COMMENTARIES

Maternal Famine, De Novo Mutations, and Schizophrenia Jack M. McClellan, MD Ezra Susser, MD, DrPH Mary-Claire King, PhD

S

CHIZOPHRENIA IS A DEBILITATING NEUROPSYCHIATRIC

disorder that likely stems from multiple genetic and environmental factors.1 Identifying molecular mechanisms underlying schizophrenia offers the promise of improved treatment and prevention strategies. Finding culprit mutations and the genes that harbor them is therefore one of the great challenges of human genomics. Studying populations who survived in utero exposure to maternal starvation may reveal clues regarding the genetic bases of schizophrenia. For example, epidemiological investigations of 2 famines in the 20th century—the Naziinduced 1944-1945 Dutch Hunger Winter2 and the Chinese famine of 1959-1961 following the failure of the Great Leap Forward3—demonstrated an increased risk for schizophrenia among offspring conceived in famine conditions. A possible molecular basis for this risk may be the occurrence of new mutations in genes critical for brain development. Furthermore, folate deficiency, which could occur in famine, may be a mediator of this risk by impairing capacity for DNA repair. Once identified, disease-causing de novo mutations in exposed individuals may be informative for the larger population of individuals with schizophrenia. A gene that harbors one disease-associated mutation likely harbors other different mutations associated with the illness in other cases. Famine and Schizophrenia In October 1944, a Nazi blockade of the western Netherlands precipitated a famine that became increasingly severe until liberation in May 1945.4 In the 2 to 3 months before liberation, the daily food ration fell to its nadir (⬍500 calories in April 1945), supplemental food was scarce, and the population was nutritionally depleted. Mortality more than doubled and fertility (reflected in birth rates 9 months later) was less than half that of the previous year. In a series of epidemiological studies, the health outcomes of a birth cohort conceived during the height of the famine were compared with the health outcomes of unexposed persons born in the same cities but in gestation before or after the fam582 JAMA, August 2, 2006—Vol 296, No. 5 (Reprinted)

ine. This birth cohort experienced increased risk of congenital anomalies of the central nervous system, including neural tube defects, increased risk of schizophreniaspectrum personality disorders diagnosed in 18-year-old males at military induction, and increased risk of schizophrenia in adulthood (relative risk [RR], 2.0; 95% confidence interval [CI], 1.2-3.4; P⬍.01).2,5,6 Epidemiological analysis of the Chinese famine of 19591961 revealed remarkably similar effects on the risk of schizophrenia.3 By some estimates, the massive famine that followed the Great Leap Forward caused 30 to 40 million deaths.7 Anhui province was one of the most affected. Birth rates in the Wuhu region of Anhui province in 1960-1961 were less than one third the average birth rate for 19561959. As adults, the 1960-1961 birth cohorts had higher rates of schizophrenia than persons conceived before or after the famine period (for those adults born in 1960: RR, 2.30; 95% CI, 1.99-2.65; and for those adults born in 1961: RR, 1.93; 95% CI, 1.68-2.23). These concordant results from studies of the Dutch and Chinese famines suggest that maternal starvation in the early months of pregnancy (possibly as early as the time of conception) is associated with an increased risk of schizophrenia in offspring. The Dutch cohort exposed prenatally to famine had higher rates of both neural tube defects and schizophrenia. A shared risk factor for both outcomes may be prenatal folate deficiency, which is related to neural tube defects and perhaps also to schizophrenia.5,6 It is possible that maternal folate levels influence risk of schizophrenia in offspring, and genotypes in pathways related to folate or homocysteine metabolism may modify such influences.8,9 However, there may be an additional mechanism. Prenatal folate deficiency may lead to vulnerability to de novo mutation, either germline or somatic in the developing fetus. Such mutations could occur in any gene, and many would be lethal. Those occurring in genes critical to brain development could increase the risk of schizophrenia late in life. Author Affiliations: Department of Psychiatry, University of Washington, Seattle (Dr McClellan); Department of Epidemiology, Mailman School of Public Health, Columbia University, and New York State Psychiatric Institute, New York (Dr Susser); and Departments of Genome Sciences and Medicine (Medical Genetics), University of Washington, Seattle (Dr King). Corresponding Author: Jack M. McClellan, MD, Department of Psychiatry, University of Washington, Box 356560, Seattle, WA 98195 ([email protected] .edu).

©2006 American Medical Association. All rights reserved.

Downloaded from www.jama.com at N/A, on August 5, 2006

COMMENTARIES

Folate, De Novo Mutation, and Schizophrenia Folate is a key component of DNA synthesis, methylation, and repair.10,11 Deficiencies in folate and related pathways have been implicated in birth defects, vascular disorders, and cancer.11 Folate deficiency can lead to chromosomal instability and aberrations of DNA repair, and thereby to increased rates of mutation.12 New mutations or new epigenetic effects that disrupt genes related to neurodevelopment could lead to schizophrenia. In the past, de novo mutations were assumed to be too rare to explain a major proportion of schizophrenia, which may affect approximately 1% of a population. However, new information regarding the human genome suggests otherwise. Germline mutations are more common than previously thought.13 The intricacies of the human genome, with complex gene regulation, splicing, and epigenetic processes, result in an enormous number of different potential mechanisms by which mutations and epigenetic effects may disrupt gene functioning.14 It is possible that new mutations may be more frequent in persons exposed prenatally to famine, and that some of these mutations may lead to schizophrenia in later life. Several features of schizophrenia are consistent with a role for such de novo individually rare mutations. First, schizophrenia is familial, in that close relatives of patients with schizophrenia are at increased risk of the illness. However, most patients have no close relatives with schizophrenia.15 This pattern is consistent with severe de novo mutations predisposing to schizophrenia, with different mutations arising in different families, and generally not persisting for many generations. Second, schizophrenia has been associated with significantly decreased fertility.16 If mutations associated with schizophrenia were ancient, reduced fertility over many generations would have led to a steep decline in the frequency of the illness. The persistence of schizophrenia over time may be explained by an ongoing contribution of new mutations. Third, schizophrenia is associated with older paternal age,17 and older paternal age is associated with increased rates of de novo germline mutations.13 Thus, new individually rare severe mutations in genes related to brain functioning or neural development may be responsible for a portion of cases of schizophrenia. Maternal starvation is rare and does not play a role in most cases of schizophrenia. However, de novo mutations may play an important role for schizophrenia. Populations exposed to famine are at higher risk for de novo events, and thus at higher risk for the illness. Detection of De Novo Mutations in Schizophrenia The normal function of any gene can be disrupted in many different ways. Individually rare mutations serve as indicators of genes likely to be responsible for multiple cases, each one of which may harbor a different mutation. That is, any gene with one disease-associated mutation likely harbors ©2006 American Medical Association. All rights reserved.

other disease-associated mutations in other patients. Birth cohorts exposed to famine may be enriched for de novo events and therefore are potentially informative for finding these indicator mutations. State-of-the-art genomic technologies can be exploited to identify certain classes of individually rare de novo mutations. It is now possible to detect deletions, duplications, and other chromosomal aberrations of several thousand or more base pairs.18,19 After an initial mutation is identified, efficient resequencing technology permits the gene harboring it to be screened for different mutations in large numbers of cases. At this step, it is increasingly possible to identify mutations in protein coding sequences as well as mutations in regulatory regions, noncoding RNA, and transposable elements. User-friendly bioinformatics resources are available to help characterize the structure and function of potential candidate genes. Famine may result in both germline and somatic de novo mutations. Somatic mutations refer to those arising in specific cell lines after conception. Somatic mutations occurring very early in development and present in both brain and blood would be detectable using current genomic technologies. Somatic mutations occurring later in development in neuronal cells may also lead to the illness but could not be detected if the tissue were not accessible. De Novo Epigenetic Effects Prenatal folate deficiency may also lead to epigenetic alterations20; that is, to stable changes in gene expression that do not depend on mutations in the affected gene. Such changes may predispose to later development of schizophrenia. The best understood mechanism for epigenetic effects is methylation of DNA in the regulatory regions of genes. Folate is necessary for normal DNA methylation. Expression of many genes involved with neurodevelopment of fetal mice, including transcription factors, growth factors, G proteins and methyltransferases, appears to be folate regulated.21 Epigenetic changes that silence expression of candidate genes in cells of accessible tissues can now be screened with genomic tools. Screening methods to identify epigenetic changes anywhere in the genome are currently in development.22 Implications for Genetic Studies A practical advantage to evaluating persons likely to harbor de novo mutations is that studies of large numbers of cases and controls are not necessary or even desirable. Rather, the focus is on individually informative cases that may reveal indicator mutations and therefore novel genes. If one mutation is identified, other mutations in the same gene may lead to the illness among other individuals. Complex illnesses are almost universally characterized by allelic heterogeneity (multiple different mutations in the same gene leading to the disease) and by locus heterogeneity (mutations in multiple different genes leading to the (Reprinted) JAMA, August 2, 2006—Vol 296, No. 5

Downloaded from www.jama.com at N/A, on August 5, 2006

583

COMMENTARIES

same disease).23,24 Allelic and locus heterogeneity play a role in hereditary forms of cancer, deafness, inherited forms of epilepsy, lipid disorders, and familial early-onset Alzheimer disease.25 For each of these conditions, the discovery of rare mutations in informative cases led to the identification of other mutations in larger populations of affected individuals. In conclusion, the risk for schizophrenia associated with maternal starvation may be mediated by an increased rate of de novo mutations. Evidence suggests folate deficiency as one component of this risk, although other factors also are likely involved. Many mutations associated with schizophrenia are likely to be rare, even specific to single cases or families. However, at-risk populations, such as the Dutch or Chinese birth cohorts exposed to famine, may reveal these mutations. Once identified, a gene harboring one mutation can be screened for other disease-associated mutations, which may range from common to rare, with effects from modest to severe. Any gene harboring one disease-related mutation is likely to harbor more than one mutation. Financial Disclosures: Dr McClellan reports receiving grants from the National Institute of Mental Health (U01MH61464) and the Stanley Medical Foundation. Dr Susser reports receiving grants from the Lieber Center for Schizophrenia Research, Columbia University. Dr King reports receiving the National Alliance for Research on Schizophrenia and Depression Distinguished Investigator Award. REFERENCES 1. Murray R, Jones P, Susser E, et al. The Epidemiology of Schizophrenia. New York, NY: Cambridge University Press; 2002. 2. Susser E, Neugebauer R, Hoek HW, et al. Schizophrenia after prenatal famine. Arch Gen Psychiatry. 1996;53:25-31. 3. St Clair D, Xu M, Wang P, et al. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959-1961. JAMA. 2005;294:557-562. 4. Stein Z, Susser M, Saenger G, et al. Famine and Human Development: The Dutch Hunger Winter of 1944-1945. New York, NY: Oxford University Press; 1975. 5. Susser E, Hoek HW, Brown A. Neurodevelopmental disorders after prenatal famine: the story of the Dutch Famine Study. Am J Epidemiol. 1998;147:213-216.

584 JAMA, August 2, 2006—Vol 296, No. 5 (Reprinted)

6. Hoek H, Brown A, Susser E. The Dutch Famine Studies: prenatal nutritional deficiency and schizophrenia. In: Susser E, Brown A, Gorman J, eds. Prenatal Exposures in Schizophrenia. Washington, DC: American Psychiatric Press Inc; 1999. 7. Smil V. China’s great famine: 40 years later. BMJ. 1999;319:1619-1621. 8. Lewis SJ, Zammit S, Gunnell D, Smith GD. A meta-analysis of the MTHFR C677T polymorphism and schizophrenia risk. Am J Med Genet B Neuropsychiatr Genet. 2005;135:2-4. 9. Picker JD, Coyle JT. Do maternal folate and homocysteine levels play a role in neurodevelopmental processes that increase risk for schizophrenia? Harv Rev Psychiatry. 2005;13:197-205. 10. Lucock M. Folic acid: nutritional biochemistry, molecular biology, and role in disease processes. Mol Genet Metab. 2000;71:121-138. 11. Lucock M. Is folic acid the ultimate functional food component for disease prevention? BMJ. 2004;328:211-214. 12. Beetstra S, Thomas P, Salisbury C, et al. Folic acid deficiency increases chromosomal instability, chromosome 21 aneuploidy and sensitivity to radiationinduced micronuclei. Mutat Res. 2005;578:317-326. 13. Crow JF. The origins, patterns and implications of human spontaneous mutation. Nat Rev Genet. 2000;1:40-47. 14. International Human Genome Sequencing Consortium. Finishing the euchromatic sequence of the human genome. Nature. 2004;431:931-945. 15. Gottesman II, Shields J. Schizophrenia: The Epigenetic Puzzle. Cambridge, England: Cambridge University Press; 1982. 16. Nimgaonkar VL. Reduced fertility in schizophrenia: here to stay? Acta Psychiatr Scand. 1998;98:348-353. 17. Malaspina D, Harlap S, Fennig S, et al. Advancing paternal age and the risk of schizophrenia. Arch Gen Psychiatry. 2001;58:361-367. 18. Sebat J, Lakshmi B, Troge J, et al. Large-scale copy number polymorphism in the human genome. Science. 2004;305:525-528. 19. Sharp AJ, Locke DP, McGrath SD, et al. Segmental duplications and copynumber variation in the human genome. Am J Hum Genet. 2005;77:78-88. 20. McKay JA, Williams EA, Mathers JC. Folate and DNA methylation during in utero development and aging. Biochem Soc Trans. 2004;32:1006-1007. 21. Spiegelstein O, Cabrera RM, Bozinov D, et al. Folate-regulated changes in gene expression in the anterior neural tube of folate binding protein-1 (Folbp1)deficient murine embryos. Neurochem Res. 2004;29:1105-1112. 22. Callinan PA, Feinberg AP. The emerging science of epigenomics. Hum Mol Genet. 2006;15:R95-R101. 23. Botstein D, Risch N. Discovering genotypes underlying human phenotypes: past successes for Mendelian disease, future approaches for complex disease. Nat Genet. 2003;33:228-237. 24. Goldstein DB, Cavalleri GL, Ahmadi KR. The genetics of common diseases: 10 million times as hard. Cold Spring Harb Symp Quant Biol. 2003;68:395401. 25. King MC, Ahsan H, Susser E. Designs for the genomic era. In: Susser E, Schwartz S, Morabia A, Bromet EJ, eds. Psychiatric Epidemiology: Searching for the Causes of Mental Disorders. New York, NY: Oxford University Press; 2006.

©2006 American Medical Association. All rights reserved.

Downloaded from www.jama.com at N/A, on August 5, 2006