TL;DR
This study leverages complete genome sequences from 30 herpesviruses to build phylogenetic trees and analyze gene gain events. It reveals that herpesviruses have a surprisingly high rate of gene gain, significantly exceeding their nucleotide mutation rate, and suggests that many "core" genes previously identified might be too broadly defined.
What Problem Does This Paper Address?
Herpesviruses are a diverse group of viruses with complex genomes, and understanding their evolutionary history and the dynamics of their gene content is crucial for developing antiviral therapies and diagnostics. While previous studies have focused on specific genes or smaller datasets, a comprehensive analysis of gene gain and loss across a broad range of herpesviruses using complete genomic data was lacking. Specifically, the paper aims to:
- Reconstruct herpesvirus phylogeny using both sequence and gene content data.
- Identify ancestral "core" genes and infer the number of genes present in the last common ancestor of all herpesviruses.
- Quantify the rate of gene gain and understand its evolutionary patterns.
- Investigate the extent of gene transfer from vertebrate hosts.
Methods
The researchers employed a multi-pronged approach:
- Ortholog Definition: They used an all-against-all BLASTP similarity search to identify homologous genes (Clusters of Orthologous Groups - COGs) across 33 complete herpesvirus genomes. This involved careful selection of BLAST parameters (e-value and alignment length) to balance the number of identified COGs with their reliability.
- Phylogenetic Inference:
- Sequence-based: A subset of 10 highly conserved and alignable COGs were concatenated to build a phylogenetic tree using Neighbor-Joining (NJ) and parsimony methods. This approach leverages the evolutionary information encoded in protein sequences.
- Gene Content-based: A binary matrix representing the presence or absence of COGs across the 30 herpesvirus genomes was constructed. This matrix was then used to build a phylogenetic tree using NJ and unweighted parsimony. This method captures evolutionary history based on the gain and loss of entire genes.
- Gene Gain Mapping: Gene gain events were mapped onto the sequence-based phylogenetic tree, assuming each COG gain was a unique event. The origin of each COG was inferred to be its most recent common ancestor.
- Horizontal Transfer Search: Singleton genes (genes unique to a specific herpesvirus or small group) were searched against vertebrate databases to identify potential horizontal gene transfer events from hosts.
Datasets
The study utilized complete genome sequences from 33 completely sequenced herpesviruses, downloaded from GenBank in August 2005. Protein sequences were obtained from NCBI. The presence/absence matrix of COGs and alignment data are available at titus.bio.uci.edu/data.htm.
Key Insights and Ideas
- High Rate of Gene Gain: Herpesviruses exhibit a remarkably high rate of gene gain, estimated to be about an order of magnitude higher than their nucleotide mutation rate. This suggests that gene acquisition and loss are major drivers of herpesvirus evolution.
- Revisiting "Core" Genes: The study infers that the last common ancestor of all herpesviruses likely possessed around 30 genes, a number smaller than previously hypothesized. This implies that many genes previously considered "core" might be more labile than assumed.
- Gene Gain is Non-Random: Gene gain is not uniformly distributed across the genome or through time. It shows both taxonomic and temporal biases, with external branches of the phylogeny (representing more recent divergences) exhibiting higher rates of gain.
- Limited Evidence for Host Gene Transfer: Despite searching singleton genes for homologs in vertebrates, only a small fraction showed significant hits, suggesting that recent gene gain in herpesviruses primarily arises from other sources like viral recombination or duplication, rather than direct transfer from hosts.
- Genome Collinearity: Despite gene gain and loss, the overall arrangement of shared genes across the 30 herpesviruses remained largely collinear, with some notable exceptions like inversions and translocations, particularly in the Betaherpesvirinae.
Key Results
- Phylogenetic analyses based on concatenated protein sequences and gene content data yielded largely concordant topologies, supporting the monophyly of Alpha-, Beta-, and Gammaherpesvirinae.
- The protein sequence-based phylogeny had significantly higher bootstrap support than the gene content-based phylogeny.
- 30 COGs were inferred to be present in the last common ancestor of all herpesviruses.
- Gene gain rates were significantly correlated with branch length (proxy for time), with external branches showing higher rates.
- The Betaherpesvirinae showed particularly high rates of gene gain.
- An estimated average rate of gene gain was ~3.5 × 10⁻⁷ genes gained per year.
- Only five singleton genes showed significant homology to vertebrate genes, suggesting limited recent host gene transfer.
Limitations and Future Work
- Homology Definition Sensitivity: The study acknowledges that the definition of orthologous genes (COGs) can be sensitive to parameter choices, although their chosen criteria led to robust results.
- "Single-Gain" Assumption: The assumption that gene gain events are unique might underestimate rates at the tips of the phylogeny and overestimate them at the base.
- Proxy for Time: Using branch length as a proxy for time is an imperfect measure.
- Limited Scope of Host Transfer Search: The search for horizontal gene transfer was limited to vertebrate sources and singleton genes.
Future work could involve refining homology definitions, exploring more sophisticated models of gene gain and loss, and investigating gene transfer from a wider range of sources. Incorporating additional data, such as three-dimensional protein structure and predicted function, could also help in defining a more comprehensive set of ancestral "core" genes.
Summary updated