This topic is an intriguing gene-environment (GxE) interactions example. The “environment” in this case is maternal obesity, or maternal diet. The “genes affected” include Dppa3 (Stella) which was previously known to participate as a maternal-effect gene in mouse embryos. DPPA3 was remarkable because of its interesting expression pattern –– in which it is strictly confined to the germ line and early preimplantation embryo. Interestingly, whereas Dppa3(–/–) knockout mice of both sexes are overtly healthy, and males have normal fertility, Dppa3(–/–) oocytes are severely compromised in their ability to support normal development, even when fertilized with wild-type sperm. (This sex difference strongly implies that epigenetic imprinting is involved.)
Investigators have begun to elucidate the molecular mechanism of this defect. DPPA3 protein binds H3K9me2 in the maternal pronucleus of the zygote (fertilized egg), blocking access to TET3 (an enzyme, tet methylcytosine dioxygenase-3) and thereby preventing oxidation of maternally-inherited 5-methylcytosine (5mC) to form 5-hydroxymethylcytosine (5hmC). Consequently, in the absence of DPPA3, there is a loss of 5mC at critically important maternally methylated sequences in the zygote and a notable transcriptional and epigenetic dysregulation (remember: “EPIGENETICS” includes DNA methylation, and is distinct from “genetics,” i.e. mutations in DNA sequence) of transposable elements. Thus, epigenetic compromise –– leading to aberrant precocious demethylation of the maternal genome and altered transposon activation (“transposons” can be considered as “small jumping genes”) are likely contributors to the frequency of preimplantation failure when DPPA3 is absent in the zygote.
Authors [see attached study & editorial] found a similar degree of early embryonic lethality in mouse embryos from mothers fed a high-fat diet (HFD). Although ovulated oocytes from these mothers appeared morphologically normal, very few could support development –– even to the blastocyst stage (blastocyst = a thin-walled hollow structure in early embryonic development that contains a cluster of cells called the ‘inner cell mass’ –– from which the embryo arises). Using a proteomics approach, authors found 111 proteins that were misregulated in oocytes from HFD-fed mothers, and intriguingly, DPPA3 was prominent on this list. Notably, overexpression of DPPA3 in oocytes of HFD-fed mice not only restored the epigenetic remodeling in zygotes, but also partly improved the maternal-obesity-associated developmental defects in early embryo and fetal growth. These interesting data suggest that DPPA3 (Stella) insufficiency in oocytes may represent a critical mechanism that mediates the effects of the “signal” of maternal obesity which is transmitted to the embryos and offspring.
“DPPA3” is an abbreviation for the “pluripotency-associated-3″ gene. I was reminded of another recent article [last one, attached above] in which liver cells were shown to secrete DPP4 under conditions of obesity, and this process promotes adipose inflammation and insulin resistance.” Naturally, I presumed that these two genes would be evolutionarily related (i.e. in the same gene family) –– because of their common “DPP” root symbol. To my surprise, DPP4 stands for the “dipeptidyl peptidase-4” gene, (which codes for an enzyme) and the gene is located on human chromosome (Chr) 2q24.2. In contrast, the human DPPA3 gene (which encodes a transcriptional regulatory factor) is located on Chr 12p13.31 –– so it can be concluded that the DPPA3 and DPP4 genes are not evolutionarily related. “Using the same root symbol for two different gene families” is inconsistent with the standardized gene nomenclature system (https://www.genenames.org/), which “Yours Truly” helped spearhead in getting launched back in the late 1980s/erly 1990s (i.e. in the genomes of virtually all species on the planet, all genes evolutionarily related to a common ancestor long ago have a common root symbol, followed by letters and numbers that represent evolutionary divergence within families and subfamilies).
Nature Genet Mar 2o18; 50: 432–442 [article] & pp 318–319 [edtorial]; Also Nature 29 Mar 2o18; 555: 673–677