Complementary Medicine Products Used in Autism - Evidence for Rationale

4.1.1.1. Intestinal hyper-permeability in autism

Functional changes have been reported in the GI tracts of children with autism including increased intestinal permeability (Horvath&Perman 2002). Intestinal permeability, as measured by the urinary excretion of metabolically inert sugars, is a surrogate marker of mucosal integrity and gut barrier function (de Magistris et al. 2010). It has been hypothesised that impaired GI function may not only be a symptom of autism but may also contribute to the phenotypic presentation by increasing absorption of chemicals from the GI tract. There is a body of thought that increased absorption of opioid-like peptides that are derived from gluten and casein (i.e. the ‘opioid excess theory’) may disturb neurological function, as may GI overgrowth of neurotoxin-producing bacteria (Shattock&Whiteley 2002).

4.1.1.2. Altered gut fermentation products in autism

A growing appreciation for the role of intestinal microflora in health and disease has emerged over the last few years, however the beneficial and potentially adverse contributions of bacterial fermentation by-products have not been well established and are largely uncharacterised in children with ASD. A recent metabonomic study revealed metabolic phenotype (metabotype) differences were observed between autistic and control children, which were associated with perturbations in the relative patterns of urinary mammalian microbial co-metabolites including dimethylamine, hippurate, and phenyacetylglutamine (Yap et al. 2010). Another study by Altieri et al. (2011) found higher levels of p-cresol in urine of young children with autism than controls and also reported a positive correlation between urinary p-cresol and autism severity. P-cresol is a toxic metabolite of tyrosine catabolism by gut bacteria such as clostridial species and Pseudomonas stutzeri (Altieri et al. 2011). Whether the observed differences in urinary metabolites observed contribute to, or reflect, GI dysfunction in individuals with ASD requires further investigation.

4.1.1.3. Altered GI microbiota profiles in autism

Several studies have found some bacterial species, particularly some Clostridium species, are present in higher numbers in children with autism experiencing GI disturbance (Finegold et al. 2002; Song et al. 2004; Parracho et al. 2005; Finegold et al. 2010). Some workers have speculated that the differences in the GI microbiotica could be contributing to the pathophysiology of ASD (Bolte 1998; Finegold 2008).

4.1.1.4. Altered digestive enzyme capacity in autism

It has been hypothesised that digestion of dietary gluten and casein in the small intestine by pancreatic and intestinal peptidases releases short chain peptide molecules (exorphins) which are structurally similar to endogenous opioid substances (White 2003). Other hypotheses suggest that excessive opioid activity linked with dietary peptides from gluten and casein have an aetiological role in the pathogenesis of autism (Reichelt&Knivsberg 2003). The ‘opioid excess’ theory of autism proposes that many of the behaviours found in individuals with ASD mimic the influence of opioids on human brain function (White 2003). In addition Horvath & Perman (2002) demonstrated that 44 of 90 (49%) of children with autism showed at least one deficient disaccharidase enzyme activity. Lactase and maltase were the enzymes most commonly measured to have deficient activity, followed by sucrase, palatinase and glucoamylase. They found that all of the children with reduced enzyme activity had flatulence and/or loose stools. It is hypothesised that disaccharide malabsorption may cause chronic diarrhoea and flatulence in children with ASD and may contribute to abnormal behaviour (Patel et al. 2002). Hence, supplementation with these enzymes may cause a reduction in autism related symptoms. For example supplementation with lactase may resolve the symptoms related to lactose malabsorption (Horvath&Perman 2002).

It is speculated that GI symptoms exacerbate the behavioural manifestations of autism contributing to the severity of the disorder (Buie et al. 2010). Abdominal pain, constipation, and/or diarrhoea are unpleasant and can be painful and likely to produce frustration, behavioural problems, and possibly sleep disturbance, aggression and self abuse, especially in children unable to communicate their discomfort.

Various CAMs are used to ameliorate the GI symptoms in children with autism as outlined in Table 2. Probiotics and prebiotics are used to promote gut health. Dietary interventions include the exclusion of gluten and casein containing foods together with dietary supplementation with peptidase enzymes. Peptidase enzymes are implemented in children with autism in an attempt to remove the opioid-like peptides that may exert a neurotoxic effect on the brain (Brudnak et al. 2002). Less obvious candidates are PUFAs which are used with the rationale that they can decrease GI inflammation and improve a 'leaky gut'.

4.1.2.1. Lipid peroxidation

It has also been found that lipid peroxidation (an oxidative biomarker) in plasma is significantly increased in children with autism when compared to their non-autistic siblings reflecting increased oxidative stress in autism (Chauhan et al. 2004).

4.1.2.2. Antioxidants

Reduced endogenous antioxidant capacity i.e. low levels of the plasma antioxidant enzymes glutathione peroxidase and superoxide dismutase have been found in autistic individuals (Sogut et al. 2003). Furthermore, Yorbik et al. (2002) found that the activities of erythrocyte superoxide dismutase and erythrocyte and plasma glutathione peroxidase were significantly lower in 45 autistic children compared with 41 controls. Chauhan et al. (2004) also found a significant reduction of the major endogenous antioxidants transferrin and ceroplasmin in the serum of children with autism as compared to their typically developing siblings. An excess of toxic free radicals e.g. nitric oxide has also been reported in children with autism as compared to age and sex matched controls (Sogut et al. 2003). Ming et al. (2005) also describe increased nitrite concentrations, thiobarbituric acid reactive substances and xanthine oxidase activity in red blood cells in children with autism compared to controls. Notably, decreased plasma levels of the antioxidant vitamins A, C and E were reported in the same autistic cohort (Ming et al. 2005).

4.1.2.3. Mitochondrial disease

Mitochondria serve a primary role in energy production during normal physiological function and generate high levels of ROS which are normally neutralised by free radical scavengers. In the event there is excess ROS relative to the antioxidant defence, mitochondrial dysfunction occurs exacerbating oxidative stress. Mitochondrial disease is associated with impaired neurodevelopment. Two recent studies have reported mitochondrial dysfunction in autism. Post-mortem samples showed increased mitochondrial metabolism and oxidised mitochondrial proteins in the brains of six people with autism compared with controls (Palmieri et al. 2010). Another study showed that 11/21 patients with ASD had definite mitochondrial disease while the rest had probable mitochondrial disease (Weissman et al. 2008).

4.1.2.4. Abnormalities in the trans-methylation and trans-sulphuration pathways

There is emerging evidence that a deficient trans-methylation (i.e. folate/methionine) pathway has a role in the aetiology of autism (Boris et al. 2004; James et al. 2004; James et al. 2006). The folate/methionine pathway (Figure 1) is responsible for the synthesis of the deoxynucleotide triphosphate (dNTP) pools required for DNA synthesis and repair, the establishment and maintenance of stable DNA methylation patterns for tissue-specific gene expression and chromatin conformation and maintenance of the redox balance within each cell.

As shown in Figure 1, the methionine cycle involves the re-methylation of homocysteine to methionine either by methionine synthase (MS) which is folate and vitamin B12 dependent or by the betaine homocysteine methyltransferase (BHMT) reaction. The methyl group is donated by 5-methyl tetrahydrofolate synthesised by methylene tetrahydrofolate (MTHFR) from 5,10 methyl tetrahydrofolate.

Methionine is activated by methionine adenosyl transferase (MAT) to S-adenosyl-methinine (SAM), the major methyl donor for cellular methyltransferase reactions (MTases). Following methyl transfer, SAM is converted to S-adenosylhomocysteine (SAH). This is further metabolised in a reversible reaction to homocysteine and adenosine. Adenosine is either phosphorylated to adenosine nucleotides by adenosine kinase (AK) or catabolised to inosine by adenosine deaminase (ADA). Homocysteine may be permanently removed from the methionine cycle by irreversible conversion to cystathionine by cystathione-β-synthase (CßS). Cystathionine is converted to cysteine which is the rate-limiting amino acid for the synthesis of glutathione.

A decrease in turnover of the folate/methionine pathway will lead to decreased synthesis of SAM which is vital for normal methylation activity and decreased synthesis of cysteine and glutathione required for normal antioxidant activity.

A systematic literature review identifying studies reporting metabolites, co-factors or genes of the folate/methionine pathway in autism found there are significant differences in the levels of various metabolites of the methionine/folate cycle in individuals with autism compared with controls although there are some inconsistencies between studies which may be due to different methodologies (Main et al. 2010).

Elevated levels of oxidised GSH and subsequent reduction in GSH:GSSG ratio together with a decrease in cysteine (a rate limiting amino acid for glutathione synthesis) and increased lipid peroxidation suggests that oxidative stress may play a significant role in the aetiology of autism (James et al. 2004; Wu et al. 2004). The conclusion of a review of oxidative pathways as potential drug targets in autism by Villagonzalo et al. (2010) was that although there is significant evidence demonstrating that oxidative pathways are disturbed in autism, there is insufficient evidence to decide whether oxidative stress is the cause of autism, or contributes to the illness, or is simply a consequence of the illness. They stated that further research is required to determine if children will benefit from antioxidant treatment and that longitudinal studies exploring oxidative biomarkers and autism symptomatology over time may be a methodology for investigation.

Thus as outlined in Table 2, oxidative stress in autism provides a rationale for the use of CAM products that are antioxidants in their own right, metabolites or co-factors of the trans-methylation or trans-sulphuration pathways, co-factors of plasma antioxidant enzymes and include: vitamins A, C & E, vitamin B12, DMG, magnesium, selenium, zinc, melatonin and glutathione.

4.1.4.1. Immune deficiencies and autism

Immune deficiencies in children with ASD have been reported due to frequently encountered medical problems including recurrent ear infection/rhinosinusitis/upper respiratory tract infection, adverse reactions to multiple medications, allergies, GI problems and prolonged courses of illness as compared to typically developing siblings (Jyonouchi et al. 2005; Levy&Hyman 2005). Various immune system deficits including abnormalities in cell-mediated immunity have been reported in autism. Abnormalities of macrophages, B cells, T cells and natural killer cells have been reported in individuals with ASD which may compromise defence against infection (Gupta 2000).

4.1.4.2. Autoimmune disease and autism

A survey administered to families of 61 children with autism and 46 control families with typically developing children discovered that the mean number of autoimmune disorders was greater in families who had children with autism (Comi et al. 1999). In this study, Comi et al. (1999) found that 46% of the families of children with autism had two or more family members with autoimmune disorders (e.g. type 1 diabetes, adult rheumatoid arthritis and hypothyroidism). In addition, this finding may further suggest that genetic predisposition plays an important role in autism. However, Micali, Chakrabarti & Fombonne (2004) did not have the same findings following administration of a semi-structured questionnaire to 79 parents of children with pervasive developmental disorders (PDDs) and 61 controls (parents with typically developing children). They found the rates for any autoimmune disorder for both mothers and fathers combined was 22.4% for parents of controls and 30.9% for parents of children with PDDs which was not significantly different.

It is hypothesised that autoimmune disease in autism may lead to neurodevelopmental damage. Autoantibodies against proteins associated with the central nervous system (CNS) has been reported in some children with autism (Singh et al. 1988; Plioplys et al. 1994; Vojdani et al. 2004). In recent studies, antibodies against the fetal brain have been detected in some mothers of children with autism; these antibodies have the ability to alter behavioural outcomes in the offspring of animal models (Enstrom et al. 2009).

A review by Theoharides et al (2008) proposes a relationship between GI factors, oxidative stress and immune dysregulation and proposes potential drug targets in autism.

4.1.4.3. Vaccination and autism

Although an array of epidemiological studies do not support causality (De Stefano 2007; Baker 2008), the alleged link between autism and vaccination has been debated extensively and many parents and parent advocacy groups continue to suspect that vaccines cause autism. Putative mechanisms for vaccine associated autism include: “1) immune response directed towards a vaccine that cross reacts with host antigens, 2) host response to a vaccine that would result in the production of cytokines and a subsequent autoimmune reaction and 3) toxic components of a vaccine that directly impact on the immune or nervous system” (Levy&Hyman 2005, p.134).

Therefore as outlined in Table 2, immune dysfunction in autism provides a rationale for the use of CAM products that are claimed to be immunomodulators/immunoadjuvants and include: vitamin A, DMG, vitamin C, zinc, colostrum and glutamine.

4.1.5.1. Serotonin and metabolites

5-HT is a monoamine neurotransmitter that plays an important role in the developing brain by directing both neuronal proliferation and maturation (McDougle et al. 2005). CNS 5-HT activity has been involved in a range of physiological functions, such as sleep, sensory perception and appetite, which are often disrupted in autism (Young et al. 1982). High levels of 5-HT during early development may cause a loss of 5-HT receptors and therefore impact on subsequent neuronal development (Whitaker-Azmitia 2001). Neuroimaging studies suggest altered developmental regulation of 5-HT synthesis may be associated with the pathogenesis of autism (Chugani et al. 1999). Hyperserotonemia has been consistently reported in people with autism in more than 25 published studies (Lam et al. 2006).

4.1.5.2. Dopamine and metabolites

Dopamine (DA) is a catecholamine which acts as a major neurotransmitter in the brain. Generally, the dopaminergic system is thought to affect a wide range of functions, including cognition and attention (Nieoullon 2002), motor function (Niimi et al. 2009), predictive reward signal mechanisms (Schultz 1998) and immunity (Basu&Dasgupta 2000). Some animal research has shown that stereotypies and hyperactivity can be induced by increasing dopaminergic functioning suggesting dopaminergic neurons may be overactive in autism (Miller et al. 2010).

4.1.5.3. Noradrenaline

Noradrenaline (NA) is synthesized from DA by dopamine β-hydroxylase and released from noradrenergic neurons as well as from the adrenal medulla into the bloodstream. NA plays a critical role in attention, the stress response (i.e. the ‘‘fight or flight’’ response), anxiety, and memory (Amaral&Sinnamon 1977; Fitzgerald 2009), which are frequently observed to be impaired in individuals with autism. Previous studies have shown that measurements of NA (i.e. in plasma and urine) are generally well correlated with measurements in the CNS (Roy et al. 1988). A range of neurochemical studies have attempted to examine excretion of urinary NA and / or adrenaline (A) in individuals with autism compared with controls and have yielded inconsistent findings. Three studies found higher levels of NA and/or A in autism compared to controls (Barthelemy et al. 1988; Herault et al. 1993; Martineau et al. 1994), while four studies found no differences (Launay et al. 1987; Martineau et al. 1992; Minderaa et al. 1994; Croonenberghs et al. 2000).

Therefore as outlined in Table 2, the range of neurotransmitter abnormalities that have been shown to occur in autism provide a rationale for the use of CAM products that are claim to normalise neurotransmitter levels and function such as: high dose pyridoxine and magnesium, metabolites and co-factors of the transmethylation and trans-sulphuration cycles (see Figure 1), vitamin C, zinc and secretin.