Does ‘imprinting’ with low prenatal vitamin d contribute to the risk of various adult disorders?

Medical Hypotheses (2001) 56(3), 367–371
doi: 10.1054/mehy.2000.1226, available online at http://www.idealibrary.com on Does ‘imprinting’ with low prenatal
vitamin D contribute to the risk of
various adult disorders?
J. McGrath
Queensland Centre for Schizophrenia Research, Wolston Park Hospital, Wacol, Queensland, Australia Hypovitaminosis D is a candidate risk-modifying factor for a diverse range of disorders apart from rickets and osteoporosis. Based on epidemiology, and on in vitro and animal experiment, vitamin D has been linked tomultiple sclerosis, certain cancers (prostate, breast and colorectal), insulin-dependent diabetes mellitus andschizophrenia. I hypothesise that low pre- and perinatal vitamin D levels imprint on the functional characteristics ofvarious tissues throughout the body, leaving the affected individual at increased risk of developing a range of adult-onset disorders. The hypothesis draws from recent advances in our understanding of the early origin of adult diseaseand proposes a ‘critical window’ during which vitamin D levels may have a persisting impact on adult health outcomes.
Methods to test the hypothesis are outlined. If correct, the hypothesis has important implications for public health.
Careful attention to maternal vitamin D status could translate into diverse improvements in health outcomes for thefollowing generation. 2001 Harcourt Publishers Ltd PRE- AND PERINATAL EXPOSURES AND
imprinting may be mediated by clonal selection and METABOLIC IMPRINTING
metabolic differentiation. Clonal selection may havelong-lasting effects if early nutritional or hormonal expo- The hypothesis that environmental factors may ‘imprint’ sures differentially advantage certain cell lines: the more on the fetus and contribute to adult health has been stim- numerous facilitated daughter cells may have persisting consequences for the adult organism. Imprinting may Metabolic imprinting has two key features; (a) there is a also operate by ‘metabolic differentiation’: the process of critical window during fetal development or early life cells acquiring a stable quantitative pattern of basal and when the fetus is particularly sensitive to exposures; and inducible gene expression. These mechanisms may relate (b) the exposure leads to changes that persist throughout to enzymes, hormones and their receptors and other components of cellular molecular biology. Metabolic dif- posed several mechanisms for metabolic imprinting.
ferentiation includes epigenetic mechanisms related to Apart from induced variations in organ structure (e.g.
chromatin structure, DNA methylation, and autoregula- vascularization and/or innervation during organogenesis) and alterations in cell numbers (e.g. changes in neuronalcount/density after prenatal malnutrition), metabolic HEALTH AND VITAMIN D: CLUES FROM
BIOLOGY AND EPIDEMIOLOGY
Evidence linking vitamin D and various disorders is basedon in vitro, animal and ecological-level research. None of Correspondence to: J. McGrath MBBS, PhD, FRANZCP, Queensland Centre
these features in isolation would be sufficient ‘proof’ that for Schizophrenia Research, Wolston Park Hospital, Wacol, Queensland, vitamin D is involved in causal pathways; however, the Australia. Phone: +617 3271 8595; Fax: +617 2171 8567; E-mail: coherence of the data suggests that low vitamin D is a candidate risk-modifying factor for a range of diseases Diseases that have urban excess include breast other than rickets and osteoporosis. The table displays some of the evidence linking vitamin D and diseases such is also strongly linked to air pollution, a local factor as multiple sclerosis, breast cancer, prostate cancer, col- that influences the availability of UVB. Colorectal orectal cancer, insulin dependent diabetes and schizo- phrenia. The epidemiological similarities between some of these diseases have already been noted for prostate Migrant studies. The offspring of dark-skinned migrants to cold climates are prone to low vitamin D. Those with dark skin require slightly longer exposure to UVB in order to produce previtamin D, disorders can be summarised under several headings: and their behaviour (e.g. dress, outdoor activity,diet) may amplify the risk of hypovitaminosis D.
In vitro and animal experiments. Vitamin D can Multiple sclerosis has a lower incidence in first ameliorate the expression of animal models of generation Asian migrants to the UK than second generation migrants (those born in the UK) Vitamin D can lead to increased differentiation in schizophrenia in the second but not the first Incidence/prevalence associations with latitude.
Oral Vitamin D intake. Apart from sunshine Many diseases have gradients in incidence, exposure, Vitamin D can be derived from certain prevalence and outcome that are correlated with foods in the diet, cod liver oil intake and vitamin latitude. Latitude acts as a risk indicator (or proxy) supplementation. Case-control and cohort studies for the population distribution of serum vitamin D have found links between low oral vitamin D intake levels. A negative correlation between latitude and and increased risk of prostate and colorectal cancer disease incidence is found for disorders such as HYPOTHESIS
Ultraviolet B radiation (UVB). The geographicaldistribution of several diseases has been linked to While the incidence of rickets has declined markedly measures related to the availability of UVB. This over the last century, hypovitaminosis D is still relatively exposure is strongly correlated with latitude and A large US survey reported that 12% of women aged associated with measures related to sunshine 20 to 39 (peak ages for child-bearing) had serum 25- hydroxyvitamin D levels below the threshold defined for are at risk of hypovitaminosis D because of the increased needs of the fetus and the potential for these women to schizophrenia birth rates have also been linked to reduce their outdoor activity, leading to diminished sup- Season of birth. The amount of ultraviolet radiation I propose that low pre- and perinatal vitamin D ‘im- fluctuates across the seasons such that individuals prints’ on a range of tissues, leaving the affected individ- born in winter and early spring tend to be exposed ual at increased risk of developing a range of disorders to lower levels of vitamin D than those born in including osteoporosis, multiple sclerosis, breast cancer, other months. Disorders that have an excess of prostate cancer, colorectal cancer, insulin-dependent dia- winter/spring births suggest that early life betes mellitus and schizophrenia. The early life exposure exposures to low vitamin D may be a risk- contributes to risk status in addition to other factors such modifying exposure. There is robust evidence from as genetic susceptibility and adult exposures (including hypovitaminosis during adult life). With respect to the putative mechanisms of metabolic imprinting, it is pro- similar but weaker evidence for multiple sclerosis posed that vitamin D mediated alterations in neuronal proliferation, differentiation, migration and apoptosis Urban-rural gradient. Urban residence is associated may be implicated in schizophrenia. Vitamin D receptors have been found in differentiating zones of the central Medical Hypotheses (2001) 56(3), 367–371
Biological factors
related to
vitamin D
Season of birth
Urban-rural gradient
Other risk indicators
Migrant studies
Epidemiological clues linking diseases to vitamin D and ultraviolet radiation
Latitude
Duration of sunshine
Evidence linking vitamin D and selected disorders Medical Hypotheses (2001) 56(3), 367–371
If the hypothesis gains support then a randomized controlled trial of vitamin D supplementation during Factors related to clonal selection and metabolic differen- pregnancy would be indicated; however, the assessment tiation may be particularly important in the vitamin D of outcomes would require several decades of observa- related cancers (breast, prostate, colorectal), multiple scle- tions. Animal experiments could examine the impact on rosis and insulin-dependent diabetes mellitus. low prenatal vitamin D on a range of health outcomes; This hypothesis gains biological plausibility from the however, not all human diseases have robust animal demonstration of persisting alterations in cellular respons- models. The hypothesis also suggests new directions for ivity and organ differentiation after early-life exposure to genetic research: as the hypothesis implicates a prenatal vitamin D deficiency, maternal genes related to vitamin D Interestingly, low prenatal vitamin D has recently been metabolism warrant consideration in addition to those of implicated in the pathogenesis of syndrome ‘X’ (increased insulin resistance, hyperlipidaemia, hypertension, centralobesity, increased fibrinogen and increased risk of non- IMPLICATIONS FOR PUBLIC HEALTH
that low prenatal vitamin D might contribute to low birth Programs that aim to reduce the prevalence of hypovita- weight and program aspects of pancreatic islet beta cell minosis D in pregnant women could translate into a function. These features would then amplify the con- lower incidence of candidate disorders in their offspring.
sequences of adult hypovitaminosis, thus leading to in- Just as folate supplementation has been shown to reduce creased risk of syndrome ‘X’. A diverse range of evidence suggests that low prenatal vitamin D is a candidate expo- vitamin D status (diet, sunlight exposure) could reduce sure for a range of adverse adult health outcomes. the burden of a range of diseases. The population attrib-utable fraction of a particular disease that could be linkedto low prenatal vitamin D is not clear. In addition, the TESTING THE HYPOTHESIS
specificity of the outcomes is weak. However, this feature One could measure vitamin D levels in a large cohort of is a distinct advantage from the public health perspective.
pregnant women, and then follow-up their offspring for If an intervention designed to reduce one particular ex- several decades in order to search for a dose-response posure translates into reduced incidence of several com- relationship (a biological gradient) between low Vitamin mon disorders, then these exposures are deemed more D and increased risk of the candidate disorders. If sera from pregnant women and/or cord blood from their off-spring were ‘banked down’ during past decades, maternal ACKNOWLEDGMENT
vitamin D levels and rates of candidate disorders in the This project was supported by the Stanley Foundation From an ecological perspective, ‘natural’ and oppor- tunistic experiments that impact on Vitamin D may pro- REFERENCES
vide exposed and non-exposed cohorts for birthrate 1. Barker D. J. P. Fetal and infant origins of adult disease. 1ed.
comparisons (e.g. famine, clusters of rickets, the introduc- tion of vitamin D supplementation). Between-year fluctu- 2. Waterland R. A., Garza C. Potential mechanisms of metabolic ations in birth rates of candidate disorders could be imprinting that lead to chronic disease. American Journal of examined for associations with duration of sunshine. For Clinical Nutrition 1999; 69: 179–197.
3. Schwartz G. G. Multiple sclerosis and prostate cancer: what do example, geographically-defined birth cohorts (born over their similar geographies suggest? Neuroepidemiology 1992; 11:
several decades) could be divided according to quartiles of perinatal sunshine exposure, and rates of diseases 4. Templer D. I., Regier M. W., Corgiat M. D. Similar distribution of could be compared between these quartiles. It would be schizophrenia and multiple sclerosis [letter]. J Clin Psychiatry predicted, for example, that rates of schizophrenia, dia- 1985; 46: 73.
5. Templer D. I., Cappelletty G. G., Kauffman I. Schizophrenia and betes mellitus, multiple sclerosis, breast cancer, prostate multiple sclerosis. Distribution in Italy. Br J Psychiatry 1988; cancer and colorectal cancer would be higher in those 153: 389–390.
born during those periods with less sunshine. The 6. Wolfgram F. Similar geographical distribution of multiple hypothesis predicts that vitamin D related disorders will sclerosis and cancer of the colon. Acta Neurol Scand 1975; 52:
cosegregate in birth cohorts that have been exposed to 7. Hayes C. E., Cantorna M. T., DeLuca H. F. Vitamin D and low prenatal vitamin D. If possible, case-control studies of multiple sclerosis. Proc Soc Exp Biol Med 1997; 216: 21–27.
candidate disorders should ask mothers about prenatal 8. Cantorna M. T. Vitamin D and autoimmunity: is sunshine behaviour, and vitamin D intake. vitamin D status an environmental factor affecting autoimmune Medical Hypotheses (2001) 56(3), 367–371
disease prevalence? Proc Soc Exp Biol Med 2000; 223:
31. Gorham E. D., Garland C. F., Garland F. C. Acid haze air pollution and breast and colon cancer mortality in 20 Canadian 9. Schwartz G. G., Hulka B. S. Is vitamin D deficiency a risk factor cities. Can J Public Health 1989; 80: 96–100.
for prostate cancer? (Hypothesis). Anticancer Res 1990; 10:
32. Elian M., Nightingale S., Dean G. Multiple sclerosis among United Kingdom-born children of immigrants from the Indian 10. Christakos S. Vitamin D and breast cancer. Adv Exp Med Biol subcontinent, Africa and the West Indies. J Neurol Neurosurg 1994; 364: 115–118.
Psychiatry 1990; 53: 906–911.
11. Thomas M. G., Tebbutt S., Williamson R. C. Vitamin D and its 33. Pritchard R. S., Baron J. A., Gerhardsson D, V. Dietary calcium, metabolites inhibit cell proliferation in human rectal mucosa vitamin D, and the risk of colorectal cancer in Stockholm, and a colon cancer cell line. Gut 1992; 33: 1660–1663.
Sweden. Cancer Epidemiol Biomarkers Prev 1996; 5: 897–900.
12. Norman J. E., Jr., Kurtzke J. F., Beebe G. W. Epidemiology of 34. Garland C. F., Garland F. C., Gorham E. D. Can colon cancer multiple sclerosis in U.S. veterans: 2. Latitude, climate and the incidence and death rates be reduced with calcium and vitamin risk of multiple sclerosis. J Chronic Dis 1983; 36: 551–559.
D? Am J Clin Nutr 1991; 54: 193S–201S.
13. Weinshenker B. G. Epidemiology of multiple sclerosis. Neurol 35. The EURODIAB Substudy 2 Study Group. Vitamin D Clin 1996; 14: 291–308.
supplement in early childhood and risk for Type I (insulin- 14. Lipkin M., Newmark H. L. Vitamin D. calcium and prevention of dependent) diabetes mellitus. Diabetologia 1999; 42: 51–54.
breast cancer: a review. J Am Coll Nutr 1999; 18: 392S–397S.
36. Lawson M., Thomas M. Vitamin D concentrations in Asian 15. Hanchette C. L., Schwartz G. G. Geographic patterns of prostate children aged 2 years living in England. British Medical Journal cancer mortality. Evidence for a protective effect of ultraviolet 1999; 318: 28.
radiation. Cancer 1992; 70: 2861–2869.
37. Utiger R. D. The need for more Vitamin D. New Engl J Med 16. LaPorte R. E., Tajima N., Akerblom H. K., et al. Geographic 1998; 338: 828–829.
differences in the risk of insulin-dependent diabetes mellitus: 38. Looker A. C., Gunter E. W. Hypovitaminosis D in medical the importance of registries. Diabetes Care 1985; 8 Suppl 1:
inpatients. New Engl J Med 1998; 339: 344–345.
39. Hilman L. S., Haddad J. G. Perinatal vitamin D metabolism. III.
17. Akerblom H. K., Reunanen A. The epidemiology of insulin- Factors influencing late gestational human serum 25- dependent diabetes mellitus (IDDM) in Finland and in northern hydroxyvitamin D. Am J Obstet Gynecol 1976; 125: 196–200.
Europe. Diabetes Care 1985; 8 Suppl 1: 10–16.
40. Markestad T. Effect of season and vitamin D supplementation 18. Welham J., Davies G., Auliciems A., McGrath J. Climate, on plasma concentrations of 25-hydroxyvitamin D in geography and the search for candidate nongenetic risk factors Norwegian infants. Acta Paediatr Scand 1983; 72:
for schizophrenia. International Journal of Mental Health in 41. Veenstra T. D., Prüfer K., Koenigsberger C., Brimijoin S. W., 19. McGrath J. Hypothesis: is low prenatal vitamin D a risk- Grande J. P., Kumar R. 1,25-Dihydroxyvitamin D receptors in modifying factor for schizophrenia? Schizophr Res 1999; 40:
the central nervous system of the rat embryo. Brain Research 1998; 804: 193–205.
20. Holick M. F. Environmental factors that influence the cutaneous 42. Musiol I. M., Feldman D. 1,25-Dihydroxyvitamin D-3 induction production of vitamin D. Am J Clin Nutr 1995; 61: 638S–645S.
of nerve growth factor in L929 mouse fibroblasts: Effect of 21. Garland F. C., Garland C. F., Gorham E. D., Young J. F.
vitamin D receptor regulation and potency of vitamin D-3 Geographic variation in breast cancer mortality in the United analogs. Endocrinology 1997; 138: 12–18.
States: a hypothesis involving exposure to solar radiation. Prev 43. Konety B. R., Nangia A. K., Nguyen T. S., Thomas A., Getzenberg Med 1990; 19: 614–622.
R. H. Effect of prenatal vitamin D (calcitriol) exposure on the 22. Gorham E. D., Garland F. C., Garland C. F. Sunlight and breast growth and development of the prostate. Prostate 1999; 41:
cancer incidence in the USSR [see comments]. Int J Epidemiol 1990; 19: 820–824.
44. Csaba G., Inczefi-Gonda A. Effect of Vitamin D(3) Treatment in 23. Emerson J. C., Weiss N. S. Colorectal cancer and solar radiation.
the Neonatal or Adolescent Age (Hormonal Imprinting) on the Cancer Causes Control 1992; 3: 95–99.
Thymic Glucocorticoid Receptor of the Adult Male Rat. Horm 24. Garland C. F., Garland F. C. Do sunlight and vitamin D reduce the Res 1999; 51: 280–283.
likelihood of colon cancer? Int J Epidemiol 1980; 9: 227–231.
45. Boucher B. J. Inadequate vitamin D status: does it contribute to 25. Nystrom L., Dahlquist G., Ostman J., et al. Risk of developing the disorders comprising syndrome ‘X’? Br J Nutr 1998; 79:
insulin-dependent diabetes mellitus (IDDM) before 35 years of age: indications of climatological determinants for age at onset.
46. Czeizel A. E., Dudas I. Prevention of the first occurrence of Int J Epidemiol 1992; 21: 352–358.
neural tube defects by periconceptional vitamin 26. McGrath J., Welham J., Davies G., Chant D., Auliciems A.
supplementation. New Engl J Med 1992; 327: 1832–1835.
Between-year fluctuations in schizophrenia birth rates and 47. Rose G. The Strategy of Preventive Medicine. Oxford: Oxford associations with the southern oscillation index, cloud cover and sunshine. Schizophrenia Research 2000; 41: 63.
48. Gale C. R., Martyn C. N. Migrant studies in multiple sclerosis.
27. Davies G., Welham J., Torrey E. F., McGrath J. Season of birth Prog Neurobiol 1995; 47: 425–448.
and latitude: a systematic review and meta-analysis of northern 49. Dean G., Elian M. Age at immigration to England of Asian and hemisphere studies. Schizophrenia Research 2000; 41: 62.
Caribbean immigrants and the risk of developing multiple 28. Templer D. I., Trent N. H., Spencer D. A., et al. Season of birth in sclerosis. J Neurol Neurosurg Psychiatry 1997; 63: 565–568.
multiple sclerosis. Acta Neurol Scand 1992; 85: 107–109.
50. Elian M., Dean G. Multiple sclerosis among the United 29. Doll R. Urban and rural factors in the aetiology of cancer. Int J Kingdom-born children of immigrants from the West Indies. Cancer 1991; 47: 803–810.
J Neurol Neurosurg Psychiatry 1987; 50: 327–332.
30. McGrath J. Hypothesis: is low prenatal vitamin D a risk- 51. Lemire J. 1,25-dihydroxyvitamin D3 – a hormone with modifying factor for schizophrenia? Schizophr Res 1999; 40:
immunomodulatory properites. Z Rheumatol 2000; 59 Suppl 1:
Medical Hypotheses (2001) 56(3), 367–371

Source: http://www.awsc07100.pwp.blueyonder.co.uk/Papers/McGrath-D-Imprinting.pdf