Mutator System Derivatives Isolated from Sugarcane Genome Sequence

M. E. Manetti 0 M. Rossi 0 G. M. Q. Cruz 0 N. L. Saccaro Jr. 0 M. Nakabashi 0 V. Altebarmakian 0 M. Rodier-Goud 0 D. Domingues 0 A. D'Hont 0 M. A. Van Sluys 0 0 Communicated by: Blake Meyers 1 ) Departamento de Botnica-IB-USP , GaTE Lab, Brasil , Rua do Mato, 277, 05508-900 So Paulo , SP, Brazil Mutator-like transposase is the most represented transposon transcript in the sugarcane transcriptome. Phylogenetic reconstructions derived from sequenced transcripts provided evidence that at least four distinct classes exist (IIV) and that diversification among these classes occurred early in Angiosperms, prior to the divergence of Monocots/ Eudicots. The four previously described classes served as probes to select and further sequence six BAC clones from a genomic library of cultivar R570. A total of 579,352 sugarcane base pairs were produced from these Mutator system BAC containing regions for further characterization. The analyzed genomic regions confirmed that the predicted structure and organization of the Mutator system in sugarcane is composed of two true transposon lineages, each containing a specific terminal inverted repeat and two transposase lineages considered to be domesticated. Each Mutator transposase M. E. Manetti, M. Rossi and G. M. Q Cruz these authors contributed equally to this work. - The forces that shape the organization of plant genomes are relevant to eukaryotic evolution. Tolerance to polyploidy is a recurrent event and genome size may vary considerably within closely related species. Under this view, the Poaceae species are rather interesting to compare since genome size can vary over 40 fold (245 Mbp to 25,456 Mbp) and ploidy levels range from basic diploids to decaploids (Arumuganathan and Earle 1991; Paterson et al. 2009). Even with this large genome size and ploidy level variation, syntenic regions are detected which show maintenance of gene order which suggest their ancestral origin and conservation of gene function. The C paradox proposed by Thomas (1971) and reviewed in (Jones and Pasakinskiene 2005) somewhat links plasticity of the genome through its association with organism complexity. Among the genetic elements capable of altering genome structure, transposable elements (TEs), originally described by McClintock (1956), are responsible for plant genome size variation as has been repeatedly demonstrated for plants, particularly in Poaceae (Shirasu et al. 2000; Bennetzen and Ramakrishna 2002; Jannoo et al. 2007; Wicker et al. 2001; Vitte et al. 2007). Recent studies are revealing that not only are TEs powerful mechanisms for genome expansion and retraction (Kalendar et al. 2000; Piegu et al. 2006), they also remodel gene content through the generation of new genes (Kazazian 2004; Cordaux et al. 2006) and provide new regulatory networks for altering gene expression (Kashkush et al. 2003; Muotri et al. 2007; Feschotte 2008). TEs are classically sorted into two groups: transposons and retrotransposons. DNA transposons are elements that mostly propagate through a cut and paste mechanism dependent on the presence of a specific transposase protein that recognizes the sub-terminal and terminal portion of the element. The transposase encoded by a particular autonomous element preferentially acts on several related copies that share nucleotide similarity at their ends. Movement of retrotransposons depends on active transcription, which provides the substrate to be reverse transcribed into a DNA copy that is reinserted into the genome to increase its copy number after each propagation cycle. Long terminal repeat (LTR) retrotransposons, the most abundant retroelements in plants, are classified on the basis of polyprotein domain order into two major families: the Ty1/ copia-like, and Ty3/gypsy-like elements. The replicative potential of these elements has been associated with genome size variation (Kalendar et al. 2000; Du et al. 2006; Ammiraju et al. 2007; Vitte et al. 2007). Expanding the genome-based knowledge from well-studied model plants such as Arabidopsis, rice and maize is fundamental to reveal the impact of these elements in other plant genomes. Studies on barley (Shirasu et al. 2000) and hexaploid wheat (Devos et al. 2005) support the close association of TEs with genome structure. The present work focuses on sugarcane, a Poaceae member that is cultivated through clonal cuttings, not sexual seeds, of hybrids selected from crosses between two species having different ploidy levels. Modern cultivars were obtained from crosses between the domesticated sugar storing species Saccharum officinarum and the wild non-sugar species Saccharum spontaneum, followed by several generations of back-crossing and clonal selection (Grivet and Arruda 2002). Both Saccharum species have autopolyploid origin, S. spontaneum (X 0 8) with 2n 0 40 to 128, and S. officinarum (X 010) with 2n 0 80. The corresponding monoploid genome sizes are 760 Mb for S. spontaneum and 930 Mb for S. officinarum (DHont and Glaszman 2001). Thus, modern sugarcane cultivars are highly polyploid and aneuploid, with a chromosome number ranging from 100 to 130, of which 7080 % comes from S. officinarum, 1020 % from S. spontaneum, and few chromosomes are derived from inter-specific recombination. Modern sugarcane monoploid genome size is roughly 1Gb based on previous studies (DHont 2005), while its close relatives Sorghum bicolor and Oryza sativa present 730 Mb and 430 Mb, respectively. Analysis of the sugarcane transcriptome revealed a diverse collection of transposable elements being expressed, among which Mutator-like transposases and Hopscotch-like retrotransposons were the most abundant TE transcripts (Rossi et al. 2001; Arajo et al. 2005). Phylogenetic reconstructions based on 173 amino acids (aa) of the Mutator-like transposases provided evidence that at least four distinct classes (IIV) exist, and suggest that diversification of the Mutator system occurred early in the evolution of Angiosperms, prior to the divergence of Monocots/Eudicots (Rossi et al. 2004). Saccaro et al. (2007) performed an in silico study on the rice genome and proposed, based on copy number and structural features such as presence of flanking terminal inverted repeats (TIRs), that Classes I and II correspond to bona fide transposons. These authors reported that not only are the rice Mutator-like transposases flanked by TIRs, but also they capture host DNA surrounded by TIRs. These host genome carriers were originally named PackMULES (Jiang et al. 2004). The host DNA surrounded by TIRs also contains a transposase-like domain, previously called Transduplicated-MULEs (Juretic et al. 2005), was also found. Classes III and IV clusters in rice and Arabidopsis correspond to previously described classes called MUSTANGs, which represent domesticated transposases (Cowan et al. 2005). Saccaro et al. (2007) also showed that copy number differs greatly between all four classes and, at least in grasses, there was a (...truncated)