LTR (long terminal repeat) retrotransposons are highly related to retroviruses, but lack envelope proteins which allow horizontal spreading between hosts. These retroelements are enormously abundant in plant and animal genomes, including the human genome. The retroelements have the ability to copy and paste themselves into different genomic locations by reverse transcribing their RNA back into DNA. In turn, this DNA can then insert itself into its host’s genome. As you can imagine, if these elements were to frequently pop in and out of DNA they would wreak havoc on our genomes; thus, host systems exist that control their expression. Conversely, some retroelements appear to have been co-opted for host functions, including memory transfer. Dr. James Priess’ lab in the Basic Sciences Division previously discovered that the LTR retrotransposon Cer1 in the nematode Caenorhabditis elegans, somehow escapes silencing and is “expressed at very high levels in adult hermaphrodite germ cells, and that Cer1 contributes to some late-stage infertility,” Dr. Priess stated. He added, “given the discovery of RNAi in C. elegans,” which helps regulate gene expression and protect genes from jumping around, this safeguard mechanism would be “expected to prevent expression of retroelements like Cer1.” So why would C. elegans evolve to specifically permit high expression of Cer1 in germ cells? To address some of these questions, Dr. Priess and collaborator Dr. Craig Mello analyzed two poorly understood Cer1 proteins called GAG and CERV. These results were published recently in Plos Genetics, and showed that CERV is required for nuclear export of viral genomic RNA and can form stage-specific mysterious giant rods in the C. elegans nucleus.
The Cer1 GAG protein shows almost no sequence similarity to retroviral GAG proteins, and is much larger. The Priess team used AlphaFold, a powerful AI system developed by DeepMind, to predict the structure of Cer1 GAG. This analysis identified a domain with clear structural similarity to the capsid domain of retroviral GAG proteins, which forms the protein shell around the viral genomic RNA (gRNA). The Priess team then used single molecule fluorescent in situ hybridization (FISH) to follow the nuclear exportation of Cer1 genomic RNA to understand how it interacts with GAG protein. The researchers discovered that viral-like GAG proteins rapidly associate with newly exported Cer1 genomic RNA, finding perinuclear colocalization of Cer1 GAG and genomic RNA. Furthermore, they uncovered that GAG-associated genomic RNA was protected from RNAi-mediated degradation, demonstrating the close association between GAG and genomic RNA. These findings highlight the structural and functional similarity Cer1 GAG shares with retroviral GAG, suggesting that it too works to package and protect cytoplasmic viral genomic RNA.
The researchers then investigated CERV structure and function. Although Cer1 retrotransposon closely resembles retroviruses, what separates it and other LTR retrotransposons from retroviruses is that it lacks an additional gene encoding a viral envelope (Env) which would facilitate cellular entry. Interestingly, the DNA region encoding the CERV protein is located in the same position as Env in retroviruses, suggesting that there may be some ties between CERV and envelope proteins. Here the authors find “that CERV is not a conventional envelope protein, and instead functions in allowing the viral genomic RNA to be exported from the nucleus,” Priess stated. The research team found that CERV nuclear localization and CERV-dependent nuclear export of genomic RNA is regulated by phosphorylation of CERV, creating an ON/OFF switch that could control this process and dictate the ability of CERV to associate with viral genomic RNA. Unexpectedly, they discovered that CERV undergoes a remarkable transition from tiny nuclear foci to giant rods that can span the entire nucleus. The rods are much larger than previously known viral structures, and can contain viral genomic RNA and possibly host proteins or RNA. Using a combination of immunofluorescence and transmission electron microscopy to visualize these rods and their formation, the authors discovered that rods appear to form from flattened plaques or streaks of CERV that localize to the periphery of the nucleolus before rolling up into rods. The transition from small foci to giant rods occurs at a specific time in adult development when self-fertilizing C. elegans hermaphrodites have depleted their stores of sperm and depend on cross-fertilization with males to produce offspring. These results suggest that rod formation might have some role in targeting viral spread to cross-progeny rather than self-progeny. Priess remarked that “the giant rods appear to be widespread in wild strains of C. elegans. Retroelements similar to Cer1 comprise large amounts of animal and plant genomes, including over 70% of some plant genomes. Several of these retroelements encode novel proteins in the same position as CERV, and it would be interesting to learn whether they have related roles.” Current and future work in the Priess Lab aims to solve these unknowns and work to “understand the role(s) of the rods in Cer1 biology and its relationship to germ cells.”
This work was funded by the National Institutes of Health, the National Institute of General Medical Sciences and the Howard Hughes Medical Institute.
UW/Fred Hutch/Seattle Children’s Cancer Consortium member James Priess contributed to this work.
Sun B, Kim H, Mello CC, Priess JR. The CERV protein of Cer1, a C. elegans LTR retrotransposon, is required for nuclear export of viral genomic RNA and can form giant nuclear rods. PLoS Genet. 2023 Jun 29;19(6):e1010804. doi: 10.1371/journal.pgen.1010804. PMID: 37384599