Theor Appl Genet (1989) 78:153-159 9Springer-Verlag 1989 Selection of somatic hybrids between diploid clones of potato (Solanum tuberosum L.) transformed by direct gene transfer J. Masson, D. Lancelin, C. Bellini, M. Lecerf, P. Guerche, and G. Pelletier Laboratoire de Biologie Cellulaire, INRA, Centre de Versailles, F-78026 Versailles Cedex, France Received February 27, 1989; Accepted April 12, 1989 Communicated by G. Wenzel Summary. Five diploid potato clones have been trans- formed by electroporation of protoplasts with different selectable markers. The resulting diploid regenerated plants have been used in somatic hybridization. It has been shown that hybrid cell selection on the basis of antibiotic or herbicide resistances brought by the two parents of fusion is an efficient method for the recovery of tetraploid somatic hybrids. Key words: Solanum tuberosum - Protoplast - Electro- poration Somatic hybrids Introduction Diploid clones of Solanum tuberosum L. have been used for crossing with diploid wild species. Introduction of resistance genes and increasing heterozygosity of the potato genus are the major goals of this breeding pro- gramme. However, tetraploid clones of potato are more vigorous and produce more tubers than diploid clones. Supplementary increase of heterozygosity during chro- mosome doubling for tetraploid creation is critical (Wen- zel et al. 1979). One way to accomplish this is to fuse protoplasts from two different diploid clones and regen- erate a tetraploid hybrid combining both parental genomes. Different approaches have been followed for protoplast fusion and recovery of somatic hybrids. Hy- brid vigour in calli resulting from protoplast fusion of diploid protoplasts has been shown to have no relation to any real hybrid nature of the regenerated plants (Wenzel et al. 1979). After protoplast fusion and mass culture of the cells, certain particularly favourable types of hybrid cells have been recovered, such as interspecific hybrids between Solanurn tuberosum and Solanum brevidens (Austin etal. 1985a; Ehlenfeldt and Helgeson 1987; Helgeson et al. 1986) or between two different Solanum tuberosum clones (Austin et al. 1985b; Deimling et al. 1988; Waara et al. 1989). Fused protoplasts could be se- lected by using a cell-sorter or by micromanipulation (Puite et al. 1986). In this case, and despite the high num- ber of fusion products selected by cell-sorting, somatic hybrids were highly instable and chromosomes of one partner have been partially lost. We propose an alternative procedure making use of introduced resistance genes. Chimaeric genes carrying coding sequences of proteins conferring antibiotic or her- bicide resistance can be introduced into plants, and all their cells express this resistance (Paskowski et al. 1984). The use of such selectable markers at the cellular level could greatly simplify the recovery of somatic hybrids. For example, after fusion between protoplasts from a kanamycin-resistant diploid clone and protoplasts from a hygromycin-resistant diploid clone, colonies resistant to both antibiotics could be expected to be somatic hy- brids. The high number of hybrid colonies which can be selected should permit the regeneration of several differ- ent plants, among which a true tetraploid somatic hybrid should be present. Foreign genes have been introduced into the potato genome by using A. rhizogenes (Ooms et al. 1986), but the regenerated plants displayed pheno- typic abnormalities due to the expression of Ri T-DNA genes. Thus, these plants were not usable for agronomic purposes. Transformation using Agrobacterium tumefa- ciens (Ooms et al. 1987) is a suitable method for the introduction of cloned genes into the potato genome. But only a few clones have been transformed and a high proportion of regenerated plants exhibited somaclonal variation. Moreover, this transformation method is still restricted to'tetraploid clones of potato. We describe here an alternative method for potato transformation using