Allele Diversity at Orthologous Candidate Genes in GCP Crops (ADOC project) Dominique This1 , Brigitte Courtois1 , Romain Philippe1 , Pierre Mournet1 , Claire Billot1 , Jean-Christophe Glaszmann1 , Roland Schaftleitner2 , Reinhardt Simon2 , Percy Rojas2 , Merideth Bonierbale2 , Rajeev Varshney3 , C. Tom Hash3 , Hari Upadhyaya3 , Spurthi Nayak3 , Dominique Brunel4 , Redouane El Malki4 , Marie Christine Le Paslier4 , Kenneth McNally5 , Michael Baum6 , Wafaa Choumane6 , Maria Von Korff6 , Matthew Blair7 , Martin Fregene7 A D O C Allelic 1 Diversity 2 3 Orthologous 4 5 Candidate genes 6 7 UMR DAP 1098, CIRAD TA A-96/03, Av. Agropolis, 34398 Montpellier cedex 5, France CIP P.O. Box 1558 La Molina, Lima 12, Lima, Peru ICRISAT Patancheru, Andhra Pradesh 502 324, India INRA, UR 1279 Etude du Polymorphisme des Génomes Végétaux, CEA Institut de Génomique/ Centre National de Génotypage, 2, rue Gaston Crémieux, CP5724, Evry F-91057, France IRRI DAPO Box 7777 Manila 1301, Philippines ICARDA P.O. Box 5466 Aleppo, Syrian Arab Republic CIAT km 17 Recta Cali-Palmira, AA6713, Cali, Colombia Water Stress Signal perception and transduction Transcription control ERECTA DREB1A, DREB2A ASR Stress response mechanisms Water use efficiency indica sadri/basmati aus/boro japonica Population structure of the rice reference collection – Sugar metabolism / growth regulation K McNally (IRRI), B Courtois, R Philippe (CIRAD) (Vacuolar) invertases, SuSy, SPS Around 10 Mbp sequences were generated and compared within and across species. Some new orthologs were identified by using degenerate primers and reconciliated taxonomic trees (Nayak et al, 2009), and allelic series derived by specific PCR Allelic sequences were generated in reference collections of barley (Hv), sorghum (Sb), rice (Os), potato (Stbr), cassava (Me), bean (Pv) and chickpea (Ca) (around 300 accessions for each crop) Exposure to drought-prone environments in some regions, or different selection strategies may have shaped the diversity patterns of our target genes. Population structure also reflects domestication and selection’s history, to be taken into account. Poaceae ancestor anAsr4 anAsr3 Distribution of θ on total sites for Oryza sativa 25.0% 20.0% 0.778 / 3.98 0.138 / 0.44 0.838 / 0.9 0.72 / 2.9 1 Asr lost A2 OsDREB2A 15.0% 10.0% anAsr4 anAsr3b A4 A6 A2 0.0% Monocots ERL anAsr3a Dicots ER Class of θ (10 -3) Distribution of π on total sites for Oryza sativa 0.569 / 1.03 -50 – 70 Ma 25.0% Dicots ERL Monocots ER OsDREB2A 0.819 / 2.33 0.814 / 3.52 OsAsr3 OsSUSY1 OsAsr6 10.0% 5.0% SbAsr2 under pseudogeneisation? >20 13.75 12.75 11.75 10.75 9.75 8.75 7.75 6.75 5.75 4.75 Diversity level of our target genes differed clearly between the different genes and gene family members within a crop. and between orthologs when comparing several crops (here for ERECTA and ERECTA-like genes). A few outliers were identified, compared to average diversity data. DREB2A presented a very low diversity in all species, while barley’s vacuolar invertase-related sequences and potato’s StbrSUSY1 presented a very high diversity within the reference collections analyzed. Distribution of Fu & Li's D* in Oryza sativa Distribution of Tajima's D in Oryza sativa 25.0% 25.0% 20.0% 20.0% 15.0% STS % OsAsr3 10.0% 5.0% OsAsr3 15.0% 10.0% 5.0% Tajima's D 5 4.2 3.4 Fu & Li's D* Distribution of Fu & Li's D* in tropical japonica rice Distribution of Tajima's D in tropical japonica rice 40.0% 40.0% 30.0% 30.0% STS % 20.0% OsAsr3 10.0% 20.0%OsAsr3 10.0% >4 3.2 2.8 2.4 2 1.6 1.2 -0.8 -1.2 -2 -1.6 -2.4 -2.8 2.4 2 1.6 1.2 0.8 0 0.4 -0.4 -0.8 -2 -1.2 -1.6 -2.4 -3.2 0.0% 0.0% <4 STS % 2.6 -3.8 -5.4 -3 0.0% 3.6 2 2.8 1.2 0.4 -0.4 -2 -1.2 -2.8 -3.6 0.0% -4.6 STS % Comparison between whole sequenced genomes allow some comparison between orthologs and paralogs within a gene family - Here the Asr (ABA- stress-ripening) family. Some hypothesis can be proposed, following Salse et al. (2008) model for Poaceae genome evolution, to infer a minimal gene number in the poaceae ancestor (6 for Asr). Several Asr were lost following whole genome duplication and in the rice lineage. A translocation event disrupted the microsyntheny between rice and sorghum around Asr1-Asr2 tandem. NB: in agreement with data on Brachypodium and maize (not shown here) 3.75 Class of π (10 -3) Candidate genes outliers in rice, based on data at the genome level from Caicedo et al (2007) (111 randomly chosen STS). Two diversity indices (Π and Θ ) are shown. Sorghum This evolution model for Asr gene family is based on Salse et al (2008) model of poaceae genome evolution, simplified in order to represent only chromosomal regions harbouring Asr genes, in a comparative analysis between rice and sorghum. Orthology relationship between rice and sorghum were infered from phylogenetic and microsynteny analysis (Philippe et al, in preparation). 2.75 Phylogenetic tree of the ERECTA and ERECTA-like gene family (established in collaboration with J. Masle, ANU), showing diversity levels (Hd: haplotype diversity and Π: nucleotide diversity) for several orthologs or paralogs. SbERL diversity is much lower than orthologs (OsERL and HvERL) and paralogs (SbER1 and SbER2) 1 s5 s8 A6 A2 SbAsr5b A4 SbAsr3a SbAsr4 s10 s4 SbAsr3b s6 1.8 s3 s9 A11A12 0.2 SbAsr6 SbAsr5a 0 A1 A5 1.75 0.75 0.0% SbAsr1 SbAsr2 0.8 Translocation of (Asr1+Asr2) tandem OsAsr3 r2 0.4 Rice r6 r4 15.0% 0 r11 r12 20.0% -0.6 A4 OsAsr4 r1 r5 0.777 / 1.02 -0.4 OsAsr5 A6 A2 -1.4 OsAsr1 OsAsr2 OsAsr6 A11 A12 Percentage of STS 0.806 / 1.4 A1 A5 OsSUSY1 9.75 anAsr5b 8.75 A11 A12 OsAsr3 7.75 anAsr5a 6.75 A1 A5 0 anAsr1 anAsr2 anAsr6 -2.2 2 Asr lost OsAsr6 5.0% 0.66 / 2.4 5.75 3 Asr lost Chromosomal rearrangement Hd / Π (10−3) Rice Sorghum Barley Chickpea Bean A4 A6 4.75 A11 A12 3.75 ~ 90 Ma A4 2.75 A1 A5 A8 anAsr5 … … Whole genome duplication A11 1.75 A7 A5 0.75 anAsr1 anAsr2 anAsr6 Percentage of STS Six gene families were selected as the initial subset of target genes. They act at different levels of the drought stress response. Fu & Li's D* Tajima's D indica aus boro sadri/basmati Tropical japonica Tests for selection were performed to estimate whether the considered genes followed the model of neutral evolution: D (Tajima, 1959) and D* (Fu and Li’s, 1993) are shown here for OsAsr3, and compared to randomly chosen data from Caicedo et al (2007), either for Oryza sativa globally (up), or for the temperate japonica subgroup (down). The haplotype network of OsAsr3 in cultivated rice shows a much higher diversity in indica than japonica, and a very distinct haplotype (H4) for a few accessions. (Philippe et al, submitted) Temperate japonica Haplotype networks are constructed for 4 rice sucrose synthase genes using NETWORK v4.5. Each node represents an haplotype and segments correspond to mutations (SNP or Indel). The different colors represent the varietal groups. Haplotype networks are sometimes influenced by population structure (An example here for OsSUSY1 and OsSUSY4, where a clear distinction can be made between japonica and indica accessions, except putative introgressions), but the diversity pattern is less clear for some other genes like OsSUSY3. A few non-synonymous changes are highlighted here for OsSUSY2 and OsSUSY4. When computing indices for selective pressure on our data, a few candidate genes that could play a role in drought tolerance were highlighted (here OsAsr3, suggesting a balanced selection at the species level, and a directional selection within the tropical japonica group). The haplotype network of this gene may provide some clues to explain this feature. Other candidate outliers include HvAsr5, OsSUSY1, MeSUSY3 and PvERECTA (Positive D), StbrAsr1 and SbAsr1 (Negative D) This study highlights the complexity of evolutionary and diversity patterns of candidate genes within gene families. Genome evolution as well as population structure and some selection pressure have shaped genes’ diversity. Functional inference based solely on orthology relationship should therefore be considered with caution. Functional analysis of haplotypes identified in this research work and eco-geographical data are now required in order to define potential candidate genes and favourable alleles for drought tolerance in GCP crops. References: