Abstract

Introduction

One of the striking and consistent features of autism and autistic spectrum disorders (ASD) is the preponderance of males versus females. For example, the male to female ratio in autism has been reported at 2-1(Levy, Mandell, & Schultz, 2009) and 6-1(Windham et al., 2010), with an average of 4.2-1 (Fombonne, 2002). The basis for this male bias is unknown with theories including the “extreme male brain” (Baron-Cohen, 2002), hormonal theories (Baron-Cohen, Knickmeyer, & Belmonte, 2005), stress and anxiety theory(Pfaff, Rapin, & Goldman, 2011) and genetic influences contributing to male sex bias. Similarly, a male bias is common in pediatric autoimmune disorders, with evidence for hormonal and genetic contributions to male sex bias in autoimmunity and certain autoimmune diseases. Given the growing evidence of immune dysregulation in autism and ASDs, it is suggested that the origins of the male bias in autism may share etiological features with male gender bias in pediatric autoimmune disease.

Autism and autoimmune disease

Autism and ASDs have been shown to have numerous characteristics in common with autoimmune and inflammatory disorders at the genetic, molecular, cellular, histological, and epidemiological level(Atladottir et al., 2009; Comi, Zimmerman, Frye, Law, & Peeden, 1999; Croen, Grether, Yoshida, Odouli, & Van de Water, 2005; Enstrom, Van de Water, & Ashwood, 2009; Keil et al., 2010; Mouridsen, Rich, Isager, & Nedergaard, 2007; Sweeten, Bowyer, Posey, Halberstadt, & McDougle, 2003). These include shared genetic associations(Becker, 2007), altered immune pathways at the RNA level(Lintas, Sacco, & Persico, 2010; Voineagu et al.), imbalances in cytokine, chemokine, and antibody profiles(Ashwood et al., 2010; Grether, Croen, Anderson, Nelson, & Yolken, 2010), variations in T cells, macrophages, and mast cells(Grigorenko et al., 2008; Mostafa, Al Shehab, & Fouad, 2010; Theoharides et al., 2010), as well as activation of glial cells in the brain(Vargas, Nascimbene, Krishnan, Zimmerman, & Pardo, 2005). Immune allergic response has been reported to be increased in Asperger syndrome as well(Magalhaes et al., 2009).

One hallmark of autoimmune disorders is an increase in the frequency of other autoimmune and inflammatory disorders, not necessarily the patient’s specific disease, in family members of the patient(Anaya, Gomez, & Castiblanco, 2006; Barcellos et al., 2006; Prahalad, Shear, Thompson, Giannini, & Glass, 2002). This is thought to represent an underlying shared genetic susceptibility and a broad dysregulation of immune pathways in autoimmunity and inflammation(Anaya et al., 2006; Becker et al., 1998). While many genes have been associated with autoimmune disease, they are almost exclusively genes of immune regulation. There is little evidence that disease target organ specificity (i.e. MS, brain; type 1 diabetes, pancreas; etc.) in autoimmune disease is genetically determined while there is evidence that infection, lack of early immune challenge, or other environmental factors drive tissue disease specificity in the context of a genetically determined susceptible immune system.

Importantly, autism and ASDs have been shown to have an increased frequency of autoimmune disorders in family members of autistic patients in multiple studies in different populations(Atladottir et al., 2009; Becker 2012; Comi et al., 1999; Croen et al., 2005; Keil et al., 2010; Mouridsen et al., 2007; Sweeten et al., 2003), suggesting an underlying genetic basis of immune dysregulation. In particular, both type 1 diabetes and autoimmune thyroid disease have been shown to be significantly increased in families with autism and ASDs (Atladottir et al., 2009; Comi et al., 1999), and a link between type 1 diabetes and autism has been suggested(Freeman, Roberts, & Daneman, 2005).

Male gender bias in pediatric autoimmune disease

Classical adult autoimmune disorders are well known to often have a strong female sex bias. This female gender bias generally begins at puberty and is most pronounced in early adulthood and may decline after menopause. However, what is not well appreciated is that prior to puberty, in pediatric or juvenile autoimmune disease, sex bias is often reversed with a clear male sex bias.

Table 1 shows reported male female ratios and age ranges for selected autoimmune disorders. A male bias has been shown in juvenile type 1 diabetes in Denmark (Christau et al., 1977), Sardinia (Songini & Muntoni, 1992), and Sweden (Ostman et al., 2008) and in 21 other populations, with no evidence for female predominance in type 1 diabetes in any population(Ostman et al., 2008). Moreover, a male sex bias has been shown in asthma with a preponderance of boys before puberty followed by a reversal of the sex ratio during puberty, with girls having more asthma and atopy throughout the reproductive years(Osman, 2003). In Crohn’s disease, a skewed male-female ratio has been described in children and adults (Ishige et al., 2010). Although early onset juvenile autoimmune disease is rare, male bias has been noted in immune thrombocytopenia(Sutor, Harms, & Kaufmehl, 2001), Goodpasture’s syndrome(Beeson, 1994), IgA nephropathy(Wyatt et al., 1998), pediatric primary sclerosing chloangitis(Miloh, Arnon, Shneider, Suchy, & Kerkar, 2009), and juvenile multiple sclerosis(Banwell, Ghezzi, Bar-Or, Mikaeloff, & Tardieu, 2007; Haliloglu et al., 2002; Hanefeld, 1995; Ruggieri, Polizzi, Pavone, & Grimaldi, 1999) Interestingly, the peak of the highest male bias often tends to be at approximately ages 2–6, the typical age of onset for autism and ASDs. This bias tends to decline and invert at puberty, leading to a strong female preponderance post puberty and into adulthood. A male bias in pediatric autoimmune disorders is not universal, with a female bias having been demonstrated for juvenile psoriatic arthritis (Stoll & Nigrovic, 2006), juvenile dermatomyositis (Pachman et al., 2005), and juvenile thyroid autoimmunity (Kaloumenou et al., 2008).

Hormones and sex bias in autoimmune disease and autism

Hormonal influences on adult female sex bias in autoimmunity are complex and highly variable between different autoimmune disorders(Fairweather, Frisancho-Kiss, & Rose, 2008; Lockshin, 2002; McCombe, Greer, & Mackay, 2009) and little is known about the role of hormonal mechanisms in pediatric autoimmune disease. Estrogen has generally been shown to enhance immune responses including altering cytokine and chemokine levels as well as the levels of important immune transcriptional regulators such as NFkB and FOXP3. Androgens tend to have immunosuppressive effects(Cutolo et al., 2006; Rubtsov, Rubtsova, Kappler, & Marrack, 2010; Voskuhl, 2011) with hormonal changes during puberty and young adulthood contributing to sex bias in autoimmunity. These effects on disease are often altered by infection, pregnancy, and menopause(Cutolo et al., 2006; Rubtsov et al., 2010; Voskuhl, 2011). Testosterone has been shown to have an immunosuppressive effect experimentally in animal models(Voskuhl, 2011), as well as clinically. In type 1 diabetes and asthma, it has been suggested that the immunosuppressive and immunomodulatory effects of testosterone may play a role in altering disease course and influencing the male-female disease ratio(Osman, 2003). Moreover, androgen anomalies have been found in adolescents and adults with type 1 diabetes.(Codner & Escobar-Morreale, 2007; Danielson, Drum, & Lipton, 2008; Meyer et al., 2000)

Similarly, testosterone is thought to play a key role in autism during fetal and postnatal development. Increased levels of testosterone have been correlated with autistic traits in toddlers and young children(Auyeung et al., 2009; Auyeung, Taylor, Hackett, & Baron-Cohen, 2010) and in women with autistic traits(Ingudomnukul, Baron-Cohen, Wheelwright, & Knickmeyer, 2007), with the suggestion that testosterone may have a direct effect on brain development and may influence male gender bias(Baron-Cohen et al., 2005; Chura et al., 2010). Other reports of testosterone and testosterone precursor anomalies have been described in the context of autism and autistic traits as well(Manning, Baron-Cohen, Wheelwright, & Sanders, 2001) (Croonenberghs et al., 2010; Ingudomnukul et al., 2007; Ruta, Ingudomnukul, Taylor, Chakrabarti, & Baron-Cohen, 2011; Takagishi et al., 2010).

At the level of molecular regulation, the effects of testosterone have been associated with both autism and immune regulation through the same signaling molecules. Recently, a new candidate gene in autism, retinoic acid related orphan receptor alpha (RORA) has been shown to be under positive and negative feedback by androgens and estrogen through its gene target aromatase (CYP19A1), which is responsible for the conversion of testosterone to estrogen(Sarachana, Xu, Wu, & Hu, 2011). RORA has also been shown to be reduced in cells lines and in the brains of autism patients(Nguyen, Rauch, Pfeifer, & Hu, 2010). Importantly, RORA, which is phosphorylated by ERK2 (see below), is a key regulator in autoimmune and inflammatory processes and is required for TH17 differentiation and IL17 expression through a mechanism involving STAT3 and interaction with FOXP3(Du, Huang, Zhou, & Ziegler, 2008; Jetten, 2009) (see below).

In addition, the extracellular signal-regulated kinases (ERKs) have been suggested to be central to neurodevelopmental anomalies in autism(Levitt & Campbell, 2009) and have been shown to be upregulated in both a mouse model of autism(Zou H, 2011) as well as in the brains of autistic patients(Yang K, 2011). Anomalous regulation of ERKs lead to altered immune regulation(Gorelik & Richardson, 2010) while testosterone activation of ERKs has been shown to be differentially regulated in neutrophils between males and females, leading to gender dependent production of 5-Lipoxygenase (Pergola et al., 2008), a key enzyme in leukotriene biosynthesis and inflammation(Radmark & Samuelsson, 2010). This has been suggested as a molecular basis for differences in male sex bias in inflammation and asthma.

The influence of the X chromosome in sex bias in autism and autoimmune disease

The influence of the X chromosome in autism and male sex bias in autism has been described by multiple groups. Genetic evidence of loci on the X chromosome has been shown for phenotypes including autism, autistic features, intellectual disability, and speech delay. Recently, genetic mutations in the gene PTCHD1 at chromosome Xp22.11 have been suggested as a basis for this male bias in autism and intellectual disability(listed, 2010; Noor et al., 2010) although mutations at this locus comprised less than 1% of affected patients tested(Noor et al., 2010).

Likewise, differences in immune function and immune disorders between males and females are profoundly influenced by the X chromosome(Libert, Dejager, & Pinheiro, 2010). Anomalies in X inactivation, haploinsuficiency, hormonal variability regulated by the X chromosome, as well as X chromosome microRNAs may play a role in male-female immune differences at different developmental stages and in different disease contexts(Libert et al., 2010).

The short segment of the X chromosome, Xp11.21-23, has been a focus of mapping in autism and intellectual disability as well as in autoimmunity and autoimmune disease. Giorda(Giorda et al., 2009) reported segmental duplications at Xp11.22-p11.23 associated with intellectual disability, speech delay and EEG anomalies in males and females. Additionally, complex rearrangements at Xp11.23-p11.3(Bonnet et al., 2006; El-Hattab et al., 2010; Giorda et al., 2009; Marshall et al., 2008; Monnot et al., 2008; Qiao et al., 2008) (Chung et al., 2011; Edens, Lyons, Duron, Dupont, & Holden; Holden ST, 2010; Marshall et al., 2008) have been noted in individuals with autism, autistic like features, speech delay, and/or intellectual disability. Similarly, type 1 diabetes and autoimmune thyroid disease, both of which have been shown to be overrepresented in families with autistic children(Keil et al., 2010; Molloy et al., 2006), have both been linked to Xp11.23 (Villano et al., 2009)and have been suggested to play a role in male gender bias(Cucca et al., 1998).

Importantly, the gene FOXP3, in Xp11.23, maps to the small region described for autism, autistic features and intellectual disability (Bonnet et al., 2006; Giorda et al., 2009). FOXP3 is a central regulator of immune function and autoimmunity(Buckner, 2010; Campbell & Koch, 2011; Gambineri, Torgerson, & Ochs, 2003; Lourenco & La Cava, 2011) with FOXP3⁺ regulatory T cells playing a key role in T cell homeostasis(Mai, Wang, & Yang, 2010). Skewed X inactivation has been shown to alter FOXP3 levels in the autoimmune disease systemic sclerosis(Broen et al., 2010). Polymorphisms in FOXP3 have been associated with multiple autoimmune disorders including type 1 diabetes(Bassuny et al., 2003), autoimmune thyroid disease(Inoue N, 2010), lupus(Lin et al., 2011), vitiligo(Birlea et al., 2011), psoriasis(Gao et al., 2010), and allergic rhinitis(Zhang et al., 2009), although other studies found no association(Zavattari et al., 2004). Moreover, a mouse model of brain inflammation has been demonstrated to be mediated through Hepatocyte Growth Factor (HGF) and c-MET (the HGF receptor), an autism associated gene, resulting in induction of FOXP3⁺ regulatory T cells(Benkhoucha et al., 2010).

Limitations and Conclusions

The suggestions made here regarding male gender bias in autism and autoimmunity are speculative. They reflect the complex and variable nature of molecular, cellular, and hormonal processes in development and disease. Moreover, attributing disease relevance to individual features of specific genetic regions, as in the example of FOXP3, is quite speculative given that CVNs, deletions, and amplifications, may affect many genes or regulatory elements in those regions.

However, evidence is mounting that immune dysregulation and autoimmunity are fundamental aspects of autism and ASDs. Prenatal and postnatal hormonal fluctuations that have been associated with both autism and autoimmunity are in the context of a nascent and rapidly developing brain and immune system and both autism and autoimmune disease may be influenced by hormonal fluctuations through shared transcriptional and regulatory interactions. Likewise, the molecular and genetic basis of the male sex bias in both autism and autoimmune disorders is unclear. It is suggested here that the underlying determinants of male sex bias in autism and pediatric autoimmune disorders overlap and may manifest themselves through shared molecular mechanisms and common hormonally driven immune processes, in the context of genetic background, and may provide clues to the etiology of both types of disorders.

Acknowledgements

References