🧬 biology

Genome-wide identification and comparative analysis of endogenous retroviruses across five breed-representative pig assemblies

This study establishes a comprehensive, breed-resolved pan-genome framework for porcine endogenous retroviruses (ERVs) across five pig assemblies, revealing that while most ERVs are degraded, a core set of breed-shared, tissue-specific ERV-associated genes overlaps with key agricultural QTLs and offers critical benchmarks for improving xenotransplantation safety.

Original authors: Xiaotong Zhao, Yongping Hu, Zhidong Zhang, Xun Xiang, Hua Chang

Published 2026-06-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Xiaotong Zhao, Yongping Hu, Zhidong Zhang, Xun Xiang, Hua Chang

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The Big Picture: A Genetic "Fossil Hunt" Across Five Pig Families

Imagine the genome of a pig as a massive library. Most of the books are the pig's own instruction manual for building a body. But scattered throughout the library are thousands of old, tattered copies of ancient viruses that infected the pig's ancestors millions of years ago. These viral remnants are called Endogenous Retroviruses (ERVs). They are like "fossils" stuck in the DNA.

Usually, scientists only look at one specific "edition" of the pig library (a single reference genome) to count these fossils. This study, however, decided to compare five different editions of the pig library, representing five distinct pig breeds:

Bama Miniature Pig (Small, used for medical research).
Duroc (A popular meat breed).
Meishan (Known for having huge litters).
Ningxiang (A traditional Chinese breed known for tasty, fatty meat).
USMARC (A mix of Western breeds used for research).

The goal was to see if the "virus fossils" looked different depending on which pig family you were looking at.

The Findings: Most Fossils Are Broken

1. The "Broken Book" Reality
The researchers found a total of nearly 24,000 viral fossils across the five breeds. However, the vast majority (98%) were like books that had been chewed up by rats, missing pages, and torn apart.

The Analogy: Imagine trying to find a complete, readable copy of an ancient novel in a junkyard. You find 23,370 piles of shredded paper, 322 books with missing chapters, and only 75 books that are still whole.
The Result: Only 75 of these viral remnants were "complete" enough to study in detail. The rest were too damaged to tell much of a story.

2. The Surprise: Who Has the "Newer" Fossils?
Scientists usually think that breeds with a long, isolated history (like the Ningxiang pig) would have older, more ancient fossils, while mixed breeds (like the USMARC) would have a mix of ages.

The Twist: The study found the opposite. The Ningxiang pigs (an isolated Chinese breed) had the youngest complete fossils (on average, inserted about 0.8 million years ago). The USMARC pigs (a mixed Western breed) had the oldest complete fossils (about 7.8 million years ago).
The Explanation: Think of it like a filter. In small, isolated populations (Ningxiang), nature isn't as strict at "cleaning out" new, slightly harmful viral insertions. So, the "younger" fossils that recently jumped in are still hanging around. In the large, mixed USMARC population, natural selection has had more time to weed out the new ones, leaving mostly the very old, ancient fossils that have been there forever.

3. Where Do They Live?
The viruses weren't spread evenly. They tended to cluster on specific "shelves" in the library (Chromosomes 1, 13, and X). It's like finding that all the old, broken magazines in a house are piled up in the living room and the attic, but never in the kitchen.

The Connection: Viruses and Pig Traits

The researchers asked: "Do these viral fossils sit near important pig genes?"

1. The "Neighborhood" Effect
They found that these viral fossils often sit right next to genes that control how the pig's body communicates (specifically in the brain and nerves) and how it moves materials around inside cells.

The Analogy: Imagine the viral fossils are like graffiti tags on the wall. The researchers found that these tags are almost always painted right next to the "Do Not Disturb" signs or the "Main Power Switch" of the house. This suggests the viruses might be influencing how the house runs, even if they aren't the ones turning the lights on.

2. The "Meat and Baby" Overlap
Crucially, the genes sitting next to these viral fossils overlap significantly with QTLs (Quantitative Trait Loci). Think of QTLs as "Hot Spots" on a map where the pig's most important traits live:

Meat Quality: How tasty and tender the pork is.
Growth: How fast the pig gets big.
Reproduction: How many babies the pig has.
Fatness: How much fat the pig stores.

The study found that the viral fossils are frequently parked right in the middle of these "Hot Spots." This suggests that these ancient viruses might have accidentally helped shape how pigs grow, reproduce, and store fat over thousands of years.

What This Means (According to the Paper)

It's a Map, Not a Cure: The paper creates a detailed map of where these viral fossils are in different pig breeds. It does not prove that the viruses cause the meat to be tasty or the pigs to have more babies, but it shows they are often found in the same neighborhood.
Safety for the Future: The authors mention that this map is useful for xenotransplantation (using pig organs for humans). Before we can safely use pig organs, we need to know exactly which viruses are hiding in which breeds so we can remove them.
No Magic Bullet: The study emphasizes that finding the virus next to a trait doesn't mean the virus is the boss. It just means they are neighbors. More experiments are needed to see if the virus is actually pulling the strings.

Summary in One Sentence

This study took a tour of five different pig families to map out their ancient viral "fossils," discovering that most are broken, that the "youngest" fossils are found in the most isolated breed, and that these fossils often hang out in the genetic neighborhoods that control meat quality and reproduction.

Technical Summary: Genome-wide Identification and Comparative Analysis of Endogenous Retroviruses Across Five Breed-Representative Pig Assemblies

Problem Statement
Porcine xenotransplantation is a potential solution to human organ shortages, yet the risk of cross-species transmission of endogenous retroviruses (ERVs) remains a critical safety barrier. Current risk assessments rely heavily on a single reference genome, creating a "blind spot" regarding breed-specific ERV diversity, structural variations, and insertion ages across diverse genetic backgrounds. Previous studies have largely focused on biomedical safety (e.g., PERV inactivation) or specific retrotransposon surveys, lacking a unified, multi-assembly description of ERV abundance, structural completeness, and genomic context in pigs. This study addresses the gap in breed-specific variation by shifting the analytical framework from a single reference to a comparative, multi-breed landscape.

Methodology
The researchers conducted a comparative genomic analysis across five chromosome-level pig genome assemblies representing distinct genetic backgrounds: Bama miniature pig, Duroc, Meishan, Ningxiang, and USMARC.

Identification and Annotation: ERV loci were identified using the LTRharvest/LTRdigest pipeline. Candidates were filtered based on paired long terminal repeat (LTR) boundaries (100–2,000 bp length, ≥80% similarity) and target site duplications (4–6 bp).
Structural Classification: Loci were classified as complete, near-complete, or partial based on the retention of paired LTRs, primer binding sites (PBS), polypurine tracts (PPT), and internal retroviral domains (e.g., gag, pol, env, RVT_1).
Phylogenetic and Age Analysis: Complete ERVs were aligned (MAFFT) and subjected to maximum-likelihood phylogenetic inference (IQ-TREE 2) against a panel of 24 retroviral reference sequences. Relative insertion ages were estimated using Kimura 2-parameter divergence between paired LTRs ( $T = K / 2r$ ).
Functional and Genomic Context: ERV-associated genes (within 10 kb flanking regions) were analyzed for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment. Tissue expression profiles were derived from the GSE93855 dataset (six tissues), and colocalization with quantitative trait loci (QTLs) was assessed using Animal QTLdb.

Key Results

ERV Landscape and Structure: A total of 23,767 ERV loci were identified across the five breeds. The landscape is dominated by structurally degraded elements: 98.33% were partial, 1.35% near-complete, and only 0.32% (75 loci) were classified as complete.
Phylogenetic Diversity: Complete ERVs clustered into four retroviral clades: Betaretrovirus, Gammaretrovirus, Spumaretrovirus, and Epsilonretrovirus. Breed-specific differences were observed; for instance, Betaretrovirus and Spumaretrovirus candidates were more frequent in Ningxiang, while Gammaretrovirus candidates were more common in Duroc and USMARC.
Insertion Age Anomalies: Relative insertion age estimates revealed a counter-intuitive pattern. The Ningxiang breed (an indigenous Chinese breed with a small effective population size) showed the lowest median insertion age (0.81 Myr), whereas the USMARC composite outbred line showed the highest (7.85 Myr). This suggests that reduced purifying selection efficiency in small effective population sizes ( $N_e$ ) may allow younger complete elements to persist, rather than older integration histories being exclusive to diverse commercial lines.
Genomic Context and Function:
- Shared Genes: 278 ERV-associated genes were shared across all five assemblies.
- Enrichment: These shared genes were enriched in synaptic signaling, vesicle transport, and cell junction pathways (GO) and pathways such as Glutamatergic synapse and Oxytocin signaling (KEGG).
- Expression: The shared gene set exhibited tissue-specific expression, with peak expression often observed in the spleen, kidney, or lung.
- QTL Colocalization: There was substantial overlap between ERV-associated genes and annotated QTLs for meat quality, growth, fatness, and reproduction. Notably, 56.5% of the 278 shared genes overlapped with QTL intervals, a higher rate than in breed-specific sets.

Significance and Claims
The authors position this study as the establishment of a comprehensive, breed-resolved pan-genome framework for porcine ERVs. The significance of the work lies in:

Moving Beyond PERV-Centric Views: The study extends the understanding of porcine retroviruses beyond the classical PERV (Porcine Endogenous Retrovirus) framework, providing a broader catalog of ERV candidates and their structural states across diverse breeds.
Demographic Insights: The findings on insertion age profiles offer a new perspective on how breed demographic history (founder events, selection intensity, admixture) shapes the detectable ERV complement, specifically highlighting the impact of effective population size on the retention of complete elements.
Regulatory Potential: The colocalization of ERV-associated genes with QTLs for economically important traits (meat quality, reproduction) and their enrichment in synaptic/transport pathways suggest that integrated ERV loci may act as functional regulatory elements rather than passive genomic remnants.
Practical Application: The generated catalog provides practical benchmarks and target lists for donor strain selection and precise CRISPR-mediated viral inactivation in the context of xenotransplantation.

Limitations and Modesty
The paper explicitly maintains a modest stance regarding causality. It clarifies that:

The analysis is based on reference assemblies and does not capture population-level insertion polymorphism (requiring multi-individual sampling).
The detection method (paired-LTR) creates an ascertainment bias favoring younger, less-diverged candidates; thus, age estimates are relative, not absolute.
Colocalization with QTLs and functional enrichment indicate candidate genomic contexts but do not constitute direct evidence that specific ERV loci regulate nearby genes or phenotypes. Experimental validation (e.g., epigenomic profiling, targeted perturbation) is required to establish functional activity.