Maternal Microdeletion at the H19/Igf2 ICR in Mice Increases Offspring Susceptibility to In Utero Environmental Perturbation

Abstract

Deficiency of methyl donor nutrients folate, choline, and methionine (methyl deficiency) during gestation can impair fetal development and perturb DNA methylation. Here, we assessed genetic susceptibility to methyl deficiency by comparing effects in wildtype C57BL/6J (B6) mice to mutant mice carrying a 1.3 kb deletion at the H19/Igf2 Imprinting Control Region (ICR) (H19^ICRΔ2,3). The H19^ICRΔ2,3 mutation mimics microdeletions observed in Beckwith-Wiedemann syndrome (BWS) patients, who exhibit epimutations in cis that cause loss of imprinting and fetal overgrowth. Dams were treated during pregnancy with 1 of 4 methyl sufficient (MS) or methyl deficient (MD) diets, with or without the antibiotic commonly used to deplete folate producing gut microbes. As expected, after ~9 weeks of treatment, dams in MD and MD + antibiotic groups exhibited substantially reduced plasma folate concentrations. H19^ICRΔ2,3 mutant lines were more susceptible to adverse pregnancy outcomes caused by methyl deficiency (reduced birth rate and increased pup lethality) and antibiotic (decreased litter size and litter survival). Surprisingly, pup growth/development was only minimally affected by methyl deficiency, while antibiotic treatment caused inverse effects on B6 and H19^ICRΔ2,3 lines. B6 pups treated with antibiotic exhibited increased neonatal and weanling bodyweight, while both wildtype and mutant pups of heterozygous H19^ICRΔ2,3/+ dams exhibited decreased neonatal bodyweight that persisted into adulthood. Interestingly, only antibiotic-treated pups carrying the H19^ICRΔ2,3 mutation exhibited altered DNA methylation at the H19/Igf2 ICR, suggesting ICR epimutation was not sufficient to explain the altered phenotypes. These findings demonstrate that genetic mutation of the H19/Igf2 ICR increases offspring susceptibility to developmental perturbation in the methyl deficiency model, maternal and pup genotype play an essential role, and antibiotic treatment in the model also plays a key independent role.

Keywords

Methyl donor deficiency Beckwith-Wiedemann Syndrome DNA methylation epigenetics H19 Igf2 genomic imprinting succinylsulfathiozole antibiotic pregnancy

Introduction

Optimal dietary intake of methyl donor nutrients such as folate, choline, and betaine during pregnancy is essential to healthy fetal growth and development, as well as long-term health outcomes.^1-7 Although there is widespread fortification of folate in the United States, in many other countries worldwide there is a high prevalence of suboptimal methyl donor nutrient intake among women of reproductive age, particularly in economically challenged communities.^8-11 Methyl donor nutrients are essential substrates in the 1 carbon metabolism pathway that generates S-adenosylmethionine (SAM),¹² the primary methyl group donor for methylation of DNA, and histone residues.¹³ Therefore, availability of methyl donor nutrients has the potential to modulate the DNA methylome and related health outcomes.¹²

DNA methylation occurs primarily at CG dinucleotides and is generated by DNA methyltransferase (DNMT) activity through covalent attachment of methyl groups from SAM to the fifth carbon position of the DNA base cytosine to generate 5-methylcytocine (5mC). One of the primary functions of DNA methylation is to act in combination with other epigenetic modifications to regulate gene expression via transcriptional activation or repression depending on the proximity of the methylation mark to specific regulatory elements such as promoters, insulators, enhancers, transcription factor binding sites, etc.¹⁴ During the susceptible window of gestation, germ, and somatic cells of the developing mammalian embryo are epigenetically reprogrammed such that DNA methylation patterns are erased and reestablished in a locus and cell-specific manner.^15-17 Once established, many methylation marks must be maintained and transmitted from parent to daughter cells to maintain normal cellular function.¹⁸ Therefore, perturbation of DNA methylation by suboptimal nutrient availability during development can substantially perturb development and cause disease.¹³

Deficiency of methyl donor nutrients folate, methionine, and choline during pregnancy have been shown to perturb DNA methylation. For example, offspring born to C57BL/6J mouse dams fed low folate diet during mating, gestation, and lactation, exhibited significant loss of global DNA methylation of the small intestine compared to dams belonging to the control/high folate group.¹⁹ Importantly, supplementation postweaning did not rescue these effects demonstrating persistent epimutations can occur as a result of in utero deficiency.¹⁹ Other studies that have assessed methylation in a locus specific manner show that methyl deficiency in utero can lead to loss of methylation at some loci and gain of DNA methylation at other loci.^20,21 One study in particular showed gain of methylation at Igf2 in response to choline deficiency during gestation.²²

The primary objective of this study was to use the in utero methyl deficiency model of environmental perturbation to determine whether genetic mutations at the H19/IGF2 imprinting control region (ICR) found in Beckwith Wiedemann Syndrome (BWS) patients, increases sensitivity to epigenetic perturbation during development. Human and animal model studies suggest imprinted domains may be particularly sensitive to environmental influences during gestation.^23-26 The ICR is differentially methylated and acts as the key regulatory element of “imprinted” parent of origin dependent monoallelic gene expression.²⁷ The differing methylation status at the maternal versus paternal ICR alleles is established during germ cell development (maternal allele in oocytes and paternal allele in sperm) and maintained after fertilization in the soma of the developing embryo and throughout the lifespan.²⁷ Thus, ICRs represent a rigorous and reproducible model locus to investigate the role of DNA sequence context in regulating offspring epigenetic susceptibility to environmental perturbation. The H19/IGF2 ICR regulates imprinted expression.^28-31 of insulin-like growth factor 2, IGF2, a protein coding gene with key roles in growth and metabolism,³² and H19, a non-coding RNA.³³ Both genes have essential roles in embryonic and placental growth and development.²⁷ The H19/IGF2 ICR is the most well-characterized ICR, located in the intergenic region between H19 and IGF2, with conserved synteny and function between mouse and human.²⁷ The maternal ICR allele is normally hypomethylated and bound by the insulator protein CCCTC-binding factor (CTCF), while the paternal allele is hypermethylated and CTCF-free.²⁷ A combination of epigenetic markings and CTCF binding at the H19/IGF2 ICR regulates promoter-enhancer interactions required to maintain allele-specific expression of H19 and IGF2.²⁷ Aberrant gain of methylation at the maternal H19/IGF2 ICR during development causes biallelic overexpression of IGF2 and subsequent fetal and postnatal overgrowth in BWS patients.^34,35 This epimutation is found in approximately 5% to 10% of BWS patients and is currently used to diagnose BWS.^35-37

The cause of gain of methylation epimutations at the H19/IGF2 ICR that lead to BWS are unknown. However, it has been proposed that environmental insults and/or changes in spacing or number of CTCF sites caused by ICR microdeletions may epigenetically destabilize the locus.^28,30,35,38 H19/IGF2 ICR microdeletions have been identified that coincide with these epimutations,^28,30,38-41 and targeted mutant mouse models made to mimic or directly insert patient deletions (humanized models) have been shown to partially recapitulate the overgrowth phenotype observed in BWS.^31,42

Here, we tested whether maternal H19/Igf2 ICR microdeletion makes the locus more sensitized to environmental epigenetic perturbation either during somatic cell development when the ICR methylation status must be maintained, or during male germcell (sperm) development when the embryo’s maternally hypomethylated allele must become hypermethylated. We investigated if mice carrying a targeted 1.3 kb deletion at the H19/Igf2 ICR (H19^ICRΔ2,3),³¹ that removes ~75% of the ICR including 2 of the 4 CTCF sites, were more susceptible to developmental and epigenetic perturbation caused by in utero methyl deficiency compared to wildtype C57BL/6J (B6) mice. We have previously shown that maternal transmission of the H19^ICRΔ2,3 mutation alone causes progeny to exhibit BWS-like growth phenotypes including reduced CTCF binding on the maternal allele, aberrant maternal Igf2 expression, and overgrowth.³¹ However, the H19^ICRΔ2,3 mutation alone did not alter ICR methylation when either maternally or paternally transmitted.³¹

The secondary objective of this study was to determine whether the antibiotic routinely used in methyl donor deficiency mouse models contributes to any of the molecular or developmental phenotypes observed. Standard methyl deficiency mouse models include 1% succinylsulfathiozole (SS), an antibiotic used to kill folate-producing gut microbes, and thus further exacerbate folate deficiency.⁴³ Although Burgoon and colleagues previously reported that diets containing 1%SS alone did not lead to major developmental defects,⁴³ minor developmental disturbances and molecular effects such as epigenetic responses have not been extensively investigated. In light of what we now know about the seemingly pervasive role of the microbiome in health, and findings that antibiotic disruption of the intestinal microflora causes a wide range of developmental and metabolic disorders,^44-47 it is important to reassess the impact of 1%SS in this model. Therefore, in this study, we have assessed the effect of methyl deficiency and the independent effect of 1%SS antibiotic treatment in perturbing development and methylation status of offspring.

Results

Maternal dietary depletion of methyl donors reduced maternal plasma folate levels but not SAM availability

To determine whether mice carrying a BWS-like microdeletion at the H19/Igf2 ICR are more susceptible to developmental and epigenetic perturbation caused by methyl donor deficiency, we compared the response of wild-type C57BL/6 (B6) mice to mice carrying a targeted 1.3 kb deletion at the H19/Igf2 Imprinting Control Region (ICR) (Figure 1).³¹ To induce methyl donor deficiency but also allow for distinguishing potentially unrelated effects of antibiotic, we compared 4 diet groups (1) methyl sufficient (MS), (2) MS1%SS, (3) methyl deficient (MD), and (4) MD1%SS (Figure 1A). Dams from 3 genotypes (Figure 1B) were treated for 5 weeks before pregnancy to ensure pup deficiency at conception, throughout gestation, and until postnatal day 5 (PND 5). Thereafter, dams and pups were placed on standard rodent chow (Harlan 8604) (Figure 1A).

Figure 1.

Dietary treatment scheme and experimental crosses: (A) dietary treatments: (1) methyl sufficient (MS), (2) MS with 1%SS (MS1%SS), (3) methyl deficient (MD), and (4) MD with 1%SS (MD1%SS). Solid arrow above indicates dietary treatment windows. Arrows below indicate the timepoint of experimental steps shown below, and (B) crosses used to generate G1 pups shown as dam x sire. Arrows indicate progeny genotype outcomes shown below. Wildtype C57BL/6 (B6, +/+), heterozygous (∆2,3/+), or homozygous (∆2,3/∆2,3) mutant dams were crossed with wildtype C7 sires to produce wildtype B6 (+/+) offspring, wildtype offspring from the ∆2,3 lineage (∆2,3(+/+)), and heterozygous offspring from the ∆2,3 lineage (∆2,3/+).

To evaluate the effect of the diets on maternal methyl donor metabolite availability to the offspring at birth and weaning, we measured maternal plasma folate levels after dietary depletion (~9 weeks, equivalent to status at birth) and after dietary repletion (~16 days restored to folate sufficient diet, equivalent to status at weaning). Since there was no significant effect of dam genotype on plasma folate levels expected or detected for the status after dietary depletion (P = .53) or repletion (P = .56), we combined the data for all genotypes. As expected, dietary depletion caused a significant reduction in maternal plasma folate concentrations such that concentrations in the MD and MD1%SS diet groups were significantly lower compared to the MS diet group (Figure 2A). The antibiotic treatment alone (MS1%SS) was not sufficient to significantly reduce plasma folate levels, however, the combined methyl donor deficiency and antibiotic treatment group (MD1%SS) exhibited the lowest concentration of folate (83% lower than MS) although not significantly lower than the MD group (Figure 2A). Plasma folate concentrations were completely restored in dams treated with 16 days of dietary repletion post-parturition using standard chow (Figure 2A).

Figure 2.

Maternal folate, SAM, and SAH response to methyl donor dietary depletion and dietary repletion: (A) box and whiskers plot of maternal plasma folate levels for dams after dietary depletion. Sample sizes in order from left to right, (n = 6, 4, 7, 8) or dietary repletion (n = 4, 4, 5, 2), and (B) box and whiskers plot of maternal liver SAM, SAH, and SAM/SAH ratio (n = 5 for all groups). Each dot represents an individual sample, Black (MS base diet groups), Red (MD base diet groups).

Methyl donor nutrient availability is a key determinant of the concentrations of the primary methyl donor liver metabolite S-adenosylmethionine (SAM) and its metabolic byproduct S-adenosylhomocysteine (SAH).⁴⁸ The ratio of SAM/SAH has been proposed to be an indicator of the cellular methylation potential.⁴⁹ Folate-deficient diets have been reported to decrease SAM and SAM/SAH ratio and increase SAH concentrations.^50-53 Therefore, we determined whether maternal folate depletion altered maternal availability of SAM, SAH, and the ratio of SAM/SAH. Surprisingly, the substantial decrease in maternal plasma folate observed in our model did not cause any significant changes in liver concentrations of SAM, SAH, or the SAM/SAH ratio. Although there was a slight decrease in average SAM levels in the MD1%SS group (23.0 ± 4.2 nmol/g, ±error represents standard error of the mean) compared to the MS group (30.8 ± 3.4 nmol/g), it was not statistically significant (Figure 2B).

Dams carrying the Δ2,3 mutation are more susceptible to adverse pregnancy outcomes caused by methyl donor depletion and antibiotic treatment

Methyl donor deficiency during the periconceptual period has been associated with poor reproductive and birth outcomes.⁵⁴ To determine whether mice carrying the BWS-like Δ2,3 microdeletion at the H19/Igf2 ICR are more susceptible to adverse pregnancy outcomes of methyl donor deficiency and whether the antibiotic treatment plays a role, for each dam genotype (+/+; Δ2,3/+, and Δ2,3/Δ2,3) we measured the (i) Methyl donor depletion effect (Diet) by comparing MS + MS1%SS versus MD + MD1%SS samples; and (ii) Antibiotic treatment effect determined by comparing MS + MD versus MS1%SS + MD1%SS samples.

Methyl donor depletion did not significantly alter litter size at birth for any of the genotypes (Figure 3A). However, for Δ2,3/+ depleted dams, there were significantly more females with that did not produce a litter, 32% (8/25) compared to 6% (1/15) of methyl sufficient Δ2,3/+ dams (Figure 3B). Methyl donor depletion also did not have a significant effect on litter survival to weaning for any of the genotypes (Figure 3C), but did result in a significantly higher rate of pup lethality for Δ2,3/Δ2,3 dams only, with an average of 24% of pups dying before weaning compared to 0% of methyl sufficient Δ2,3/Δ2,3 dams (Figure 3D). A similar trend of pup lethality was observed for methyl donor depleted Δ2,3/+ dams but was not statistically significant (Figure 3D). These data demonstrate that Δ2,3 pregnancies are more susceptible to adverse effects of methyl donor depletion compared to wild-type B6 pregnancies.

Figure 3.

H19^ICRΔ2,3 genetic susceptibility to pregnancy outcomes of methyl donor depletion and antibiotic treatment: (A) box and whiskers plot of litter sizes at birth for each treatment group and cross [Sample sizes in order from left to right, B6(+/+) n = 4, 7, 5, 8; ∆2,3/+ n = 10, 5, 12, 5; ∆2,3/∆2,3 n = 6, 4, 6, 4]. Each dot represents a litter, (B) proportion of matings resulting in live birth [B6(+/+) n = 4, 9, 6, 12; ∆2,3/+ n = 11, 5, 16, 9; ∆2,3/∆2,3 n = 8, 4, 6, 4], (C) pup fitness as determined by proportion of litter that survived to weaning [B6(+/+) n = 4, 7, 5, 9; ∆2,3/+ n = 10, 5, 12, 5; ∆2,3/∆2,3 n = 6, 4, 6, 4], and (D) for litters that survived to weaning [B6(+/+) n = 4, 5, 5, 6; ∆2,3/+ n = 10, 3, 11, 3; ∆2,3/∆2,3 n =3, 3, 3, 1], the percent of pups for each litter that died before weaning. Error bars indicate standard error of the mean (SEM).

Unlike methyl donor depletion, antibiotic treatment did not alter number of females that gave birth to a litter or partial litter survival to weaning (Figure 3B and D). Antibiotic treatment did cause significantly lower litter size for Δ2,3/+ dams with ~5 pups per litter on average (from 14 matings, MS1%SS + MD1%SS combined) compared to ~7 pups per litter for antibiotic-free pregnancies (from 22 matings, MS + MD combined) (Figure 3A). Although we did not measure statistical significance for each group separately, both methyl sufficient and methyl deficient 1%SS Δ2,3/+ pregnancies were substantially skewed towards smaller litter sizes (Figure 3A). B6 and Δ2,3/Δ2,3 litter sizes were not significantly affected (Figure 3A). Antibiotic treatment also resulted in significantly more whole litters dying before weaning for wild-type B6 pregnancies, with 25% (4/16) of litters dying compared to 0% (0/9) of antibiotic-free pregnancies (Figure 3C). Δ2,3/+ pregnancies followed a similar trend, while Δ2,3/Δ2,3 pregnancies exhibited high litter mortality for all treatment groups despite a high rate of females that produced a litter (Figure 3C). These data demonstrate that both wild-type B6 and Δ2,3/+ pregnancies are susceptible to adverse effects of antibiotic treatment during pregnancy but homozygous Δ2,3/Δ2,3 pregnancies may not be similarly responsive. Antibiotic seemingly had a greater effect on litter size and subsequent litter survival while methyl donor deficiency had a greater effect on dam litter production and partial litter survival.

Due to the high rate of pup lethality that led to very few pups surviving past birth, we did not further analyze developmental or epigenetic outcomes for Δ2,3/Δ2,3 pregnancies.

Offspring of Δ2,3 dams are more susceptible to developmental delay/growth restriction response to antibiotic treatment

Suboptimal methyl donor intake during pregnancy has been linked to reduced neonatal weight.⁴⁸ On the other hand, we have shown that mutant offspring of Δ2,3/+ dams exhibit classic BWS-like increased neonatal weights compared to their wildtype littermates, likely due to loss of imprinting at the H19/Igf2 locus and increased expression of the growth factor Igf2. To determine whether offspring of Δ2,3/+ dams exhibit increased susceptibility to impaired growth/development due to methyl donor deficiency or antibiotic treatment, we measured offspring bodyweight at PND5 (neonate), PND21 (weaning, juvenile), and PND56 (young adult) and compared to that of wildtype B6 offspring. For each mouse lineage (B6 or Δ2,3, G1 lineage) and pup genotype (+/+ or −/+) we measured the (i) Methyl donor depletion effect (Diet) by comparing MS+MS1%SS versus MD + MD1%SS samples; and (ii) Antibiotic treatment effect determined by comparing MS + MD versus MS1%SS + MD1%SS samples.

Methyl donor depletion in utero did not alter neonatal bodyweight of B6 offspring (Figure 4A). On the other hand, Δ2,3 neonates (combined genotypes as not genotyped at this stage) were significantly affected by methyl donor depletion and weighed on average ~6% (~180 mg) less than methyl sufficient pups (2.64 ± 0.03 g vs. 2.82 ± 0.04 g, respectively), with male and female pups similarly affected (Figure 4A). This is unlikely biased by combining genotypes since genotyping at PND21 revealed similar rates of transmission for each treatment group (MS, n=28:39; MS1%SS, n = 7:8; MD, n = 24:27, MD1%SS, n = 7:7). At weaning, methyl depleted B6 offspring weighed on average ~7% (~0.64 g) more than sufficient pups (9.29 ± 0.16 g vs. 8.65 ± 0.16 g, respectively), with male and female juvenile pups similarly affected (Figure 4B). MD1%SS B6 weanlings weighed the most on average at 9.45 ± 0.25 g compared to 8.27 ± 0.21g for MS pups, (~1.2 g more on average) (Figure 4B). For Δ2,3 juvenile pups, there was no significant effect of methyl depletion alone for either (+/+) or (−/+) offspring genotypes (Figure 4B). Developmental effects of methyl donor depletion did not last into adulthood for any of the genotypes (Figure 4C).

Figure 4.

Offspring growth response to in utero methyl donor depletion and antibiotic treatment: (A) top panel: Bar graph of offspring PND5 (neonatal) bodyweight separated by wildtype (B6, left panel) versus mutant (∆2,3, right panel) dam lineage and pup genotype (+/+ or +/− in parentheses). Mice not genotyped at PND5 therefore, data for both ∆2,3 pup genotypes are combined [sample sizes in order from left to right, B6(+/+) n = 19, 29, 20, 19; ∆2,3(+/+ and −/+) n = 80, 9, 64, 14]. Bottom panel: Individual pup weights (dots) separated by sex, male (M, black dots), female (F, grey dots); lines connect means between groups for males (solid line) or females (dashed line), (B) top panel: Bar graph of offspring PND21 (weaning) bodyweight separated by dam lineage and pup genotype [B6(+/+) n = 19, 25, 19, 19; ∆2,3(+/+) n = 23, 7, 23, 6; ∆2,3(−/+) n = 38, 7, 27, 6]. Bottom panel: Individual pup weights (dots) separated by sex, and (C) bar graph of offspring PND56 (adult) bodyweight (males only) separated by dam lineage and pup genotype. [B6(+/+) n = 9, 14, 7, 11; ∆2,3(+/+) n = 17, 2, 18, 4; ∆2,3(−/+) n = 19, 2, 16, 2].

Antibiotic treatment in utero, compared to untreated mice, led to a significant increase in B6 offspring bodyweight for neonatal and juvenile stages (neonatal, ~10%/~260 mg increase, 2.97 ± 0.0.05 g vs. 2.71 ± 0.05 g; juvenile, ~5%/0.46 g increase, 9.16 ± 0.16 g vs. 8.70 ± 0.16 g, respectively), with both males and females similarly affected (Figure 4A and B). However, this developmental effect did not last into adulthood (Figure 4C). In contrast, Δ2,3 lineage offspring exposed to antibiotic, compared to untreated mice, exhibited a significant decrease in bodyweight at neonatal, juvenile, and adult stages (Figure 4A to C). On average, antibiotic treated (+/+) and (−/+) neonates of Δ2,3/+ dams (combined as not genotyped at this stage) weighed ~12% (340 mg) less than untreated mice (2.45 ± 0.05 g vs. 2.79 ± 0.03 g, respectively) (Figure 4A). Antibiotic treated (+/+) juveniles of Δ2,3/+ dams weighed ~20% (1.9 g) less on average (7.31 ± 0.27 g vs. 9.17 ± 0.13 g, respectively), while adults weighed ~7% (1.6 g) less on average (22.27 ± 0.26 g vs. 23.90 ± 0.33 g) (Figure 4B and C). However, the effect of antibiotic treatment was seemingly more severe for males compared to female (+/+) juveniles of Δ2,3/+ dams (Figure 4B). Antibiotic treated (−/+) juveniles of Δ2,3/+ dams weighed ~17% (1.6 g) less on average (7.49 ± 0.27 g vs. 9.07 ± 0.13 g, respectively), while adults weighed only ~5% (1.4 g) less on average (24.82 ± 0.39 g vs. 26.18 ± 0.56 g), but the latter was not statistically significant (Figure 4B and C).

Similar effect of methyl donor depletion and antibiotic on both wildtype (+/+) and mutant (−/+) offspring of Δ2,3 dams implicates that these developmentally-induced changes in bodyweight are caused by a maternal effect.

Offspring from H19^ICRΔ2,3 dams are more susceptible to subtle epigenetic perturbation in cis

Methyl donor nutrients play a key role in several transmethylation pathways essential for metabolic and epigenetic regulation, especially during gestational reprogramming events.⁵⁵ To determine whether increased susceptibility of the Δ2,3 lineage to methyl deficiency and antibiotic treatment is caused by epigenetic perturbation of the mutated H19/Igf2 locus, we assessed the DNA methylation state across key regulatory elements within the domain in offspring liver and sperm. We measured methylation in liver at : (i) CTCF binding sites at the beginning and end of the imprinting control region (ICR) that regulates H19 and Igf2 expression (H19CTS1 and H19CTS4); (ii) the region immediately upstream of the H19 promoter that regulates H19 expression (H19PromProx); and (iii) a differentially methylated region at Igf2 that regulates Igf2 expression (Figure 5A).

Figure 5.

Male offspring liver DNA methylation status at the H19/Igf2 domain in response to in utero methyl donor depletion and antibiotic treatment: (A) schematic of H19/Igf2 domain showing expected allelic expression and methylation status for liver. Blue arrows indicate regions assessed for DNA methylation changes, (B) bar graph of methylation levels at the H19/Igf2 ICR at CTCF site 1 (H19CTS1), (C) bar graph of methylation levels at CTCF site 4, (H19CTS4) (D) bar graph of methylation levels at the H19 promoter proximal region, and (E) bar graph of methylation levels at the Igf2DMR1. Sample sizes in order from left to right, B6(+/+) n = 8, 10, 10, 9; ∆2,3(+/+) n = 3, 2, 5, 4; ∆2,3(−/+) n = 3, 2, 1, 2.

In adult offspring liver, which is normally partially methylated (~50%) at the regions tested, due to gain of methylation on the paternal allele during gestation (Figure 5A), we found that methyl donor depletion in utero did not significantly alter liver DNA methylation status at any of the key regulatory regions tested at the H19/Igf2 domain for any of the genotypes (Figure 5B to E). However, a trend of increased DNA methylation was detected for Δ2,3 (+/+) offspring at the ICR CTCF binding site 1 (Figure 5B) and for Δ2,3 (+/+) and (−/+) offspring at the Igf2DMR1 (Figure 5E). Antibiotic treatment in utero similarly had little effect on liver DNA methylation at the H19/Igf2 domain for any of the genotypes (Figure 5B to E). We did detect a subtle but significant decrease in DNA methylation at the ICR CTCF binding site 4 that was induced by antibiotic treatment but only in Δ2,3 (−/+) offspring (63.0 ± 1.6% vs. 60.4 ± 1.7%) (Figure 5C). B6 and wildtype Δ2,3 (+/+) offspring were not similarly affected (Figure 5C).

In adult offspring sperm, which is normally fully methylated (90% to 100%) at the regions tested, due to gain of methylation on both maternal and paternal alleles during gestation (Figure 6A), we found that methyl donor depletion in utero caused a subtle (<2%) but significant decrease in DNA methylation at the ICR CTCF binding site 1 in offspring from B6 (+/+) dams (Figure 6B). However, this effect was not observed for any other regions tested at the H19/Igf2 domain or for any other genotypes(Figure 6B to E). In contrast, a subtle but significant increase in DNA methylation was observed for methyl donor depleted Δ2,3 (+/+) offspring at the Igf2DMR1 (Figure 6E). Other than a subtle significant increase in methylation for Δ2,3 (−/+) offspring (Figure 6B), antibiotic treatment in utero had little effect on sperm DNA methylation at the H19/Igf2 domain for any of the genotypes (Figure 5B to E).

Figure 6.

Male offspring sperm DNA methylation status at the H19/Igf2 domain in response to in utero methyl donor depletion and antibiotic treatment: (A) schematic of H19/Igf2 domain showing expected allelic expression and methylation status for sperm. Blue arrows indicate regions assessed for DNA methylation changes, (B) bar graph of methylation levels at the H19/Igf2 ICR at CTCF site 1 (H19CTS1), (C) bar graph of methylation levels at CTCF site 4 (H19CTS4), (D) bar graph of methylation levels at the H19 promoter proximal region, and (E) bar graph of methylation levels at the Igf2DMR1. Sample sizes in order from left to right, B6(+/+) n = 8, 10, 7, 9; ∆2,3(+/+) n = 3, 2, 5, 3; ∆2,3(−/+) n = 4, 3, 1, 2.

Interestingly, we found that methylation status at the H19/Igf2 domain also differed by mouse line and genotype independent of diet or antibiotic treatment. After adjustment for diet and antibiotic treatment effects in the regression model, Δ2,3 (−/+) mutant offspring liver exhibited significantly higher levels of DNA methylation than B6 (+/+) wildtype offspring at all regulatory regions tested across the H19/Igf2 domain, and significantly higher levels than Δ2,3 (+/+) wildtype offspring at all regions except the distal portion of the H19/Igf2 ICR (H19CTS4) (Figure 7A). On the other hand, Δ2,3 (+/+) wildtype offspring were only significantly higher than B6 (+/+) wildtype offspring at H19CTS4 and the region upstream of the H19 promoter (H19PromProx), while similar methylation levels were observed at the proximal portion of the ICR and Igf2 (H19CTS1 and Igf2DMR1) (Figure 7A). After adjustment for treatment effects, adult body weight was significantly lower in Δ2,3 (−/+) and Δ2,3 (+/+) offspring compared to B6 (+/+) (Figure 7B) and adult body weight was significantly correlated with H19CTS4 methylation levels (Figure 7C). In contrast to liver, genotype had only a subtle effect on the methylation status of sperm at the H19/Igf2 domain, with only a significant increase of <2% at the H19CTS1 locus (Figure 8A) that was not associated with bodyweight (Figure 8B).

Figure 7.

Effect of genotype on adult liver methylation status at the H19/Igf2 domain independent of diet: (A) box and whiskers plot of offspring adult liver methylation levels at regulatory regions across the H19/Igf2 domain, (B) adult (PND56) bodyweight stratified by genotype, and (C) relationship between adult (PND56) bodyweight and DNA methylation at H19CTS4. Each dot represents an individual sample.

Figure 8.

Effect of genotype on adult sperm methylation status at the H19/Igf2 domain: (A) box and whiskers plot of offspring adult sperm methylation levels at regulatory regions across the H19/Igf2 domain, and (B) relationship between adult (PND56) bodyweight and DNA methylation at H19CTS1. Each dot represents an individual sample.

Discussion

Here, we have demonstrated that the H19^ICRΔ2,3 mutation makes mice more susceptible to environmentally-induced developmental effects including adverse pregnancy outcomes, restricted pup growth/development, and subtle loss of DNA methylation in cis. The H19^ICRΔ2,3 mutation mimics microdeletions at the H19/IGF2 ICR observed in BWS patients³⁹ and was previously shown to mimic BWS overgrowth but only in fetal/neonatal stages and only when pups carried the mutation.³¹ Therefore, it was surprising to find that in utero antibiotic treatment caused reduced bodyweight persisting into adulthood for offspring of H19^ICRΔ2,3 dams and that both wildtype and mutant offspring were similarly affected (Figure 4). We also found that antibiotic treatment seemingly had a greater effect on pregnancy outcomes and postnatal development than methyl donor deficiency.

Antibiotic treatment more severely and consistently impaired pregnancy and pup development in a manner that was seemingly independent of methyl deficiency. Both B6 and H19^ICRΔ2,3 dams exhibited reduced littersize at birth and litter survival to weaning (Figure 3A and C). However, similar to methyl deficiency effects, B6 and H19^ICRΔ2,3 offspring exhibited inverse response in pup development, with B6 pups exhibiting increased neonatal and weanling bodyweight and wildtype and mutant H19^ICRΔ2,3 littermates exhibiting decreased bodyweight from neonatal stages into adulthood (Figure 4). Although it is unclear what causes this inverse effect in susceptibility, the similar response of wildtype and mutant H19^ICRΔ2,3 littermates suggests a maternal/intrauterine effect. Furthermore, the finding that mutant H19^ICRΔ2,3 pups exhibited significantly altered ICR methylation (Figure 5), demonstrates that this deletion in cis does sensitize the locus to epigenetic perturbation by in utero environment although the magnitude of effect is very small (<2%) and unlikely to alter gene expression. The fact that wildtype littermates were not similarly affected, suggests this epimutation alone is not sufficient to explain the pup growth/developmental phenotype.

Independent of treatment, adult mice carrying the H19^ICRΔ2,3 mutation inherited from the dam exhibited increased DNA methylation in cis in the liver at regions both upstream and downstream of the deletion (Figure 7A). Interestingly, their wildtype littermates also exhibited higher methylation at the ICR than wildtype B6 mice but only at regions downstream of the deletion (Figure 7A). This finding of altered adult ICR methylation changes, not previously found at neonatal stages, strongly suggests the H19^ICRΔ2,3 mutation has delayed onset effects on methylation and possibly even imprinting at the locus. Furthermore, ICR epigenetic perturbation of both wildtype and mutant pups of heterozygous H19^ICRΔ2,3/+ dams suggests the potential for a maternal/intrauterine effect of this mutation on imprinting defects at this locus. This could explain why both wildtype and mutant offspring of heterozygous H19^ICRΔ2,3/+ dams in this study exhibited similar susceptibility to phenotypic response to methyl deficiency and antibiotic treatment. Interestingly, adult H19^ICRΔ2,3/+ offspring actually weighed less than B6 controls despite ~10% higher Igf2DMR1 methylation levels (Figure 7). Higher perinatal Igf2DMR1 methylation is usually associated with increased Igf2 expression and subsequent growth in fetal and neonatal stages.³¹ The role of Igf2 in postnatal bodyweight remains poorly understood, however, previous studies show upregulation of postnatal Igf2 can result in reduced bodyweight.⁵⁶

We were surprised to find that despite more than 9 weeks of dietary depletion of methyl donor nutrients folate, choline, and methionine, and a more than 80% reduction in maternal plasma folate levels, methyl deficiency in this model seemingly had no effect on methylation potential (SAM availability) and only subtle effects on pregnancy and pup development, primarily observed for H19^ICRΔ2,3 offspring. It is possible that a longer feeding time is necessary or further depletion of redundant methyl donors is necessary to alter methylation potential. Based on the minimal phenotypic response to methyl deficiency, it was not surprising that there was no significant effect of methyl deficiency on DNA methylation at the locus (Figure 5E). Although there was a nonsignificant trend of gain of methylation at the Igf2DMR1 for methyl deficient H19^ICRΔ2,3 offspring, this epimutation is usually associated with increased Igf2 expression and increased fetal growth.³¹ In contrast, methyl deficient H19^ICRΔ2,3 offspring actually exhibited decreased neonatal bodyweight (Figure 4A) suggesting the 2 outcomes are likely not linked. On the other hand, methyl deficient wildtype B6 offspring that also showed no change in methylation at the H19/Igf2 domain, exhibited normal neonatal bodyweight but had increased bodyweight at weaning (PND21). This finding suggests the interactive gene x diet effects detected here are more complex than just the result of imprinting defects in the offspring implicating a potential role for maternal/intrauterine effects.

Taken together, this data supports the role of genetic alteration at the H19/Igf2 ICR in susceptibility to in utero environmental perturbation. Furthermore, adverse developmental effects detected here likely occur as a result of maternal and possibly intrauterine effects of these mutations. Although the exact mechanism of maternal influence remains unclear, previous studies show that reduced Igf2 expression in the placenta leads to reduced potential for placental nutrient transport to the fetus and subsequent growth restriction of the fetus.⁵⁷ Likewise, H19 expression in the placenta also plays an important role in placental growth and subsequent function, in part by restricting growth of the placenta.⁵⁸ Future studies will test whether these mechanisms are at play in this model. This data also show for the first time that treatment with 1% succinylsulfathiozole antibiotic during mouse pregnancy restricts pup growth/development beyond the perinatal period. Growth restriction has been linked to increased mortality,⁵⁹ neurodevelopmental disorders,⁶⁰ and metabolic disorders⁶¹ later in life. This model can also be used to further understand the role of antibiotic in these outcomes.

Materials and Methods

Animals – sources, housing dietary treatments, and breeding scheme

Animal handling was performed in accordance with the Guide for the Care and Use of Laboratory Animals under the corresponding animal use protocol at the University of North Carolina at Chapel Hill and housed at the David H. Murdock Institute Vivarium in Kannapolis, NC. The mouse lines used were (i) wild-type fully inbred C57BL/6 (B6) mice, (ii) wild-type mice homozygous for CAST/EiJ across chromosome 7 and mixed CAST/EiJ-B6 genetic background (C7), and (iii) previously³¹ described mutant H19^ICRΔ2,3 mice carrying a targeted deletion of 1.3 kb at the H19/Igf2 imprinting control region on a mostly (>98%) B6 genetic background. As previously described,⁶² C7 sires used here carry CAST genetic sequence at the ICR versus the B6 sequence of the dams and were used in the experimental design for the purpose of distinguishing parental origin of the ICR alleles, should substantial epigenetic changes suggest an imprinting error has occurred. B6 mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and C7 and H19^ICRΔ2,3 mice were derived in Dr. Marisa Bartolomei’s lab at the University of Pennsylvania. All were bred in-house at the David H. Murdoch Institute Vivarium in Kannapolis to generate experimental animals. Sterilized water and rodent chow were fed ad libitum. Vivarium temperature was maintained between 21 and 23 °C with a 12-h light cycle. Crosses and treatment schemes are illustrated in Figure 1. Dams were trio bred with stud males. Dams were treated 5 weeks before pregnancy, during pregnancy, and to postnatal day 5 (PND 5) with 1 of 4 purified pelleted diets purchased from Envigo (formerly Harlan/Teklad): (1)Methyl Sufficient (MS, TD.130225, ~0.6% methionine; ~2.5 g/kg choline; ~3 ppm folic acid), (2)MS + 1%SS (modified TD.130225 with addition of 1% succinylsulfathiozole), (3)Methyl Deficient (MD, TD.130224, ~0.4% methionine, no added choline, and ~0.2 ppm folic acid), and (4)MD + 1%SS (modified TD.130224 with addition of 1% succinylsulfathiozole).

Litter handling and sample collections

To reduce maternal stress and improve fecundity and pup survival, breeding cages, and litters were minimally disturbed until weaning. Therefore, we waited until PND5 to sex pups (using sexually dimorphic genitalia) and alter the diet. Since the goal of this study is to study developmentally programmed effects that persist after restoration of normal environmental conditions, at PND5 dams (and pups) were switched to standard chow diet (TD. 8604, ~0.4% methionine, ~2.5 g/kg choline, 3 ppm folate, Envigo, formerly Harlan/Teklad) (Figure 1). All offspring were weaned onto TD. 8604 and remained until the end of the study Litter size was measured at birth and weaning. Pups were genotyped at weaning using previously described assays,⁶³ and ear tagged for identification into adulthood. Adult G1 pups were euthanized at 8 weeks of age in accordance with current recommendations by the American Veterinary Medical Association (AVMA) guidelines. Fresh tissues and plasma were collected immediately after euthanasia and flash frozen in liquid nitrogen and stored at −80°C.

Body weight

Offspring bodyweight was measured by analytical balance at PND5, PND21, and PND56.

Plasma folate, SAM, and SAH measurements

Plasma were collected from wild-type and mutant dams by cardiac puncture immediately after CO₂ exposure in heparin coated tubes. Plasma levels after dietary depletion, equivalent to maternal status at birth, was obtained from dams that did not produce a live litter and were on the diet for ~9 weeks (Figure 2A). Plasma levels after dietary repletion, equivalent to status after 16 days of methyl sufficient diet, was obtained from dams with litters that were euthanized at weaning (Figure 2A). Folate concentrations were measured by microbiological assay, which measures growth response of the bacteria Lactobacillus casei (ATCC 7469) to folic acid.⁶⁴ S-adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH) were measured in 50 mg of liver first pulverized in liquid nitrogen with mortar and pestle and stored at −80°C. SAM and SAH concentrations were measured in liver homogenates using Ultra Performance Liquid Chromatography. All assays were performed at the UNC Nutrition Obesity Research Center Metabolic Molecular Phenotyping Core.

DNA extraction and methylation detection by pyrosequencing

Genomic DNA was isolated from frozen liver and sperm using an automated magnetic-bead based nucleic acid extraction system (Anaprep 12). DNA quality and concentration was assessed by NanoDrop 2000 spectrometer (Thermo Scientific, DE). Double stranded DNA was measured by Quant-it PicoGreen dsDNA assay (#P7589, Life technologies, NY). Genomic DNA was bisulfite-converted using the EZ DNA methylation Gold kit according to the manufacturer’s protocol (#D5006, Zymo research, CA). All pyrosequencing PCR and sequencing primers and sequenced regions were devoid of SNPs or deletions. Pyrosequencing was performed using the Pyromark Q96MD instrument according to the manufacturer’s instructions (Qiagen, MD). Assays for all loci have been previously described, including DMR1 at the insulin-like growth factor 2 gene (Igf2DMR1),⁵³ the H19/Igf2 ICR H19 CTCF site 1 (H19CTS1, previously H19ICR_Rp1),⁶⁵ and CTCF site 4 (H19CTS4, previously H19Cbs4)⁶⁶ and a DMR upstream of the H19 gene promoter (H19 promoter proximal H19PromProx).⁵³ The average methylation of all CpGs combined within the assayed region are presented as mean methylation (%). The number of CpGs assayed for each region was 3 (Igf2DMR1), 6 (H19CTS1), 5 (H19CTS4), and 3 (H19PromProx).

Statistical analysis

Statistical analyses were performed in JMP Pro software version 12.2.0 (SAS, NC). Where applicable, normality of data distribution was confirmed by Shapiro-Wilk goodness-of-fit test and variance tested by Bartlett’s test. Data that met assumptions of normality were analyzed by parametric tests (t-test, ANOVA, linear regression), whereas data that did not meet assumptions of normality were analyzed by nonparametric tests (Wilcoxon/Kruskal Wallis test) as indicated in each figure legend. For all comparisons, P-value <.05 are considered statistically significant. All error bars and “± values” represent standard error of the mean. Where figures indicate “Diet” and “Anti” we measured the (i) Methyl donor depletion effect (Diet) determined by comparing MS + MS1%SS versus MD + MD1%SS samples; and (ii) Antibiotic treatment effect determined by comparing MS + MD versus MS1%SS + MD1%SS samples.

Footnotes

Acknowledgements

We thank Dr. Marisa Bartolomei for the H19^ICRΔ2,3 mouse lines and the C7 mice; Dr. Mihai Niculescu (UNC Nutrition Research Institute) for advice on the methyl donor deficiency mouse model; and Dr. Steve Orena at the Nutrition Obesity Research Center Metabolic Molecular Phenotyping Core for measuring folate, SAM, and SAH levels.

Funding:

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: NIEHS Transition to Independent Environmental Health Research Career Development Award [KES023849A] and NIEHS funded UNC Center for Environmental Health and Susceptibility Award [5P30ES010126] to F.Y.I.

Declaration of conflicting interests:

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Author Contributions

Study design was conceived and supervised by FI. Animal and molecular experiments and preliminary data analyses were performed by JO and AP (with sperm methylation data generated by ME). Final data analyses were performed by FI. Manuscript was drafted and revised by AP and FI.

ORCID iD

Folami Y Ideraabdullah

References

Cavalcante-Silva

Fernandes

Haseyama

Agamme

Muniz

D’Almeida

Long-term consequences of methyl donor deficiency during in utero and early life development on markers of the metabolic syndrome. J Nutr Food Sci. Published online September 16, 2016. doi:10.4172/2155-9600.1000555

McGee

Bainbridge

Fontaine-Bisson

A crucial role for maternal dietary methyl donor intake in epigenetic programming and fetal growth outcomes. Nutr Rev. 2018;76:469-478.

Waterland

Michels

KB.

Epigenetic epidemiology of the developmental origins hypothesis. Annu Rev Nutr. 2007;27:363-388.

Yajnik

Deshmukh

US.

Maternal nutrition, intrauterine programming and consequential risks in the offspring. Rev Endocr Metab Disord. 2008;9:203-211.

Korsmo

Jiang

Caudill

MA.

Choline: exploring the growing science on its benefits for moms and babies. Nutrients. Published online August 7, 2019. doi:10.3390/nu11081823

Zeisel

SH.

Importance of methyl donors during reproduction. Am J Clin Nutr. 2009;89:673S-677S.

McKee

Reyes

TM.

Effect of supplementation with methyl-donor nutrients on neurodevelopment and cognition: considerations for future research. Nutr Rev. 2018;76:497-511.

Tulchinsky

TH.

Micronutrient deficiency conditions: global health issues. Public Health Rev. 2010;32:243-255.

Qvist

Abdulla

Jägerstad

Svensson

Iron, zinc and folate status during pregnancy and two months after delivery. Acta Obstet Gynecol Scand. 1986;65:15-22.

10.

Fekete

Berti

Trovato

, et al. Effect of folate intake on health outcomes in pregnancy: a systematic review and meta-analysis on birth weight, placental weight and length of gestation. Nutr J. Published online September 19, 2012. doi:10.1186/1475-2891-11-75

11.

Blumfield

Hure

Macdonald-Wicks

Smith

Collins

CE.

A systematic review and meta-analysis of micronutrient intakes during pregnancy in developed countries. Nutr Rev. 2013;71:118-132.

12.

Obeid

The metabolic burden of methyl donor deficiency with focus on the betaine homocysteine methyltransferase pathway. Nutrients. 2013;5:3481-3495.

13.

Ideraabdullah

Zeisel

SH.

Dietary modulation of the epigenome. Physiol Rev. 2018;98:667-695.

14.

Jones

PA.

Functions of DNA methylation: islands, start sites, gene bodies and beyond. Nat Rev Genet. 2012;13:484-492.

15.

Messerschmidt

Knowles

Solter

DNA methylation dynamics during epigenetic reprogramming in the germline and preimplantation embryos. Genes Dev. 2014;28:812-828.

16.

Morgan

Santos

Green

Dean

Reik

Epigenetic reprogramming in mammals. Hum Mol Genet. 2005;14 Spec No 1:R47-R58.

17.

Reik

Dean

Walter

Epigenetic reprogramming in mammalian development. Science. 2001;293:1089-1093.

18.

Chen

Dent

SYR

. Chromatin modifiers and remodellers: regulators of cellular differentiation. Nat Rev Genet. 2014;15:93-106.

19.

McKay

Waltham

Williams

Mathers

JC.

Folate depletion during pregnancy and lactation reduces genomic DNA methylation in murine adult offspring. Genes Nutr. 2011;6:189-196.

20.

Sinclair

Allegrucci

Singh

, et al. DNA methylation, insulin resistance, and blood pressure in offspring determined by maternal periconceptional B vitamin and methionine status. Proc Natl Acad Sci USA. 2007;104:19351-19356.

21.

Niculescu

Craciunescu

Zeisel

SH.

Dietary choline deficiency alters global and gene-specific DNA methylation in the developing hippocampus of mouse fetal brains. FASEB J. 2006;20:43-49.

22.

Kovacheva

Mellott

Davison

, et al. Gestational choline deficiency causes global and Igf2 gene DNA hypermethylation by up-regulation of Dnmt1 expression. J Biol Chem. 2007;282:31777-31788.

23.

Pembrey

Imprinting and transgenerational modulation of gene expression; human growth as a model. Acta Genet Med Gemellol (Roma). 1996;45:111-125.

24.

Thompson

Konfortova

Gregory

Reik

Dean

Feil

Environmental effects on genomic imprinting in mammals. Toxicol Lett. 2001;120:143-150.

25.

Waterland

RA.

Epigenetic mechanisms affecting regulation of energy balance: many questions, few answers. Annu Rev Nutr. 2014;34:337-355.

26.

Dolinoy

DC.

The agouti mouse model: an epigenetic biosensor for nutritional and environmental alterations on the fetal epigenome. Nutr Rev. 2008;66 Suppl 1:S7-S11.

27.

Ideraabdullah

Vigneau

Bartolomei

MS.

Genomic imprinting mechanisms in mammals. Mutat Res. 2008;647:77-85.

28.

Sparago

Russo

Cerrato

, et al. Mechanisms causing imprinting defects in familial Beckwith-Wiedemann syndrome with Wilms’ tumour. Hum Mol Genet. 2007;16:254-264.

29.

Nativio

Sparago

Ito

Weksberg

Riccio

Murrell

Disruption of genomic neighbourhood at the imprinted IGF2-H19 locus in Beckwith-Wiedemann syndrome and Silver-Russell syndrome. Hum Mol Genet. 2011;20:1363-1374.

30.

Beygo

Citro

Sparago

, et al. The molecular function and clinical phenotype of partial deletions of the IGF2/H19 imprinting control region depends on the spatial arrangement of the remaining CTCF-binding sites. Hum Mol Genet. 2013;22:544-557.

31.

Ideraabdullah

Thorvaldsen

Myers

Bartolomei

MS.

Tissue-specific insulator function at H19/Igf2 revealed by deletions at the imprinting control region. Hum Mol Genet. 2014;23:6246-6259.

32.

Agrogiannis

Sifakis

Patsouris

Konstantinidou

AE.

Insulin-like growth factors in embryonic and fetal growth and skeletal development (Review). Mol Med Rep. 2014;10:579-584.

33.

Gabory

Ripoche

Yoshimizu

Dandolo

The H19 gene: regulation and function of a non-coding RNA. Cytogenet Genome Res. 2006;113:188-193.

34.

Choufani

Shuman

Weksberg

Beckwith-Wiedemann syndrome. Am J Med Genet C Semin Med Genet. 2010;154C:343-354.

35.

Soejima

Higashimoto

Epigenetic and genetic alterations of the imprinting disorder Beckwith-Wiedemann syndrome and related disorders. J Hum Genet. 2013;58:402-409.

36.

Weksberg

Smith

Squire

Sadowski

Beckwith-Wiedemann syndrome demonstrates a role for epigenetic control of normal development. Hum Mol Genet. 2003;12:R61-R8.

37.

Reik

Brown

Slatter

Sartori

Elliott

Maher

ER.

Allelic methylation of H19 and IGF2 in the Beckwith-Wiedemann syndrome. Hum Mol Genet. 1994;3:1297-1301.

38.

De Crescenzo

Coppola

Falco

, et al. A novel microdeletion in the IGF2/H19 imprinting centre region defines a recurrent mutation mechanism in familial Beckwith-Wiedemann syndrome. Eur J Med Genet. 2011;54:e451-e454.

39.

Sparago

Cerrato

Vernucci

Ferrero

Silengo

Riccio

Microdeletions in the human H19 DMR result in loss of IGF2 imprinting and Beckwith-Wiedemann syndrome. Nat Genet. 2004;36:958-960.

40.

Prawitt

Enklaar

Gärtner-Rupprecht

, et al. Microdeletion of target sites for insulator protein CTCF in a chromosome 11p15 imprinting center in Beckwith-Wiedemann syndrome and Wilms’ tumor. Proc Natl Acad Sci USA. 2005;102:4085-4090.

41.

Prawitt

Enklaar

Gärtner-Rupprecht

, et al. Microdeletion and IGF2 loss of imprinting in a cascade causing Beckwith-Wiedemann syndrome with Wilms’ tumor. Nat Genet. 2005;37:785-786.

42.

Freschi

Hur

Valente

, et al. Tissue-specific and mosaic imprinting defects underlie opposite congenital growth disorders in mice. PLoS Genet. 2018;14:e1007243.

43.

Burgoon

Selhub

Nadeau

Sadler

TW.

Investigation of the effects of folate deficiency on embryonic development through the establishment of a folate deficient mouse model. Teratology. 2002;65:219-227.

44.

Baothman

Zamzami

Taher

Abubaker

Abu-Farha

The role of gut microbiota in the development of obesity and diabetes. Lipids Health Dis. 2016;15:108.

45.

Carding

Verbeke

Vipond

Corfe

Owen

LJ.

Dysbiosis of the gut microbiota in disease. Microb Ecol Health Dis. 2015;26:26191.

46.

Hur

Lee

M-S.

Gut microbiota and metabolic disorders. Diabetes Metab J. 2015;39:198-203.

47.

Zhang

Y-J

Gan

R-Y

Zhou

D-P

H-B.

Impacts of gut bacteria on human health and diseases. Int J Mol Sci. 2015;16:7493-7519.

48.

Mahmoud

Ali

MM.

Methyl donor micronutrients that modify DNA methylation and cancer outcome. Nutrients. Published online March 13, 2019. doi:10.3390/nu11030608

49.

Xia

Zhang

, et al. Daily variation in global and local DNA methylation in mouse livers. PLoS One. 2015;10:e0118101.

50.

Duthie

Narayanan

Brand

Grant

DNA stability and genomic methylation status in colonocytes isolated from methyl-donor-deficient rats. Eur J Nutr. 2000;39:106-111.

51.

Wasson

McGlynn

McNulty

, et al. Global DNA and p53 region-specific hypomethylation in human colonic cells is induced by folate depletion and reversed by folate supplementation. J Nutr. 2006;136:2748-2753.

52.

Poirier

LA.

The effects of diet, genetics and chemicals on toxicity and aberrant DNA methylation: an introduction. J Nutr. 2002;132:2336S-2339S.

53.

Pogribny

Basnakian

Miller

Lopatina

Poirier

James

SJ.

Breaks in genomic DNA and within the p53 gene are associated with hypomethylation in livers of folate/methyl-deficient rats. Cancer Res. 1995;55:1894-1901.

54.

Zeisel

da Costa

K-A.

Choline: an essential nutrient for public health. Nutr Rev. 2009;67:615-623.

55.

Lee

H-S.

Impact of maternal diet on the epigenome during in utero life and the developmental programming of diseases in childhood and adulthood. Nutrients. 2015;7:9492-9507.

56.

Wolf

Hoeflich

Lahm

What is the function of IGF-II in postnatal life? Answers from transgenic mouse models. Growth Horm IGF Res. 1998;8:185-193.

57.

Sibley

Coan

Ferguson-Smith

, et al. Placental-specific insulin-like growth factor 2 (Igf2) regulates the diffusional exchange characteristics of the mouse placenta. Proc Natl Acad Sci USA. 2004;101:8204-8208.

58.

Keniry

Oxley

Monnier

, et al. The H19 lincRNA is a developmental reservoir of miR-675 that suppresses growth and Igf1r. Nat Cell Biol. 2012;14:659-665.

59.

Hack

Merkatz

IR.

Preterm delivery and low birth weight – a dire legacy. N Engl J Med. 1995;333:1772-1774.

60.

Kjellmer

Liedholm

Sultan

Wennergren

Wallin Götborg

Thordstein

Long-term effects of intrauterine growth retardation. Acta Paediatr Suppl. 1997;422:83-84.

61.

Bertram

CE.

Animal models and programming of the metabolic syndrome. Br Med Bull. 2001;60:103-121.

62.

Mann

MRW

Chung

Nolen

Verona

Latham

Bartolomei

. Disruption of imprinted gene methylation and expression in cloned preimplantation stage mouse embryos. Biol Reprod. 2003;69:902-914.

63.

Ideraabdullah

Abramowitz

Thorvaldsen

, et al. Novel cis-regulatory function in ICR-mediated imprinted repression of H19. Dev Biol. 2011;355:349-357.

64.

Horne

Patterson

Lactobacillus casei microbiological assay of folic acid derivatives in 96-well microtiter plates. Clin Chem. 1988;34:2357-2359.

65.

Pietryk

Clement

Elnagheeb

, et al. Intergenerational response to the endocrine disruptor vinclozolin is influenced by maternal genotype and crossing scheme. Reprod Toxicol. 2018;78:9-19.

66.

Xue

Schoenrock

Valdar

Tarantino

Ideraabdullah

FY.

Maternal vitamin D depletion alters DNA methylation at imprinted loci in multiple generations. Clin Epigenetics. 2016;8:107.

Maternal Microdeletion at the H19/Igf2 ICR in Mice Increases Offspring Susceptibility to In Utero Environmental Perturbation

Abstract

Keywords

Introduction

Results

Maternal dietary depletion of methyl donors reduced maternal plasma folate levels but not SAM availability

Dams carrying the Δ2,3 mutation are more susceptible to adverse pregnancy outcomes caused by methyl donor depletion and antibiotic treatment

Offspring of Δ2,3 dams are more susceptible to developmental delay/growth restriction response to antibiotic treatment

Offspring from H19ICRΔ2,3 dams are more susceptible to subtle epigenetic perturbation in cis

Discussion

Materials and Methods

Animals – sources, housing dietary treatments, and breeding scheme

Litter handling and sample collections

Body weight

Plasma folate, SAM, and SAH measurements

DNA extraction and methylation detection by pyrosequencing

Statistical analysis

Footnotes

Acknowledgements

Funding:

Declaration of conflicting interests:

Author Contributions

ORCID iD

References

Offspring from H19^ICRΔ2,3 dams are more susceptible to subtle epigenetic perturbation in cis