II diabetes Maternal thalidomide use Maternal valproate use Congenital virus infection Early childhood vaccines (MMR)/ thimerosal Organophosphate pesticide exposure Excess vitamins Low vitamin D levels Lead exposure Genetic factors Maternal cannabis use Early childhood infectious disease Mercury exposure

No Yes Yes No No No No

Maternal obesity

No No Uncertain No No Yes Yes Dependent on fish consumption rates Yes

S. Crawford / Medical Hypotheses 85 (2015) 798–806

increased significantly over the past fifty years. Epidemiologists have identified many risk factors for ASD whose levels have changed significantly over the past half century, and it is very likely that, taken together, these may account for a significant fraction of the increased incidence rates for ASD over the past quarter century. Many of these risk factors involve the association of immunological function with brain development. The QTE hypothesis is consistent with many identified etiological risk factors associated with ASD and provides a substantive explanation for its dramatic increase in incidence over the past several decades. For example: 1. The QTE hypothesis accounts for disorder heterogeneity, that ASD severity and anatomic deficits correlate with the level and timing of exposure to a broad spectrum of etiological factors that directly affect the severity and attributes of the disorder phenotype. 2. The QTE hypothesis attributes ASD causation to the cumulative exposure to diverse etiological factors that impact IS/CNS pathways during critical developmental windows and, as such, accounts for the fact that, despite extensive research on its etiology, no specific, individual genetic or environmental cause of ASD has been identified. 3. The QTE hypothesis accounts for the recent dramatic increase in ASD incidence over the past several decades as the disorder incidence rates correlate with increasing levels of exposure to etiological risk factors that impact threshold occurrence rates. How does this hypothesis stand up to the epidemiological profile for ASD? First and most important, the hypothesis argues against a single environmental culprit-such as vaccines – in the causation of ASD. Rather, it argues that the sum total of antigenic exposure coupled with an at-risk genetic profile is responsible for generating a threshold boundary for ASD development. This hypothesis accounts for the failure of epidemiologists to identify a single specific risk factor for ASD or a single genetic profile that is associated with enhanced risk. Rather, the sum total of epidemiological data supports the notion of the QTE hypothesis, as it implicates a plethora of infectious disease, autoimmune phenomena linked to ASD, all of which increase the risk of ASD, but none of which has been implicated as a single agent causative factor. Thus the QTE hypothesis integrates a diverse group of genetic and environmental factors with the timing of exposure, such that this complex and unstable environmental interface with critical periods of IS/CNS development may be responsible for the increased incidence of ASD. The QTE hypothesis suggests a positive correlation between excessive antigenic exposure and disordered brain development when it occurs during critically sensitive periods of CNS development. The identification of a causal link requires an in-depth assessment of pertinent etiological factors that may affect brain development at these critical stages. A causal link is suggested by clinical data showing that brain development is influenced by antigenic stimuli that elicit immune activation and modulatory mechanisms that interact physiologically with the developing brain. In this regard, the number and quantity of antigenic stimuli to which the developing brain is exposed may be critical. The development of the immune system is largely shaped by its physiological and environmental connections, in the recognition of self versus non-self, in the development of memory cells providing an historical record of an individual’s history of environmental antigenic exposure and the highly reactive immune responses that often occur following exposure to a previously unseen, unrecognized antigen. Likewise, the brain develops and matures along pathways critically sensitive to both physiological and environmental input that affect social, cognitive and sensory development.

805

Both systems require lifelong interaction and high-level sensitivity to changing components of the environmental milieu. Herein lies the absolute uniqueness of these systems: the uniqueness of the IS written by the history of its exposure to self/non-self antigens; the uniqueness of the brain is written by the uniqueness of its interactions with the sensory world and the internal physiology of each individual. The extraordinary sensitivity and plasticity of these highly evolved organ systems may ultimately provide the physiological foundation for disorders such as ASD as the consequence of dysregulatory interactions with the external world that impair the maturation of parts of the brain most sensitive to the developmental impacts of the early life environment. Conclusion The Quantitative Exposure Threshold hypothesis for ASD is consistent with our current understanding of the epidemiological risk factors and etiological mechanisms connecting immune system dysfunction with abnormal brain development seen in ASD. The model has important predictive value, as it suggests that, rather than attempting to identify a specific causative agent linked to ASD, the combined risk factor profile should be evaluated in population studies in order to specify more accurately the limits of prethreshold exposure. It is critical to emphasize that the QTE does not implicate any single causative factor linked to the development of autism; rather, it proposes that the level of exposure to the sum total of immune-stimulating factors at critical CNS developmental windows may be implicated in ASD. Conflict of interest No. Acknowledgments The author would like to acknowledge Hannah Noelle Lord for research reference assistance. This paper is dedicated to Christina Jennette, who knows more about autism than most of us. References [1] Kanner L. Autistic disturbances of affective contact. Nervous Child 1943;2:217–50. [2] Division of Birth Defects, National Center on Birth Defects and Developmental Disabilities 2012. Autism spectrum disorders: data and statistics. Centers for Disease Control and Prevention (CDC); 2014. [3] Currenti SA. Understanding and determining the etiology of autism. Cell Mol Neurobiol 2010;30:161–71. [4] Murphy CM, Christakou A, Daly EM. Abnormal functional activation and maturation of fronto-striato-temporal and cerebellar regions during sustained attention in autism spectrum disorder. Am J Psychiatry 2014;171 (10):1107–16. [5] Stoltenberg C, Schjølberg S, Bresnahan M, et al. ABC Study Group. The autism birth cohort: a paradigm for gene-environment-timing research. Mol Psychiatry 2010;15:676–80. [6] Comi AM, Zimmerman AW, Frye VH, et al. Familial clustering of autoimmune disorders and evaluation of medical risk factors in autism. J Child Neurol 1999;14(6):388–94. [7] Wei H, Zou H, Sheikh AM, Malik M, et al. IL-6 is increased in the cerebellum of autistic brain and alters neural cell adhesion, migration and synaptic formation. J Neuroinflammation 2011;8:52. http://dx.doi.org/10.1186/17422094-8-52. [8] Gardener H, Spiegelman D, Buka SL. Prenatal risk factors for autism: comprehensive meta-analysis. Br J Psychiatry 2009;195:7–14. [9] Libbey JE, Sweeten TL, McMahon WM, Fujinami RS. Autistic disorder and viral infections. J Neurovirol 2005;11:1–10. [10] Hwang SJ, Chen YS. Congenital rubella syndrome with autistic disorder. J Chin Med Assoc 2010;73:10–47. [11] Atlado´ ttir HO, Thorsen P, Ostergaard L, et al. Maternal infection requiring hospitalization during pregnancy and autism spectrum disorders. J Autism Dev Disord 2010;40:142–330. [12] Dietert RR. Distinguishing environmental causes of immune dysfunction from pediatric trigger of disease. Open Pediatr Med J 2009;3:38–44.

806

S. Crawford / Medical Hypotheses 85 (2015) 798–806

[13] Bilder D, Pinborough-Zimmerman J, Miller J, McMahon W. Prenatal, perinatal, and neonatal factors associated with autism spectrum disorders. Pediatrics 2009;123:1293–300. [14] Elamin NE, Al-Ayadhi LY. Brain autoantibodies in autism spectrum disorder. Biomark Med 2014;8(3):345–52. [15] Hviid A, Stellfeld M, Wohlfahrt J, et al. Association between thimerosalcontaining vaccine and autism. JAMA 2003;290(13):1763–6. [16] Jain A, Marshall J, Buikema A, et al. Autism occurrence by MMR vaccine status among US children with older siblings with and without autism. JAMA 2015;313(15):1534–40. [17] Shaw W. Evidence that increased acetaminophen use in genetically vulnerable children appears to be a major cause of the epidemics of autism, attention deficit with hyperactivity, and asthma. J Restor Med 2013;2:1–16. [18] Shaw GM, Yang W, Roberts E, et al. Early pregnancy agricultural pesticide exposures and risk of gastroschisis among offspring in the San Joaquin Valley of California. Birth Defects Res A Clin Mol Teratol 2014;100(9):686–94. [19] Zhou SS, Zhou YM, Li D, Ma Q. Early infant exposure to excess multivitamin: a risk factor for autism? Autism Res Treat 2013;9:636–97. [20] Cannell JJ. Autism and vitamin D. Med Hypotheses 2008;70:750–9. [21] Brown J, Bianco JJ, Mcrath JJ, Eyles DW. 1,25-Dihydroxyvitamin D3 induces nerve growth factor, promotes neurite overgrowth and inhibits mitosis in embryonic rat hippocampal neurons. Neurosci Lett 2003;343:139–43. [22] Grant WB, Cannell JJ. Autism prevalence in the United States with respect to solar UV-B doses. Dermato Endocrinol 2013;5:159–64. [23] Cannell JJ. Autism, will vitamin D treat core symptoms? Med Hypotheses 2013;81:195–8. [24] Bittker S. Oral vitamin D, infants, toddlers, and autism in the United States: 1980 to 2010. J Autism 2015;2:1–8. [25] Schultz ST, Klonoff-Cohen HS, Wingard DL, et al. Breastfeeding, infant formula supplementation, and autistic disorder: the results of a parent survey. Int Breastfeed J 2006;1–16. [26] Pioggia G, Tonacci A, Tartarisco G, et al. Autism and lack of D3 vitamin: a systematic review. Res Autism Spectrum Disord 2014;8(12):1685–98. [27] Siniscalco D, Bradstreet JJ, Cirillo A, Antonucci N. The in vitro GcMAF effects on endocannabinoid system transcriptionomics, receptor formation, and cell activity of autism-derived macrophages. J Neuroinflammation 2014;78–89. [28] Stilp RL, Gernsbacher MA, Schweigert EK, Arneson CL, Goldsmith HH. Genetic variance for autism screening items in an unselected sample of toddler age twins. J Am Acad Child Adolesc Psychiatry 2010;49:267–76. [29] Neale BM, Kou Y, Liu L, Ma’Ayan A, Samocha KE. Patterns and rates of exonic de novo mutations in autism spectrum disorders. Nature 2012;485(7397):242–5. [30] Hallmayer J, Cleveland S, Torres A, et al. Genetic heritability and shared environmental factors among twin pairs with autism. Arch Gen Psychiatry 2011;68(11):1095–102. [31] Luo R, Sanders SJ, Yuan Tian Y, et al. Genome wide transcriptome profiling reveals the functional impact of rare de novo and recurrent CNVs in autism spectrum disorders. Am J Human Gen 2012;9:38–55. [32] Onore C, Careaga M, Ashwood P. The role of immune dysfunction in the pathophysiology of autism. Brain Behav Immun 2012;26(3):383–92. [33] Kostovic I, Judas M. The development of the subplate and thalamocortical connections in the human foetal brain. Acta Paediatr 2010;99:1119–27. [34] Dietert RR, Dietert JM. Potential for early-life immune insult including developmental immunotoxicity in autism and autism spectrum disorders: focus on critical windows of immune vulnerability. J Toxicol Environ Health B Crit Rev 2008;11:660–80. [35] Jeong J-W, Tiwari VN, Chugani DC, et al. In vivo detection of reduced Purkinje cell fibers with diffusion MRI tractography in children with autistic spectrum disorders. Front Hum Neurosci 2014;8:1–10. [36] Lange N, Travers BG, Bigler ED, et al. Longitudinal volumetric brain changes in autism spectrum disorder ages 6–35 years. Autism Res 2014. http://dx.doi.org/ 10.1002/aur.1427. [37] Yunginger JW, Reed CE, O’Connell EJ, Melton 3rd LJ, O’Fallon WM, Silverstein MD. A community-based study of the epidemiology of asthma. Incidence rates, 1964–1983. Am Rev Respir Dis 1992;146(4):888–94. [38] Osman M, Tagiyeva N, Wassall HJ, et al. Changing trends in sex specific prevalence rates for childhood asthma, eczema, and hay fever. Pediatr Pulmonol 2007;42(1):60–5. [39] Williams JG, Higgins JP, Brayne CE. Systematic review of prevalence studies of autism spectrum disorders. Arch Dis Child 2006;91(1):8–15.

[40] Nafstad P, Brunekreef B, Skrondal A, Nystad W. Early respiratory infections, asthma, and allergy: 10-year follow-up of the Oslo birth cohort. Pediatrics 2005;116(2):e255–62. [41] Wu P, Hartert TV. Evidence for a causal relationship between respiratory syncytial virus infection and asthma. Expert Rev Anti Infect Ther 2011;9 (9):731–45. [42] Jackson DJ, Gangnon R, Evans MD, et al. Wheezing rhinovirus illnesses in early life predict asthma development in high-risk children. Am J Resp Crit Care Med 2008;178(7):667–72. [43] Isgaard Hans. The Copenhagen Prospective Study on Asthma in Childhood (COPSAC): design, rationale, and baseline data from a longitudinal birth cohort study. Ann Allergy Asthma Immunol 2004;93(4):381–9. [44] Galley J. Stay-at-home mothers through the years. Monthly Labor Rev 2014. [45] Bell DM, Gleiber DW, Mercer AA, et al. Illness associated with child day care: a study of incidence and cost. Am J Public Health 1989;4:479–84. [46] Hurwitz ES et al. Risk of respiratory illness associated with day-care attendance: a nationwide study. Pediatrics 1991;87(1):62–9. [47] Shay DK, Holman RC, Newman RD, et al. Bronchiolitis-associated hospitalizations among US children, 1980–1996. JAMA 1999;282(15):1440–6. [48] Hagerhed-Engman L, Bornehag CG, Sundell J, Aberg N. Day-care attendance and increased risk for respiratory and allergic symptoms in preschool age. Allergy 2006;61(4):447–53. [49] Dogaru CM, Nyffenegger D, Pescatore AM, et al. Breastfeeding and childhood asthma: systematic review and meta-analysis. Am J Epidemiol 2014;179 (10):1153–67. [50] Heberling CA, Dhurjati PS, Sasser M. Hypothesis for a systems connectivity model of autism spectrum disorder pathogenesis: links to gut bacteria, oxidative stress, and intestinal permeability. Med Hypotheses 2013;80 (3):264–70. [51] Jo H, Schieve LA, Sharma AJ, et al. Maternal prepregnancy body mass index and child psychosocial development at 6 years of age. Pediatrics 2015. http://dx. doi.org/10.1542/peds.2014-3058. [52] Feig DS, Hwee J, Shah BR, et al. Trends in incidence of diabetes in pregnancy and serious perinatal outcomes: a large, population-based study in Ontario, Canada, 1996–2010. Diabetes Care 2014;37(6):1590–6. [53] Substance abuse and mental health services administration mental health, United States; 2012 [HHS Publication No. (SMA)12-4681]. [54] Passey ME, Sanson-Fisher RW, D’Este CA, Stirling JM. Tobacco, alcohol and cannabis use during pregnancy: clustering of risks. Drug Alcohol Depend 2014;134:44–50. [55] Shabani M, Mahnam A, Sheibani V, Janahmadi M. Alterations in the intrinsic burst activity of Purkinje neurons in offspring maternally exposed to the CB1 Cannabinoid agonist. J Membr Biol 2014;247:63–72. [56] Siniscalco D, Sapone A, Giordano C, et al. Cannabinoid receptor type 2, but not type 1, is up-regulated in peripheral blood mononuclear cells of children affected by autistic disorders. J Autism Dev Disord 2013;43(11):2686–95. [57] Clarke G, O’Mahony SM, Dinan TG, Cryan JF. Priming for health: gut microbiota acquired in early life regulates physiology, brain and behavior. Acta Paediatr 2014;103(8):812–9. [58] Muhl D, Kathmann M, Hoyer C, et al. Increased CB2 mRNA and anandamide in human blood after cessation of cannabis abuse. Naunyn Schmiedebergs Arch Pharmacol 2014;387(7):691–5. [59] Auguet T, Berlanga A, Guiu-Jurado E, et al. Endocannabinoid receptors gene expression in morbidly obese women with nonalcoholic fatty liver disease. BioMed Res Int 2014. http://dx.doi.org/10.1155/2014/502542. [60] Deveaux V, Cadoudal T, Ichigotani Y, et al. Cannabinoid CB2 receptor potentiates obesity-associated inflammation, insulin resistance and hepatic steatosis. PLoS ONE 2009;4(6) [article ID e5844]. [61] Horcajadas FA. Cannabinoids in eating disorders and obesity. Mol Neurobiol 2007;36(1):113–28. [62] Harrolda J, Elliotta JC, Kinga PJ, et al. Down-regulation of cannabinoid-1 (CB-1) receptors in specific extrahypothalamic regions of rats with dietary obesity: a role for endogenous cannabinoids in driving appetite for palatable food? Brain Res 2002;952(2):232–8. [63] Cutando L et al. Microglial activation underlies cerebellar deficits produced by repeated cannabis exposure. J Clin Invest 2013;123(7):2816–31. [64] Tortoriello G, Morris CV, Alpar A. Miswiring the brain: D9tetrahydrocannabinol disrupts cortical development by inducing an SCG10/ stathmin-2 degradation pathway. EMBO J 2014;33:668–85.