We’ve recently been talking about our new video in the Secrets of the Human Body series, “Building a Baby,” which, as David Klinghoffer said, encourages us to “respect — and more, to stand in awe of — the process that gave life to each of us.” On that subject, here is fresh evidence that so-called “junk DNA” is in fact crucial for human development. According to a new paper in Science, “Endogenous retroviruses synthesize heterologous chimeric RNAs to reinforce human early embryo development.”
The paper uncovers vital roles played by endogenous retroviruses (ERVs), repetitive DNA sequences that were long thought to be relics of ancient viral infections in our ancestors’ genomes — the ultimate example of supposed genetic junk. ERVs were at one time iconic for evolution activists. As I observed here some year ago, one notably uncivil writer, Abbie Smith, even made a mascot of them with the name of her popular science blog, ERV. The debate has progressed a lot since then. We keep discovering ERVs have function (see here, here, here, here, and here for some examples). In fact, we covered functions of ERVs in a Long Story Short video last year, “ERVs, Pseudogenes, and Onions.”
If these ERVs really are just blindly introduced viral insertions in our ancestor’s genomes, there’s no particular reason why they should be functional, much less absolutely “essential” at a “widespread” scale in the earliest stages of human development. Yet this is exactly what this new paper in Science has found.
“Essential” for Development at a “Widespread” Scale
The editor’s summary nicely explains the paper’s findings:
Zygotic genome activation (ZGA) marks the critical moment when the embryo’s genome is awakened from dormancy. Xiang et al. now show that an ancient viral remnant, the endogenous retrovirus MLT2A1, plays an essential role in human ZGA. MLT2A1 produces chimeric RNAs during ZGA by fusing with various downstream sequences. These chimeras interact with numerous ZGA genes and form an auto-amplifying network promoting widespread, de novo gene activation during ZGA. Reducing MLT2A1 activity halts early human embryo development. Thus, ancient viral sequences have been repurposed to orchestrate the very beginning of human life, a finding that may have relevance for reproductive medicine.
The research article summary explains the rationale behind the study. It is the fact that various types of repetitive DNA elements are already known to be “highly active during early embryonic development”:
Retro–transposable elements (retro-TEs) consist of long interspersed nuclear elements (LINEs), short interspersed nuclear elements (SINEs), and endogenous retroviruses (ERVs). They are highly active during early embryonic development and constitute a substantial proportion of the species-specific transcriptome during ZGA.
Thousands of Copies
It’s important to understand that the ERV they studied, MLT2A1, is not simply a single instance of an ERV in the genome. It’s a subfamily of ERV of which there are thousands of copies in the human genome. So this evidence potentially implicates thousands of individual ERV copies — typically each a few hundred base pairs in length — as having function:
The MLT2A1 subfamily consists of 3838 annotated copies in the human genome. For most of the copies, both their expression level and genomic loci accessibility peaked at ZGA stage. Different MLT2A1 loci possessed diverse transcriptional activity during early embryogenesis. We classified MLT2A1 copies into two categories: “embryo-expressed” (n = 1322 copies) and “embryo-unexpressed” (n = 2516 copies). Among the expressed copies, the most active ones showed comparable normalized expression levels to those of key ZGA regulators.
They found that “MLT2A1 was specifically up-regulated during ZGA” and this led to a complex mechanism which facilitated the recruitment of RNA polymerase to further transcribe key sections of the genome, including genes, at this crucial developmental stage:
Long-read transcriptome profiling revealed that the majority of MLT2A1 copies generated diverse chimeric transcripts by fusion with downstream sequences, including coding and noncoding sequences, and, most prominently, with other retro- TE subfamilies. Chromatin isolation by RNA purification demonstrated that the variable 3′ fusion regions of these chimeric RNAs increased the sequence complexity of MLT2A1, thereby expanding their genome- targeting potential. On the other hand, RNA pull-down coupled with mass spectrometry showed that the conserved 5′ MLT2A1 region of the chimeric RNAs binds to heterogeneous nuclear ribonucleoprotein U (HNRNPU), recruiting RNA polymerase II (RNAP II) to facilitate transcription. This dual strategy conferred by MLT2A1 chimeric RNAs ensured global induction of a broad spectrum of ZGA genes.
And not only did these MLT2A1 chimeric RNAs activate other genes, they also activated other instances of MLT2A1 ERVs, leading to a “positive feedback loop to cross-activate one another”:
MLT2A1 subfamily members formed a self-amplifying, positive feedback loop to cross-activate one another. Collectively, these chimeric RNAs orchestrated a robust and decisive transition through ZGA, which is essential for securing early human embryo development.
They concluded that these ERVs are “essential regulators coordinating global transcriptional activation during human ZGA” at a level that is “widespread”:
Our study established MLT2A1-derived chimeric RNAs as essential regulators coordinating global transcriptional activation during human ZGA. The distinctive features of MLT2A1 RNAs allowed them to operate as an extensive regulatory network, specifically tailored to drive the widespread de novo transcription that defines ZGA.
Which Model Better Explains the Data?
Again, all of this raises a crucial question: What sort of model better anticipates this evidence?
The standard evolutionary view would have us believe that ERVs are viral insertions, emplaced in our ancestor’s genomes yielding massive amounts of junk DNA that has no reason to be useful to the organism. Sure, defenders of the evolutionary junk-DNA viewpoint will sometimes admit that an ERV here or there might accidentally be “co-opted” to perform some function. But what we see is something completely different: thousands of ERVs from this entire ERV subfamily are “essential” at a “widespread” scale in the earliest stages of human development — and if they don’t activate properly then you quickly get a dead embryo. ID fully anticipates this discovery of widespread essential functions for ERVs. The accidental junk-filled genome view just doesn’t. As a mascot for evolution activists, ERVs need to be retired.
Serious Ethical Questions
Now I have to comment on one disturbing part of this research: This research destroyed human embryos for study at the 8-cell stage that had been previously created for potential use in in vitro fertilization. Now they argue that these embryos were “ethically approved” for research because they “are clinically considered unsuitable for implantation” because they were “zygotes with observable 3 pronuclei (3PN),” meaning they have a triploid genome (69 total chromosomes) and would not survive development. Perhaps that’s true but many serious ethical questions are raised about destroying human embryos for research, even if they could not develop to full term. How ironic that researchers who probably believe that humanity is a mere accident are now, through ethically questionable research justified under that viewpoint, finding evidence that humanity is no accident at all.