I've discussed the nutritional epigenetics work done in Gambia before here. New work following up on the previous Gambian study from Robert Waterland and colleagues was recently published (1) that expanded upon their previous findings, looking at newly established metastable epialleles.
As a quick reminder before I get into the study findings, rural Gambia is characterized by a rainy and dry season that leads to variation in nutrient availability and workload/physical activity. The rainy season occurs from July to November, during which pregnant mothers in rural areas gain 400-500g/mo less than weight gain during the dry season, and birth weight is, on average, 90g lower in the rainy than dry season. Seasonal patterns of maternal weight gain closely parallel the rate of Small for Gestational Age (SGA) births; LBW parallels with increases in agricultural labor and malarial infections (2). The original hypothesis in the last Gambian study stated that there should be higher percentage methylation at candidate metastable epialleles (MEs) during the dry harvest season (less PEM) - the opposite was found, however. This led to the suggestion that methyl donor availability, which is higher during the rainy season, contributes to greater average levels of methylation, despite reduced calorie intake.
In this newest publication, The authors attempted to understand:
1. whether methylation of 7 previously identified ME's varied according to season
2. whether plasma biomarkers associated with 1-c metabolism were predictive of ME methylation
3. whether the changes in the availability of methyl donors explain the seasonal difference in ME methylation.
Interestingly (and for feasibility's sake), the study did not actually take blood samples or collect dietary intakes from the mothers of infants directly. An indicator group of non-pregnant women (n=30) had dietary intake (48hr) and plasma biomarkers assessed monthly over a full calendar year (June-June). These biomarkers were methionine, choline, betaine, folate, B2 (riboflavin),6 (pyridoxine),12, and Dimethylglycine/DMG (biproduct of betaine donating a methyl group to homocysteine), SAM, SAH, homocysteine, and cysteine. At the same time, another group of women (n=2040) were followed while non-pregnant and visited monthly until a missed menses was reported. Women were classified as either having conceived at the peak of the rainy season (July-Sep) or the peak of the dry season (Feb-April). 166 women conceived during this time.
Maternal periconceptional concentrations of Folate, b2, methionine, betaine, SAM:SAH, and betaine:DMG were higher in the rainy season, whereas b12, DMG, homocysteine and SAH were lower. This is interesting that you have higher betaine concentrations yet lower DMG - is betaine not being used as a methyl donor? It's been shown that there are compensatory changes in methyl donor contribution depending on relative availability of each (12). The fact that DMG was lower in the rainy than dry season likely points to lower folate intakes and increased need for using choline/betaine as a methyl donor.
The methylation sites, identified, previously, as ME's, were BOLA3, LOC654433, EXD3, ZFYVE28, RBM46, PARD6G, and ZNF678. The percentage methylation was looked at in peripheral blood lymphocytes (PBL) and hair follicle (HF) samples. PARD6G was consistently hyper-methylated in HF and was ultimately excluded. The effects of season of conception on DNA methylation were highly correlated in the two tissues. Offspring of the rainy season conceptions had higher levels of CpG methylation at the other 6 ME's in PBL than dry season conceptions.
After adjusting for sex effects, maternal BMI, vitamin B2, homocysteine and cysteine concentrations at the time of conception predicted mean ME methylation in both PBL and HF DNA of infants. Increases in maternal homocysteine and cysteine predicted decreased systemic methylation in both tissue types, whereas b2 predicted increase methylation. This is as one would expect - having high homocysteine levels would be reflective of low methyl donor availability to remethylate homocysteine to methionine.
This data is exciting because, for the first time, a significant effect of maternal diet and methyl donor supply was suggested to predict methylation at several loci. What we can't determine yet is if there is any phenotypic outcome due to methyl donor availability. There are noted phenotypic effects in the offspring, related to the SGA and its potential metabolic outcomes, but we can't parse out the effects of methyl donor supply vs reduced calorie intake/IUGR.
I couldn't seem to tell from the wording whether folate was tested from plasma or red blood cells - plasma reflects recent consumption and RBCs tell us about long term consumption. It sounds like they used plasma, which would be disappointing.
If you're a fan of stats, the statistical analyses are a good read - the authors performed some interesting transformations to back extrapolate the biomarker concentrations from the indicator group of women to predict those of the women who actually gave birth.
One thing I find interesting in all of this, that i would like to see looked at in the future, is dietary intakes of methionine and serine. It's been discussed before that serine is an underappreciated methyl donor (11), and serum markers of serine metabolism would add an interesting understanding to this picture. For anyone interested in a good read on dietary methyl donors and the relative contributions of different methyl donors, see here.
If you're curious like I was, the gene functions are as follows:
BOLA3 - involved in assembly of mitochondrial respiratory chain complexes; mutations lead to mitochondrial dysfunction (3)
LOC654433 - unknown function; has 25 mRNA variants. it's closest homolog is Pax8 in mice, which has been studied for its role in a huge number of cellular processes and pathologies. (4)
EXD3 - exhibits exonuclease activity, associated with Werner syndrome (5)
ZFYVE28 - encodes a zinc finger protein thought to negatively regulate epidermal growth factor pathways (6,7)
RBM46 - encodes an RNA binding motif (8), which has been shown to be upregulated in omental adipose tissue (9)
ZNF678 - this gene encodes a zinc finger protein that was identified in a GWAS as being 1 of 20 loci influencing adult height (10)
My mind isn't painting any stellar pictures, maybe abstract art at best ^^
1. http://www.nature.com/ncomms/2014/140429/ncomms4746/full/ncomms4746.html. Dominguez-Salas, P. et al. Maternal nutrition at conception modulates DNA methylation of human metastable epialleles. Nat. Commun. 5:3746 doi: 10.1038/ncomms4746 (2014).
2. http://ajcn.nutrition.org/content/81/1/134.full
3. http://www.genecards.org/cgi-bin/carddisp.pl?gene=BOLA3
4. http://www.ncbi.nlm.nih.gov/IEB/Research/Acembly/av.cgi?db=human&l=LOC654433
5. http://www.ncbi.nlm.nih.gov/gene/54932
6. http://www.ncbi.nlm.nih.gov/gene/57732
7. http://www.ncbi.nlm.nih.gov/pubmed/19460345
8. http://www.ncbi.nlm.nih.gov/gene/166863
9. http://www.ncbi.nlm.nih.gov/geoprofiles/64961202
10.http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2681221
11. http://www.asbmb.org/asbmbtoday/asbmbtoday_article.aspx?id=18036
12. http://jn.nutrition.org/content/132/8/2333S.long
As a quick reminder before I get into the study findings, rural Gambia is characterized by a rainy and dry season that leads to variation in nutrient availability and workload/physical activity. The rainy season occurs from July to November, during which pregnant mothers in rural areas gain 400-500g/mo less than weight gain during the dry season, and birth weight is, on average, 90g lower in the rainy than dry season. Seasonal patterns of maternal weight gain closely parallel the rate of Small for Gestational Age (SGA) births; LBW parallels with increases in agricultural labor and malarial infections (2). The original hypothesis in the last Gambian study stated that there should be higher percentage methylation at candidate metastable epialleles (MEs) during the dry harvest season (less PEM) - the opposite was found, however. This led to the suggestion that methyl donor availability, which is higher during the rainy season, contributes to greater average levels of methylation, despite reduced calorie intake.
In this newest publication, The authors attempted to understand:1. whether methylation of 7 previously identified ME's varied according to season
2. whether plasma biomarkers associated with 1-c metabolism were predictive of ME methylation
3. whether the changes in the availability of methyl donors explain the seasonal difference in ME methylation.
Interestingly (and for feasibility's sake), the study did not actually take blood samples or collect dietary intakes from the mothers of infants directly. An indicator group of non-pregnant women (n=30) had dietary intake (48hr) and plasma biomarkers assessed monthly over a full calendar year (June-June). These biomarkers were methionine, choline, betaine, folate, B2 (riboflavin),6 (pyridoxine),12, and Dimethylglycine/DMG (biproduct of betaine donating a methyl group to homocysteine), SAM, SAH, homocysteine, and cysteine. At the same time, another group of women (n=2040) were followed while non-pregnant and visited monthly until a missed menses was reported. Women were classified as either having conceived at the peak of the rainy season (July-Sep) or the peak of the dry season (Feb-April). 166 women conceived during this time.
Maternal periconceptional concentrations of Folate, b2, methionine, betaine, SAM:SAH, and betaine:DMG were higher in the rainy season, whereas b12, DMG, homocysteine and SAH were lower. This is interesting that you have higher betaine concentrations yet lower DMG - is betaine not being used as a methyl donor? It's been shown that there are compensatory changes in methyl donor contribution depending on relative availability of each (12). The fact that DMG was lower in the rainy than dry season likely points to lower folate intakes and increased need for using choline/betaine as a methyl donor.
The methylation sites, identified, previously, as ME's, were BOLA3, LOC654433, EXD3, ZFYVE28, RBM46, PARD6G, and ZNF678. The percentage methylation was looked at in peripheral blood lymphocytes (PBL) and hair follicle (HF) samples. PARD6G was consistently hyper-methylated in HF and was ultimately excluded. The effects of season of conception on DNA methylation were highly correlated in the two tissues. Offspring of the rainy season conceptions had higher levels of CpG methylation at the other 6 ME's in PBL than dry season conceptions.
After adjusting for sex effects, maternal BMI, vitamin B2, homocysteine and cysteine concentrations at the time of conception predicted mean ME methylation in both PBL and HF DNA of infants. Increases in maternal homocysteine and cysteine predicted decreased systemic methylation in both tissue types, whereas b2 predicted increase methylation. This is as one would expect - having high homocysteine levels would be reflective of low methyl donor availability to remethylate homocysteine to methionine.
This data is exciting because, for the first time, a significant effect of maternal diet and methyl donor supply was suggested to predict methylation at several loci. What we can't determine yet is if there is any phenotypic outcome due to methyl donor availability. There are noted phenotypic effects in the offspring, related to the SGA and its potential metabolic outcomes, but we can't parse out the effects of methyl donor supply vs reduced calorie intake/IUGR.
I couldn't seem to tell from the wording whether folate was tested from plasma or red blood cells - plasma reflects recent consumption and RBCs tell us about long term consumption. It sounds like they used plasma, which would be disappointing.
If you're a fan of stats, the statistical analyses are a good read - the authors performed some interesting transformations to back extrapolate the biomarker concentrations from the indicator group of women to predict those of the women who actually gave birth.
One thing I find interesting in all of this, that i would like to see looked at in the future, is dietary intakes of methionine and serine. It's been discussed before that serine is an underappreciated methyl donor (11), and serum markers of serine metabolism would add an interesting understanding to this picture. For anyone interested in a good read on dietary methyl donors and the relative contributions of different methyl donors, see here.
If you're curious like I was, the gene functions are as follows:
BOLA3 - involved in assembly of mitochondrial respiratory chain complexes; mutations lead to mitochondrial dysfunction (3)
LOC654433 - unknown function; has 25 mRNA variants. it's closest homolog is Pax8 in mice, which has been studied for its role in a huge number of cellular processes and pathologies. (4)
EXD3 - exhibits exonuclease activity, associated with Werner syndrome (5)
ZFYVE28 - encodes a zinc finger protein thought to negatively regulate epidermal growth factor pathways (6,7)
RBM46 - encodes an RNA binding motif (8), which has been shown to be upregulated in omental adipose tissue (9)
ZNF678 - this gene encodes a zinc finger protein that was identified in a GWAS as being 1 of 20 loci influencing adult height (10)
My mind isn't painting any stellar pictures, maybe abstract art at best ^^
1. http://www.nature.com/ncomms/2014/140429/ncomms4746/full/ncomms4746.html. Dominguez-Salas, P. et al. Maternal nutrition at conception modulates DNA methylation of human metastable epialleles. Nat. Commun. 5:3746 doi: 10.1038/ncomms4746 (2014).
2. http://ajcn.nutrition.org/content/81/1/134.full
3. http://www.genecards.org/cgi-bin/carddisp.pl?gene=BOLA3
4. http://www.ncbi.nlm.nih.gov/IEB/Research/Acembly/av.cgi?db=human&l=LOC654433
5. http://www.ncbi.nlm.nih.gov/gene/54932
6. http://www.ncbi.nlm.nih.gov/gene/57732
7. http://www.ncbi.nlm.nih.gov/pubmed/19460345
8. http://www.ncbi.nlm.nih.gov/gene/166863
9. http://www.ncbi.nlm.nih.gov/geoprofiles/64961202
10.http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2681221
11. http://www.asbmb.org/asbmbtoday/asbmbtoday_article.aspx?id=18036
12. http://jn.nutrition.org/content/132/8/2333S.long
Comments
Post a Comment