Association of maternal and nutrient supply line factors with DNA methylation at the imprinted IGF2/H19 locus in multiple tissues of newborn twins

Figures & data

Figure 1. Structure of the IGF2/H19 locus with DMRs. The IGF2/H19 locus is shown with direction of transcription, exon locations (gray boxes) and DMRs (white boxes). DMRs assayed for DNA methylation in this study are indicated in bold.

Table 1. Subject characteristics of twin pairs

Variable
Pair-specific data	Mean (SD) or %
Gestational age at birth, weeks (median, SD)	37.0 (1.94)
Birth weight, g (median, SD)	2433.9 (479.7)
Sex: both male	40.5%
Sex: both female	50%
Sex: one male, one female	9.5%
Supply line factors
Placental weight, mean (g)	419.5 (85.4)
Central cord insertion	38.8%
Non-central (velamentous or peripheral) cord insertion	53%
Maternal factors
Periconceptional^a folate supplements, > 400mg/day	90.5%
Third trimester serum vitamin B12 (pmol/L)	174.5 (69.2)
Third trimester serum homocysteine (µ-mol/L)	4.55 (2.13)
Carbohydrate intake in the first 2 trimesters (grams/day)	219.9 (82.2)
Protein intake in the first 2 trimesters (grams/day)	93.8 (43.0)
Energy intake in the first 2 trimesters (kilojoules/day)	8373.9 (3207.7)
Periconceptional* alcohol consumption^†	50%
Periconceptional* smoking^‡	24.1%
Maternal stress score, mean (range of possible scores 0–46)	21.3 (7.23)
Gestational diabetes	7.7%

* Defined as “before you knew you were pregnant (from conception to week 12 of your pregnancy”). ^†Women who consumed any quantity of alcohol on a weekly basis before knowing they were pregnant. ^‡Women who consumed any quantity of cigarettes on a weekly basis before knowing they were pregnant.

Table 2. Linear regression of IGF2/H19 methylation jointly on maternal and supply line factors over all five cell types combined

Factor	All Assays Combined		H19 promoter DMR		IGF2/H19 ICR		IGF2 DMR0		IGF2 DMR2
	Coeff	p-val	Coeff	p-val	Coeff	p-val	Coeff	p-val	Coeff	p-val
Had folate	0.50%	0.44	-1.70%*	0.024*	0.40%	0.69	0.90%	0.46	2.90%*	0.035*
Vitamin B12 (z-score)	-0.23%	0.24	-0.97%*	0.002*	-0.23%	0.54	0.23%	0.55	0.27%	0.63
Homocysteine (z-score)	0.27%	0.29	0.10%	0.75	0.40%	0.29	0.37%	0.30	0.17%	0.72
Macronutrients (z-score)	0.37%	0.17	0.80%*	0.049*	0.20%	0.50	0.43%	0.27	0.10%	0.77
Had alcohol	0.50%	0.27	0.00%	0.98	-0.70%	0.18	1.30%	0.14	1.90%	0.062
Smoked	0.90%	0.059	2.10%*	0.005*	1.20%*	0.043*	0.30%	0.74	-0.10%	0.90
Stress (z-score)	-0.10%	0.68	0.00%	0.98	-0.27%	0.42	0.30%	0.58	0.17%	0.65
Gestational diabetes	-1.30%*	0.012*	-1.50%	0.14	-0.80%	0.44	-1.50%	0.14	-1.60%	0.17
Central cord insertion (MC)	3.00%*	< 0.001*	3.20%*	0.046*	2.40%*	0.005*	2.70%*	0.001*	4.30%*	0.002*
Central cord insertion (DC)	-0.30%	0.47	0.80%	0.28	1.80%*	0.001*	-1.80%*	0.045*	0.00%	0.98
Placenta Weight	0.53%*	0.044*	0.53%	0.18	0.87%*	0.010*	0.30%	0.57	0.27%	0.60

Coefficients measure difference in absolute percentage methylation. The “All Assays Combined” columns display the data from all assays and tissues are combined, the rest of the table displays the data resulting from associations analyzed separately for each assay. Coefficient for placenta weight, vitamin B12, homocysteine, macronutrients and stress score represents difference in mean methylation for a 1 standard deviation increase in the corresponding factor. Central cord insertion indicates central vs. non-central (peripheral and velamentous) insertion. *Coefficients with P-values less than 0.05. Confidence intervals are shown in Table S3.

Figure 2. Association of IGF2/H19 methylation and nutrition factors across all cell types and DMRs. Light gray bars represent methylation changes in the IGF2 DMRs, and the dark gray bars represent changes in H19 DMRs. 95% confidence intervals of the coefficient change are represented by a horizontal black line through each bar. The mean methylation differences corresponding to serum vitamin B12, serum homocysteine, and macronutrient are for a 1 standard deviation increase in the respective factor’s z-score. Macronutrient score is obtained as the average of protein, carbohydrate and energy z-scores.

Figure 3. Association of IGF2/H19 methylation and lifestyle factors across all cell types and DMRs. Legend as for Figure 2. The mean methylation difference corresponding to stress is for a 1 standard deviation increase in the stress z-scores.

Figure 4. Association of IGF2/H19 methylation and supply line factors across all cell types and DMRs. Legend as for Figure 2. Mean methylation differences are regression coefficients obtained from multivariable regression analysis, whereby regression coefficients are converted to percentage difference by multiplying each regression coefficient with average standard deviation. The mean methylation corresponding to placenta weight is for a 1 standard deviation increase in placenta weight z-score.

Table 3. Comparisons with previously published data

	Published studies					Our study
Exposure/ reference	Model	Region	Tissue/Age studied	Exposure timing	Methylation result	Tissue	Exposure timing	Methylation result
Folic acid supplements^40	Human	IGF2 DMR0	Whole blood/ 17 mo	Periconception	+4.5%	Granulocytes CBMCs	Periconception	+5.2% ns* +2.7% ns
Folic acid supplements^39	Human	IGF2/H19 ICR (CTCF1)	Whole blood /birth	Before pregnancy	+2.2% to +2.4%	Granulocytes	Periconception	+2.7% ns
				During pregnancy	-2.9% to -3.7%	CBMCs	Periconception	+1.5% ns
Alcohol^28	Mouse	IGF2/H19 ICR	Placenta	Pre-implantation	Average -12% in paternal clones	Placenta	Periconception	-0.6%
Famine (lack of macronutrients)^37^,^72	Human	IGF2 DMR0	Whole blood/ 60 y old	prenatal	-5.2%, -2.0%^e	Granulocytes CBMCs	Periconception	+1.0%^† ns +1.7%^† ns
		H19 promoter DMR			-0.5% ns	Granulocytes CBMCs	Periconception	+0.5%^† ns +0.5%^† ns
Smoking^36	Human	IGF2 DMR0	Whole blood/birth	Periconception	~4% ^‡,§	Granulocytes CBMCs	Periconception	+4.9% ns +0.2% ns

Previous studies that looked at association between the factors examined in this study and IGF2/H19 methylation shown on the left are compared with the data from our study (right). *ns, not significant; ^†the value indicates the mean methylation difference in every 1 standard deviation decrease in macronutrient score;^‡results presented for males and females and average shown; ^§significant for males only; ^e corresponding to a standard effect size of -0.6 SD units.

Steegers-Theunissen RP, Obermann-Borst SA, Kremer D, Lindemans J, Siebel C, Steegers EA, et al. Periconceptional maternal folic acid use of 400 microg per day is related to increased methylation of the IGF2 gene in the very young child. PLoS One 2009; 4:e7845; http://dx.doi.org/10.1371/journal.pone.0007845; PMID: 19924280

 PubMed Web of Science ®Google Scholar

Hoyo C, Murtha AP, Schildkraut JM, Jirtle RL, Demark-Wahnefried W, Forman MR, et al. Methylation variation at IGF2 differentially methylated regions and maternal folic acid use before and during pregnancy. Epigenetics 2011; 6:928 - 36; http://dx.doi.org/10.4161/epi.6.7.16263; PMID: 21636975

 PubMed Web of Science ®Google Scholar

Haycock PC, Ramsay M. Exposure of mouse embryos to ethanol during preimplantation development: effect on DNA methylation in the h19 imprinting control region. Biol Reprod 2009; 81:618 - 27; http://dx.doi.org/10.1095/biolreprod.108.074682; PMID: 19279321

 PubMed Web of Science ®Google Scholar

Heijmans BT, Tobi EW, Stein AD, Putter H, Blauw GJ, Susser ES, et al. Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proc Natl Acad Sci U S A 2008; 105:17046 - 9; http://dx.doi.org/10.1073/pnas.0806560105; PMID: 18955703

 PubMed Web of Science ®Google Scholar

Tobi EW, Slagboom PE, van Dongen J, Kremer D, Stein AD, Putter H, et al. Prenatal famine and genetic variation are independently and additively associated with DNA methylation at regulatory loci within IGF2/H19. PLoS One 2012; 7:e37933; http://dx.doi.org/10.1371/journal.pone.0037933; PMID: 22666415

 PubMed Web of Science ®Google Scholar

Murphy SK, Adigun A, Huang Z, Overcash F, Wang F, Jirtle RL, et al. Gender-specific methylation differences in relation to prenatal exposure to cigarette smoke. Gene 2012; 494:36 - 43; http://dx.doi.org/10.1016/j.gene.2011.11.062; PMID: 22202639

 PubMed Web of Science ®Google Scholar