Euchromatin and Pericentromeric Heterochromatin: Comparative Composition in the Tomato Genome

Ying Wang,*,†,1 Xiaomin Tang,‡ Zhukuan Cheng,‡ Lukas Mueller,*,† Jim Giovannoni§,** and Steve D. Tanksley*,†,2 *Department of Plant Breeding and Genetics, †Department of Plant Biology, §Boyce Thompson Institute for Plant Research and **U.S. Department of Agriculture–Agricultural Research Service, Plant, Soil, and Nutrition Lab, Cornell University, Ithaca, New York 14853 and ‡Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, People’s Republic of China Manuscript received October 18, 2005 Accepted for publication February 6, 2006

ABSTRACT Eleven sequenced BACs were annotated and localized via FISH to tomato pachytene chromosomes providing the first global insights into the compositional differences of euchromatin and pericentromeric heterochromatin in this model dicot species. The results indicate that tomato euchromatin has a gene density (6.7 kb/gene) similar to that of Arabidopsis and rice. Thus, while the euchromatin comprises only 25% of the tomato nuclear DNA, it is sufficient to account for 90% of the estimated 38,000 nontransposon genes that compose the tomato genome. Moreover, euchromatic BACs were largely devoid of transposons or other repetitive elements. In contrast, BACs assigned to the pericentromeric heterochromatin had a gene density 10–100 times lower than that of the euchromatin and are heavily populated by retrotransposons preferential to the heterochromatin—the most abundant transposons belonging to the Jinling Ty3/gypsy-like retrotransposon family. Jinling elements are highly methylated and rarely transcribed. Nonetheless, they have spread throughout the pericentromeric heterochromatin in tomato and wild tomato species fairly recently—well after tomato diverged from potato and other related solanaceous species. The implications of these findings on evolution and on sequencing the genomes of tomato and other solanaceous species are discussed.

ANY plant and animal species possess chromo- well as the structure and function of the heterochro- M somes differentiated into highly condensed, matin. The Ty3/Gypsy class retrotransposons and their pericentromeric heterochromatin and largely decon- relatives have been found in pericentromeric regions, densed euchromatic arms. In animals, heterochroma- especially in grasses (Cheng et al. 2002; Jiang et al. 2003; tin is often associated with transcriptional inactivity and Mroczek and Dawe 2003; Wu et al. 2004). Tandem suppressed genetic recombination and contains a large repeats and retroelements are also keys for centromere number of repetitive sequences (Dean and Schmidt recognition by kinetochore proteins. Repeats in the 1995; Renauld and Gasser 1997; Myster et al. 2004). pericentromeric regions also play important roles in In plants, more is known about the gene-rich euchro- sufficiently initiating the recruitment of histone modi- matin, but less about heterochromatin (De Jong 1998; fication enzymes and promoting the formation and Cheng and Murata 2003; Koornneef et al. 2003). maintenance of heterochromatin by RNAi machinery Studies thus far suggest that plant heterochromatin, (Cheng et al. 2002; Hall et al. 2002; Volpe et al. 2002; while gene poor compared with euchromatin, still con- Zhong et al. 2002; Bender 2004; Sun et al. 2004). This tains transcriptionally active genes, at least at certain still leaves open several important questions: What are times during the life cycle (Martin et al. 1993; Li et al. the global differences between euchromatin and het- 2004; Lippman et al. 2004; Martienssen et al. 2004; erochromatin? Which of these differences are consis- Nagaki et al. 2004). tent among species and which are species specific? What Advances in the genomic sequencing of model plants evolutionary forces have molded the composition and and animals have provided great insight into the com- function of heterochromatin vs. euchromatin? In hopes position and organization of genes in euchromatin as of shedding light on these questions, we have conducted a series of molecular cytogenetic experiments to char- acterize the nature of heterochromatin vs. euchromatin 1Present address: Wuhan Botanical Garden, Chinese Academy of in the model crop plant—tomato. Science, Wuhan, Hubei, People’s Republic of China, 430074. The tomato (Solanum lycopersicum) is a diploid species 2Corresponding author: Department of Plant Breeding and Genetics, 248 Emerson Hall, Cornell University, Ithaca, NY 14853. with a genome composed of 12 chromosomes (2n ¼ E-mail: [email protected] 2x ¼ 24) totaling 950 Mb of DNA (Arumuganathan et al.

Genetics 172: 2529–2540 (April 2006) 2530 Y. Wang et al.

1991). Along with maize, it was an early model system repetitive elements in two ways. First, each was compared for genetics and cytogenetics studies in plants (for re- against two repeat databases: Solanaceae repeat database, view, see Rick 1971). Like many other plant species, such which contains all known repeats from solanaceous species, and Repbase, which is a comprehensive database of repetitive as Arabidopsis and Medicago truncatula, tomato chromo- element consensus sequences in eukaryotic genomes, includ- somes contain long, contiguous stretches of euchromatin ing transposable elements (TEs) and tandem repeats of at the distal ends of most chromosomes and hetero- diverse origins ( Jurka 2000; Ouyang and Buell 2004). If chromatic regions flanking the centromeres (De Jong the gene model was homologous with known TEs (TIGR_ 1998; Kulikova et al. 2001; Fransz et al. 2003). Ap- Sol_repeat, or Repbase repeats, or Arabidopsis TEs), then it was annotated as a TE-related gene. Second, the BACs were proximately 25% of the tomato genome is contained in computationally screened against each other (blastn; E-value the euchromatin and 75% is contained in the pericen- ,1020 and bits score .100) in an effort to identify repetitive tromeric heterochromatin (Peterson et al. 1998). Al- sequence motifs shared by two or more BACs. Segments of though much work has been done on Arabidopsis, which BACs containing putative repetitive elements were aligned has relatively little heterochromatin, tomato would be and manually examined in an effort to determine the length and boundaries of each repetitive element. more similar to the majority of plants, with less known Southern hybridization: A subset of the repetitive elements on the global organization of euchromatin and hetero- identified in this study was screened against the genomes of chromatin, especially crop plants that have large ge- other solanaceous species to determine the taxonomic distri- nomes and more heterochromatin. We report herein bution of these elements. For these studies, PCR-amplified 32 the annotation of 11 sequenced BAC clones assigned, repeats were amplified, labeled with P, and used as probes on the genomic Southern blots containing restriction-digested via in situ hybridization, to both euchromatin and het- genomic DNA from S. pimpinellifolium L. (LA1589), S. chmie- erochromatin. From a comparative analysis of these BACs lewskii (C. M. Rick, E. Kesicki, J. Fobes, M. Holle, D. M. Spooner, emerges a general picture of the global organization G. J. Anderson, and R. K. Jansen) (LA1316), S. peruvianum L. and evolution of euchromatin vs. heterochromatin in (LA1708), S. chilense (Dunal) Reiche (LA1959), S. pennellii this model dicot plant species. Correll (TA56), S. neorickii (D. M. Spooner, G. J. Anderson, and R. K. Jansen) (LA2133), S. habrochaites (S. Knapp and D. M. Spooner) (LA1777), S . lycopersicum L. (TA209), S. tuberosum L., MATERIALS AND METHODS Capsicum annuum L. (garden pepper), Petunia 3 hybrida hort. ex E. Vilm., and S. melongena L. (eggplant) (Spooner et al. 2005). BAC clones and annotation: Eleven sequenced tomato Blots were hybridized at 60° and washed in 23 SSC for 20 min nuclear BAC clones were included in the analyses described and in 13 SSC for 10 min. The tomato 45S rDNA (R45S) herein. The sequences of all BACs have been submitted to ribosomal gene probe was used as a positive control for GenBank (see Table 1 for accession numbers). Three BACs estimating the relative strength of hybridization signals (181K1, 181O9, and 181C9) were randomly selected from a (Ganal et al. 1988). tomato BAC library, whereas the other 8 were isolated from the For estimating the methylation status of cytosine in the same library by virtue of screening with known genes or single- retrotransposons, a Southern blot was made using tomato copy probes (Budiman et al. 2000; van der Hoeven et al. 2002; (TA56 and TA209) genomic DNA digested with methylation- Y. Wang, unpublished data). All BACs were subjected to sensitive restriction enzymes, HpaII and EcoRI, and methyla- annotation according to guidelines utilized for the rice and tion insensitive isoschizomers, MspI and BstNI (Messeguer Arabidopsis genome (Mao et al. 2001; Goff et al. 2002; Yuet al. et al. 1991). The LTR and polyprotein regions of the retro- 2002; International Rice Genome Sequencing Project transposon were labeled with 32P-dCTP and used as probes on 2005). Genes were predicted with four computational gene- the genomic Southern blots. Since cytosines in the tomato finder programs: FGENESH (Salamov and Solovyev 2000), chloroplast are not methylated, PCR fragments of tomato GenemarkHMM (Borodovsky and McIninch 1993), Genscan1 chloroplast genes (AY216521, AF397080, and AF263101) were (Burge and Karlin 1997; http://genes.mit.edu/GENSCAN. used as the methylation negative control (Fojtova et al. 2001). html), and GlimmerM (Salzberg et al. 1998), using an Arabi- Genetic mapping: DNA fragments were amplified from dopsis training data set. In addition, a BAC segment was also each BAC clone using primers designed from annotated exons considered to represent a coding region if it showed a sig- or BAC ends and then mapped as cleaved amplified polymorphic nificant match with the Arabidopsis proteome (E-value ,1010 sequence or RFLP markers onto the high-density tomato map 10 for tblastx) and/or plant ESTs (E-value ,10 for blastn). on the basis of a population of 80 F2 individuals from the Genes thus identified, but that showed strong homology to cross S. lycopersicum LA925 3 S. pennellii LA716 (Fulton et al. transposon-related genes (e.g., reverse transcriptase), were not 2002; http://www.sgn.cornell.edu/cgi-bin/mapviewer) (sup- considered to be part of the tomato gene repertoire. plemental Table 1 at http://www.genetics.org/supplemental/). Tomato BAC library screening: High-density BAC filters Copy number reconstruction experiments: To estimate the were prepared using a tomato HindIII BAC library with copy numbers of the tomato repetitive sequences, PCR prod- 129,024 BAC clones (Budiman et al. 2000). These filters were ucts representing the repeats were used in the reconstruction screened with probes randomly labeled with 32P-dCTP as de- experiments. Denatured tomato genomic DNA extracted scribed previously (Feinberg and Vogelstein 1983). Hybrid- from S. lycopersicum cv. TA209 (400 and 1000 ng) was spotted ization was carried out overnight in plastic boxes at 65°. Filters onto Hybond N1 membrane (Amersham, Arlington Heights, were washed at 65° progressively with 23 SSC 1 0.1% SDS for IL) according to Ganal et al. (1988). Different quantities of 30 min, 13 SSC 1 0.1% SDS for 20 min, and 0.53 SSC 1 0.1% PCR product, representing different copy numbers of the SDS for 10 min. Phosphorimaging screens were exposed repeat in the tomato genome, were also spotted on the same overnight and were scanned on a Storm PhosphorImager membrane. Hybridization of the 32P-dCTP-labeled repeats, (Molecular Dynamics, Sunnyvale, CA). analysis of the autoradiographs, and estimation of the copy Identifying highly repetitive sequences in BACs: The 11 number, from a plot of densities vs. copy numbers, were sequenced BAC clones were computationally screened for carried out as described by Bernatzky and Tanksley (1986). Euchromatin and Pericentromeric Heterochromatin in the Tomato Genome 2531

The mean results of three independent reconstruction esti- elements common to heterochromatin on all chromo- mates were reported for each repeat. somes (Figure 1, Table 1). In addition, the hybridization Pachytene chromosome preparation and fluorescence in signals of these 3 BACs were not evenly distributed, with situ hybridization: Immature tomato flower buds, 3mmin length, were harvested and fixed in Carnoy’s solution (3:1 some regions stained more intensively than nearby ethanol:glacial acetic). Anthers containing pachytene stage regions, which indicated the uneven distribution of microsporocytes were squashed in acetocarmine solution. repeat sequences or different condensation patterns in Slides were frozen in liquid nitrogen, with coverslips removed, tomato chromosomes. dehydrated through a series of ethanol washes (70, 90, and BAC annotation: In an effort to determine composi- 100%), and subjected to fluorescence in situ hybridization (FISH) using the method of Jiang et al. (1995). DNA of tional differences between the euchromatin and het- individual BACs was isolated using a standard alkaline extrac- erochromatin, each of the 11 BACs was subjected to tion and labeled by nick translation with digoxigenin-16-dUTP computational and manual annotation. Putative genes (Roche Diagnostics, Indianapolis). Slides with meiotic chro- were identified by virtue of ab initio gene prediction mosomes were denaturated at 80° for 2 min in a buffer con- programs, significant matches to ESTs from tomato or taining 70% formamide in 23 SSC and immediately placed in a series of precooled ethanol baths (70, 90, and 100%, for other related solanaceous species, and significant 5 min each). Denaturated probe mixture (20 mlcontaining50– matches to predicted proteins from Arabidopsis or rice 100 ng labeled probe, 23 SSC, 50% deionized formamide, and (Figure 2B). Gene models homologous with known 10% dextran sulfate) was applied to each slide and covered transposable elements in solanaceous species, rice, or with a coverslip. Hybridization was carried out at 37° overnight Arabidopsis were annotated as TE-related genes. Thus in a humid chamber. After removing the coverslips, slides were washed in 23 SSC at 42° for 10 min and 23 SSC at room the BACs could be assigned into three categories: (1) temperature for 5 min. Biotin-labeled probes were detected those containing only genes associated with transpo- by fluorescein isothiocyanate (FITC)-conjugated sheep- sons, (2) those containing transposon genes as well as antidigoxigenin antibody (Roche Diagnostics). Chromosomes nontransposon genes, and (3) those containing only were counterstained with 49,6-diamidino-2-phenylindole (DAPI) nontransposon genes. The 6 BACs derived from het- in an antifade solution (Vector Laboratories, Burlingame, CA). Chromosomes and FISH signal images were captured erochromatin (as determined by FISH, see previous using an Olympus BX61 fluorescence microscope with a section) fell into categories 1 and 2 (Figure 2B). More- micro-CCD camera. Grayscale images were captured for each over, the 3 random BACs (181K1, 181C9, and 181O9) color channel and then merged using Image-Pro Plus software. from tomato heterochromatic regions (i.e., not selected using a genic or single-copy probe) all fell into category 1—containing no genes other than those related to RESULTS AND DISCUSSION transposons (Figure 2B). Of the 5 BACs assigned to Genetic and physical localization of BACs in tomato euchromatin, 4 fell into the third category (only non- chromosomes: Eleven sequenced tomato nuclear BAC transposon genes). The remaining BAC fell into category clones were analyzed—3 BACs (181K1, 181O9, and 2—containing both transposons and nontransposon 181C9) were randomly selected from a tomato BAC genes (Table 1, Figure 2). library and the other 8 were isolated by screening with Gene composition of heterochromatin vs. euchro- known genes or single-copy probes (van der Hoeven matin: The six BACs assigned to heterochromatin et al. 2002; Y. Wang, unpublished data). These 11 se- together comprise 678 kb with a gene density (not count- quenced tomato BAC clones were genetically mapped ing transposon-related genes) corresponding to approx- to their corresponding positions in 7 of the 12 tomato imately one gene every 56 kb. However, this estimate for chromosomes in the tomato high-density genetic map the gene density of heterochromatin has to be tem- (Figure 1) (Fulton et al. 2002, http://www.sgn.cornell. pered with the fact that only three of the heterochro- edu). Five of these BACs mapped to regions distal to matic BACs were randomly drawn from the BAC library. the genetically defined centromeres and hence are The rest were selected with genic or other single-copy likely to be in euchromatic regions, which were con- probes, hence biasing for BACs containing genes. The firmed by FISH on tomato (S. lycopersicum) pachytene three truly randomly sampled heterochromatic BACs chromosomes (Figure 1, Table 1). The remaining 6 comprise 370 kb and do not contain any nontransposon BACs were all mapped to regions proximal to centro- genes. Thus, the estimate of one gene every 56 kb in the meres (Figure 1). Of these, 2 BACs (181O9 and 47I13) heterochromatin is most likely an overestimate with the hybridized to single loci clearly contained within the actual gene density in the heterochromatin likely being heterochromatin of chromosomes 8 and 9, respectively much lower than this value. The situation is dramatically (Figure 1, Table 1). A third BAC (2O7) also hybridized different for the euchromatin. The five BACs derived to the heterochromatin, but near the heterochromatin– from euchromatin together comprised 518 kb and euchromatin boundary of chromosome 7 (Figure 1). contain 77 nontransposon genes. Therefore, we estimate The remaining 3 BACs (181K1, 40B13, and 181C9) that the euchromatin contains, on average, one gene hybridized to multiple sites throughout the chromo- every 6.7 kb. somes with most of the signals concentrated in hetero- The fully sequenced genomes of Arabidopsis and rice chromatic regions, suggesting that these BACs contain provide reference points with which to compare the 2532 Y. Wang et al.

Figure 1.—FISH with BACs located in euchromatin and heterochromatin. The distance of the FISH signals in a relative scale from centromere to telomere were correlated to the genetic mapping positions on the highly saturated tomato F2.2000 map (Fulton et al. 2002; www.sgn.cornell.edu). BAC19 and 62O11 were mapped with high mapping conﬁ- dence of LOD .3, and other BACs were mapped with LOD ,2 as interval markers. Chromosomes were counterstained with 49,6-diamidino-2-phenylindole (DAPI) and BAC DNA was labeled with digoxigenin- 16-dUTP. Euchromatin and Pericentromeric Heterochromatin in the Tomato Genome 2533

Figure 1.—Continued. organization of tomato euchromatin and heterochro- and rice, 6.9 kb/gene (Copenhaver et al. 1999; Kumekawa matin. For all three species, the estimated nontransposon et al. 2001; Jiao et al. 2005). However, striking differ- (non-TE) gene densities for euchromatin are remarkably ences are revealed in the heterochromatin. Rice het- similar: tomato, 6.7 kb/gene; Arabidopsis, 4.5 kb/gene; erochromatin has gene density only slightly lower than

TABLE 1 Eleven sequenced tomato BACs that were mapped to the tomato genome with FISH

GenBank BAC No. of No. of BAC accession length Chromosomal Euchromatin/ nontransoposon transposon-related clone no. (bp) location heterochromatin genes genes Jinling LARD1 LARD2 181O9a AY881150 136,132 08.016 Heterochromatin 0 10 0 0 0 181C9a AY881151 118,553 10.028 Heterochromatin 0 6 2 1 0 181K1a AY881152 115,428 08.028 Heterochromatin 0 10 9 1 1 47I13 AF411804, 100,810 09.050 Heterochromatin 0 6 1 0 1 AF411805 2O7a AF411806 92,221 07.016 Heterochromatin 2 4 1 0 1 40B13 AC174607 115,283 02.010 Heterochromatin 10 4 2 0 0 62O11 AF411808 70,347 07.002 Euchromatin 7 0 0 0 0 127E11 AF411807 95,848 04.019 Euchromatin 19 0 0 0 0 FW2.2 AF411809 127,892 02.116 Euchromatin 20 0 0 0 0 BAC19 AF273333 105,308 02.089 Euchromatin 18 0 0 0 0 240K4 AF275345 118,813 11.032 Euchromatin 13 2 0 0 0 Chromosomal locations are depicted as two digits of chromosome number followed by three digits of the chromosome length after the decimal. The GenBank accession numbers for Jinling from BAC 40B13 and LARDs DQ445619–DQ445624. a These BACs have unknown-size sequence gaps. 2534 Y. Wang et al.

Figure 2.—(A) Structure of a regular Ty3/Gypsy retroelement and Jinliing elements found in tomato BAC 40B13 and 2O7. (B) The structural annotation of 11 tomato BACs, indicating the putative genes and transposable elements. The number in the putative gene indicates the supporting ev- idence: 1, signiﬁcant match to Solanaceae EST; 2, significant match to Arabidopsis gene; 3, both Solanaceae EST match and Arabidopsis match; 4, computational prediction only with no signi- ﬁcant matches found in Solanaceae ESTs, Arabidop- sis, or GenBank. The three random BACs (181K1, 181C9, and 181O9) from tomato heterochromatic regions all fell into category 1—containing no genes other than those related to transposons. The other three BACs derived from heterochromatin were assigned into categories 1 and 2 (containing both transposons and nontransposon genes). BACs from euchromatin, except 240K4, fell into category 3, containing only nontransposon genes.

the gene density in euchromatin (11 kb/gene, Jiao heterochromatin. If the heterochromatin contains 1 et al. 2005). In contrast, Arabidopsis and tomato het- gene every 56 kb, one would estimate that the hetero- erochromatin have non-TE gene densities dramat- chromatin contains 12,700 genes and that the tomato ically lower than that in euchromatin: Arabidopsis, genome contains a total of 48,000 genes, which is 256 kb/gene on chromosomes 2 and 4; and tomato, dramatically higher than the estimate from the EST .56 kb/gene (Copenhaver et al. 1999). As already data. This discrepancy is likely due to the biased estimate noted, the higher gene density estimate for tomato of the heterochromatic gene density (see previous heterochromatin (in comparison with Arabidopsis) may section). If the unbiased heterochromatin non-TE gene reﬂect a bias in the tomato estimate due to the non- density from Arabidopsis (256 kb/gene) is used for the random sampling of heterochromatic BACs (see previous estimation, the tomato heterochromatin would be pre- section). dicted to contain only 2800 genes. This would also lead Thetomatogenomeiscomposedof950 Mb of DNA, to a predicted gene content of 38,240 genes in the tomato 25% of which is euchromatin (Arumuganathan et al. genome (35,440 in the euchromatin 1 2800 in the 1991; De Jong 1998; Peterson et al. 1998). Thus, at a heterochromatin)—a value much closer to that pre- gene density of 6.7 kb/gene, we estimate that the eu- dicted from EST data (van der Hoeven et al. 2002). If chromatin contains 35,440 genes. On the basis of a these estimates are valid, then the tomato heterochro- large EST database, the entire tomato genome has been matin, which comprises 75% of the tomato nuclear ge- previously estimated to encode 35,000 genes (van der nome, likely accounts for ,10% of the non-TE genes—an Hoeven et al. 2002). Thus the euchromatin is sufﬁcient important observation to be taken into account when to encode most of tomato non-TE genes. The remaining contemplating sequencing of the genome of tomato or 75% (712 Mb) of the tomato genome is composed of other solanaceous species (see next section). Euchromatin and Pericentromeric Heterochromatin in the Tomato Genome 2535

Repeat composition of heterochromatin vs. euchro- reverse transcriptase (RT) domains, and one integrase matin: In an effort to identify repeat elements in the (IN) domain (Figure 2A, supplemental Table 1 at http:// tomato genome, all BAC sequences were compared with www.genetics.org/supplemental/). All but one member each other using blastn. A sequence element was in BAC 2O7 also lack the GAG domain found in many classified as a ‘‘repeat’’ if it was shared by two or more other Ty3 retrotransposons. The LTRs of this retro- BAC clones. Using this criterion, a total of three distinct transposon family also show homology (80% identity) repeat families were identified, Jinling, LARD1, and with a previously described tomato repeat—tomato ge- LARD2, involving five different BACs—all assigned to nome repeat II (TGRII) (Ganal et al. 1988). The TGRII the heterochromatin (Table 1). No shared repeats were repeat family was determined to consist of .4000 mem- identified in the euchromatic BACs (Table 1). ber repeats, making it the highest-copy, interspersed Tom-LARD1 and Tom-LARD2: Two repeat families as- element in the tomato genome (Ganal et al. 1988). On sociated with retrotransposons (deemed large retro- the basis of the high level of homology, we propose transposon derivatives, LARDs) (Kalendar et al. 2004) a family name of Jinling (meaning ‘‘golden bell’’ in were identified due to the hallmarks of long terminal Chinese and indicating the heterochromatin vicinity) to repeats (LTRs), ranging from 750 to 1100 bp. LARD1 represent all the related elements reported herein as and LARD2 had two- and three-member elements, well as the previously reported PCRT1a and TGRII respectively. The sizes of the internal sequences of LARD1 elements (Ganal et al. 1988; Yang et al. 2005). and LARD2 elements varied from 2 to 5 kb, but none of Copy number of Jinling: The copy number of Jinling them contained any coding sequences, and hence they in the tomato genome was estimated via a genomic are not likely to be functional elements. Elements in the reconstruction experiment using various components same family only share sequence similarity in the LTRs, of the Jinling element from BAC 40B13 as probes but not in the internal noncoding sequences, suggest- (supplemental Table 1 at http://www.genetics.org/ ing that LARDs are nonautonomous retroelements. supplemental/). The results indicate that the tomato LTRs at opposite ends of each Tom-LARD1 and Tom- genome contains 2080, 1800, 1120, and 1450 copies of LARD2 element shared 90–96% identity (sequence the LTR, RT, GAG, and IN domains, respectively. Thus, alignments are in the supplemental material at http:// on the basis of Southern hybridization, we estimate a www.genetics.org/supplemental/). A comparison of or- minimum of 2000 Jinling elements in the tomato thologous, noncoding intergenic nuclear DNA segments genome. Previous reports had estimated that TGRII is from tomato and potato has revealed a sequence present in 4000 copies (Ganal et al. 1988). The dis- identity of 86% (Y. Wang, unpublished data for BAC crepancy in these estimates may be due to the fact that comparison). Using this value and assuming that the TGRII shows 80% homology with the canonical Jinling transposon elements had identical LTRs upon inser- element. This level of sequence divergence is nearing tion, we estimate that these LARDs integrated into the limits of detection by Southern hybridization. Thus, their existing sites after the divergence of tomato and it is likely that additional, more diverged, copies of the potato—an event estimated to have occurred 10 MYA Jinling element were not detected by the Southern (Chaw et al. 1997; Wikstrom et al. 2001). reconstruction experiments described herein. Using Jinling—a high-copy retrotransposon family preferential to 2000 and 4000 as the upper and lower copy-number tomato pericentromeric heterochromatin: A third repeat limits and 8.8 kb as the upper size limit, we estimate that family was identified that contains 15 elements (Table Jinling comprises 17.5–35 Mb of genomic DNA in the 1). The elements were found in multiple copies in five of tomato genome (or 2.5% of the heterochromatin). the six BACs assigned to heterochromatin and hence Thus, Jinling is the largest family of retrotransposons may represent an abundant and heterochromatin- thus far identified in the tomato genome and is ap- preferential repeat. On the basis of alignments and parently specific to the pericentric heterochromatin. annotation of the 15 repeat members, this element is Jinling is nonrandomly distributed in pericentromeric het- deduced to be a member of the Ty3/gypsy retrotrans- erochromatin: While Jinling is apparently specific to the poson family and to share homology (84% identity) heterochromatin, it is not necessarily uniformly distrib- with the LTRs of two pericentromeric retrotransposon uted throughout the heterochromatin. To address this (PCRT)1a elements reported by Yang et al. (2005) issue, various domains of the Jinling element (LTR, RT, (Figure 2A; Su and Brown 1997; Marin and Llorens IN, and GAG) as well as TGRII (Ganal et al. 1988) were 2000) (data not shown). Eight of the 15 members used as probes for FISH on tomato pachytene chromo- appear to be complete, containing all of the structural somes. The results show that all probes hybridized and coding elements predicted for the Ty3 retrotrans- almost entirely to pericentromeric heterochromatin—a poson family. However, none of these elements appear result consistent with FISH experiments conducted to be capable of autonomous transposition since one or with complete BACs containing Jinling (Figures 1 and more of the coding regions always contained premature 3). However, nonuniform hybridization signals suggest stop codons. A complete element from BAC 40B13 is that Jinling may not be randomly distributed in the annotated to comprise 8.8 kb, including 2-kb LTRs, two heterochromatin (Figure 3). To further investigate this 2536 Y. Wang et al.

Figure 3.—FISH with repetitive elements (LTR, RT, and IN from the Jinling element in 40B13; the GAG domain from the Jinling element in BAC 2O7; and TGRII).

possibility, the Jinling LTR and the entire TGRII ele- sequence cannot be affected by selection—i.e.,the ments were used as probes on tomato BAC ﬁlters of the principle of random genetic drift applies. HindIII library (Budiman et al. 2000). If evenly distrib- 3. The terminal repeats, at both ends of the element, uted in the heterochromatin, one Jinling element would have been retained and can be aligned, allowing be predicted every 240 kb (using a 3000-copy-number generation of sequence divergence statistics. estimate and 710 Mb of pericentric heterochromatin—- 4. Neutral nucleotide divergence rates for noncoding see previous section). Considering an average BAC regions are available for the organism under study, insert size of 120 kb one would predict that 37% of allowing molecular clock estimates to be used in heterochromatic BAC clones (28% of all BACs) would calculating the amount of time that has passed since contain at least one Jinling element. However, actual the element inserted and the terminal repeats began hybridization results indicated that only 13% of the to mutate. BACs contain Jinling elements—a result consistent with LTRs at the opposite ends of Jinling members share Jinling occurring in nonrandom clusters in the hetero- sequence similarities ranging from 90 to 97%. Consid- chromatin. However, we cannot rule out the possibility ering the intergenic sequence identity of 86% between that the BAC library is not biased against heterochro- tomato and potato (Y. Wang, unpublished data), these matic clones that are more likely to be rich in repeat results indicate that Jinling spread throughout the elements. However, to further test the hypothesis that tomato heterochromatin subsequent to the divergence Jinling is not randomly distributed in the heterochro- of tomato and potato from their last common an- matin, the frequency and distribution of Jinling in the cestor—estimated to be 10 MYA (Chaw et al. 1997; six heterochromatic BACs was tested against a uniform Wikstrom et al. 2001). Moreover, an unrooted tree distribution model using the Kolmogorov–Smirnov test suggests that the Jinling LTRs began diverging from each of uniform distribution. Using this test, we could reject other at a very similar time (Figure 4). Using neutral the hypothesis that Jinling is randomly distributed across substitution rates reported by Nesbitt and Tanksley the six sequenced heterochromatic BACs (P , 0.001). (2002), we estimate that Jinling began spreading through- This clustering effect may be caused in part by localized out the tomato genome from 5 MYA—a period corre- movement of Jinling (see next section). sponding to the radiation of the Solanum tomato clade Estimating the timing of Jinling transposition during (formerly Lycopersicon) containing tomato and its evolution of the Solanaceae: The time of insertion of any closely related wild species (Miller and Tanksley 1990; given transposable element in a genome can be esti- Nesbitt and Tanksley 2002). Thus, Jinling elements mated, provided that the following conditions are met are young TEs in the tomato genome, similar to the (SanMiguel et al. 1998; Jiang et al. 2002; Ma and Dasheng elements found in the rice centromeric regions Bennetzen 2004): (Jiang et al. 2002). 1. The inserted element must have terminal repeats Occurrence of Jinling in the genomes of other solanaceous that were identical in sequence at the time of the species: Southern hybridization using a Jinling LTR probe insertion. showed strong signals in tomato and the closely related 2. Random mutations begin to accumulate in the ter- wild tomato species (S. pimpinellifolium, S. chmielewskii, minal repeats after insertion. This mutational drift in S. peruvianum, S. chilense, S. pennellii, and S. habrochaites), Euchromatin and Pericentromeric Heterochromatin in the Tomato Genome 2537

Figure 4.—A phylogenetic tree il- lustrating the evolutionary relation- ships among Jinling elements based on the LTRs at the opposite ends. The alignment of all Jinling LTRs is in supplemental ﬁle 1 at http:// www.genetics.org/supplemental/.

but the hybridization signals were not detected in tain gene density similar to that of the euchromatin of potato, eggplant, pepper, and petunia (Figure 5). The species with smaller genomes such as rice and Arabi- confinement of Jinling to tomato and its closest Solanum dopsis. In contrast, the pericentromeric heterochroma- relatives is consistent with the estimated transposition tin has a gene density 10–100 times lower than that of date and the time that this Solanum tomato clade euchromatin and is largely occupied by retrotrans- diverged from its last common ancestor (see previous posons—the largest family being the Ty3/gypsy-type section). Jinling family. LTR divergence data suggest that Jinling Jinling elements are highly methylated and rarely tran- originated and spread rapidly in the pericentromeric scribed: The methylation status of Jinling elements heterochromatin of tomato and closely related wild was tested via digestion with methylation-sensitive and tomato species 5 MYA—well after the divergence -insensitive isoschizomers and Southern hybridization from potato and other solanaceous species that lack using Jinling fragments as probes. Chloroplast probes this element. However, the fact that tomato and other were used as negative controls as cpDNA is normally solanaceous species have very similar chromosome unmethylated (Fojtova et al. 2001). The results in- architecture with respect to euchromatin and pericen- dicate that Jinling elements are highly methylated at tromeric heterochromatin raises the possibility that both of the CNG and CG sites (Figure 6). DNA meth- heterochromatin regions in these species are occupied ylation is often associated with reduced gene expression by other retrotransposon families, which are common and previous studies in maize and Arabidopsis have to all Solanaceae and might have radiated at differ- also shown that repetitive sequences in heterochro- ent times during Solanaceae evolution (Gottschalk matin are highly methylated (Bennetzen et al. 1994; 1954). The fact that pericentromeric heterochromatin Lippman et al. 2004; Rabinowicz et al. 2005). A blastn appears to be deficient in genes and is evolving rather search against .180,000 tomato ESTs (www.sgn.cornell. rapidly with respect to repeat composition may explain edu) revealed no matches to Jinling. Hence, despite its why chromosome pairing and meiotic recombination high copy number, the Jinling element must be rarely are often repressed (up to 1000-fold) in heterochroma- transcribed. tin vs. euchromatin—especially in interspecific hybrids Conclusions: Sequence composition of euchromatin and (Tanksley et al. 1992). pericentromeric heterochromatin: The results presented Implications of this study to the sequencing of the tomato herein paint the picture of the tomato genome in which genome and the genomes of other solanaceous species: The at least 90% of the nontransposon genes are seques- results from this study indicate that as much as 90% of tered in the contiguous stretches of euchromatin found the estimated 38,000 nonretrotransposon genes in at the distal portions of most chromosomes and tomato can be discovered by sequencing the tomato comprising only 25% of the total DNA in the tomato euchromatin, which accounts for only 25% of the total genome. These euchromatic regions are largely devoid DNA. Moreover, the high gene density and the lack of repetitive elements (e.g., retrotransposons) and con- of repetitive sequences should make computational 2538 Y. Wang et al.

Figure 6.—Autoradiograms derived from S . lycopersicum (TA209) and S. Pennellii (TA56) digested with MspI (C*CGG), HpaII (C*C*GG), EcoRII (CC*(A/T)GG), and BstNI (CC(A/ T)GG). Southern hybridizations were performed using unmethylated chloroplast genes (AY216521, AF397080, and AF263101) (A) and Jinling in 40B13 (B) as probes. The en- Figure 5.—Autoradiograms derived from HindIII-digested zyme digestion will be blocked if the cytosines with asterisks genomic DNA of S. pimpinellifolium (LA1589), S. chmielewskii are methylated. (LA1316), S. peruvianum (LA1708), S. chilense (LA1959), S. pennellii (TA56), S. neorickii (LA2133), S. Habrochaites (LA1777), S . lycopersicum (TA209), S. tuberosum (potato), Cap- family (Doganlar et al. 2002; Paran et al. 2004; Y. sicum annuum (garden pepper), Petunia 3 hybrida hort. ex E. Wang, unpublished data). Vilm., and Solanum melongena (eggplant), probed with Jinling LTR (A) and 45S rDNA (R45S) (B). Lanes of TA209 and We thank Wojtek Pawlowski, Zach Lippman, Amy Frary, and Steve M. LA2133 were adjusted to display a similar amount of DNA Stack for critical reading of the manuscript. This work was partially as the other Solanum species for better comparison. (A) Po- supported by National Science Foundation grant DBI-0116076, tato showed weak hybridization signals, which is due to the ‘‘exploitation of tomato as a model for comparative and functional strong background caused by overloading of potato DNA. genomics,’’ and 0421634, ‘‘sequence and annotation of the euchro- To further test the availability of LTR sequences in potato, ge- matin of tomato,’’ and by a Binational Agricultural Research and nomic DNAs of potato, TA209, and TA56 were digested with Development grant, IS-3337-02, ‘‘saturationmutagenesis in tomato: an EcoRI, EcoRV, DraI, and HaeIII and hybridized with LTR (C) integrated infrastructure for functional genomics.’’ and R45S (D).

LITERATURE CITED assembly of euchromatin sequences relatively straight- Arumuganathan, K., J. P. Slattery,S.D.Tanksleyand E. D. Earle, forward. Currently, an international project to sequence 1991 Preparation and flow cytometric analysis of metaphase the tomato euchromatin on a BAC-by-BAC basis is being chromosomes of tomato. Theor. Appl. Genet. 82: 101–111. conducted by a consortium of 10 countries (http:// Bedell, J. A., M. A. Budiman,A.Nunberg,R.W.Citek,D.Robbins et al., 2005 Sorghum genome sequencing by methylation filtra- www.sgn.cornell.edu/). Because of the low gene density, tion. PloS Biol. 3: 103–115. sequencing the 75% of the genome comprising hetero- Bender, J., 2004 Chromatin-based silencing mechanisms. Curr. chromatin would be expensive and may result in Opin. Plant Biol. 7: 521–526. Bennetzen, J. L., K. Schrick,P.S.Springer,W.E.Brown and discovery of a relatively small fraction of genes. The P. Sanmiguel, 1994 Active maize genes are unmodified and small portion of genes embedded in the heterochro- flanked by diverse classes of modified highly repetitive DNA. matin could be more efficiently discovered through EST Genome 37: 565–576. Bernatzky, R., and S. D. Tanksley, 1986 Majority of random cDNA databases (currently .500,000 ESTs exist for tomato clones correspond to single loci in the tomato genome. Mol. and related solanaceous species; http://www.sgn.cornell. Gen. Genet. 203: 8–14. edu/) and/or methyl filtration sequencing (Palmer Borodovsky, M., and J. Mcininch, 1993 GeneMark: parallel gene edell recognition for both DNA strands. Comput. Chem. 17: 123–133. et al. 2003; B et al. 2005). Finally, because of the Budiman, M. A., L. Mao,T.C.Wood and R. A. Wing, 2000 A deep- high level of conserved chromosomal macrosynteny coverage tomato BAC library and prospects toward development and microsynteny among the genomes of solanaceous of an STC framework for genome sequencing. Genome Res. 10: 129–136. species, the sequence of the tomato euchromatin could Burge, C., and S. Karlin, 1997 Prediction of complete gene struc- be utilized in genetic/genomic studies across the entire tures in human genomic DNA. J. Mol. Biol. 268: 78–94. Euchromatin and Pericentromeric Heterochromatin in the Tomato Genome 2539

Chaw, S., A. Zharkikh,H.Sung,T.Lau and W. Li, 1997 Molecular Kumekawa,N.,T.Hosouchi,H.Tsuruoka and H. Kotani,2001 The phylogeny of extant gymnosperms and seed plant evolution: size and sequence organization of the centromeric region of Arabi- analysis of nuclear 18S rRNA sequences. Mol. Biol. Evol. 14: dopsis thaliana chromosome 4. DNA Res. 8: 285–290. 56–68. Li, W. L., P. Zhang,J.P.Fellers,B.Friebe and B. S. Gill, Cheng, Z. J., and M. Murata, 2003 A centromeric tandem repeat 2004 Sequence composition, organization, and evolution of family originating from a part of Ty3/gypsy-retroelement in the core Triticeae genome. Plant J. 40: 500–511. wheat and its relatives. Genetics 164: 665–672. Lippman, Z., A. V. Gendrel,M.Black,M.W.Vaughn,N.Dedhia Cheng, Z. K., F. G. Dong,T.Langdon,O.Y.Shu,C.R.Buell et al., et al., 2004 Role of transposable elements in heterochromatin 2002 Functional rice centromeres are marked by a satellite re- and epigenetic control. Nature 430: 471–476. peat and a centromere-specific retrotransposon. Plant Cell 14: Ma, J., and J. L. Bennetzen, 2004 Rapid recent growth and diver- 1691–1704. gence of rice nuclear genomes. Proc. Natl. Acad. Sci. USA 101: Copenhaver, G. P., K. Nickel,T.Kuromori, M.-I. Benito,S.Kaul 12404–12410. et al., 1999 Genetic definition and sequence analysis of Arabi- Mao, L., D. Begum,S.A.Goff and R. A. Wing, 2001 Sequence and dopsis centromeres. Science 286: 2468–2474. analysis of the tomato JOINTLESS locus. Plant Physiol. 126: 1331– Dean, C., and R. Schmidt, 1995 Plant genomes—a current molec- 1340. ular description. Annu. Rev. Plant Physiol. Plant Mol. Biol. 46: Marin, I., and C. Llorens, 2000 Ty3/Gypsy retrotransposons: de- 395–418. scription of new Arabidopsis thaliana elements and evolutionary De Jong, J. H., 1998 High resolution FISH reveals the molecular perspectives derived from comparative genomic data. Mol. Biol. and chromosomal organization of repetitive sequences in to- Evol. 17: 1040–1049. mato. Cytogenet. Cell Genet. 81: 104. Martienssen, R. A., P. D. Rabinowicz, A. O’Shaughnessy and W. R. Doganlar, S., A. Frary,M.C.Daunay,R.N.Lester and S. D. Mccombie, 2004 Sequencing the maize genome. Curr. Opin. Tanksley, 2002 Conservation of gene function in the Solana- Plant Biol. 7: 102–107. ceae as revealed by comparative mapping of domestication traits Martin, G. B., S. H. Brommonschenkel,J.Chunwongse,A.Frary, in eggplant. Genetics 161: 1713–1726. M. W. Ganal et al., 1993 Map-based cloning of a protein-kinase Feinberg, A. P., and B. Vogelstein, 1983 A technique for radiolab- gene conferring disease resistance in tomato. Science 262: 1432– eling DNA restriction endonuclease fragments to high specific 1436. activity. Anal. Biochem. 132: 6–13. Messeguer, R., M. W. Ganal,J.C.Steffens and S. D. Tanksley, Fojtova, M., A. Kovarik and R. Matyasek, 2001 Cytosine methyl- 1991 Characterization of the level, target sites and inheritance ation of plastid genome in higher plants. Fact or artifact? Plant of cytosine methylation in tomato nuclear-DNA. Plant Mol. Biol. Sci. 160: 585–593. 16: 753–770. Fransz, P., W. Soppe and I. Schubert, 2003 Heterochromatin in Miller, J. C., and S. D. Tanksley, 1990 Effect of different restric- interphase nuclei of Arabidopsis thaliana. Chromosome Res. 11: tion enzymes, probe source, and probe length on detecting 227–240. restriction-fragment-length-polymorphism in tomato. Theor. Appl. Fulton,T.M.,R.Van Der Hoeven,N.T.Eannetta and S. D. Tanksley, Genet. 80: 385–389. 2002 Identification, analysis, and utilization of conserved or- Mroczek, R. J., and R. K. Dawe, 2003 Distribution of retroelements tholog set markers for comparative genomics in higher plants. in centromeres and neocentromeres of maize. Genetics 165: Plant Cell 14: 1457–1467. 809–819. Ganal, M. W., N. L. V. Lapitan and S. D. Tanksley, 1988 A molec- Myster, S. H., F. Wang,R.Cavallo,W.Christian,S.Bhotika et al., ular and cytogenetic survey of major repeated DNA-sequences in 2004 Genetic and bioinformatic analysis of 41C and the 2R het- tomato (Lycopersicon esculentum). Mol. Gen. Genet. 213: 262–268. erochromatin of Drosophila melanogaster: a window on the hetero- Goff, S. A., D. Ricke,T.H.Lan,G.Presting,R.L.Wang et al., chromatin–euchromatin junction. Genetics 166: 807–822. 2002 A draft sequence of the rice genome (Oryza sativa L. Nagaki, K., Z. Cheng,S.Ouyang,P.B.Talbert,M.Kim et al., ssp. japonica). Science 296: 92–100. 2004 Sequencing of a rice centromere uncovers active genes. Gottschalk, W., 1954 Die chromosomenstruktur der Solanaceen Nat. Genet. 36: 138–145. unter Berucksichtigung phylogenetischer Fragestellungen. Nesbitt, T. C., and S. D. Tanksley, 2002 Comparative sequencing Chromosoma 6: 539–626. in the genus Lycopersicon: implications for the evolution of fruit Hall, I. M., G. D. Shankaranarayana,K.I.Noma,N.Ayoub, size in the domestication of cultivated tomatoes. Genetics 162: A. Cohen et al., 2002 Establishment and maintenance of a het- 365–379. erochromatin domain. Science 297: 2232–2237. Ouyang, S., and C. R. Buell, 2004 The TIGR plant repeat data- International Rice Genome Sequencing Project, 2005 The bases: a collective resource for the identification of repetitive map-based sequence of the rice genome. Nature 436: 793–800. sequences in plants. Nucleic Acids Res. 32: D360–D363. Jiang, J., B. Gill,G.Wang,P.Ronald and D. Ward, 1995 Meta- Palmer, L. E., P. D. Rabinowicz, A. L. O’Shaughnessy,V.S.Balija, phase and interphase fluorescence in situ hybridization mapping L. U. Nascimento et al., 2003 Maize genome sequencing by of the rice genome with bacterial artificial chromosomes. Proc. methylation filtrations. Science 302: 2115–2117. Natl. Acad. Sci. USA 92: 4487–4491. Paran, I., J. R. Van Der Voort,V.Lefebvre,M.Jahn,L.Landry Jiang, N., Z. Bao,S.Temnykh,Z.Cheng,J.Jiang et al.,2002 Dasheng: et al., 2004 An integrated genetic linkage map of pepper (Cap- a recently amplified nonautonomous long terminal repeat ele- sicum spp.). Mol. Breed. 13: 251–261. ment that is a major component of pericentromeric regions in Peterson, D. G., W. R. Pearson and S. M. Stack, 1998 Char- rice. Genetics 161: 1293–1305. acterization of the tomato (Lycopersicon esculentum) genome using Jiang,J.M.,J.A.Birchler,W.A.Parrott andR.K.Dawe,2003 A in vitro and in situ DNA reassociation. Genome 41: 346–356. molecular view of plant centromeres. Trends Plant Sci. 8: 570–575. Rabinowicz, P. D., R. Citek,M.A.Budiman,A.Nunberg,J.A. Jiao, Y., P. Jia,X.Wang,N.Su,S.Yu et al., 2005 A tiling microarray Bedell et al., 2005 Differential methylation of genes and re- expression analysis of rice chromosome 4 suggests a chromo- peats in land plants. Genome Res. 15: 1431–1440. some-level regulation of transcription. Plant Cell 17: 1641–1657. Renauld, H., and S. M. Gasser, 1997 Heterochromatin: A meiotic Jurka, J., 2000 Repbase update—a database and an electronic jour- matchmaker? Trends Cell Biol. 7: 201–205. nal of repetitive elements. Trends Genet. 16: 418–420. Rick, C. M., 1971 Some cytogenetic features of the genome in dip- Kalendar, R., C. M. Vicient,O.Peleg,K.Anamthawat-Jonsson,A. loid plant species. Stadler Genet. Symp. 1: 153–174. Bolshoy et al., 2004 Large retrotransposon derivatives: abun- Salamov, A. A., and V. V. Solovyev, 2000 Ab initio gene finding in dant, conserved but nonautonomous retroelements of barley Drosophila genomic DNA. Genome Res. 10: 516–522. and related genomes. Genetics 166: 1437–1450. Salzberg, S., A. Delcher,S.Kasif and O. White, 1998 Microbial Koornneef, M., P. Fransz and H. De Jong, 2003 Cytogenetic tools gene identification using interpolated Markov models. Nucleic for Arabidopsis thaliana. Chromosome Res. 11: 183–194. Acids Res. 26: 544–548. Kulikova, O., G. Gualtieri,R.Geurts, D.-J. Kim,D.Cook et al., Sanmiguel, P., B. S. Gaut,A.Tikhonov,Y.Nakajima and J. L. 2001 Integration of the FISH pachytene and genetic maps of Bennetzen, 1998 The paleontology of intergene retrotranspo- Medicago truncatula. Plant J. 27: 49–58. sons of maize. Nat. Genet. 20: 43–45. 2540 Y. Wang et al.

Spooner, D. M., I. E. Peralta and S. Knapp, 2005 Comparison of Wikstrom, N., V. Savolainen and M. W. Chase, 2001 Evolution of AFLPs with other markers for phylogenetic inference in wild toma- the angiosperms: calibrating the family tree. Proc. R. Soc. Lond. toes [Solanum L. section Lycopersicon (Mill.) Wettst.] Taxon 54: 43–61. Ser. B Biol. Sci. 268: 2211–2220. Su, P. Y., and T. A. Brown, 1997 Ty3/gypsy-like retrotransposon se- Wu, J. Z., H. Yamagata,M.Hayashi-Tsugane,S.Hijishita, quences in tomato. Plasmid 38: 148–157. M. Fujisawa et al., 2004 Composition and structure of the cen- Sun, F. L., K. Haynes,C.L.Simpson,S.D.Lee,L.Collins et al., tromeric region of rice chromosome 8. Plant Cell 16: 967– 2004 Cis-acting determinants of heterochromatin formation 976. on Drosophila melanogaster chromosome four. Mol. Cell. Biol. Yang, T. J., S. Lee,S.B.Chang,Y.Yu,H.Jong et al., 2005 In-depth 24: 8210–8220. sequence analysis of the tomato chromosome 12 centromeric Tanksley, S. D., M. W. Ganal,J.P.Prince,M.C.Devicente,M.W. region: identiﬁcation of a large CAA block and characterization Bonierbale et al., 1992 High-density molecular linkage maps of pericentromere retrotranposons. Chromosoma 114: 103– of the tomato and potato genomes. Genetics 132: 1141–1160. 117. van der Hoeven, R., C. Ronning,J.Giovannoni,G.Martin and S. Yu, J., S. N. Hu,J.Wang,G.K.S.Wong,S.G.Li et al., 2002 A draft Tanksley, 2002 Deductions about the number, organization, sequence of the rice genome (Oryza sativa L. ssp indica). Science and evolution of genes in the tomato genome based on analysis 296: 79–92. of a large expressed sequence tag collection and selective geno- Zhong, C. X., J. B. Marshall,C.Topp,R.Mroczek,A.Kato et al., mic sequencing. Plant Cell 14: 1441–1456. 2002 Centromeric retroelements and satellites interact with Volpe, T. A., C. Kidner,I.M.Hall,G.Teng,S.I.S.Grewal et al., maize kinetochore protein CENH3. Plant Cell 14: 2825–2836. 2002 Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by RNAi. Science 297: 1833–1837. Communicating editor: J. Tamkun