Disordered Methylation in Autism
Home » Understanding Autism » Disordered Methylation in Autism facebook

Video:  Mothers, Methylation and Mitochondria, S. Jill James, Ph.D.

Effects of Mercury on Methionine Synthase: Implications for Disordered Methylation in Autism

Richard Deth, PhD and Mostafa Waly, PhD
Northeastern University, Boston, MA 02115

Methionine synthase carries out the vitamin B12-dependent methylation of homocysteine, using a methyl group from 5-methyltetrahydrofolate (5-methylTHF). In doing so it provides a crucial link between two important metabolic systems, the single-carbon folate pathway and the methionine cycle (Fig. 1, right). Recently our laboratory discovered that methionine synthase is also required for dopamine-stimulated methylation of membrane phospholipids, a unique signaling activity of the D4 dopamine receptor (1-3). In the latter case, a homocysteine residue in the D4 receptor, generated by donation of a methyl group to the phosphatidylethanolamine (PE), is re-methylated to methionine, with 5-methylTHF again serving as a co-factor (Fig. 1, left). Thus methionine synthase also links dopaminergic neurotransmission to metabolism.

Figure 1. Pathways of single-carbon metabolism. The folate pathway (middle) supplies formate-derived single-carbon groups for purine and thymidine synthesis and methyl groups (as 5-methylTHF) to methionine synthase (Met Syn). The methionine cycle (lower right) provides S-adenosylmethionine (SAM), the methyl donor for many methylation reactions. MET313 in the D4 receptor also serves as a donor of methyl groups for methylation of the phospholipid phosphatidylethanolamine (PE).

Sufficient methylation of homocysteine is essential not only to supplement the diet-derived supply of methionine, but also to maintain a low level of homocysteine and its precursor, S-adenosylmethionine (SAH), which otherwise inhibits methylation reactions by competing with the methyl donor S-adenosylmethionine (SAM). Increased methionine synthase activity can therefore promote methylation by lowering SAH, while impaired activity will impede methylation. In addition, THF released by methionine synthase is needed for other folate-dependent reactions, avoiding a “methyl-folate trap”.

D4 receptor-mediated phospholipid methylation (PLM) can be very robust, up to 50 methylations/receptor/sec, increasing the spacing between phospholipid headgroups and altering the fluid properties of the membrane in the region surrounding the receptor. The activity of membrane proteins located near the D4 receptor can be modulated by PLM and this “solid-state signaling” mechanism has been implicated in the molecular mechanism of attention (4). Proline-rich segments present in the cytoplasmic portion of the receptor in all species allow it to serve as a docking site for signaling proteins that become targets for PLM-based modulation. In humans and other primates, the D4 receptor possesses anywhere from 2 to 11 additional proline-rich repeat segments (Fig. 2), and a higher number of repeats (i.e. seven) brings an increased risk of attention-deficit hyperactivity disorder (ADHD) (5). Thus methionine synthase activity is important for normal attention while a decrease in its activity may contribute to ADHD.

Figure 2. Structural features of the D4 dopamine receptor. Methionine 313 serves as donor of methyl groups to the headgroups of surrounding phospholipids. Proline-rich segments allow other proteins to bind to the D4 receptor.
Methionine Synthase is Regulated by IGF-1 and Dopamine

Recently our laboratory has showed that the enzymatic activity of methionine synthase in human neuroblastoma cells is increased by stimulation of either D4 dopamine receptors or insulin-like growth factor-1 (IGF-1) receptors (6). In intact cells, this stimulation is evident as an increase in the rate of folate-dependent phospholipid methylation (Fig. 3), while pretreatment of cells with IGF-1 or dopamine or their combination increased the enzyme activity of methionine synthase by more than five-fold (Table 1).


Figure 3. Dose-dependent stimulation of folate-dependent phospholipid methylation (PLM) by IGF-I in SY5Y neuroblastoma cells.

Further investigation revealed that stimulation of methionine synthase by dopamine and IGF-1 involves activation of the PI3-kinase signaling pathway. Via different mechanisms, D4 and IGF-1 receptor activation leads to increased phosphorylation of plasma membrane inositol phospholipids by PI3-kinase. Numerous studies have shown that activation of this pathway by IGF-1 promotes cell survival while activation by other growth factors (e.g. nerve growth factor) leads to cellular differentiation. Moreover, interference with the methionine cycle blocks the ability of nerve growth factor to induce differentiation (7). Inhibition of PI3-kinase not only blocked stimulation of methionine synthase by dopamine and IGF-1, but also reduced enzyme activity to zero (Table 1), indicating an absolute dependence on PI3-kinase in the human neuroblastoma cells we used. This is the first report that extracellular signaling molecules can regulate methionine synthase.

By increasing methionine synthase activity, IGF and dopamine will decrease the levels of both homocysteine and SAH, increasing the SAM to SAH ratio and thereby promoting methylation reactions. Methylation of DNA leads to the formation of nucleosomes (stable complexes of DNA and histones), causing gene silencing. We measured global DNA methylation after a 6-hour treatment with IGF-1 or dopamine and found increases of 101% and 71% respectively, which were blocked by an inhibitor of PI3-kinase (6). Furthermore, PI3-kinase inhibition decreased methylation of the cyclin D2 gene, and increased its transcription. Together these observations indicate that PI3-kinase-dependent activation of methionine synthase provides a mechanism by which IGF-1 and dopamine can regulate gene expression via changes in DNA methylation.

Neurodevelopmental Toxins Inhibit Methionine Synthase

Environmental exposure to heavy metals, such as lead and mercury, causes neurotoxicity and leads to developmental disorders. It has been proposed that the recent dramatic rise in the incidence of autism is linked to the increased number of required vaccinations containing the ethylmercury derivative thimerosal (8,9). Since developmental disorders such as Rett syndrome and fragile-X syndrome include a crucial role for DNA methylation, we investigated the effects of various metal ions on methionine synthase activity and folate-dependent PLM.

As shown in Table 1, treatment of neuroblastoma cells with mercury (1 mM) or thimerosal (10 nM) for 60 min caused a complete loss of measurable methionine synthase activity, while lead significantly reduced enzyme activity. Ethanol, a well-recognized neurodevelopmental toxin, also eliminated activity at a concentration of 0.1%. Folate-dependent PLM studies (Fig. 4) showed that mercury and lead produced dose-dependent inhibition with a threshold near 1 nM, while thimerosal was at least 10-fold more potent. It is notable that a single dose of thimerosal-containing vaccine produces blood levels between 10 and 100 nM (10). These results clearly demonstrate the ability of neurodevelopmental toxins and thimerosal to inhibit PI3-kinase-dependent methionine synthase activity. Additional studies indicated that heavy metal-induced inhibition is likely caused by competition with Cu2+, which is required for PI3-kinase activity (6). In contrast, ethanol acts by interfering with IGF-1 receptor activation.

Table 1: Table 1 Effect of IGF-1, dopamine and neurodevelopmental toxins on methionine synthase activity. IGF-1 and dopamine increase activity by 2.3- and 2.8-fold respectively. The PI3-kinase inhibitor wortmannin reduces activity to undetectable levels and blocks stimulation.The neurodevelopmental toxins ethanol, mercury, lead and thimerosal either inhibit or eliminate methionine synthase activity.







Figure 4. Inhibition of folate-dependent PLM by mercury, lead and thimerosal. Mercury and lead exhibit a threshold near 1 nM for inhibition of IGF-1-stimulated PLM (Left panel). Thimerosal inhibits both basal PLM and stimulation by either IGF-1 or dopamine at concentrations of 0.1 nM and higher (Right panel).

Implications for Autism

It has been proposed that the recent dramatic increase in the incidence of autism is due to neurodevelopmental effects of thimerosal, associated with an increase in the number of required vaccines containing this preservative (8,9). However, this proposal has met with considerable skepticism, since there was no evidence that thimerosal could produce adverse effects on a relevant biochemical process at the concentrations produced by vaccination. A single vaccination produces blood levels between 10 and 100 nM (10). Our findings clearly demonstrate that thimerosal inhibits PI3-kinase-dependent methionine synthase at concentrations well below these levels, raising the possibility that this inhibition might contribute to the pathology of autism.

Several important questions arise: 1. How might lower methionine synthase activity account for the symptoms of autism? 2. What are the risk factor(s) for developing autism?

As noted above, methionine synthase has two substrates: the homocysteine state of the D4 dopamine receptor and homocysteine itself. Reduced activity will therefore decrease D4 receptor-mediated PLM and increase homocysteine levels. Since D4 receptor-mediated PLM appears to be important for the molecular mechanism of attention (4), its impairment could lead to a reduced capacity for attention, which is a primary symptom of autism. Indeed, autism shares several features with ADHD, including a 3-4-fold higher prevalence in males vs. females and both conditions have shown a markedly higher incidence over the past several decades. Reduced activity of methionine synthase, caused by exposure to heavy metals and/or thimerosal, could therefore impair the molecular mechanism of attention, leading to symptoms of autism. ADHD may represent a milder form of autism, associated with moderate inhibition of D4 receptor-mediated PLM.

Since SAH hydrolase (Ado HCYase in Fig. 1) is reversible, a failure of methionine synthase to efficiently convert homocysteine to methionine will lead to increased formation of SAH, to an extent dependent upon the prevailing concentration of adenosine. SAH inhibits methylation reactions, and decreased methionine synthase activity could therefore reduce DNA methylation, resulting in altered patterns of gene expression and impaired development. More specifically, heavy metals and thimerosal may interfere with the ability of growth factors like IGF-1 to promote development by impairing their control over methionine synthase.

The risk of developing autism in response to heavy metal or thimerosal exposure may depend upon genetically-transmitted risk factors that interact with methylation events. For example, previous studies showed that adenosine deaminase (ADA) activity is reduced in autistic individuals (11), associated with increased prevalence of a polymorphism in the ADA gene that reduces enzyme activity (12). As illustrated in Fig. 1, reduced ADA activity will cause elevated adenosine levels that will synergize with impaired methionine synthase activity to produce higher levels of SAH, yielding greater inhibition of methylation reactions. Increased synthesis of adenosine due to elevated 5’-nucleotidase activity has also been reported in autism (13). Mutations in the adenosylsuccinate lyase (ASL) gene are a rare cause of autism (14). These mutations divert single-carbon groups toward de novo purine synthesis and limit the availability of 5-methylTHF. Moreover, increased purine synthesis is common in autism (15). Lower availability of 5-methylTHF will synergize with the inhibitory effects of metals and thimerosal. These examples serve to illustrate how genetic and metabolic abnormalities can predispose to autism. Any impairment in the ability to excrete or detoxify heavy metals will also impose a further increased risk (16).


Our studies provide new insights into the control of methylation reactions by dopamine and by growth factors that increase PI3-kinase. By increasing methionine synthase activity and accelerating the conversion of homocysteine to methionine, they can lower SAH levels and promote methylation reactions. Neurodevelopmental toxins and thimerosal interfere with PI3-kinase-dependent methionine synthase, resulting in impaired methylation, including DNA methylation that is essential for normal development. D4 receptor-dependent PLM is an essential component of the molecular mechanism of attention, and reduced methionine synthase activity will therefore lead to impairments in attention and in attention-related learning. ADHD may reflect a milder degree of impairment in these same mechanisms. Since thimerosal has largely, but not completely, been eliminated from vaccines in the U.S., it will be of particular interest to observe whether the incidence of autism decreases during the next 3-5 years. Finally, we hope that these findings may point the way toward the discovery of new therapeutic approaches for the treatment of autism as well as new diagnostic tests that could identify individuals at high risk of developing autism in response to thimerosal or heavy metal exposure.


1. Sharma, A., Kramer, M.L., Wick, P.F., Liu, D., Chari, S., Shim, S., Tan, W., Ouellette, D., Nagata, M., DuRand, C.J. et al. (1999) Mol. Psychiatry 4, 235-46.
2. Zhao, R., Chen, Y., Tan, W., Waly, M., Sharma, A., Stover, P., Rosowsky, A., Malewicz, B., & Deth, R.C. (2001) J. Neurochem. 78, 788-96.
3. Sharma, A., Waly, M., & Deth, R.C. (2001) Eur. J. Pharmacol. 427, 83-90.
4. LaHoste, G.J., Swanson, J.M., Wigal, S.B., Glabe, C., Wigal, T., King, N., & Kennedy, J.L. (1996) Mol. Psychiatry 1, 121-4.
5. Deth, R.C. (2003) in Molecular Origins of Human Attention, (Kluwer Academic Publishers, Boston).
6. Waly, M., Olteanu, H., Banerjee, R., Choi, S.-W., Mason, J., Parker, B., Sukumar, S., Shim, S., Sharma, A., Benzecry, J., Power-Charnitsky, V.-A., & Deth, R.C. (Submitted) Molec. Psychiatry.
7. Cimato, T.R., Ettinger, M.J., Zhou, X.& Aletta, J.M. (1997) J. Cell Biol. 138,1089-103.
8. Bernard, S., Enayati, A., Roger, H., Binstock, T., & Redwood, L. (2002) Mol. Psychiatry 7, S42-3.
9. Geier, M.R. & Geier, D.A. (2003) J. Am. Phys. Surg. 8, 6-11.
10. Stajich, G.V., Lopez, G.P., Harry, S.W., & Sexson, W.R. (2000) J. Pediatr. 136, 679-81.
11. Stubbs, G., Litt, M., Lis, E., Jackson, R., Voth, W., Lindberg, A., Litt, R. (1982) J. Am. Acad. Child Psychiatry 21, 71-4.
12. Persico, A.M., Militerni, R., Bravaccio, C., Schneider, C., Melmed, R., Trillo, S., Montecchi, F., Palermo, M.T., Pascucci, T., Puglisi-Allegra, S. et al. (2002) Am. J. Med. Genet. 96, 784-90.
13. Page, T., Yu, A., Fontanesi, J., Nyhan, W.L. (1997) Proc. Natl. Acad. Sci. U S A 94, 11601-6.
14. Stone, R.L., Aimi, J., Barshop, B.A., Jaeken, J., Van den Berghe, G., Zalkin, H., & Dixon, J.E. (1992) Nat. Genet. 1, 59-63.
15. Page, T., & Coleman, M. (1998) Adv. Exp. Med. Biol. 431, 793-6.
16. Blaxill, M. F. & Haley, B.E. (2003) Int. J. Toxicol. 22,