The fact that parental diet affects the metabolism of offspring goes to show that inheritance can be shaped by environmental factors. While early studies focused on mom’s contribution, we’re now starting to see the importance of paternal contribution for shaping an offspring’s epigenome.
For the past 15 years, the lab of Oliver Rando at the University of Massachusetts has been going strong examining how paternal diet can reprogram offspring metabolism via intergenerational epigenetic inheritance. Earlier on they used MeDIP-Seq to show that a paternal low-protein diet creates a profile of modest (~20%) DNA methylation changes in the livers of offspring. Interestingly, there was a reproducible (~10%) change in the enhancer of a lipid regulating gene.
However, the methylation patterns found in the somatic tissues of offspring were not present in the sperm of their fathers. This critical finding suggests that while changes to DNA methylation can influence phenotype, it’s not the primary epigenetic mechanism of intergenerational inheritance.
Epivariation Primarily Shapes DNA Methylation in Sperm and Offspring
Now, Rando’s lab has set out to tackle what shapes epigenomic variation in sperm and how this is inherited by offspring. Shea et al. examined the effects of giving mice high fat, low protein, or control diets. They sought to answer whether the observed methylation differences are inherited as a response to paternal environment or if they are a result of epivariation, which represents stochastic individual differences in the epigenome that are also inherited.
Here are the highlights from that paper:
- Whole genome bisulfite sequencing (WGBS) revealed a few modest (~10%) changes occurring in CpG shores and large scale (300bp) methylation differences occurring primarily over several repeated gene families.
- The most notable large scale difference was in ribosomal RNA (rDNA) clusters.
- They used reduced representation bisulfite sequencing (RRBS) to increase sample size to 61, which revealed that the sperm methylome from siblings on any diet was more similar than that from non-siblings on identical diets, suggesting inherited epivariation shapes the sperm methylome more so than diet.
- Honing in on the types of CpGs displaying epivariation revealed that a surprising amount were CpG shores.
rDNA Methylation Inheritance
The team also examined the inheritance of rDNA methylation from a father’s sperm to the liver of offspring. This involved using an in vitro fertilization (IVF) pipeline, which enabled the researchers to examine the epigenome of a sperm sample while leaving enough for some offspring. From a total of 25 sperm samples and 75 paired IVF offspring, they found that rDNA methylation levels showed intergenerational inheritance.
Digital droplet PCR (ddPCR) revealed that rDNA copy number correlated with rDNA methylation, which they analyzed by pyrosequencing 45S rDNA. This finding suggests that rDNA methylation levels are driven by DNA copy number. Overall, these findings indicate that DNA methylation does not have a direct role in the intergenerational inheritance of metabolic reprogramming.
Diet Responsive Sperm tRNA Fragments Regulate Embryos
Just when the door of DNA methylation closed, the door of ncRNA opened. Continuing their quest, Sharma et al. examined a ‘moonlighting’ role for tRNA fragments.
By purifying small (<40 nt) RNAs and deep sequencing, they found that small 5’ tRNA fragments represent 80% of total small RNA content in sperm. They confirmed this in bulls too, which suggests that tRNA cleavage is a conserved feature in mammals.
By examining mice on a protein-restricted diet, they found that:
- 5’ tRNA fragments (TRFs) for glycine and other amino acids accumulated at 2-3-fold higher levels than in control animals whereas levels of several let-7 miRNAs were decreased.
- Testicular tRNA fragments levels weren’t correlated with standard tRNA levels.
- tRNA fragments were scarce in testicular sperm but became abundant as sperm matured in the epididymis by vesicles called Epididymosomes, which deliver the “RNA payload” over the course of the week long voyage known as sperm maturation.
The team then carried out functional experiments in an embryonic stem cell system in which they used antisense LNA-containing oligonucleotides to interfere with specific tRFs and analyzed gene expression.
This revealed that:
- Most antisense oligos had no effect on mRNA.
- tRF-Gly-GCC resulted in a “dramatic upregulation” of ~70 genes.
- Interestingly, these genes are highly expressed in pre-implantation embryos and are regulated by a retroviral element: MERVL long terminal repeats (LTRs).
The team then showed that injecting the antisense oligos into zygotes decreased MERVL targets later in development. This triggered some experiments into IVF pre-implantation (2-cell) embryos via RNA-seq which showed that the in vitro experiments hold up in vivo.
Function and Future of Paternal Intergenerational Inheritance
Recently, the role of small RNAs in transmitting phenotypes across generations has come to light from studies on stress by the labs of Mansuy and Bale. Rando shares that, “Those papers also provide solid evidence for sperm RNAs as carriers of environmental information. One question that interests us is how specific all these systems are — if we were to look at blood brain barrier, as Bale and colleagues do, would we see an effect of paternal low protein diet? So at this point I cannot say how much “bandwidth” I think sperm carry.”
In terms of evolution, Rando thinks the impact is “Very little…in mammals the vast majority of environmentally-induced phenotypes occur in F1 offspring, with very few even persisting to the F2 generation…it is probably better thought of as a special (and very very interesting) case of “plasticity”.”
While these studies in mice reveal the molecular mechanisms of intergenerational epigenetic inheritance, they also have implications for human conditions such as for obesity or starvations like the Dutch hunger winter.
Rando shares that “we and others have documented cytosine methylation changes in offspring, so those could show up in EWAS studies. No-one knows whether those methylation changes are directly driven by sperm RNAs in early embryos, or (more likely) are secondary to altered development later on. At this point we know very little about how tRF-Gly-GCC acts mechanistically. But given that it seems to affect transcription, rather than the stability of MERVL-driven transcripts, it is easy to speculate that it interacts with other epigenetic modifications.”
Rando concludes with his future outlook about the field, “On the sperm side, I imagine some of the other epigenetic marks like adenine methylation will be measured soon by someone. I see more potential for breakthroughs on the early embryogenesis side as more and more genome-wide assays are adapted for single cells.”