Two ancient rounds of polyploidy in rice genome

Two ancient rounds of polyploidy in rice genome 1 first-author 2 1 1 2 †E-mail:fanlj@zju.edu.cn †E-mail:fanlj@zju.edu.cn fanlj@zju.edu.cn Institute of Bioinformatics/IBM Biocomputational Lab, Zhejiang University, Hangzhou 310029, China Institute of Bioinformatics/IBM Biocomputational Lab, Zhejiang University, Hangzhou 310029, China Institute of Crop Science, Zhejiang University, Hangzhou 310029, China Institute of Crop Science, Zhejiang University, Hangzhou 310029, China An ancient genome duplication (PPP1) that predates divergence of the cereals has recently been recognized. We report here another potentially older large-scale duplication (PPP2) event that predates monocot-dicot divergence in the genome of rice ( Oryza sativa L.), as inferred from the age distribution of pairs of duplicate genes based on recent genome data for rice. Our results suggest that paleopolyploidy was widespread and played an important role in the evolution of rice. Oryza sativa Polyploidy Genome evolution Age distribution of duplicate genes Monocot-dicot divergence INTRODUCTION 1 Genome duplication or polyploidy is common in flowering plants (Wendel, 2000 2000 ). An ancient polyploidy (paleopolyploidy) is difficult to detect because extensive gene loss or inactivation and chromosomal rearrangements can occur after a large-scale duplication. Genomic sequence analyses performed on yeast ( S. cerevisiae ), Arabidopsis and rice ( Oryza sativa L.) provided strong evidence for ancient polyploidies (Wolfe and Shields, 1997 1997 ; Arabidopsis Genome Initiative, 2000 2000 ; Paterson et al., 2004 2004 ; Vision et al., 2000 2000 ; Goff et al., 2002 2002 ; Simillion et al., 2002 2002 ; Blanc and Wolfe, 2004 2004 ). In these analyses, many non-overlapped duplicate regions with conserved gene order and orientation, or duplicate blocks have been detected intro-genome, and this has been considered to be evidence of polyploidy (Wolfe and Shields, 1997 1997 ). The age distributions duplicate genes have also been widely used to infer potential polyploidy events in plant species (Vision et al., 2000 2000 ; Goff et al., 2002 2002 ; Blanc and Wolfe, 2004 2004 ). Large-scale duplication events lead to a dramatic increase in the number of duplicate genes. Pairs of paralogs of a particular age that correspond to a large-scale duplication are expected to give rise to an independent peak in the age distribution. An initial peak that accounts for the most recently duplicated genes is also expected. Based on this method and data regarding unigenes, widespread paleopolyploidy has been found in many plant species, such as Arabidopsis , Zea mays etc. (Blanc and Wolfe, 2004 2004 ). A paleopolyploidy (PPP1) or “whole-genome duplication” in rice was first suggested by Goff et al.( 2002 2002 ), who based their suggestion on the age distribution of paralogous protein pairs on a genome scale and dated the event at around 40–50 million years ago. Based on their own early assembly of the unfinished genome sequence, Vandepoele et al.( 2003 2003 ) and Paterson et al.( 2003 2003 ) reported the duplication of rice chromatin. Paterson et al.( 2004 2004 ) found many clear non-overlapped duplicate blocks in the rice genome and suggested that a whole-genome duplication event occurred ~70 million years ago, which predates the divergence of the grass family based on the current genome sequence of rice. About 10 years ago, chromosomal duplications (such as chromosome 11–12 and 1–5 duplication) were reported in rice genome based on genetic maps using DNA markers (Kishimoto et al., 1994 1994 ; Nagamura et al., 1995 1995 ; Wang et al., 2000 2000 ). In this study based on the age distribution of duplicate genes and using an effective approach to control background noise, another potentially older paleopolyploidy (PPP2) that predates monocot-dicot divergence was detected in the rice genome. MATERIALS AND METHODS 1 The rice (osa1, version 2.0) and Arabidopsis thaliana (ath1, version 5.0) genome annotation databases were downloaded from The Institute for Genomic Research (TIGR) ( http://www.tigr.org http://www.tigr.org ). A total of 59712 annotated coding sequences of rice (version 2.0) and 26207 of Arabidopsis (version 5.0) encoded by their chromosomal order were compared by reciprocal BLASTN searching ( E < 10 −14 ) for any two chromosomes. Two sequences were defined as one-to-one paralogous or pairs of a duplicate gene when each was the best hit of the other. Coding sequences that show a BLASTN match ( E < 10 −10 ) with members of the rice and Arabidopsis repeat databases by TIGR should first be removed. A pair of duplicate genes identified by this method is presented as a single dot in Fig. 1 1 . Based on dot-plots of pairs of duplicate genes among chromosomes, 13 duplicate blocks corresponding to the latest ancient large-scale duplication (PPP1) and a recent duplication between chromosomes 11 and 12 were identified and duplicate blocks of PPP1 were used in the next analysis of distribution age. Amino acid substitution rates ( d A ) were estimated by the method of Goff et al.( 2002 2002 ): protein sequences of one-to-one paralogs were used to estimate their d A values using the aaml program in the Phylogenetic Analysis by Maximum Likelihood (PAML) package (Yang, 1999 1999 ) with the Dayhoff matrix. The divergence time was calculated based on a molecular clock rate of 9×10 −10 nonsynonymous substitutions per site per lineage per year and 2.25 nonsynonymous substitutions per amino acid change. RESULTS AND DISCUSSION 1 Many non-overlapping duplicate blocks, which cover almost all regions of rice chromosomes, have been reported using the first assembly (osa1, version 1.0) of the rice genome by TIGR (Paterson et al., 2004 2004 ). Here, based on the current version (version 2.0), almost the same duplicate blocks were detected (Fig. 1 1 ). An ancient large-scale duplication, or a whole-genome duplication (PPP1) and a recent segmental duplication between chromosomes 11 and 12 (11–12 duplication, data not shown) are involved in the formation of the duplicate blocks in Fig. 1 1 . The ancient whole-genome duplication was believed to have occurred about 70 million years ago, and thus predates the divergence of the cereals and also the latest detectable ancient polyploidy (PPP1) event in the rice genome. The amino acid substitution rates ( d A ) of pairs of duplicate genes on duplicate blocks of PPP1 were calculated. Three clear peaks were observed in the frequency distribution of their d A values (Fig. 2b 2b ). The duplicate gene pairs on syntenic lines of duplicate blocks that corresponded to PPP1 were further separated from the overall distribution. The results showed that PPP1 gave rise to the first peaks (Fig. 2c 2c ). The second peak was still retained and strongly suggested another older large-scale duplication, which may have been a whole-genome duplication (PPP2) for the rice genome (Fig. 2a 2a ). The peak value of the distribution of PPP2 was ~0.875, which suggests that the ancient polyploidy event occurred about 220 million years ago, which predates monocot-dicot divergence, with an assumption of a molecular clock. A relatively small initial peak was observed in our study. Only inter-chromosome pairs of duplicate genes were used in our analysis and mass intro-chromosome pairs of new duplicate genes were excluded. Therefore, the initial peak in our figures will be small. One concern in our study is the reality of the ancient peak in the age distribution detected in the rice genome. It is located at a very marginal position, which raises the question of whether this is a ghost peak. Three steps were used to prevent such mistakes in this study: (1) masking of repeat sequences (genes); (2) a rigid reciprocal BLASTN best-hit search method was used to identify pairs of duplicate genes, so-called one-to-one paralogs, in a genome; and (3) a stricter standard was used to identify one-to-one paralogs in this study: a successful pair must first have less than an 10 −14 BLASTN match. These three steps basically prevent the misidentification of un-related genes as pairs of one-to-one paralogs. Meanwhile, the age distribution of pairs of duplicate genes from Arabidopsis was estimated using the same method that we used for the rice genome. No similar ancient peak was observed in the Arabidopsis genome (Fig. 1d 1d ). The results suggest that the ancient peak in the age distribution observed in the rice genome is not an artifact of our method, but rather is an actual peak that is due to a potential ancient polyploidy. Our analysis of the Arabidopsis genome failed to show clear evidence of a polyploidy that predated monocot-dicot divergence (Fig. 1d 1d ), although a large-scale duplication near or prior to the divergence of monocots from the dicot had been reported (Vision et al., 2000 2000 ; Simillion et al., 2002 2002 ; Chapman et al., 2004 2004 ). It is possible that rapid evolution in Arabidopsis has made it difficult to detect the ancient polyploidy. In contrast, rice has been suggested to have an evolutionarily stable genome (Ilic et al., 2003 2003 ). Controlling background noise is an important part of large-scale genome analysis. Long evolutionary processes make it difficult to identify the signals of ancient events in a genome. Two key steps were used to control background noise in this study: (1) Only inter-chromosome pairs of duplicate genes were used: both small-scale (gene level) and large-scale duplication played significant roles in genome evolution (Gu et al., 2002 2002 ). Many small-scale duplications occur intro-chromosome and can significantly affect the detection of large-scale duplication events; (2) Only pairs on duplicate blocks of PPP1 were used: clear, long and intact duplicate blocks of the latest ancient polyploidy were identified in the rice genome. No similar long and intact duplicate blocks have yet been observed in other species. Such distinct blocks provide a sound basis for further searching for older evolutionary events in the rice genome. The hidden footprints of some ancient evolutionary events were uncovered and therefore became detectable. References Arabidopsis Genome Initiative Analysis of the genome sequence of the flowering plant Arabidopsis thaliana . Nature , 2000 . 408 796 - 815 . . Analysis of the genome sequence of the flowering plant Arabidopsis thaliana G Blanc , KH Wolfe . Widespread paleopolyploidy in model plant species inferred from age distributions of duplicate genes . Plant Cell , 2004 . 16 1667 - 1678 . . Widespread paleopolyploidy in model plant species inferred from age distributions of duplicate genes BA Chapman , JE Bowers , SR Schulze , AH Paterson . A comparative phylogenetic approach for dating whole genome duplication events . Bioinformatics , 2004 . 20 180 - 185 . . A comparative phylogenetic approach for dating whole genome duplication events SA Goff , D Ricke , TH Lan , G Presting , R Wang , M Dunn , J Glazebrook , A Sessions , P Oeller , PH Quail . A draft sequence of the rice genome ( Oryza sativa L. ssp japonica ) . Science , 2002 . 296 92 - 100 . . A draft sequence of the rice genome (Oryza sativa L. ssp japonica) X Gu , Y Wang , J Gu . Age distribution of human gene families shows significant roles of both large- and small-scale duplications in vertebrate evolution . Nature Genet. , 2002 . 31 205 - 209 . . Age distribution of human gene families shows significant roles of both large- and small-scale duplications in vertebrate evolution K Ilic , PJ SanMiguel , JL Bennetzen . A complex history of rearrangement in an orthologous region of the maize, sorghum, and rice genomes . Proc. Natl. Acad. Sci. USA , 2003 . 100 12265 - 12270 . . A complex history of rearrangement in an orthologous region of the maize, sorghum, and rice genomes N Kishimoto , H Higo , K Abe , S Arai , A Saito , K Higo . Identification of the duplicated segments in rice chromosomes 1 and 5 by linkage analysis of cDNA markers of known functions . Theor. Appl. Genet. , 1994 . 88 722 - 726 . . Identification of the duplicated segments in rice chromosomes 1 and 5 by linkage analysis of cDNA markers of known functions Y Nagamura , T Inoue , B Antonio , T Shimano , H Kajiya , A Shomura , S Lin , Y Kuboki , Y Harushima , N Kurata , 等 . Conservation of duplicated segments between rice chromosomes 11 and 12 . Breed Sci. , 1995 . 45 373 - 376 . . Conservation of duplicated segments between rice chromosomes 11 and 12 AH Paterson , JE Bowers , DG Peterson , JC Estill , BA Chapman . Structure and evolution of cereal genomes . Curr. Opin. Genet. Dev. , 2003 . 13 644 - 650 . . Structure and evolution of cereal genomes AH Paterson , JE Bowers , BA Chapman . Ancient polyploidization predating divergence of the cereals, and its consequences for comparative genomics . Proc. Natl. Acad. Sci. USA , 2004 . 101 9903 - 9908 . . Ancient polyploidization predating divergence of the cereals, and its consequences for comparative genomics C Simillion , K Vandepoele , MCE Van Montagu , M Zabeau , Y Van de Peer . The hidden duplication past of Arabidopsis thaliana . Proc. Natl. Acad. Sci. USA , 2002 . 99 13627 - 13632 . . The hidden duplication past of Arabidopsis thaliana K Vandepoele , C Simillion , Y Van de Peer . Evidence that rice and other cereals are ancient aneuploids . Plant Cell , 2003 . 15 2192 - 2202 . . Evidence that rice and other cereals are ancient aneuploids TJ Vision , DG Brown , SD Tanksley . The origins of genomic duplications in Arabidopsis . Science , 2000 . 290 2114 - 2117 . . The origins of genomic duplications in Arabidopsis S Wang , K Liu , Q Zhang . Segmental duplications are common in rice genome . Acta. Bot. Sin. , 2000 . 42 1150 - 1155 . . Segmental duplications are common in rice genome JF Wendel . Genome evolution in polyploids . Plant. Mol. Biol. , 2000 . 42 225 - 249 . . Genome evolution in polyploids KH Wolfe , DC Shields . Molecular evidence for an ancient duplication of the entire yeast genome . Nature , 1997 . 387 708 - 713 . . Molecular evidence for an ancient duplication of the entire yeast genome Z Yang . Phylogenetic Analysis by Maximum Likelihood (PAML) , 1999 . Version 2 : London, UK University College , . Phylogenetic Analysis by Maximum Likelihood (PAML) 10.1631/jzus.2005.B0087 Project supported by the National Natural Science Foundation of China (Nos. 30270810, 90208022 and 30471067) and IBM Shared University Research (Life Science), China 2004-10-27 2004-11-05 2005-01-20 2022-01-05 Copyright © 2005, Journal of Zhejiang University Science 2005 Journal of Zhejiang University-SCIENCE B (Biomedicine and Biotechnology) 02 6B