The evolution of transposon repeat-induced point mutation in the genome of Colletotrichum cereale: Reconciling sex, recombination and homoplasy in an ‘‘asexual” pathogen
Introduction
Mobile genetic elements such as transposons (TEs) are abundant in eukaryotes, and with the exception of Plasmodium falciparum (Gardner et al., 2002), the causative agent of human malaria, TEs populate the DNA of all well-studied organisms. TEs may occupy a substantial proportion of the host genome: 60% of the maize genome is transposon-derived (Messing and Dooner, 2006), as is 38% of the mouse genome (IMGSC, 2001) and 45% of the human genome (IHGSC, 2001). In contrast, the genomes of many eukaryotes are composed of relatively few transposons: for example, only 4.3% of the chicken genome is transposon-derived (Wicker et al., 2005). A relatively small contribution of TEs to the genomes of fungi is typical, with only 3.1% of the Saccharomyces cereviseae genome comprised of TEs (Goffeau et al., 1996), while 8.2–14% of the genome of the rice blast fungus, Magnaporthe oryzae may be derived from TEs (Dean et al., 2005, Thon et al., 2006).
Because TEs are able to move about the host genome and insert into a host’s DNA through either cut-and-paste (DNA, or Class II transposons) or copy-and-paste mechanisms via RNA intermediates (retro, or Class I transposons), these elements can exert a significant influence on the fitness and evolutionary potential of their hosts through events such as insertional mutagenesis, disrupted or enhanced gene expression or gross chromosomal rearrangements (Hua-Van et al., 2005). Given the numerous ways that transposition can impact the genome, a variety of methods have evolved to safeguard the host against the effects of potentially deleterious insertions or unsupportable transposition rates. In several organisms, highly specific targeting mechanisms have been shown to limit TE integration to non-essential genomic regions, thereby protecting host integrity. Most TEs appear to have integration “hotspots” that are dictated by nucleotide sequence, patterns of hydrogen bonds, DNA-bending proteins and/or DNA conformation (Chalmers et al., 1998, Bender and Kleckner, 1992, Ketting et al., 1997, Liu et al., 2005). Well known examples of targeted integration in fungi include the Ty retroelements of S. cerevisiae, which insert preferentially upstream of pol III transcribed genes and in silent chromatin regions (Zou et al., 1996, Devine and Boeke, 1996, Chalker and Sandmeyer, 1992) and the retrotransposon Tf1 in Schizosaccharomyces pombe, which exhibits a clear preference for integration in tandem and divergent intergenic pol II promoter regions (Singleton and Levin, 2002).
Filamentous fungi actively regulate repetitive sequences through silencing mechanisms such as quelling (RNA silencing) (Cogoni et al., 1996), meiotic silencing (Shiu et al., 2001), and repeat-induced point mutation (RIP) (Cambareri et al., 1989). The RIP mutation process is remarkably efficient in disabling transposable elements through the detection and subsequent mutation of duplicated sequences longer than ∼400 bp (Watters et al., 1999). Just prior to karyogamy, GC-to-AT transitions are induced in duplicate sequences sharing >80% similarity, with as many as 30% of GCs converted to ATs (Cambareri et al., 1989) and repetitive DNA remaining susceptible to “RIPping” through six generations (Cambareri et al., 1991). Since its initial discovery in Neurospora crassa (Selker et al., 1987), the RIP-mutation process has been identified experimentally or through sequence analysis in the ascomycetes Aspergillus fumigatus (Neuveglise et al., 1996), Aspergillus nidulans (Nielson et al., 2001, Clutterbuck, 2004), Aspergillus oryzae (Montiel et al., 2006), Fusarium oxysporum (Hua-Van et al., 2001), Leptosphaeria maculans (Attard et al., 2005), M. oryzae (Nakayashiki et al., 1999, Ikeda et al., 2002), N. tetrasperma (Bhat et al., 2004); Ophiostoma sp. (Bouvet et al., 2007) and Podospora anserina (Graia et al., 2001) and in the basidiomycete Microbotryum violacum (Hood et al., 2005), although RIP-mutation activity in these species has always been found in a much less aggressive form than that observed in N. crassa (Galagan and Selker, 2004).
We have been developing several molecular tools, including transposon-based marker systems, to increase our understanding of the recent emergence of the mitosporic ascomycete fungus Colletotrichum cereale as a pathogen of turfgrasses and its benign existence in natural grassland and agroecosystems. Beginning in the mid 1990s, C. cereale emerged as one of the most destructive pathogens of cool-season turfgrasses (Smiley et al., 2005). Outside of golf course greens, C. cereale is a common inhabitant of a wide range of C3 cereals and grasses of the grass subfamily Pooideae, where it survives without inducing noticeable levels of disease (Crouch et al., 2006, J.A. Crouch and B.I. Hillman, unpublished data). Little is known about C. cereale populations and, until recently, the fungus was thought to be conspecific with C. graminicola, a pathogen of corn (Crouch et al., 2006). Two major C. cereale lineages (clades A and B) have been described using sequences of the internal transcribed spacer (ITS) region of the ribosomal DNA (Crouch et al., 2005) and multilocus phylogenetic analyses (Crouch et al., 2006), but the evolutionary processes that shaped these lineages remain largely unexplored.
During the course of surveying the C. cereale genome for TEs suitable for use as molecular markers, we observed that many of this organism’s transposon sequences were distinguished by a pronounced A + T nucleotide bias; subsequent bioinformatics analysis demonstrated this bias reflected the characteristic patterns of RIP-like C → T and G → A transitions. In this study, five different species of transposons were identified from the two major lineages of C. cereale in RIPped and ‘‘normal”, non-mutated forms: two DNA transposons, two species of long-terminal repeat (LTR) retrotransposons and one non-LTR retrotransposon. In this paper, we describe these C. cereale transposable elements and document how the process of RIP mutation has considerably altered 21 of 35 unique transposon copies surveyed in a lineage-specific manner. We then employ the Ccret2 retrotransposon pol gene sequence to explore the impact of RIP-mutated transposons when these elements are used to generate evolutionary inferences for phylogenetic and population genetic analyses.
Section snippets
Construction of genomic DNA libraries and the identification of repetitive transposon DNA
Genomic DNA was isolated from fungal mycelia using a standard phenol:chloroform extraction protocol (Sambrook and Russell, 2001). C. cereale genomic DNA libraries were constructed from the EcoRI-digested DNA of isolates PA-50231 (clade A) and PA-50005 (clade B) in the plasmid vector pGEM3zf+ (Promega, Madison, WI); the culture and origin of these isolates was described previously (Crouch et al., 2006). To screen for repetitive sequences, insert-bearing colonies were transferred to Colony/Plaque
Identification and nomenclature of TEs from the C. cereale and C. sublineolum genomes
A total of 35 unique transposon copies were identified using a combination of methods: (1) five plasmid clones were identified as containing repetitive elements due to their strong hybridization signal when probed with C. cereale total genomic DNA; (2) 13 TEs were found during the course of random sequencing of the PA-50005 genomic DNA library; (3) six TEs were sequenced from subclones of the cosmid 9F8 identified through colony hybridization; and (4) nine C. cereale and two C. sublineolum
Discussion
From an organismal standpoint, the lineage-specific distribution of RIP mutation in C. cereale—absent in clade A, present in clade B—is an important contribution to our understanding of how this species has evolved. The widespread identification of RIPping in diverse ascomycete species, including the closely related C. falcatum (J.A. Crouch and B.I. Hillman; unpublished data), strongly discourages the conclusion that the clade A genome might be ‘‘RIP-free”, but our inability to detect a RIP
Acknowledgments
We thank Lisa Vaillancourt for providing the C. graminicola and C. sublineolum cultures used in this study. This work was funded by grants from the Rutgers Center for Turfgrass Science to B.I.H. and B.B.C and by the New Jersey Agricultural Experiment Station. We gratefully acknowledge financial support for J.A.C.’s graduate studies provided by a U.S. Environmental Protection Agency Science to Achieve Results (STAR) Graduate Fellowship, a Rutgers Excellence Fellowship, the Robert White-Stevens
References (91)
- et al.
Basic local alignment search tool
J. Mol. Biol.
(1990) - et al.
Elimination of active Tad elements during the sexual phase of Neurospora crassa life cycle
Fungal Genet. Biol.
(2001) - et al.
Truncated and RIP-degenerated copies of the LTR retrotransposon Pholy are clustered in a pericentromeric region of the Leptosphaeria maculans genome
Fungal Genet. Biol.
(2005) - et al.
Characterization of three DNA transposons in the Dutch elm disease fungi and evidence of repeat-induced point (RIP) mutations
Fungal Genet. Biol.
(2007) - et al.
IHF modulation of Tn10 transposition: sensory transduction of supercoiling status via a proposed protein/DNA molecular spring
Cell
(1998) MATE transposable elements in Aspergillus nidulans: evidence of repeat-induced point mutation
Fungal Genet. Biol.
(2004)- et al.
Global heterochromatic colocalization of transposable elements with minisatellites in the compact genome of the pufferfish Tetraodon nigroviridis
Gene
(2004) - et al.
Target-site preferences of Sleeping Beauty transposons
J. Mol. Biol.
(2005) - et al.
Organization and variability of the maize genome
Curr. Opin. Plant Biol.
(2006) - et al.
Recombination between retrotransposons as a source of chromosome rearrangements in the yeast Saccharomyces cerevisiae
DNA Repair
(2006)