Chromosomal Evolution in Potato and Tomato

Potato (Solanum tuberosum L.) and tomato (Lycopersicon esculentum) are members of the Solanaceae (nightshade family) and have the same basic chromosome number (x = 12). However, they cannot be cross-hybridized and, until now, it was unknown how conserved the gene order might be between these two species. We report herein the construction of a genetic linkage map of potato chromosomes based on genomic and cDNA clones from tomato. The potato map was drawn from segregation data derived from the interspecific cross S. phureja X (S. tuberosum X S. chacoense) (2n = 2x = 24), and consists of 135 markers defining 12 distinct linkage groups. Nearly all of the tomato probes tested hybridized to potato DNA, and in most cases, the copy number of the employed clones was the same in both species. Furthermore, all clones mapped to the same linkage group in both species. For nine chromosomes, the order of loci appears to be identical in the two species, while for the other three, intrachromosomal rearrangements are apparent, all of which appear to be paracentric inversions with one breakpoint at or near the centromere. These results are consistent with cytogenetic theory, previously untested in plants, which predicts that paracentric inversions will have the least negative effect on fitness and thus be the most likely form of chromosomal rearrangements to survive through evolutionary time. Linkage maps based on a common set of restriction fragment length polymorphism markers provide a basis for uniting the previously separate disciplines of tomato and potato genetics. Using these maps, it may now be possible to test theories about homologies or orthologies of other genes, including those coding for disease resistance and stress tolerances. T HE potato (Solanum tuberosum L.) is the most important food crop of the plant family Solanaceae, which contains some 3000 species including tomato (Lycopersicon esculentum Mill.), tobacco (Nicotiana tabacum L.), petunia (Petunia hybrida Hort. Vilm.-Andr.), eggplant (Solanum melongena L.), and garden pepper (Capsicum annuum L). The tuber-bearing Solanum are one relatively small group of a very large genus and include approximately 170 species. The basic chromosome number of potato, like most other solanaceous species, is x = 12 and wild forms range from diploids to hexaploids. Tomato and potato are members of the same tribe (Solanaceae) (D’ARcY 1976), have nearly identical karyotypes and contain comparable monoploid amounts of nuclear DNA (0.7 pg). The fact that the two species cannot be cross-hybridized has prevented detailed comparative genetic studies and, by-in-large, genetics of the two species have remained distinct and nonoverlapping. In addition, the heterozygous (outbreeding), tetraploid nature of the cultivated potato, and lack of useful genetic markers, has hindered the construction of a genetic linkage map for this species. The objectives of the research reported here were: (1) to construct a linkage map of potato based on restriction fragment length polymorphism (RFLP) Genetics 120: 1095-1 103 (December, 1988) markers, and (2) to determine the degree of similarity between potato and tomato chromosomes with respect to gene order. Genetically, tomato is one of the bestcharacterized plant species with well-populated linkage maps based on both morphological and molecular markers (TANKSLEY, MUTSCHLER and RICK 1987). By using previously mapped clones isolated from the tomato, we have been able to accomplish both of the stated goals. The construction of comparative maps based on a common set of clones provides a basis for uniting tomato and potato genetics, an accomplishment mutually advantageous for research in both species. MATERIALS AND METHODS Plant materials for segregation analysis: Parents and tuber progeny of an interspecific, diploid Solanum cross were obtained from C. QUIROS and D. DOUCHES, University of California, Davis. The pistillate parent of the cross (84S10) was an accession of the diploid species Solanum phureja Juz. et Buk. The staminate parent (T704) was an interspecific clone obtained by the hybridization of a dihaploid Solanum tuberosum (2n = 2x = 24) and a member of the diploid species Solanum chacoense Bitt. Attractive features of this population were its diploid constitution, prior characterization of isozyme loci (DOUCHES and QUIROS 1987), and the high likelihood that it would exhibit polymorphism at the DNA level. The mapping population consisted of 65 F1 1096 M. W. Bonierbale, R. L. Plaisted and S. D. Tanksley progeny, among which segregation of alleles from the interspecific parent was monitored for genetic mapping. Plants were propagated clonally in the greenhouse at Cornell University. The S. phureja clone and the S. tuberosum X S. chacoense hybrid will be referred to hereafter as phu and tbrchc, respectively (HUAMAN and ROSS 1985). DNA isolation, restriction digests, electrophoresis and blotting: Isolation of potato genomic DNA from leaf tissue was as described for tomato (BERNATZKY and TANKSLEY 1986a), except that sodium bisulfite was used instead of mercaptoethanol. Digestions with the following endonucleases (BRL) were performed according to the manufacturer's instructions using 2 units of enzyme per fig of DNA: EcoRI, EcoRV, HindIII, DraI, XbaI, BstNI, HaeIII, TaqI, MboI , MspI and HinfI. Seven micrograms of digested potato DNA were loaded and separated on 0.9% agarose gels. Electrophoresis and Southern blotting were conducted as described by BERNATZKY and TANKSLEY (1 986a). The only modifications of the above-mentioned protocols were the substitution of Genescreen plus (NEN) filters for Zetabind, and the use of capillary blots instead of dry blotting. On survey filters, the parents were represented in each of 11 pairs of restriction digests. Four to six replicated progeny filters, including one lane of digested tomato DNA (L. esculentum cv. VF36), 65 potato progeny, and one lane of each of the potato parental DNAs, were prepared from bulk digests with each enzyme. DNA probes, labeling and hybridization: Both cDNA and genomic clones were used as probes for mapping the potato genome. Construction of the cDNA and size-selected (0.5-3.0 kb) PstI or EcoRI genomic libraries from tomato have been described elsewhere (BERNATZKY and TANKSLEY 1986a; TANKSLEY et al. 1987). The loci detected by hybridization of cDNA clones were designated C D l , CD2, etc., and those detected by the genomic clones are designated TGI, TG2, etc. Duplicate loci homologous to a single probe were suffixed with designations A, B , etc. Clones of several known genes were also used in the comparative mapping study: small subunit ribulose bisphosphate carboxylase (Rbcs), chlorophyll a/b binding protein (Cab), and the 45s major ribosomal repeat (R45s). The Rbcs and Cab clones were from tomato (PICHERSKY et al. 1985, 1986) and the ribosomal clone was from pea (JORGENSEN et al. 1982). The clones used in this study had previously been mapped to orthologous loci in tomato, and are distributed throughout 94% of the 1237 cM tomato RFLP map at intervals averaging 11 cM (TANKSLEY et al. 1987; Figure 1B). Whole plasmids were either nick-translated or randomhexamer-labeled with "P-dCTP (RIGBY et al. 1977, FEINBERG and VOCELSTEIN 1983). Potato parental survey or progeny filters were hybridized overnight as previously described (BERNATZKY and TANKSLEY 1986b). Filters were washed to medium stringency (0.5 X SSC, 65") and placed on X-ray film for 1-5 days. Data from eleven isozyme markers and Y, a morphological marker determining tuber flesh color (HOWARD 1970), segregating in the same diploid population, were kindly supplied by DAVE DOUCHES. These data were included in the mapping analysis by their co-segregation with the DNA markers. RFLP analysis and map construction: Enzyme-probe combinations were selected for which the tbr-chc parent appeared to be heterozygous, while the other parent (phu) was either homozygous, or heterozygous but with different alleles. Potential heterozygosity with a given probe was inferred by the presence of more than one hybridizing restriction fragment per digestion on parental survey filters, and polymorphism between the parents required that at least one fragment be unique to the parent in question. Segregation of alleles from tbr-chc were verified and scored in the progeny. Analysis of segregation data was performed on an IBM PC/AT computer using SPSS software. Recombination frequencies were converted to map units according to KOSAMBI (1 944). In some instances, heterozygous loci in the phu parent were also observed to segregate in the progeny, and a second set of mapping data was collected.