Would it surprise you to learn that 17% of your genome is made up of multiple copies of something resembling a virus—a thing that, since it was first introduced to cells eons ago, has dramatically reshaped the human genome? Perhaps you’d also be interested—or alarmed—to learn that these ancient elements, called retrotransposons, are still capable of moving about in your DNA, thereby driving changes in cellular behavior, and sparking or exacerbating diseases such as cancer. Princeton researcher George Ghanim is intent upon uncovering the secrets of these ancient entities, and has already made some important new discoveries, as outlined in a paper published March 6, 2025 in the journal Science.

Ghanim joined the faculty of the Princeton Department of Molecular Biology in mid-January of 2025, after completing his postdoctoral work at the MRC Laboratory of Molecular Biology in England. Towards the end of his postdoctoral studies, while he was still in England, Ghanim decided to investigate something he’d been curious about for a long time: LINE-1 retrotransposons.

“I actually have been thinking about this project since graduate school,” said Ghanim. “As a graduate student I worked on an enzyme called P element transposase that was really important in early Drosophila genetics. Towards the end of my PhD, my advisor, Donald Rio, and I started discussing other mobile elements such as LINE-1. In fact, as a going-away gift, he gave me a codon-optimized sequence of LINE-1 genes.”

Those discussions percolated in the back of Ghanim’s mind while he pursued his postdoctoral work with Kelly Nguyen at MRC-LMB, where Ghanim studied the enzyme telomerase.

“Given that I had the opportunity to start my own research group soon, this is something I really wanted to be studying,” said Ghanim.

The human genome is absolutely littered with detritus resulting from the activity of transposable elements: defective or partial copies of LINE-1 are found throughout the stretches of so-called “junk DNA” in the human genome, but also occasionally nearby or within functional genes. However, a minority of LINE-1 copies, up to 150 of them, are complete and remain active in the human genome. That is, they have retained the ability to “jump” from one location in a person’s DNA to another. 

Cells enact stringent controls to prevent LINE-1 elements from jumping because if a LINE-1 element lands within another gene, it could alter or break that gene’s function. This could impair or kill the cell, or else throw things so wildly off-balance that it results in cancer. For reasons that are still unclear, cellular controls on LINE-1 jumping sometimes slip, allowing LINE-1 to be transcribed into an RNA version that both serves as a copy of the transposable element and contains the instructions to make two proteins. One of these proteins, ORF2p, is the enzyme responsible for inserting a copy of LINE-1 at a new site in the cell’s DNA. Very little is known about the molecular details of this insertion process, so that is what Ghanim and his colleagues at MRC-LMB focused on first.

DNA is a molecule made up of two complementary (mirror-image) strands. Current theories hold that ORF2p first cuts the backbone of one of these two strands at the insertion site, then uses an enzymatic activity called “reverse transcriptase” to make a single-stranded DNA copy of LINE-1 RNA. Next, ORF2p cuts the backbone of the second strand of the recipient DNA, and synthesizes the complimentary DNA strand of LINE-1. Finally, cellular proteins that specialize in repairing DNA damage fill in the gaps and cement the new LINE-1 copy in place. However, when Ghanim used cryo-electron microscopy to capture images of ORF2p in the act of performing a new insertion, he discovered evidence that these steps proceed in a different order than originally thought.

“The way it does this is, I think, more structurally sophisticated than we had anticipated.

The complex that forms to allow this is quite complicated. It's really torquing the DNA into two pieces, and it effectively breaks the DNA in half,” said Ghanim.

The data indicate that ORF2p snaps the second strand of the recipient DNA before or during the reverse transcription step. This puts ORF2p into a race: it must finish reverse transcribing LINE-1 before it gets interrupted by the DNA repair processes brought in to deal with the broken DNA. This could explain why human DNA contains so many partial copies of LINE-1: they result from LINE-1 losing that race. 

ORF2p is able to both break DNA and execute reverse transcription all on its own in a test tube. However, when operating within a cell, LINE-1 retrotransposons likely behave similarly to other viruses. That is, they probably co-opt normal cellular proteins to promote viral activity. Scientists have identified a handful of cellular proteins that are co-opted by LINE-1, but again, it’s not known how these interactions work. Ghanim used the machine learning algorithm Alpha Fold to predict how two co-opted human proteins interact with ORF2p. The predicted structures formed by these interactions were consistent with what Ghanim had already observed through cryo-electron microscopy. He additionally confirmed that he could disrupt these interactions by making targeted mutations in ORF2p.

“That's something that I’m following up on here at Princeton: looking at these cellular factors and what they're doing at particular stages of retrotransposition,” said Ghanim. “The structural biology, EM and computational capabilities here at Princeton will be very helpful for these sorts of studies.” 

Ghanim’s laboratory is now recruiting students and postdocs to tackle the biology of these enigmatic entities from our ancient past.

Citation: George E. Ghanim, Hongmiao Hu, Jerome Boulanger, and Thi Hoang Duong Nguyen. Structural mechanism of LINE-1 target-primed reverse transcription. Science. 2025.  DOI: https://doi.org/10.1126/science.ads8412

Funding: Jane Coffin Childs Postdoctoral Fellowship (GEG), UKRI-Medical Research Council grant MC_UP_1201/19 (THDN), EMBO Young Investigator Program Award (THDN)