Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 18 (1), 177

Evolutionary Diversity and Novelty of DNA Repair Genes in Asexual Bdelloid Rotifers

Affiliations

Evolutionary Diversity and Novelty of DNA Repair Genes in Asexual Bdelloid Rotifers

Bette J Hecox-Lea et al. BMC Evol Biol.

Abstract

Background: Bdelloid rotifers are the oldest, most diverse and successful animal taxon for which males, hermaphrodites, and traditional meiosis are unknown. Their degenerate tetraploid genome, with 2-4 copies of most loci, includes thousands of genes acquired from all domains of life by horizontal transfer. Many bdelloid species thrive in ephemerally aquatic habitats by surviving desiccation at any life stage with no loss of fecundity or lifespan. Their unique genomic diversity and the intense selective pressure of desiccation provide an exceptional opportunity to study the evolution of diversity and novelty in genes involved in DNA repair.

Results: We used genomic data and RNA-Seq of the desiccation process in the bdelloid Adineta vaga to characterize DNA damage reversal, translesion synthesis, and the major DNA repair pathways: base, nucleotide, and alternate excision repair, mismatch repair (MMR), and double strand break repair by homologous recombination (HR) and classical non-homologous end joining (NHEJ). We identify multiple horizontally transferred DNA damage response genes otherwise unknown in animals (AlkD, Fpg, LigK UVDE), and the presence of genes often considered vertebrate specific, particularly in the NHEJ complex and X family polymerases. While 75-100% of genes involved in MMR and HR are present in 0-2 copies, genes involved in NHEJ, which are present in only a single copy in nearly all other animals, are retained in 3-8 copies. We present structural predictions and expression evidence of neo- or sub-functionalization of multiple copy genes involved in NHEJ and other repair processes.

Conclusion: The horizontally-acquired genes and duplicated genes in BER and NHEJ suggest resilience to oxidative damage is conferred in part by increased DNA damage recognition and efficient end repair capabilities. The pattern of gene loss and retention in MMR and HR may facilitate recombination and gene conversion between divergent sequences, thus providing at least some of the benefits of sex. The unique retention and divergence of duplicates genes in NHEJ may be facilitated by the lack of efficient selection in the absence of meiotic recombination and independent assortment, and may contribute to the evolutionary success of bdelloids.

Keywords: APLF; AlkD; Blm; Fpg; Ku 70/80; Ligase K; NHEJ; Polymerase lambda; UVDE; XRCC4.

Conflict of interest statement

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures

Fig. 1
Fig. 1
Gene Copy Number per Gene by DDR Category. Metazoan genes are indicated with open circles; non-metazoan genes by solid green diamonds. DR: Direct Reversal; BER: Base Excision Repair; NER: Nucleotide Excision Repair; AER: Alternate Excision Repair; MMR: Mismatch Repair; HR: Homologous Recombination; NHEJ: Non-homologous End Joining; TLS: Translesion Synthesis; Pol: Polymerase
Fig. 2
Fig. 2
Fpg. a Phylogeny of Fpg and Nei genes; Adineta vaga Fpg is in red. Clades with greater than 70% RAxML bootstrap support or 90% MrBayes posterior probability are marked with red and blue asterisks, respectively. Complete trees and accession numbers and species names of OTUs are available in Additional file 2. b The 8-oxoG capping loop region of Fpg (DNA shown in black). LeftA. vaga (AvFpg) in bronze threaded onto Geobacillus stearothermophilus Fpg (BstFpg, from Bacillus basonym, PDB 1R2Y) in blue. The BstFpg αF-β9/10 loop (purple) extends down to cover and trap the 8-oxoG, but AvFpg αF-β9/10 loop (red), is predicted to be too short to fully cover an 8-oxoG in the binding pocket. Right, Arabidopsis thaliana Fpg (AthFpg, PDB: 3TWK) in bronze overlaying BstFpg in blue, as shown in [43]. Here, the much shorter AthFpg αF-β9/10 loop (orange) cannot trap 8-oxoG in the binding pocket as the BstFpg loop (purple) can
Fig. 3
Fig. 3
Phylogeny of UVDE. Adineta vaga and Trichuris ssp. Fpg are in red. Clades with greater than 70% RAxML bootstrap support or 90% MrBayes posterior probability are marked with red and blue asterisks, respectively. Complete trees and species names and accession numbers of OTUs are available in Additional file 2
Fig. 4
Fig. 4
Phylogeny of AlkD. Adineta vaga AlkD is in red, well-separated from the clade of metazoan sequences, shaded in orange. Clades with greater than 70% RAxML bootstrap support or 90% MrBayes posterior probability are marked with red and blue asterisks, respectively. Complete trees and species names accession numbers of OTUs are available in Additional file 2
Fig. 5
Fig. 5
Ligase K. a Simplified phylogenetic tree of Ligase K with Ligase III as an outgroup; Adineta vaga Ligase K is in red. Clades with greater than 70% RAxML bootstrap support or 90% MrBayes posterior probability are marked with red and blue asterisks, respectively. Complete trees and species names and accession numbers of OTUs are available in Additional file 2. b Domain models of A. vaga Ligase K copies A1 and B1, with domain models of Ligase K peptides from other species (Mortierella verticillata KFH62561.1, Rhizophagus irregularis ESA15105.1, Capitella teleta ELT89513.1, Salpingoeca rosetta XP_004993722.1, Lottia gigantean XP_009061413.1, Aplysia californica XP_005100834.1, Blastopirellula marina WP_002650560.1, Tetrahymena thermophila XP_001011861.1). c) Comparison of A. vaga Ligase K PBZ domains with sequence logo of the Pfam model. d, e Dotplots generated with EMBOSS dotmatcher of copies B1 vs A1 and B1 to itself; lines along the diagonal indicate regions of similarity between the compares sequences. f Differential expression of A. vaga Ligase K ohnologs entering and recovering from desiccation, compared to hydrated controls. Values are log2 fold change of normalized counts, significance test values are listed in Additional file 1
Fig. 6
Fig. 6
Bloom helicases. a Domain models of a Bloom helicase from Homo sapiens (XP_011520183) and Bloom-like helicases from Dictyostelium purpureum (XP_003287311.1, “hypothetical protein”), Triticum monococcum (AGH18689.1, “PHD-finger family protein”), Symbiodinium microadriaticum (OLP97093.1, “ATP-dependent RNA helicase DHH1”), Stentor coeruleus (OMJ79001.1, “hypothetical protein”), Daldinia sp. (OTB17292.1, “hypothetical protein”). b Domain models of the five copes of the Blm-like helicase from A. vaga. c Sliding window analysis of nonsynonymous (Ka) difference (solid line) and ratio of nonsynonymous to synonymous differences (Ka/Ks, dashed line) between copies C1 and C2. d Differential expression of A. vaga Blm ohnologs entering and recovering from desiccation, compared to hydrated controls. Values are log2 fold change of normalized counts, significance test values are listed in Additional file 1
Fig. 7
Fig. 7
Ku70 and Ku80. a Domain model of A. vaga Ku70 A and B ohnologs, with sliding window analysis of nonsynonymous (Ka) difference (solid line) and ratio of nonsynonymous to synonymous differences (Ka/Ks, dashed line) between AvKu70A1 and B1. The alignment on the upper left shows the region where Ka/Ks > 1 near the N-terminus; the alignment to the lower right shows the SAP domains compared to human Ku70. Predicted sumoylation sites are in red, predicted acetylation sites are highlighted in blue. b Domain model of A. vaga Ku80 A and B ohnologs, with sliding window analysis of Ka and Ka/Ks. The alignment in the upper left shows the Q3E4Q8 track at the terminus of the α/β domain present in copy A and not in B. c Crystal structure PDB 1JEY, human Ku70 (yellow) Ku80 (red) heterodimer complexed with DNA (grey). d Three views of the superposition of the predicted structure of A. vaga Ku70A1 (purple) and A. vaga Ku70B1 (green). The N terminal region under putative positive selection and the SAP domain are indicated in red and blue for Ku70 A1 and B1, respectively. First orientation is the same as in (c), second is an elevated view (45° rotation along horizontal axis), third is a top view (90° rotation along horizontal axis). e Superposition of the predicted structures of A. vaga Ku80A1 (blue) and Ku80B1 (copper) in the same orientation as in (c). f Differential expression of A. vaga Ku70 A and B and Ku80 A and B ohnologs entering and recovering from desiccation, compared to hydrated controls. Values are log2 fold change of normalized counts, significance test values are listed in Additional file 1
Fig. 8
Fig. 8
DNAPKcs. Domain model of A. vaga DNAPKcs A (top) and DNAPKcs B (bottom) ohnologs with sliding window analysis of nonsynonymous difference (Ka, solid line) and ratio of nonsynonymous to synonymous differences (Ka/Ks, dashed line) between A1 and B1
Fig. 9
Fig. 9
Artemis. a Domain models of A. vaga Artemis A (top) and B (bottom) ohnologs showing predicted DNAPKcs phosphorylation sites (blue circles indicate sites conserved between A and B peptides, yellow circles indicate unique sites on each peptide) with sliding window analysis of nonsynonymous difference (Ka, solid line) and ratio of nonsynonymous to synonymous differences (Ka/Ks, dashed line) between A1 and B1. b Differential expression of A. vaga Artemis A and B ohnologs entering and recovering from desiccation, compared to hydrated controls. Values are log2 fold change of normalized counts, significance test values are listed in Additional file 1. c Normalized read counts of the two ohnologs under hydrated, entering, and recovering conditions
Fig. 10
Fig. 10
XRCC4. a Domain models of AvXRCC4 A (top) and B (bottom) ohnologs showing the three regions of XRCC4, the total charge of each region, and predicted DNAPKcs phosphorylation sites (blue circles indicate sites conserved between A and B peptides, yellow circles indicate unique sites on each peptide) with sliding window analysis of nonsynonymous difference (Ka, solid line) and ratio of nonsynonymous to synonymous differences (Ka/Ks, dashed line) between copies A1 and B1. b Superposition of the predicted structures of A. vaga XRCC4A1 (blue) and XRCC4B1 (copper) showing the conserved structure of the N-terminal head region, the central helix with Ligase 4 binding regions shown in purple (XRCC4A1) and red (XRCC4B1), and the poorly conserved C terminus with 22 residues present in A1 but not in B1 shown in magenta. c Alignment of the Ligase 4 binding region in A1, B1, and human XRCC4; colons (:) indicate residues involved in binding [116]. d Differential expression of A. vaga XRCC4 A and B ohnologs entering and recovering from desiccation, compared to hydrated controls. Values are log2 fold change of normalized counts, significance test values are listed in Additional file 1
Fig. 11
Fig. 11
Ohnologs of Polλ. a Domain structure of the three types of polymerase λ in A. vaga. Boundaries of domains defined by hmmscan of PfamA are shown above (start) and below (end); the disordered Ser/Pro-rich region (SP) is not a defined domain. The three residues that make up the phosphate binding pocket in the 8kD domain are shown (RRK or RSK). The position of residues encoded by codons determined to be under positive selection in the lineage leading to AvPolA are shown above the A. vaga A structure clustered by diamonds for each domain. b Secondary structure of the BRCT domain in polymerase λ as determined by Phyre. Beta sheets are shown as blue arrows, alpha helices as pink cylinders. The region identified as the Pfam domain BRCT_2 by hmmscan is shown with domain-specific expectation value; positions of structure boundaries outside of the predicted BRCT domain are indicated. c Alignment of the disordered SP region in A. vaga copies of polymerase λ. Numbering is to AvPolLA1. Serine and proline residues are highlighted in blue and yellow, respectively. d Unrooted gene tree of the six copies of Pol λ in A. vaga and three additional rotifer species used for codeml tests of selection. 1, Seison sp.; 2, Brachionus manjavacas; 3, Brachionus calyciflorus. e Differential expression of the three paralogs entering and recovering from desiccation, compared to hydrated controls. Values are log2 fold change of normalized counts, significance test values are listed in Additional file 1
Fig. 12
Fig. 12
Ohnologs of APLF. a Domains and interactions of H. sapiens APLF assigned using Uniprot Q81W19 as template with domain annotation and function refined with reference to [60,83,84,93,119]. b Domain models and phylogenetic relationship of the two pairs of A. vaga APLF ohnologs, with cladograms showing the relationship of gene copies and ohnologs. c Alignment of the tandem PBZ domains. Each PBZ domain has a conserved C(M/P)Y and CRY motif, highlighted in aqua along with nearby conserved residues, and these form a basic, hydrophobic pocket for ADP-ribose binding. APLF binds multiple ADP-ribose residues within PAR, and Y381/Y386 and Y423/Y428 are critical for interactions with the adenine ring, and R387/R429 coordinate the interactions with the phosphates [91]. All are marked with (*). The Y423F difference in D is found in some other metazoans. The C and D ohnologs substitute Q for P in the first PBZ domain, which would not be expected to maintain the characteristics of the basic, hydrophobic binding pocket. Both also terminate before the final H of the second PBZ domain, which undoubtedly alters the domain’s binding properties. d Alignment of the Ku80 Binding Domain (KBD) regions. Copies A, C and D retain R184 and W189, the residues found critical for Ku binding in mammals [117]; copy A lacks one of the conserved positively charged residues found in most KBD domains. e Differential expression of all four paralogs entering and recovering from desiccation, compared to hydrated controls. Values are log2 fold change of normalized counts, significance test values are listed in Additional file 1

Similar articles

See all similar articles

References

    1. Hsu WS. Oogenesis in the Bdelloidea rotifer Philodina roseola Ehrenberg. Cellule. 1956;57:283–296.
    1. Hsu WS. Oogenesis in Habrotrocha tridens (Milne) Biol Bull. 1956;111:364–374. doi: 10.2307/1539143. - DOI
    1. Signorovitch A, Hur J, Gladyshev E, Meselson M. Allele sharing and evidence for sexuality in a mitochondrial clade of Bdelloid rotifers. Genetics. 2015;200:581–590. doi: 10.1534/genetics.115.176719. - DOI - PMC - PubMed
    1. Maynard Smith J. Contemplating life without sex. Nature. 1986;324:300–301. doi: 10.1038/324300a0. - DOI - PubMed
    1. Flot J-F, Hespeels B, Li X, Noel B, Arkhipova I, et al. Genomic evidence for ameiotic evolution in the bdelloid rotifer Adineta vaga. Nature. 2013;500:453–457. doi: 10.1038/nature12326. - DOI - PubMed

Publication types

LinkOut - more resources

Feedback