Mapping global and local coevolution across 600 species to identify novel homologous recombination repair genes

Abstract

The homologous recombination repair (HRR) pathway repairs DNA double-strand breaks in an error-free manner. Mutations in HRR genes can result in increased mutation rate and genomic rearrangements, and are associated with numerous genetic disorders and cancer. Despite intensive research, the HRR pathway is not yet fully mapped. Phylogenetic profiling analysis, which detects functional linkage between genes using coevolution, is a powerful approach to identify factors in many pathways. Nevertheless, phylogenetic profiling has limited predictive power when analyzing pathways with complex evolutionary dynamics such as the HRR. To map novel HRR genes systematically, we developed clade phylogenetic profiling (CladePP). CladePP detects local coevolution across hundreds of genomes and points to the evolutionary scale (e.g., mammals, vertebrates, animals, plants) at which coevolution occurred. We found that multiscale coevolution analysis is significantly more biologically relevant and sensitive to detect gene function. By using CladePP, we identified dozens of unrecognized genes that coevolved with the HRR pathway, either globally across all eukaryotes or locally in different clades. We validated eight genes in functional biological assays to have a role in DNA repair at both the cellular and organismal levels. These genes are expected to play a role in the HRR pathway and might lead to a better understanding of missing heredity in HRR-associated cancers (e.g., heredity breast and ovarian cancer). Our platform presents an innovative approach to predict gene function, identify novel factors related to different diseases and pathways, and characterize gene evolution.

Figure 1.

Evolution of HRR proteins. (A) Normalized phylogenetic profiles (NPPs) of all human protein coding genes after hierarchical clustering and dendrogram leaf order optimization. Each row represents the NPP of a single gene across 578 eukaryotes ordered by their phylogenetic distance from Homo sapiens. The colors in the heat map indicate the relative degree of conservation between a human protein and its ortholog in a certain species (column). When zero, this means that the ortholog is conserved at the average conservation level of orthologs in the species, relative to human; negative values mean less conserved than average, and positive values mean more conserved than average (the values are in Z-scores) (for further details, see Tabach et al. 2013a). White indicates poor conservation, and dark blue indicates highly conserved genes (blue). The bars on the right side represent clusters enriched for known HRR genes, in which the score represents the fraction of known HRR genes in each cluster (Supplemental Methods), and with an FDR-adjusted P-value indicating the significance of the enrichment (hypergeometric test). The colors of the bars indicate the functional module within the HRR pathway (see legend above heat map and in Supplemental Fig. S1). (B) Examples for HRR genes clustered together. Note that only the known HRR genes in each cluster are shown. (C) Detailed view of three couples of genes that coevolved in different clades. (Top) BRCA1 and PALB2 are locally coevolved in animals but not in plants, fungi, and protists. (Middle) RAD52 and FANCA are locally coevolved only in animals. (Bottom) BLM and WRN are locally coevolved in fungi and protists but not in animals or plants. The clades are indicated by colored bars with the Pearson correlation coefficient of the two PPs within the respective clade. The y-axis indicates the NPP score as in A. Red rectangles show regions of coevolution.

Figure 2.

Coevolution of HRR genes is clade-specific. (A) Number of cell cycle and DNA repair genes among the top 50 genes coevolved with BLM and RAD51 in seven different clades. The P-values were calculated using hypergeometric tests. (B) ROC curve of CladePP compared to normalized phylogenetic profiling on individual clades or all eukaryotes, with the HRR gold standard genes used as positives. Numbers in the legend indicate area under the curve (AUC) for each clade. (***) P < 10⁻⁵, (**) P < 10⁻³, (*) P < 0.05.

Figure 3.

NPPs of SMG1, RAN, and NME1. HRR genes show coevolution within certain clades: (A) SMG1 coevolved with ATR in protists and fungi; (B) RAN coevolved with CHEK2 in plants and across all 578 eukaryotes; and (C) NME1 highly coevolved with H2AFX in plants and animals.

Figure 4.

Screen for germline radiation sensitivity in C. elegans. L1 worms were subjected to RNAi of the indicated genes by feeding. Adults were exposed to 50 Gy. After 2 h of recovery, 15 worms were replated with one worm per plate, and 15 nonirradiated worms were plated as controls. After 36 h, the number of eggs and larvae were recorded. The number of progeny is presented relative to worms fed empty-vector containing bacteria. rad-51, a known HRR gene in C. elegans, served as a positive control.

Figure 5.

(Top) The DR-GFP reporter assay. SceGFP is a modified version of the GFP gene, which contains an I-Sce site and an in-frame premature stop codon that can be removed by HR-mediated repair of the I-SceI–induced DSB using the internal GFP fragment (iGFP) placed downstream from the SceGFP cassette. (Middle) The experimental setup used to monitor HRR in the HeLa DR-GFP cell line. (Bottom) The percentage of GFP+ cells for each experimental condition (siRNA +/− I-SceI) was measured by flow cytometry (FACS) following the experimental procedure detailed in the top right panel and normalized to the siCtrl + I-SceI conditions, in two human cell lines: HeLa and U2OS. (***) P < 0.0001, (**) P < 0.001, (*) P < 0.01, (ns) nonsignificant.