CHG and CHH methylation of most sequences was also higher in embryo (Fig. 1, C to F Fig. 3, B, C, and E and fig. S5). The dme mutation uniformly restored CG methylation, while uniformly reducing CHG and CHH methylation (Fig. 1 and Fig. 3, A to C). Methylation in all contexts was higher in embryo than in aerial tissues (Fig. 1 and fig. S4), withparticularlyextensiveCHH hypermethylation: We identified 10,858loci covering21.88millionbp with an absolute change in CHH methylation of at least 5% (P 0.0001, Fisher���s exact test), 96.8% ofwhich(10,510)weremoremethylatedinembryo (table S3). Virtually, genome-wide CG demeth- ylation of the maternal endosperm genome is thus accompanied by similarly extensive CHH hyper- methylation in the embryo. We investigated the source of the substantial non-CG hypermethylation in wild-type endosperm compared with the embryo (table S3) by examining methylation differences between embryo and dme endospermofsequencesthat weremoremethylated in wild-type endosperm than in embryo (Fig. 3, B and C, green trace). If endospermhypermethylation were random, we would expect to see no correla- tion between hypermethylation in wild-type and dme endosperm. Our analysis showed that for both CHG and CHH contexts, loci hypermethylated in Fig. 2. Associations be- tween endosperm methyla- tion, siRNAs, and expression. (A) Box plots showing siRNA abundance within 50-bp win- dows in the entire Arabidopsis genome(All)andinsequences hypermethylated in WT endo- sperm compared with the em- bryo in the CHG and CHH contexts. (B) Box plots show- ing differences in gene ex- pression between embryo and endosperm for all genes (n = 21,021), genes with 5��� hypo- methylation in endosperm (n = 1097), and genes with 3��� hypomethylation in endo- sperm (n = 505). Each box encloses the middle 50% of the distribution, with the horizontal line marking the median and the dot marking the mean. The lines extending from each box mark the minimum and maximum values that fall within 1.5 times the height of the box. Fig. 3. Genome-wide demeth- ylation of endosperm. (A to C) Kernel density plots of the differ- ences between embryo and WT endosperm methylation (blue trace) and the differences be- tween embryo and dme endo- sperm methylation (red trace). The green trace in (B) and (C) represents methylation differ- ences between embryo and dme endosperm for windows with absolute fractional methylation increase in WT endosperm com- pared with embryo of at least 0.4 in the CHG context (B) (n = 135) or at least 0.2 in the CHH context (C) (n = 6168). Methyl- ation differences for the 3��� MEA repeats, FWA, FIS2, PHE1, and MPC are indicated specifics are listed in table S2. (D and E) All TAIR8-annotated genes (28,244) were aligned at the 5��� end and stacked from the top of chromo- some 1 to the bottom of chro- mosome 5. Embryo methylation is displayed as a heat map in the left panel, differences between embryo and WT endosperm in the right panel. CG methylation is shown in (D), CHG in (E). www.sciencemag.org SCIENCE VOL 324 12 JUNE 2009 1453 REPORTS on June 11, 2009 www.sciencemag.org Downloaded from

wild-type endosperm had a strong tendency to be hypermethylated in dme endosperm as well (Fig. 3, B and C, green trace), despite the overall reduction of non-CG methylation caused by the dme muta- tion. Endosperm hypermethylation is thus a highly specific, RNAi-targeted process. We calculated methylation levels of sequences either known or strongly inferred to cause imprinted expression of five Arabidopsis genes (3, 5���8): the MEA 3��� repeats, the FWA promoter and start of transcription, the FIS2 promoter, the PHE1 3��� re- peats, and the MPC gene and flanking regions (Fig. 3, A to C, and table S2). MEA methylation was reduced from 88% CG, 39% CHG, and 42% CHH in embryo to 63% CG, 16% CHG, and 17% CHH in wild-type endosperm. MEA CG methyla- tion was restored to 87% in dme endosperm, whereas CHG (13%) and CHH (8%) methylation was further reduced. The other four genes behaved similarly (Fig. 3, A to C, and table S2), in line with the overall trends. Imprinted genes are thus not ex- ceptional sequences specifically targeted for de- methylation in the central cell but rather part of a nearly universal process that reshapes DNA methylation of the entire maternal genome in the endosperm (14). Imprinted expression of genes regulated by allele-specific DNA methylation could potentially arise whenever a transposable element insertion or a local duplication near a gene���s regulatory sequences induces methyla- tion and gene silencing in other tissues, includ- ing the paternal endosperm genome. Genomic imprinting is a fast-evolving process driven by genetic conflict between parents (1). In mammals, which exhibit virtually global CG meth- ylation (15), imprinting is orchestrated in part by differential methylation of specific sequences in the gametes (16). Arabidopsis, which targets methylation primarily to transposable elements (9), apparently adapted a radical implementation of imprinting by partially suspending its transposon suppression system and globally demethylating central cell DNA, resulting in a hypomethylated maternal endo- sperm genome. Because the endosperm genome is not transmitted to the next generation, transient transposon activation is likely to carry a fairly low cost, especially in an organism with few functional transposons, like Arabidopsis. Transposon activa- tion and siRNA accumulation in the central cell might actually contribute to enhanced methylation and silencing of elements in the egg cell (and later the embryo) through siRNA transport (17), which could be the original selective force driving the evolution of central cell demethylation. An analo- gous mechanism has recently been proposed to operate between the vegetative and reproductive cells of pollen (18). It is an open question whether other plants, particularly those with more aggres- sive transposable elements, have adopted a similar strategy. References and Notes 1. R. Feil, F. Berger, Trends Genet. 23, 192 (2007). 2. Y. Choi et al., Cell 110, 33 (2002). 3. M. Gehring et al., Cell 124, 495 (2006). 4. J. H. Huh, M. J. Bauer, T.-F. Hsieh, R. L. Fischer, Cell 132, 735 (2008). 5. P. E. Jullien, T. Kinoshita, N. Ohad, F. Berger, Plant Cell 18, 1360 (2006). 6. T. Kinoshita et al., Science 303, 521 (2004). 7. G. Makarevich, C. B. R. Villar, A. Erilova, C. Kohler, J. Cell Sci. 121, 906 (2008). 8. S. Tiwari et al., Plant Cell 20, 2387 (2008). 9. I. R. Henderson, S. E. Jacobsen, Nature 447, 418 (2007). 10. S. J. Cokus et al., Nature 452, 215 (2008). 11. R. Lister et al., Cell 133, 523 (2008). 12. A. Meissner et al., Nature 454, 766 (2008). 13. Materials and methods are available as supporting material on Science Online. 14. P. E. Jullien et al., PLoS Biol. 6, e194 (2008). 15. M. G. Goll, T. H. Bestor, Annu. Rev. Biochem. 74, 481 (2005). 16. A. Munshi, S. Duvvuri, J. Genet. Genomics 34, 93 (2007). 17. Y. Z. Han, B. Q. Huang, S. Y. Zee, M. Yuan, Planta 211, 158 (2000). 18. R. K. Slotkin et al., Cell 136, 461 (2009). 19. We thank L. Tonkin for performing Illumina sequencing, J. Shin for gene annotation, and S. Henikoff for sharing unpublished data. This work was partially funded by an NIH grant (GM69415) to R.L.F. A.Z. is a fellow of the Jane Coffin Childs Memorial Fund for Medical Research. Sequencing data are deposited in GEO with accession number GSE15922. Supporting Online Material www.sciencemag.org/cgi/content/full/324/5933/1451/DC1 Materials and Methods Figs. S1 to S6 References 17 February 2009 accepted 1 May 2009 10.1126/science.1172417 Hyper-Recombination, Diversity, and Antibiotic Resistance in Pneumococcus William Paul Hanage,1* Christophe Fraser,1 Jing Tang,2 Thomas Richard Connor,1 Jukka Corander2 Streptococcus pneumoniae is a pathogen of global importance that frequently transfers genetic material between strains and on occasion across species boundaries. In an analysis of 1930 pneumococcal genotypes from six housekeeping genes and 94 genotypes from related species, we identified mosaic genotypes representing admixture between populations and found that these were significantly associated with resistance to several classes of antibiotics. We hypothesize that these observations result from a history of hyper-recombination, which means that these strains are more likely to acquire both divergent genetic material and resistance determinants. This could have consequences for the reemergence of drug resistance after pneumococcal vaccination and also for our understanding of diversification and speciation in recombinogenic bacteria. MDNAbacteria any undergo homologous re- combination, in which short tracts of in the recipient are replaced by the corresponding tract from a donor strain, re- sulting in a mosaic of DNA from different an- cestors (1). Although this occurs mainly within species and declines markedly with increasing sequence divergence between donor and recipi- ent (2), occasional gene transfers between species do occur. Such events have the potential to intro- duce new phenotypes, such as virulence or anti- biotic resistance, into a new genetic background that may or may not be the same as the species of the donor strain (3���7) and may have considerable impacts on bacterial evolution and human health. One group in which homologous recombina- tion is frequent is the mitis group streptococci. This includes the major human pathogen Streptococcus pneumoniae, the pneumococcus, which is respon- sible for at least 1 million deaths per year worldwide (8). The closely related species S. oralis, S. mitis, and S. pseudopneumoniae (among others) have a history of taxonomic confusion, which may be partly explained by genetic diversitywithin the mitis group (9, 10). Moreover, rare but important events have led to the acquisition of antibiotic resistance by pneumococcus as a result of the transfer of resist- ance determinants across species boundaries (4, 5). The high rates of recombination within the species have the potential to shuffle resistance determinants among pneumococcal genotypes. It is not known whether or not recombination, either at resistance loci or housekeeping genes, is equally likely for all members of the species or whether some strains are more likely to be involved in this process. Although a vaccine is available for 7 of the more than 90 pneumococcal serotypes, this has not eliminated pneumococcal disease because the nonvaccine serotypes derive an ecological advan- tage from the removal of their competitors and have been increasing in carriage prevalence (11) and, concomitantly, in disease (12). Alongside 1Department of Infectious Disease Epidemiology, Imperial College London, Norfolk Place, London W2 1PG, UK. 2 Depart- ment of Mathematics, Abo �� Akademi, FI-20500, Turku, Finland. *To whom correspondence should be addressed. E-mail: w.hanage@imperial.ac.uk 12 JUNE 2009 VOL 324 SCIENCE www.sciencemag.org 1454 REPORTS on June 11, 2009 www.sciencemag.org Downloaded from