Gluten and Schizophrenia – does it all start in the womb?

Credit: Mary-Claire King, Ph.D., University of Washington [original paper]

Credit: Mary-Claire King, Ph.D., University of Washington [original paper]

  • Research is now identifying how schizophrenia starts in the womb
  • Maternal infections and dietary antigens such as gluten are implicated
  • A key immune molecule C1q has been identified which links schizophrenia, gluten and neuronal development in the unborn child
  • Evidence for dietary factors associated with healthy brain development are considered

Schizophrenia affects one in 100 people at some point in their lifetime. The causes are not understood, but what is known is that it is associated with abnormal networks  of neurones in the brain, specifically, aberrant connections within and between different brain regions. 25 years ago the neurodevelopmental hypothesis proposed that these problems begin in the womb, with disruption to the development of the foetal prefrontal cortex leading to schizophrenia.

The healthy adult brain depends on complex, synchronised neuronal migration during foetal development, involving highly orchestrated cell proliferation, signalling and transcription. Much of this happens during a key period of neuronal development in the foetus. During this time multiple network genes are switched on and many hormones and signalling molecules come into play. Dramatic neurone proliferation, synapse connections and pruning take place at an accelerated rate throughout this all-important phase. Disruption to any one of the multiple pathways involved could have profound, life-long effects on brain health.

Bars across the top indicate the period of emergence of symptoms and diagnosis. In normal subjects, spine numbers increase before and after birth; spines are selectively eliminated during childhood and adolescence to adult levels. In ASD, exaggerated spine formation or incomplete pruning may occur in childhood leading to increased spine numbers. In schizophrenia, exaggerated spine pruning during late childhood or adolescence may lead to the emergence of symptoms during these periods. In Alzheimer's disease, spines are rapidly lost in late adulthood, suggesting perturbed spine maintenance mechanisms that may underlie cognitive decline. [Nature, 2011]

Bars across the top indicate the period of emergence of symptoms and diagnosis. In normal subjects, spine numbers increase before and after birth; spines are selectively eliminated during childhood and adolescence to adult levels. In autism spectrum disorder ASD, exaggerated spine formation or incomplete pruning may occur in childhood leading to increased spine numbers. In schizophrenia (SZ), exaggerated spine pruning during late childhood or adolescence may lead to the emergence of symptoms during these periods. In Alzheimer’s disease (AD), spines are rapidly lost in late adulthood, suggesting perturbed spine maintenance mechanisms that may underlie cognitive decline. Nature, 2011

Gene variants associated with schizophrenia and autism, unsurprisingly, turn out to have roles in processes such as axon guidance, neurone mobility, synaptic function, and chromosomal remodelling. Interestingly, many of the key genes and signals involved in the initial burst of cortex networking are silent throughout childhood, only to be activated again during early adulthood when the brain undergoes extensive pruning and rewiring. This is typically the time when Schizophrenia first manifests.

Schizophrenia and Maternal Infections

One environmental factor associated with increased schizophrenia and autism risk is maternal infections during pregnancy. Intrauterine and vaginal infections, for example, increase the risk of preterm delivery and brain damage. For example, the risk is increased after prenatal infection of the mother by viruses such as influenza, rubella, measles, and polio, as well as some bacterial pathogens. Epidemiological data indicates that the risk is independent of the particular pathogen, suggesting that the immune response itself may be the causative factor. The hypothesis is that foetal pro-inflammatory cytokines and associated signalling molecules evoked as part of the immune response interfere with the neuronal development. This is based on the fact that many of these immune molecules play a crucial role in normal neural network formation in the foetus, and a sudden immune mediated increase in their activity is likely to have a consequential affect on the neuronal development.

One problem is explaining how two distinct and separate neurological conditions – schizophrenia and autism – can be triggered by a shared risk factor – prenatal infections. Urs Meyer and colleagues at the Laboratory of Behavioural Neurobiology, Swiss Federal Institute of Technology, propose that the outcome is determined primarily by the success of the maternal and/or foetal immune system in dealing with the acute inflammation of the infection [2].

They suggest that failure of the immune system to control inflammation following the acute infection may lead to chronic inflammation persisting into childhood with the kind of neuronal damage associated with autism. Successful suppression, on the other hand, may leave the infant with immune abnormalities (latent inflammation) and lead to brain development associated with schizophrenia. This model fits with the observation that in autism there are many pro-inflammatory markers, whereas schizophrenic patients have enhanced anti-inflammatory and immunosuppressive pathways.


So where does gluten fit in?

There have been many studies over the years that have found raised antibodies to gluten and bovine casein in schizophrenia, bipolar disorder and autism (good summary here [3]). For such food proteins to enter the blood and stimulate antibody production a ‘leaky gut’ is required – a commonly identified condition among sufferes of these conditions.

A superb line of research carried out by Emily Severance and colleagues at Johns Hopkins University has thrown considerable light on gluten and psychosis over the last few years. For example, in 2013 they identified raised markers of both leaky gut and antibodies to gluten (and bovine casein) among bipolar sufferers. [4]

Whilst investigating the link between schizophrenia, maternal infection and gluten, they demonstrated that gluten exposure and concurrent infection with toxoplasma gondii led to raised gluten antibodies in mice and their subsequent offspring. [5]

An emerging culprit in this research is complement protein C1q. The role of this molecule is primarily to clear antibody-antigen complexes. So, for example, when the immune system finds gluten in the blood, gluten-antibodies attach to the gluten forming a gluten antibody-antigen complex. C1q then binds to this complex as the first stage in clearing it from the system. It then stimulates phagocytosis, local inflammation and clearing of damaged cells.  What makes C1q especially interesting in respect of neurological disorders is that it has a secondary role in the brain.

Alongside its immune function, C1q is also involved in neuronal pruning in the foetus (and possibly during adult neuronal repair). In 2007 Beth Stevens, at the F.M. Kirby Center for Neurobiology at Children’s Hospital Boston, discovered that astrocytes (brain cells that outnumber neurones) stimulate neurones to express C1q during development. C1q seems to mark redundant synapses, allowing for their selective destruction [6].

The hypothesis is that if C1q is raised in response to an excess of dietary antigens (e.g. gluten) this may interfere with its function in neuronal remodelling. Severence’s team found “significantly elevated levels of complement C1q activation and large effect sizes of C1q antibody levels in individuals with schizophrenia compared to controls.” and “we established that milk casein and wheat gluten comprise the antigen component of C1q-based immune complexes in a prominent portion of individuals” [3]. In other papers they linked raised C1q to inflammatory gut processes and infections.

Which brings us to the most recent paper by Severence’s team, which pulls many of these ideas together [7] (unfortunately this is not available free, although I have access to the full text myself). In this well designed study 55 mothers who had offspring that developed schizophrenia were compared to 55 control mothers matched for age, and child’s birth date, delivery hospital and race. These mothers had taken part in the National Collaborative Perinatal Project (1959-1966), so prenatal blood samples were available enabling comparisons of C1q levels. Additionally, antibodies to food allergens and infectious agents were screened.


Raised gluten antibodies will stimulate C1q production to clear antigen-antibody pairs. In the foetus this may affect C1q’s function in synaptic clearance. This dual role provides a hypothetical link between gluten and schizophrenia.

C1q, gluten and schizophrenia: results

C1q was significantly raised among case mothers compared to controls. Mothers within the top 25% of C1q levels had three times the incidence, and in the top 5% six times the incidence of schizophrenia among their offspring. (Increased risk was also associated with antibodies to Herpes simplex 2, Influenza B and T Gondii.)

Among case mothers only was there a strong and significant correlation between C1q levels and gluten antibodies, suggesting that gluten may be driving the raised C1q among genetically susceptible mothers.

In conclusion then, pregnant mothers probably should avoid raising their C1q, especially if there is a family history of schizophrenia. That means avoiding infections and gluten. Of these two, gluten is the risk factor most in the mother’s control, and considering that grains contain no essential nutrients, there can be no harm in avoiding them altogether, especially if she replaces grains with foods containing nutrients needed for proper brain development.

Maternal diet, brain development and schizophrenia

When considering neurological development of the foetus, several nutrients are recognised as pivotal for proper brain growth: iodine – severe deficiency of which causes cretinism; folate – where deficiency is associated with neural tube defects; essential ω3-fatty acids are crucial for brain development with DHA making up 10% of the dry weight of the human brain – all of which must be supplied from the mother’s limited body stores and diet. Vitamin A is involved in neuronal differentiation, maturation, and migration whilst vitamin D play a significant role in transcription of networking genes. Low vitamin D levels have been associated with multiple sclerosis. Iron plays a central role in ensuring sufficient oxygenation: the rapidly developing hippocampus is believed to be especially vulnerable to hypoxia. Iron is also essential for myelination, affecting neural connectivity.

The fact that a nutrient is associated with healthy brain development in normal offspring, however, does not mean that it necessarily alters the risk of schizophrenia. Evidence directly linking maternal nutrition to schizophrenia is somewhat sparse due to the temporal delay of two decades between pregnancy and the emergence of schizophrenia in early adulthood. That said, there have been a number of such studies…

Vitamin A
Alan Brown’s group at Columbia University found that low retinol (vitamin A) during the second trimester was associated with a 200% increase incidence of schizophrenia. [8] The easiest way to ensure adequate retinol status is by consuming liver once per week. Inexcusably, liver is often advised against during pregnancy! (See our article on vitamin A from animal sources here).

Vitamin D 
The observation that there are more schizophrenic births in the winter months (approx. 20% more) and in urban areas suggests a role for vitamin D, although a pilot study in 2003 found no correlation between maternal vitamin D status and schizophrenia. [9] This study was, however, observational, with no attempt to increase vitamin D status, so could not detect the effects of supplementation rates.

Vitamin D supplementation in pregnancy is currently recommended by the NHS (400IU/day) to ensure proper muscular-skeletal development, and avoid rickets – cases of which have returned in the UK in the last decade.

According to Bruce Hollis, Ph.D. director of paediatric nutritional sciences at the Medical University of South Carolina, in Charleston, standard supplementation levels are far too low. In his study 500 women who were at least 12 weeks pregnant took either 400, 2,000, or 4,000 IU of vitamin D per day. The women in the 4,000 IU group were least likely to go into labour early, give birth prematurely, or develop infections. [10]

“Pregnant women need to take 4,000 IU of vitamin D a day… We didn’t see a single adverse effect. It was absolutely safe, and we saw a lot of improved outcomes. The risk of preterm labor was vastly decreased and so was the risk of other complications of pregnancy.”

The authors point out that their study does not show if it is safe to take such high levels earlier in pregnancy than the second trimester.

The fact that 4000 IU reduced infection rates would suggest a protective effect for schizophrenia.

Folate (aka vitamin B9) has been hypothesised as a risk factor for schizophrenia, especially as during the Dutch famine of 1944, there was a peak in both neural tube defects and schizophrenia. However, a 2010 study found no connection between maternal folate status and schizophrenia [11]. Folate supplementation throughout pregnancy – and even from several months before – is now a standard recommendation which reduces incidence of neural tube defects by 70%. [12]

Recent understanding of the human genome led to the discovery of the methylenetetrahydrofolate reductase gene variant (MTHFR C677T*), which leads to raised homocysteine – an independent identified risk factor for several neurological conditions including depression and dementia. People with this genotype have a need for much higher folate intake to prevent raised homocysteine. A paper published last month found a significant association between the MTHFR C677T gene and schizophrenia [13]. Furthermore, raised maternal homocysteine levels in the third trimester have been associated with increased risk of schizophrenia in offspring.

Dietary folate is present in liver and vegetables such as kale, Brussels sprout, spinach, and beans, and in lower amounts in fruit and egg yolks. Inexcusably, most lists of folate-rich foods ignore liver, despite it having the highest concentration. Another fact they rarely mention is that folate from animal sources withstands cooking significantly better than from vegetable sources which can lose 40% of their folate when cooked.

* It is notable that the MTHFR C677T genetic test is the only screening test approved by the FDA to come out of the Human Genome project as it can be simply treated with increased folate intake.

An often overlooked element, necessary for correct thyroid function, and essential for proper brain development. Frank deficiency causes seriously stunted physical and mental growth (cretinism). Perhaps surprisingly, the UK population is mildly iodine deficient as unlike most of Europe and the USA there is no mandatory salt iodisation programme. Sarah Bath at the University of Surrey recently showed that mothers with the lowest iodine status were at heightened risk of having children with lower verbal IQ, reading accuracy and comprehension. [14]

Iodine rich foods include seaweed, white fish, iodised table salt and milk. Organic milk generally has lower levels than non-organic milk.

In 2008 Alan Brown’s team found that schizophrenia incidence was associated with maternal iron status. Mothers with haemoglobin concentrations of 10.0 g/dL were 270% more likely to have offspring with schizophrenia than mothers with 12.0 g/dL. [15] Avoiding anemia also requires adequate intakes of B12 which can be achieved through consumption of liver, meat, eggs and dairy products.

The best sources of dietary iron are liver and red meat.

Omega 3 fatty acids; fish oils and fish consumption
Consumption of fish during pregnancy has been linked to increased IQ in offspring, but increased methyl mercury intake may offset this benefit. Results are mixed and the benefits and risks appear modest, but eating oily fish and avoiding tuna appears to be the best option [16,17]. Much of these benefits are laid at the door of improved maternal DHA status, however, one study found that higher maternal DHA levels were linked to an increased risk of schizophrenia. [18] Brown_CrabLassek and Gaulin, on the other hand, found that in 4000 American children, ω3 intake was positively related to cognitive test scores, but ω6 was inversely related. [19]

One of the highest food sources of DHA and other ω3 fatty acids is brown crab meat. 100g portion provides 5 x RDA of omega 3 PUFAs. Other good sources include wild Alaskan salmon, mackerel and sardines. Aiming for an ideal ω3:ω6 ratio of 1:1 necessitates complete avoidance of all vegetable oils (peanut, corn, sunflower and safflower) as ω6 intake from these sources dominates the western diet.


A gluten-free diet rich in meat and vegetables, liver, eggs and fruits, alongside sea foods, can help maintain levels of vitamin A, B12, iron, folate and omega 3 fatty acids – many of the key nutrients needed for healthy brain development throughout life.

In conclusion

All of the above nutritional considerations can be met by a paleo-style diet. Based around natural meats, vegetables, seafoods and fruits it can provide all of the nutrients needed for proper brain development, whilst avoiding gluten (and possibly casein) which are associated with increased risk of schizophrenia, autism and bipolar disorder.

Although the exact mechanisms linking gluten to neurological problems remains to be fully elucidated, the fact that grains in general are a wholly unnecessary dietary element means they can be safely avoided.

For mothers with a family history of schizophrenia or other neurological diseases a more aggressive regimen including nutrient profiling, food antibody testing and specific supplementation may be advisable.


  1. Gulsuner et al, Spatial and Temporal Mapping of De Novo Mutations in Schizophrenia to a Fetal Prefrontal Cortical Network, Cell, Aug 2013
  2. Meyer, et al, Schizophrenia and Autism: Both Shared and Disorder-Specific Pathogenesis Via Perinatal Inflammation? Pediatric Research, Nov 2010
  3. Emily Severance et al, Complement C1q formation of immune complexes with milk caseins and wheat glutens in schizophrenia, Neurobiology of Disease, Dec 2012
  4. Emily Severance et al, Seroreactive marker for inflammatory bowel disease and associations with antibodies to dietary proteins in bipolar disorder, Bipolar Disorders, 2013
  5. Emily Severance et al, Anti-gluten immune response following Toxoplasma gondii infection in mice. PlosOne, 2012
  6. Parizad Bilimoria, Immune molecules fine-tune brain circuits, Boston Children’s Hospital website
  7. Emily Severance et al, Maternal complement C1q and increased odds for psychosis in adult offspring, Schizophrenia research, Oct 2014
  8. Yuan Yuan Bao et al, Low maternal retinol as a risk factor for schizophrenia in adult offspring, Schizophrenia review, May 2013
  9. Mc Grath J et al, Low maternal vitamin D as a risk factor for schizophrenia: a pilot study using banked sera. Sep 2003
  10. Hollis BW et al, Vitamin D supplementation during pregnancy: double-blind, randomized clinical trial of safety and effectiveness. Journal of bone and mineral research, Oct 2011
  11. Muntjewerff JW et al, Effects of season of birth and a common MTHFR gene variant on the risk of schizophrenia. European neuropsychopharmacology, 2011
  12. Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. MRC Vitamin Study Research Group. The Lancet, 1991
  13. Akira Nishi, Meta-analyses of Blood Homocysteine Levels for Gender and Genetic Association Studies of the MTHFR C677T Polymorphism in Schizophrenia, Schizophrenia bulletin, Sep 2014
  14. Bath SC et al, Effect of inadequate iodine status in UK pregnant women on cognitive outcomes in their children: results from the Avon Longitudinal Study of Parents and Children (ALSPAC). The Lancet, Jul 2013
  15. Insel BJ et al, Maternal iron deficiency and the risk of schizophrenia in offspring. Archives of General psychiatry, Oct 2008
  16. Leino et al, Effects of docosahexaenoic acid and methylmercury on child’s brain development due to consumption of fish by Finnish mother during pregnancy: a probabilistic modeling approach. Food and chemical toxicology, 2011
  17. Muntjewerff JW et al, Quantitative risk-benefit analysis of fish consumption for women of child-bearing age in Hong Kong. Food additives and contaminants, 2014
  18. Harpen KN et al, Maternal serum docosahexaenoic acid and schizophrenia spectrum disorders in adult offspring. Schizophrenia research, 2011
  19. Lassek & Gaulin, Sex Differences in the Relationship of Dietary Fatty Acids to Cognitive Measures in American Children, Frontiers of evolutionary neuroscience, Nov 2011

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s