Cytogenetics and Plant Breeding Cytogenet Genome Res 120:384–395 (2008) DOI: 10.1159/000121087 The use of cytogenetic tools for studies in the crop-to-wild gene transfer scenario a a b b E. Benavente M. Cifuentes J.C. Dusautoir J. David a Departamento de Biotecnología, ETS Ingenieros Agrónomos, Universidad Politécnica de Madrid, Madrid (Spain) b Unité Mixte de Recherche Diversité et Adaptation des Plantes Cultivées (UMR-DIAPC), Supagro Montpellier-INRA-IRD, Montpellier (France) Accepted in revised form for publication by M. Schmid, 7 December 2007. Abstract. Interspecific hybridization in plants is an im- into a recipient species (i.e., the wild). Several examples in portant evolutionary phenomenon involved in the dynam- hybrids and derivatives are given to illustrate how these ap- ics of speciation that receives increasing interest in the con- proaches may be used to evaluate the potential for gene text of possible gene escapes from transgenic crop varieties. transfer between crops and wild relatives. Different tech- Crops are able to cross-pollinate with a number of wild re- niques, from classical chromosome staining methods to re- lated species and exchange chromosome segments through cent developments in molecular cytogenetics, can be used homoeologous recombination. In this paper, we review a set to differentiate genomes and identify the chromosome re- of cytogenetic techniques that are appropriate to document gions eventually involved in genetic exchanges. Some clues the different steps required for the stable introgression of a are also given for the study of fertility restoration in the in- chromosome segment from a donor species (i.e., the crop) terspecific hybrid forms. Copyright © 2008 S. Karger AG, Basel Observation of interspecific hybrids between different was kept isolated within each species and for a gene ex- taxa in nature is rare but recurrent (Ellstrand et al., 1996; changed after their evolutionary differentiation can be in- Arnold, 1997). Most often, such hybrids are almost fully consistent. These phylogenetic incongruities support a re- sterile and therefore were long considered as dead end forms. ticulate evolution pattern (Vriesendorp and Bakker, 2005; Nowadays, the new technological tools permit us to inves- Marhold and Lihova, 2006). tigate whether a majority of plant species may be derived Following hybridization, gene flow may result in the in- from past hybridization events (e.g., Stebbins, 1958) and this trogression of chromosome segments from one species to is acknowledged to be an important phenomenon in angio- another by direct and recurrent back crosses. Alternatively, sperm evolution (Rieseberg and Ellstrand, 1993; Rieseberg interspecific hybrids might give birth to new fertile species and Wendel, 2004). The reality of ancient introgression is either via spontaneous and instantaneous chromosome difficult to establish but phylogenetic studies are increas- doubling (allopolyploidy; e.g., wheat and oilseed rape) or via ingly providing indirect evidence. Interspecific gene trans- the fixation of viable recombinant chromosome sets (homo- fers cancel the accumulated nucleotide divergences between ploidy; see Rieseberg, 1997; Rieseberg and Carney, 1998). species. Then the phylogenetic trees obtained for a gene that These stable and fertile allopolyploid and homoploid forms can also constitute bridges and gene reservoirs for subse- quent gene flows back to their diploid progenitors. Introgression can then act on ecosystems by species in- The authors are funded by the ‘ACI Impact des OGM’ from the French Ministry vasion or replacement (Ellstrand, 2003) and may thus have of Higher Education and Research. an impact in biological diversity. In the crop-wild context, Request reprints from E. Benavente with or without genetically transformed varieties, evolution Departamento de Biotecnología (Genética), ETSI Agrónomos towards nasty weeds is also a major rising concern (Ell- Universidad Politécnica de Madrid, 28040–Madrid (Spain) telephone: +34 91 3365714; fax: +34 91 5434879 strand et al., 1999; Burke et al., 2002; Lu and Snow, 2005). e-mail: [email protected] Weeds incorporating transgenes carrying insect or disease Fax +41 61 306 12 34 © 2008 S. Karger AG, Basel Accessible online at: E-Mail [email protected] 1424–8581/08/1204–0384$24.50/0 www.karger.com/cgr www.karger.com resistance, or resistance to environmental stresses, are like- 1 ly to display increased overall fitness and competitiveness, Crop Wild relative and their spread may have serious agronomic and environ- mental implications (Warwick et al., 2003; Watrud et al., 2004; Reichman et al., 2006). Natural selection of a favour- Interspecific hybrid able allele in an interspecific context has long been neglect- 2 ed but while the spread of an introgressed allele is possible through neutral evolution, it can thrive much more success- Homoeologous pairing fully through natural selection even with low selective ad- Intergenomic recombination vantage (Arnold et al., 1999; Rieseberg and Burke, 2001). Wild species and crops are growing in sympatry in many 3 Reductional Non-reductional documented situations (see Ellstrand et al., 1999 for a re- meiosis meiosis view). In these complexes of species, either the direct wild ancestor of the crop or any related species could be a pos- Aneuploid gametes Unreduced gametes sible recipient of the crop genes. The probability of interspe- cific hybridization depends on various factors, such as phy- 2n 2n logenetic relatedness, mating systems, and density and spa- tial distribution of wild relatives (Warwick et al., 2003; Chevre et al., 2003). Nevertheless, the main goal is not to prove that interspecific hybridization occurs in nature and Backcross Selfing may impact the species communities, but to document the Hybrid offspring likelihood of chromosome segment transmission and to de- 4 scribe the ways of gene exchange and transfer between spe- cies. Figure 1 illustrates the various pathways that a crop nu- cleotide sequence must follow to be stably incorporated into G a wild related species. In this paper we review the utility of n cytological methodologies to investigate the critical steps in Backcross the route and to identify the mechanisms and the fate of a transferred chromosome segment. Gn T h e t o o l s Stable introgression into the wild species This section briefly describes the most common meth- ods used for chromosome analysis in interspecific hybrids Fig. 1. Steps in the crop-to-wild gene flow route. 1: Interspecific and their progenies. hybridization. 2: Chromatid exchange between crop and wild chromo- somes. 3: Production of functional gametes and fertilization. 4: Trans- mission of crop-wild recombinant chromosomes during karyotype Classical cytogenetic methods evolution. The red block represents a crop genetic sequence success- Traditional dyes (aceto-carmine, aceto-orcein, Feulgen fully introgressed, whereas the blue block represents a non-recombined and variants) bind chromatin and nicely stain chromo- sequence that will likely be lost during karyotype evolution in the hy- somes for optical microscopy. Despite the great develop- brid lineage. ment of newer alternatives, these staining methods are still commonly used to visualize chromosomes both in mitotic and meiotic cells. Their main limitation is that individual phase-I (MI) cells due to the different morphology of chro- chromosomes within a complement cannot be identified mosomes at this meiotic stage. This explains why many re- unless they differ morphologically, i.e., by size, centromere ports that document the employment of C-banding for position, presence or location of secondary constrictions, meiotic MI pairing analyses have not used the technique to etc. identify specific chromosomes but to simply discriminate Several staining methods provide distinctive and repro- genomes with sharply distinct banding patterns (e.g., ducible banding patterns for specific chromosomes or ge- Hutchinson et al., 1983). nomes based on the size and location of different classes of chromatin. Among the classical banding methods, C-band- Cytomolecular methods ing has been the most frequently used for identification of Fluorescent in situ hybridization (FISH), using different individual plant chromosomes either at mitosis or meiosis classes of DNA sequences as probes, has been extensively (e.g., Dhaliwal et al., 1977; Gill et al., 1991) ( Fig. 2 ). It is worth used for cytological discrimination of specific chromo- mentioning that a perfectly distinctive band pattern on mi- somes and individual genomes in many plant species (Jiang totic karyotypes may not be as easily resolved on meta- and Gill, 1994, 2006). The following sections will provide Cytogenet Genome Res 120:384–395 (2008) 385 numerous references illustrating the use of different FISH approaches in interspecific hybrids and derivatives from most of the crops for which gene flow to wild relatives has been documented (wheat, rice, oilseed rape, cotton or sun- flower, among others). FISH using total genomic DNA labelled probes (GISH) easily and reliably informs on the genome adscription of specific chromosomes (or chromosomal segments) in many multigenomic materials based on colour codes ( Figs. 3 –5). The technique is sometimes referred to as multicolor-GISH when differentially labelled genome probes are simultane- ously hybridized to chromosome spreads. In some instanc- es, the presence of shared sequences in the target and non ! target genomes results in cross-hybridization signals of to- Fig.