Functional divergence of gene duplicates through ectopic recombination

• DOI: 10.1038/embor.2012.157
• Publication Date: 2012-10-16T10:34:38Z
• Authors: Joaquin F Christi...; Sebastiaan E Van ...; Jorge Duitama; Chris A Brown; Maarten G Ghequire; Luc De Meester; Jan Michiels; Tom Wenseleers; Karin Voordeckers; Kevin J Verstrepen
• Keywords: Functional divergence; Divergence (linguistics)

Scientific Report16 October 2012Open Access Functional divergence of gene duplicates through ectopic recombination Joaquin F Christiaens Department Microbial and Molecular Systems, Centre Plant Genetics (CMPG), KU Leuven, Kasteelpark Arenberg 22, B-3001 Leuven (Heverlee), Belgium VIB Laboratory Systems Biology, Search for more papers by this author Sebastiaan E Van Mulders Malting Brewing Science, Faculty Bioscience Engineering, Jorge Duitama Chris A Brown Arts Sciences Center Harvard University, Cambridge, Massachusetts, 02138 USA Fathom Information Design, Boston, 02114 Maarten G Ghequire Luc De Meester Animal Ecology Systematics Section, B-3000 Jan Michiels Tom Wenseleers Karin Voordeckers Kevin J Verstrepen Corresponding Author Christiaens1,2,‡, Mulders3,‡, Duitama1,2,‡, Brown1,2,4,5, Ghequire1, Meester6, Michiels1, Wenseleers6, Voordeckers1,2 1,2 1Department 2VIB 3Department 4Faculty 5Fathom 6Department ‡These authors contributed equally to paper. *Corresponding author. Tel:[phone]; Fax:+32 91; E-mail: [email protected] EMBO Reports (2012)13:1145-1151https://doi.org/10.1038/embor.2012.157 PDFDownload PDF article text main figures. Peer ReviewDownload a summary the editorial decision process including letters, reviewer comments responses feedback. ToolsAdd favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Gene duplication stimulates evolutionary innovation as resulting paralogs acquire mutations that lead sub- or neofunctionalization. comprehensive in silico analysis Saccharomyces cerevisiae reveals cell-surface subtelomeric genes also undergo recombination, which leads new chimaeric alleles. Mimicking such intergenic events FLO (flocculation) family shows alleles confer different adhesion phenotypes than parental genes. Our results indicate between can generate large set alleles, thereby providing raw material adaptation innovation. INTRODUCTION How organisms evolve adapt environments remains central question biology. have crucial role processes, especially rapid development functions [1, 2]. Duplication might yield adaptive benefits increased dosage gene. In addition, stimulate one both subfunctionalization neofunctionalization [3, 4, 5, 6, 7, 8]. duplications are not spread evenly over genome, occuring much frequently subtelomeres, regions directly adjacent telomeres [9, 10, 11]. As result, families often large, with some carrying many several hundred Trypanosoma Plasmodium species, variable expression variants antigen allows these pathogens elude host immune system. Subtelomeric compact genome generally smaller enriched genes, well involved nutrient transport metabolism 12, 13, 14, 15, 16, 17, 18]. Recombination could sequence diversity duplicated [19]. However, apart from anecdotal examples specific VAR VSG Plasmodia Trypanosomes major histocompatibility complex (MHC) class vertebrates, occurrence biological relevance been systematically investigated [16, 20]. Here, we present all paralog model eukaryote S. cerevisiae. show recombinations occur predominantly located at subtelomeres and/or encode To verify whether altered phenotypes, mimicked shaped adhesin family. Phenotypic analyses artificial chimaera revealed they were functional conferred differed their adhesins. RESULTS AND DISCUSSION Ectopic investigate cerevisiae, first identified paralogous BLASTing reference strain S288c's proteome against itself. Next, used MCL clustering algorithm manual curation define list 210 families, each comprising least two (see methods supplementary information online details) [21]. For collected known GO categories chromosomal locations. We attributed characteristics if shared annotation (the rationale being it takes create allele). BLAST-searched database containing 24 recently published high-quality genomes (supplementary Table S1 online). family, performed tests check evidence First, SplitsTree4 programme [22[ produce reticulate phylogenetic trees absence procedure classic, unrooted tree. Alternatively, when event took place, is represented tree closed rectangle. further analyse provide statistically significant proof Pairwise Homoplasy Index (PHI) test [23]. The null hypothesis observed differences due convergent mutations, implies PHI values similar pairs (i.e., do vary physical distance residues). presence distant sites low values. These calculations then result single value details about procedures, please refer online. distribution indicates certain highly recombinogenic (Fig 1). Interestingly, almost sequences part contain either (Table Statistical distributions Komolgorov–Smirnov enrichment (P-value 3.48 × 10−6) 5.5 10−3). combination was significantly Moreover, removed analysis, still find higher vice versa. unable other subgroups depleted (e.g., P-value comparison intragenic tandem repeats, those without 0.38; Fig Figure 1.Analyses recombination. Distribution genome. Most cluster towards left hand side graph, high indicative on far right 1) inter-paralog Note group exists exclusively families. calculation values, see Bruen et al [23] online) methods. PHI, pairwise homoplasy index. Download figure PowerPoint 1. showing Family description Sequences S288c Total number retreived Cell-surface Genes coding (putative) Helicase-like proteins 31 323 Yes No HXT 213 COS 9 116 (Iso)maltases 7 86 AAD 74 6 19 Type I transmembrane sorting receptor vacuolar hydrolases 69 PHO 5 79 Pheromone-regulated protein motif COPII binding/putative integral membrane 4 27 MAL activators ENA 3 TDH 52 FRE 55 Transporters thiamine/nicotinamide riboside 44 TPO 2 FLO, flocculation. Examples representative (for full list, Dataset table contains 15 (PHI value<10−16). First column lists function characterized members ‘sequences S288c’ whereas retrieved’ represents total found classified meet characteristic. Several implications events, focused encodes lectin-like adhesins abiotic surfaces yeast cells [24, 25, 26, 27, 28, 29]. biologically important relatively easy measure quantify 29, 30, 31, 32, 33]. amino terminus rod-shaped globular domain pentapeptide carbohydrate residues surface tissues [34, 35]. formed repetitive pattern heavily glycosylated serine/threonine-rich peptide, thought act rod-like spacer helps display N-terminal environment [36[. perform an in-depth paralogs, gathered 58 (including few partial pseudogenes) NCBI S2 Phylogenetic three subclades FLO1, FLO10 FLO11 strain. line previous findings, extensive variation repeat length, even among same [36, 37]. verified signs across region. Unrooted N- 2A) carboxy-terminal 2B) domains originate region (Figs 2A–C). confirmed complete open reading frames <10−16). Whereas repeats does occur, using internal above). 2.Phylogenetic reveal A–C principle analysis. Shown blue haplotypes (A) nucleotide FLO1-like sequences. (B) C-terminal subset (C) Trees combined network overlay shown B). Such represent current ancestral provides precise visualization history Sequence (27) appears places (B), thus placing extension corner square squares generated predicted event. (D) Reticulate displaying available but sake clarity, only cause artifacts alignments, repeated specifically search recognition binding carbohydrates 35], pathogenicity Candida strains [38, 39], brewing [28, 40]. carbohydrate-binding properties therefore phenotypic consequences. <10−16), domains. subgroup 2D) confirms FLO1 frequency lack non-recombined outgroups prevent identification recombined into present-day Nevertheless, impossible discriminate possible identify groups parents product event, examine breakpoints sequences, previously described LgFLO1 Further, types appear adhesins, potential strong effect. type, outside regions, small microhomology S3 Detailed occurred in-between substrate [35]. subtly alter strength preference influence FLO. second group, domain. length turn combinations Chimaeric Previous studies yeasts Candia albicans glabrata speculated produced contribute species avoiding response [39, 41]. this, pathogenic collecting EPA (epithelial adhesins) C. (11 sequences) ALS (agglutinin-like sequences; 14 albicans. Both cases suggesting common strains. Engineered distinct assess constructing combining 5′ end 3′ another (FLO1, FLO11) classes products engineered retain traditional three-domain structure fungal while differing domains, positions glycosylation sites. functionality chimaera, expressed construct separately non-flocculent non-adherent Expression levels (determined quantitative PCR) comparable constructs, average level around 75% ACT1 transcription level, natural flocculent derived feral EM93 [33]. determined flocculation strength, diverse cell wall hydrophobicity Flo proteins. demonstrate differ 3; 2; S4 details). This demonstrates degree adhesion. 3.Chimaeric cell–cell phenotypes. (cell–cell adhesion) agar-adhesion consisting adhesin, overexpressed express any generates wide array (FLO) Strains expressing broad range depends nature central- domain, weak no measured transformants grown days subsequently washed under gentle stream water estimate propensity stick agar surface. resist washing details), intermediate plate-washing assay. flocculation; WT, wild-type. 2. Overview FLOx FLOy Length Floc. Hydrop. Agar A. Polys WT 11±0 9±5 + 1.0±0.2 — 4,614 98±1 72±6 3.0±0.3 3,510 55±2 50±10 +++ 2.1±0.3 4,104 12±5 80±8 1.5±0.1 FLO11–807 FLO10847–3510 3,471 89±6 61±29 ++ 4.0±0.3 FLO10877–3510 3,441 59±7 63±9 4.3±0.3 FLO11–819 3,453 59±6 80±11 3.9±0.2 FLO11–831 3,465 78±3 87±2 FLO11574–4104 4,338 88±4 69±23 ++++ 4.5±0.2 4,362 87±7 80±3 4.0±0.4 FLO101–864 FLO1799–4614 4,680 75±5 64±2 3.7±0.2 4,395 50±4 65±13 1.2±0.1 FLO111–579 14±6 34±6 0.8±0.1 FLO111–633 4,449 18±1 72±8 0.7±0.1 FLO111–435 3,069 17±2 54±13 3,297 12±1 83±6 0.9±0.2 wild type. origin gene, respectively (with numbers denoting frame). final adhesin. (%) respective (%). after estimated assay (+, adhesive growth; ++++, growth). adherence polystyrene relative WT. CONCLUSION elegant mechanism be molecular toolbox generation findings necessarily completely independently mutation. Instead, propels 2, 3, Intergenic harbour lifestyle-specific interactions environment. Hence, ever-changing reservoir quickly tune way interact conditions opportunities. confirm speculation 42], host. protozoans, spp. where modular coat continuously recombine form avoid system 43]. seems theme (eukaryotic) microorganisms, yielding substantial source variability clearly limited MHC likely mechanisms exist eukaryotes, humans [20]. yeast, allelles: near encoding Importantly, categories, belonging category indicating independent. One unresolved