Cobalamin, Microbiota and Epigenetics
Functional cobalamin (B12) status and assessment are inextricably intertwined with the human microbiome. Small bowel bacterial overgrowth can both cause and result from gastritis and alter dietary cobalamin absorption. Some bacterial species may produce human-inaccessible cobalamin corrinoids and may create competition for human-accessible cobalamin. Increased human-inaccessible corrinoids from bacterial production may raise the total corrinoid level assessed by the serum total cobalamin, limiting diagnostic utility and masking a deficiency of human-accessible cobalamin. Anaerobic bacteria may reverse the propionic to succinic acid pathway, converting methylmalonic acid back to propionic acid to release CO2; this could raise propionic acid and lower methylmalonic acid levels, limiting its diagnostic utility. Cobalamin deficiency limits enzymatic conversion of homocysteine to methionine and increases homocysteine levels. Increased homocysteine can be reduced by diversion into the transsulfuration pathway, limiting the diagnostic power of this metabolite. Finally, in the delicate balance between folate and cobalamin which regulates DNA synthesis, excess synthetic folate from public health policies can combine with bacterial folate production to mask the macrocytic anemia of cobalamin deficiency.
Small bowel bacterial overgrowth can increase propionic acid production and reduce cobalamin bioavailability. Both propionic acid administration and cobalamin deficiency can alter brain fatty acid levels and brain function and cause autistic symptomology. Essential fatty acid ratios can modify gut bacterial species which can, in turn, modify fatty acid composition and inflammation. Omega-3 supplementation can reverse many of the symptoms of propionic acid neurotoxicity. Cobalamin supplementation can raise omega-3 fatty acid levels in the brain and can improve autism symptomology. Therefore, there are strong epigenetic interrelationships among cobalamin and its enzymatic activity, propionic acid, essential fatty acids, folate, and the human bacterial microbiome.
KeywordsAutism spectrum disorder B12 Brain-derived neurotrophic factor Cobalamin Dysbiosis Methylation Methylmalonic acid Propionic acid Micronucleated lymphocytes Polyunsaturated fatty acid Short-chain fatty acids Small bowel bacterial overgrowth
List of Abbreviations
Attention deficit hyperactivity disorder
Autism spectrum disorder; B12 = cobalamin
Brain-derived neurotrophic factor
Micronucleated lymphocytes MTHF = methyltetrahydrofolate
Polyunsaturated fatty acid
Short-chain fatty acids
Small bowel bacterial overgrowth
Vitamin B12 (cobalamin) is essential and cannot be synthesized by humans. Cobalamin refers to a corrin ring-containing cobalt complex belonging to a larger group of cobalamin corrinoids (Krautler 2012). The richest dietary cobalamin sources include red meat, liver, and organ meat (Gille and Schmid 2015; Williams 2007). In industrialized countries, red and organ meat intake has declined amid increasing vegetarian and vegan trends (Clonan et al. 2016; Daniel et al. 2011). Generally then, there has been an overall decline in cobalamin intakes among industrialized populations, which can have negative multigenerational impacts. Low cobalamin status increases neural tube risk (Molloy et al. 2009). Cobalamin deficiency is associated with a higher rate of miscarriages (Bennett 2001). Breastfed babies of vegan or vegetarian mothers can exhibit profound neurological delays and evidence of under-myelination on MRI (Lovblad et al. 1997; Kocaoglu et al. 2014; Guez et al. 2012).
Cobalamin is a component of key enzymes that influence both the homocysteine-methionine pathway and the propionic-succinic acid pathway, potentially altering DNA and cellular energy production. Equally, propionic acid-producing small bowel intestinal bacteria can modify cobalamin absorption and host cobalamin availability as well as influence fatty acid-mediated inflammatory responses and brain fatty acids. Cobalamin, therefore, demonstrates a unique interplay with the gut microbiome, with methylation cofactors, and with short-chain (propionic acid) and polyunsaturated fatty acids (PUFA) in an epigenetic manner.
Cobalamin Absorption and Metabolism: Interrelationships with Gut Microbiota and PPA
Only 1% of cobalamin intake can be passively absorbed; 99% must be IF facilitated, and IF is produced by the gastric parietal cells (Muyshondt and Schwartz 1964). Thus gastrointestinal (GI) health can have a profound effect on the bioavailability and absorption of dietary cobalamin intakes. At the first juncture, insufficiency of gastric acid can alter the ability to cleave cobalamin from dietary protein. Gastric hypochlorhydria can reduce parietal cell secretions (McKoll et al. 1998) and contribute to atrophic gastritis (Sipponen et al. 1996) and small bowel bacterial overgrowth (SIBO) (Belitsos et al. 1992). Paradoxically, hypochlorhydria can lead to heartburn and high acid symptoms (Bredenoord et al. 2006). Treatment for heartburn symptoms includes acid-reducing medications such as proton pump inhibitors and H2 antagonists. Use of acid-reducing medication is associated with cobalamin deficiency and increased methylmalonic acid (MMA) (Ruscin et al. 2002; Saltzman et al. 1994). Proton pump inhibitor (PPI) use is also associated with SIBO (Fujimori 2015; Lewis et al. 1996); in one study, 53% of patients on omeprazole demonstrated SIBO (Thorens et al. 1996). Acid-reducing medications are increasingly prescribed to pregnant women, children, and infants (Diav-Citrin et al. 2005; Smith et al. 2013; Van der Pol et al. 2011) and may have a multigenerational impact on the human microbiome.
At the next juncture in cobalamin digestion, alterations in GI epithelial health can influence the ability to produce and release IF, a rate-limiting step in absorption. The incidence of atrophic gastritis, celiac disease, non-gluten enteropathies, and inflammatory bowel disease is rising in industrialized countries. Gastritis is associated with villous atrophy, impaired IF production, and compromised cobalamin absorption (Lebwohl et al. 2015; Meyniel et al. 1981; Wood et al. 1964). SIBO is associated with villous atrophy (Lappinga et al. 2010) and may be reversible with antibiotic treatments (Haboubi et al. 1991). Patients with atrophic gastritis and bacterial overgrowth absorb significantly less protein-bound cobalamin, which is reversed by antibiotics (Suter et al. 1991).
At the third juncture of cobalamin absorption, exocrine pancreatic insufficiency may preferentially transfer the cobalamin to alternative proteins than IF, rendering the cobalamin functionally inaccessible to the human host (Marcoullis et al. 1980). Pancreatic insufficiency can also contribute to the development of SIBO (Therrien et al. 2016).
Thus, there is considerable overlap between conditions contributing to cobalamin malabsorption and to SIBO. Further, once established, large colonies of small bowel bacteria may have unique needs for cobalamin. Although some bacterial species can produce their own cobalamin (Morita et al. 2008; LeBlanc et al. 2013), excessive bacterial growth may set up an environment of competition for cobalamin between bacterial residents and the human host (Brandt et al. 1977; Degnan et al. 2014; Schjonsby, 1989). Some small bowel bacteria can selectively couple dietary cobalamin to alternative binding proteins, forming non-IF complexes which are inaccessible to the human host and deprive the host of this essential cobalamin source. At the same time, small bowel bacteria may themselves produce cobalamin corrinoids whose molecular structure does not allow them to be used by humans (Schjonsby 1989). A decrease in IF-bound cobalamin availability to the human host amid extensive bacterial production of human-inaccessible corrinoid forms may not only contribute to cobalamin deficiency development but also mask the diagnosis of cobalamin deficiency if only serum cobalamin is tested. Serum cobalamin measures only total corrinoids and is not able to distinguish between human-specific and bacteria-specific cobalamin corrinoids (Degnan et al. 2014). Overgrowth of small bowel bacteria may therefore contribute to a deceivingly normal or even elevated total cobalamin corrinoid status while obscuring human-specific cobalamin deficiency.
Small bowel bacterial overgrowth is associated with increased bacterial production of short-chain fatty acids (SCFA) such as butyrate, propionate, and acetate. Although these SCFA s may be largely anti-inflammatory at normal physiological levels (Rios-Covian et al. 2016), excess production of propionic acid (PPA) may have negative implications in SIBO, particularly in association with the neurodevelopmental disorder of autism spectrum disorder (ASD) (Frye et al. 2015). Elevated concentrations of several propionate-producing bacterial species have been isolated in stool samples of children with ASD, as have elevated levels of PPA (Finegold et al. 2010; Song et al. 2004; Wang et al. 2012). Though genetically mediated propionic acidemia does not usually elicit symptoms of ASD, cerebral infusions of PPA in the rodent model do (Al-Owain et al. 2013; McFabe et al. 2007). A multifactorial association between elevated PPA levels and the complex symptomology of autism has been evocatively formulated by McFabe et al. Pediatric antibiotic treatment and high intakes of carbohydrates and PPA additive-rich foods have been identified as possible contributing causes to both small bowel bacteria overgrowth and elevated PPA concentrations in ASD (McFabe 2012).
Thus, human-accessible cobalamin corrinoids are critically implicated in both the production and degradation of propionic acid. Methionine synthase and methylmalonyl-CoA mutase are the only two cobalamin-dependent enzymes in human metabolism. However, within the bacterial microbiome of the human gut, there are many cobalamin corrinoid-dependent enzymes, among which MCM and MetH are the most prominent (Degnan et al. 2014). In some bacteria, the conversion of PPA to succinic acid via cobalamin-specific MMA can also be a particularly useful bidirectional pathway (Fig. 2); anaerobic bacteroides can capitalize on the production of CO2 from the decarboxylation of MMA by running the pathway in reverse, thereby also increasing the pool of PPA (Fischbach and Sonnenburg 2011).
Cobalamin and PPA in Autism: Interrelationships with MMA and the Microbiome
Much recent attention has been given to a potential association between small bowel dysbiosis and elevated propionic acid in the pathophysiology of autism. However, the production of PPA as part of a cobalamin-dependent continuum involving MMA must also be taken into account, especially given the interrelatedness of human cobalamin species with the pathway substrates, products, and essential enzyme activities. MMA is emerging as an important metabolite biomarker of functional cobalamin deficiency since the bioactive corrinoid adenosylcobalamin is a rate-limiting ingredient in the MMA conversion to succinic acid. Thus, in many conditions, MMA levels may rise before serum cobalamin levels drop (Klee 2000). It is also of note that in cobalamin-deficient patients, antibiotic treatment can lower MMA levels without altering homocysteine levels, indicating that PPA produced by anaerobic gut bacteria may be a precursor to MMA in cobalamin-deficient patients (Lindenbaum et al. 1990). Equally, however, early work on organic acidemias raised the question of whether the presence of excess PPA could mask a concurrent elevation in MMA (Duran et al. 1973). Bacteria can produce PPA from non-SCFA precursors moving forward along the PPA to succinic acid pathway. However, anaerobic bacteria are also able to produce PPA from MMA moving backward along the PPA to succinic acid pathway, in order to increase CO2 production (Fischbach and Sonnenburg 2011). This latter cobalamin-independent conversion of MMA to PPA, such as in established small intestinal bacterial overgrowth (SIBO), could conceivably reduce the hallmark elevations of MMA which are diagnostically important to the confirmation of cobalamin deficiency and, at the same time, contribute to the elevations in PPA theoretically linked to the pathophysiology of autism.
Symptoms of cobalamin deficiency resemble those of autism (Agrawal and Nathani 2009), and levels of methylcobalamin, adenosylcobalamin, and methionine are lower in the autistic brain (Zhang et al. 2016). Symptoms associated with PPA administration to rats also resemble autism (McFabe 2012). However, documentation of PPA and MMA levels in human autism research is limited. One study of 58 children with autism found unexpectedly lower stool PPA than among controls (Adams et al. 2011a), while stool PPA was significantly elevated among a group of 23 children with autism compared to controls (Wang et al. 2012). In a retracted Lancet study (Wakefield et al. 1998), urinary MMA levels were elevated among 8 children with autism and GI symptomology. By contrast, no significant elevations in plasma MMA were found among a group of 55 children with ASD; however, there were also no indications of significant GI symptomology or SIBO among this group (Adams et al. 2011b). In a case report of a young child with ASD and evidence of small bowel bacterial overgrowth, there was a high-normal MMA which was nonresponsive to oral cobalamin intervention (Fitzgerald et al. 2012). However, when subcutaneous cobalamin injections were initiated, bypassing the SIBO issue, there were dramatic improvements in behavior and development. Thus, children with autism and concurrent significant GI symptoms may have higher MMA levels than children with autism and minimal GI symptoms, potentially confirming that SIBO could contribute to elevated MMA levels. Further, in SIBO, bacteria may preferentially sequester supplemental cobalamin travelling through the GI tract and limit the effectiveness of oral cobalamin treatment. The reportedly dramatic impact of subcutaneous cobalamin treatment on behavior and development in an ASD child previously nonresponsive to oral supplementation would appear to confirm the preferential sequestering of oral cobalamin by small bowel bacteria and the need to bypass the gut for effective redress of functional cobalamin deficiency during SIBO.
Cobalamin and Epigenetics: Interrelationships with Folate and Gut Microbiota
Alterations in human-accessible cobalamin availability in conditions of small bowel bacterial overgrowth will have effects on other biochemical pathways, with potential epigenetic implications. Notably, the methylation of homocysteine to methionine is both cobalamin and folate dependent and directly impacts DNA synthesis. If there is insufficient cobalamin, or if the cobalamin is oxidized during oxidative stress, methyltetrahydrofolate (MTHF) will not be converted to tetrahydrofolate (THF) for recycling, and methionine synthase production (MetH) will drop with reduced conversion of homocysteine to methionine. This reduction in methionine synthesis will then impact the methylation of DNA via the S-adenosylmethionine (SAM-e) to S-adenosylhomocysteine (SAH) pathway (Chiang et al. 1996; Waterland). Changes in DNA methylation are principal epigenetic mediators. Rodent experiments supplementing methylation cofactors before and during pregnancy have demonstrated increased DNA methylation in both early and mid-gestation (Waterland 2006; Waterland et al. 2006). The increased DNA methylation in response to gestational supplementation does not appear to be multigenerational in nature (Waterland et al. 2007).
Targeted cobalamin and folate supplementation in humans has also been shown to alter DNA, as determined by the frequency of micronucleated lymphocytes (MNL) . In both adolescent and adult males, the frequency of MNL demonstrated a significant negative relationship with serum cobalamin levels but not folate status, despite the absence of clinical cobalamin deficiency. Although the adolescents were each supplemented with 3.5 and 10 times the RDI for both folate and cobalamin, the magnitude of serum cobalamin increase was much smaller than that of the RBC folate increase. Further, the increases in folate status were not associated with the improvements in MNL frequency seen with improved cobalamin status. Overall, the greatest reduction in MNL frequency was achieved at post-supplementation cobalamin levels >300 pmol/L (Fenech et al. 1997, 1998). The differences in cobalamin and folate response to supplementation may reflect differences in barriers to absorption between dietary folate and cobalamin and underline the challenges in addressing the epigenetic consequences of cobalamin deficiency through gut-mediated fortification or supplementation.
Children with Down syndrome exhibit altered cobalamin- and folate-mediated homocysteine-to-methionine conversion, and a higher frequency of MNL , than do controls (Youness et al. 2016). Levels of micronucleated lymphocytes are also elevated among young mothers of children with Down syndrome, which may indicate a multigenerational effect (Coppede et al. 2016). Children with autism who also have altered homocysteine-methionine function may or may not have similar increased rates of micronucleated cells: among a small number of sibling pairs (6), indicators of DNA damage (including micronuclei frequency) in response to hydrogen peroxide challenge were higher among the children with autism, but failed to reach statistical significance. However, the children with autism did demonstrate higher rates of lymphoblast necrosis (Main et al. 2013).
There is a unique dance between cobalamin and folate availability (which impacts methylation reactions throughout the human body) and their related epigenetic effects. While some human gut bacteria can produce cobalamins which are inaccessible to the human host, other flora can synthesize folates which appear to be accessible to the host (Leblanc et al. 2013; Rossi et al. 2011). Rat studies have identified significantly higher fecal folate levels among Bifidobacterium models which correlate with serum folate levels and hemoglobin and mean cell volume status (Sugahara et al. 2015). The balance between bacteria variants in the gut may thus influence the balance between cobalamin and folate in human.
On a larger scale, relative ratios between dietary intakes of cobalamin and folate may also influence methylation pathways and DNA methylation. Since the late 1990s, numerous countries have elected to add synthetic folic acid to flour to reduce neural tube defects and to encourage the use of synthetic folic acid-containing supplements, particularly among women who are pregnant or of child-bearing age. Levels of total RBC folate and unmetabolized folic acid are elevated in the populations of several folate-fortified nations. Unmetabolized folic acid has also been detected in the cord blood of infants whose mothers did not receive maternal folate supplements (Obeid et al. 2010). By contrast, serum cobalamin levels are not elevated (McFarlane et al. 2011). Among Canadian women, cobalamin levels were lowest among adolescents and women of child-bearing age; lower cobalamin status was associated with higher homocysteine levels (McFarlane et al. 2011).
The biological significance of supraphysiological total folate and high levels of unmetabolized folate, particularly in the context of cobalamin- and folate-related polymorphisms, remains to be clarified. Supraphysiological folate may mask the macrocytic anemia of cobalamin deficiency, further complicating a complicated diagnosis (Wyckoff and Ganji 2007). There may also be subpopulations who are more sensitive than others to the accumulation of synthetic folate, as evidenced by differences between Canadian women with and without detectable unmetabolized folic acid levels following supplementation (Tam et al. 2012). However, an elevated ratio of folate to cobalamin has been associated with adverse neurological effects among the elderly (Morris et al. 2007) and with altered hippocampal microstructure and memory performance (Kobe et al. 2016) in adults. An increased folate-to-cobalamin ratio is also associated with embryonic delay and growth retardation in mice (Pickell et al. 2011) and with reduced neonatal growth anthropometrics in humans (Gadgil et al. 2014). Intake of synthetic folic acid supplements during pregnancy has been positively correlated with risk of autism in analyses of the Rochester (MN) Epidemiology Project and the Centers for Disease Control and Prevention pediatric dataset, respectively (Beard et al. 2011; DeSoto et al. 2012). High folate in the presence of genetic alterations to folate/homocysteine pathways may also increase the risk of Down syndrome (Coppede 2009).
Cobalamin and Brain Function: Interrelationships with PUFA, Microbiota, and PPA
The ratio of cobalamin to folate may also interact with the composition of polyunsaturated fatty acids, which are important methyl group acceptors, in the brain. Among older persons with mild cognitive impairment, cobalamin and folic acid supplementation appears to slow the rate of brain atrophy among patients with high baseline omega-3 fatty acids. However, there was no benefit from folate and cobalamin supplements among patients with lower omega-3 levels (Jerneren et al. 2015). The ratio of folic acid to cobalamin during pregnancy can have direct impacts on fetal brain docosahexaenoic acid (DHA) and arachidonic acid (AA) accretion. During pregnancy, maternal DHA levels are primary regulators of brain DHA in offspring, while fetal DHA status influences neurogenesis and neuron survival during pregnancy (Dhobale and Joshi 2012). However, in rat studies, excess folate supplementation of cobalamin-deficient mothers decreased DHA concentrations in both mothers and offspring. Omega-3 supplementation of cobalamin-deficient rats receiving high folate improved DHA levels in both mother and offspring, but AA levels were reduced. Rates of malondialdehyde (MDA), indicative of lipid peroxidation, were also elevated in B12-deficient rats and offspring receiving high folate (Roy et al. 2012). Further study identified alterations in the mRNA of brain-derived neurotrophic factors (BDNF) in offspring of rats fed with high-folate, cobalamin-deficient diets; in rats receiving prenatal supplements of DHA and eicosapentaenoic acid (EPA), abnormal mRNA of BDNF was normalized (Sable et al. 2014).
DHA levels are lower in children with autism (Brigandi et al. 2015) and attention deficit hyperactivity disorder (ADHD) (Parletta et al. 2016). DHA supplementation of children with autism failed to demonstrate significant positive effects in core behavioral symptoms (Voigt et al. 2014). However, interactions between DHA and methylation nutrients such as folate and cobalamin were not accounted for. It is possible that correction of potential folate-to-cobalamin imbalances may be required before a therapeutic response to DHA supplementation can be demonstrated in autism. Similar limitations to study protocols and therapeutic outcomes are evident in omega-3 supplementation research with ADHD (Gillies et al. 2012).
Gut microbiota can alter the impact of omega-3 and omega-6 fatty acid ratios in mice in vivo, and vice versa (Kaliannan et al. 2015; Wall et al. 2009). Mice fed an omega-6-enriched diet demonstrate elevated levels of gastrointestinal inflammatory biomarkers; this effect is not seen in genetically altered mice that intrinsically produce omega-3 fatty acids. Treatment with antibiotics eliminates the effect. Microbiome analyses indicate omega-3-producing mice have higher counts of anti-inflammatory bacterial species including bifidobacterium and lactobacillus strains; mice with high omega-6 and inflammatory biomarker levels demonstrate bacterial overgrowth with pro-inflammatory proteobacteria. Cohousing of the mouse types leads to fecal transfer of bacteria from the omega-3 mice, which increased intestinal zinc-dependent alkaline phosphatase levels and altered bacterial growth in the omega-6 mice toward anti-inflammatory species (Kaliannan et al. 2015).
In humans, fish oil supplementation in combination with breastfeeding cessation altered the gut bacterial species of infants, compared to sunflower oil supplements (Andersen et al. 2011). Among older adults, a high omega-6 diet altered gut flora and caused gut dysbiosis ; this effect was reversed by fish oil supplementation (Ghosh et al. 2013). A further mouse study found that in utero and early life omega-3 supplementation beneficially altered not only gut microbiota but also depressive, social, and cognitive behaviors during later life (Robertson et al. 2016); however, some of the effects of DHA supplementation, and the subsequently altered gastric microbiome, appear to be gender specific and stronger among males (Davis et al. 2017). Rates of autism spectrum disorder and ADHD are also higher among males (CDC).
Dictionary of Terms
Autism – a pediatric neurological disorder affecting communication, behavior, and cognition.
Cobalamin – a cobalt-based vitamin and essential enzymatic cofactor for pathways affecting DNA and RNA.
Gut dysbiosis – bacterial imbalance causing inflammation and nutrient malabsorption.
Polyunsaturated fatty acids – double-bond fatty acids important for brain function including DHA, EPA, and AA.
Propionic acid – a short-chain fatty acid produced by bacteria, fermentation, and metabolism of some amino and fatty acids.
Key Facts About Vitamin Cobalamin
Cobalamin supports normal production of white blood cells and platelets, affecting immune responses and blood clotting.
Cobalamin assists in red blood cell and hemoglobin production to carry oxygen to tissues and organs.
Cobalamin assists in myelin formation, the insulating sheath that protects nerves throughout the brain and body.
Cobalamin influences production of neurotransmitters such as dopamine and serotonin.
Cobalamin influences hormonal regulation of bone growth and strength.
Some bacteria compete for cobalamin, divert human-accessible cobalamin into human-inaccessible forms, and alter the gut environment, influencing cobalamin absorption, bioavailability, metabolism, and assessment.
Cobalamin deficiency can raise methylmalonic acid (MMA) and alter the propionic → methylmalonic → succinic acid pathway.
Small bowel bacterial overgrowth can increase propionic acid (PPA) from precursors; some bacteria reverse MMA to PPA, increasing total PPA.
Both cobalamin deficiency and PPA can cause autistic symptoms and alter brain function.
Cobalamin-folate ratios modify PPA and MMA production, alter methylation pathways and DNA, and impact pregnancy outcomes and brain function.
Cobalamin-folate ratios alter brain polyunsaturated fatty acid (PUFA) levels and response to PUFA supplementation.
Gut bacteria alter PUFA levels; PUFA supplementation can alter gut bacterial species.
Both cobalamin deficiency and PPA administration can alter brain PUFA levels, and altered brain PUFA levels are associated with autism.
PUFA supplementation can reverse the neurotoxicity of PPA.
Cobalamin supplementation can raise PUFA levels.
Cobalamin and PUFA supplementation, with controlled cobalamin-folate ratios, may epigenetically modify the PPA model of autism.
- Adams JB, Audhya T, McDonough-Means S et al (2011b) Nutritional and metabolic status of children with autism vs. neurotypical children, and the association with autism severity. Nutr Metab (London) 8(34):1–32Google Scholar
- Agrawal S, Nathani S (2009) Neuro-regression in vitamin B12 deficiency. BMJ Cas Rep 2009:bcr06.2008.0235Google Scholar
- DeSoto MC, Hitlan RT (2012) Synthetic folic acid supplementation during pregnancy may increase risk of developing autism. J Ped Biochemist 2(4):251–261Google Scholar
- Frye RE, Rose S, Slattery J et al (2015) Gastrointestinal dysfunction in autism spectrum disorder: the role of the mitochondria and the enteric microbiome. Microb Ecol Health Dis 26:2748Google Scholar
- Gillies D, Sinn JK, Lad SS et al (2012) Polyunsaturated fatty acids (PUFA) for attention deficit hyperactivity disorder (ADHD) in children and adolescents. Cochrane Database Syst Rev 7:CD007986Google Scholar
- Lewis SJ, Franco S, Young G et al (1996) Altered bowel function and duodenal bacterial overgrowth in patients treated with omeprazole. Aliment Pharmacol Ther 10(4):557–561Google Scholar
- MacFabe DF (2012) Short-chain fatty acid fermentation products of the gut microbiome: implications in autism spectrum disorders. Microb Ecol Health Dis 23:1–24Google Scholar
- Meyniel D, Petit J, Bodin F et al (1981) Vitamin B12 deficiency in chronic atrophic gastritis. 3 cases (author’s transl). Nouvelle Presse Med 10(27):2281–2284Google Scholar
- Parletta N, Niyonsenga T, Duff J (2016) Omega-3 and omega-6 polyunsaturated fatty acid levels and correlations with symptoms in children with attention deficit hyperactivity disorder, autistic spectrum disorder and typically developing controls. PLoS One 11(5):e0156432CrossRefPubMedPubMedCentralGoogle Scholar
- Rossi M, Amaretti A, Raimondi S (2011) Folate production by probiotic bacteria. Forum Nutr 3(1):118–134Google Scholar
- Youness ER, Aly HF, El-Bassyouni HT et al (2016) Micronucleus assay and pro-oxidant status of patients with known chromosomal aneuploidy. Der Pharma Chemica 8(13):158–164Google Scholar