«Biotechnology-Assisted Participatory Plant Breeding: Complement or Contradiction? PPB Monograph No. 3 Ann Mane Thro and Charlie Spillane 1 7 ...»
Recent progress using advanced backcross QTL methods has shown that DNA marker technology can be used to extract yie ldenhancing traits from exotic germplasm sueh as wild relatives (Tanksley and McCoueh, 1997). At present the cost:benefit ralios for developing the use of molecular marker teehnology in breeding programs are in the main only favorable for high -value commercial crops. Nonetheless, it is expected that eosts wiU fall and that MAS will eventually become an integral part of modern plant breeding (D. Duvick, pers. comm.). The effeet of the anti-transgenic food lobby on research funding and objectives (e.g.) in the European Union) may steer future research in sorne regions towards the use of molecular rnarkers to Biotechnology as a Set ofTools for Formal aJ\d Informal Plant Breeding manipulate germplasm within sexually accessible crop genepools, avoiding gene tic modification.
Once the use of markers becomes routine, MAS may provide a powerful tool for promoting geneflow lO locally adapted populations, since it allows the identification of individual QTLs for a specific trait not only in the donor but also in the recipient parent (deVicente and Tanksley, 1993; Tanksley et al, 1996; Tanksley and Nelson, 1996 ;
Tanksleyand McCouch, 1997). Recent advances in the use of molecular markers to identify QTLs may mean that 'trait·enriched' populations can be developed which wiU be easier to combine with locally adapted varieties or landraces. In sum, the innovative use of molecular maps and markers is likely to alter radically the way in which exotic germplasm is used in plant breeding and genetic enhancement in the decades ahead (McCouch, 1998).
Comparative molecular mapping is opening up hitherto unknown opportunities to capitalize on the similarity between d¡fferent species in the grass family (McCouch, 1998). It may be possible to develop a unified genetic map of higher plants which spans both monocots and dicots (Paterson et al, 1996). These developments will make it possible to study the genetic basis of adaptation across difTerent crop species and to apply the knowledge gained from one crop to the introduction of new genes into another crop (Devos and Gale, 1997; McCouch, 1998;
Sasaki, 1998). The relatively small genome of rice has meant that this crop is likely to become the 'anchor genome' for the comparative mapping and isolation of a1l cereal genes. A number of public· and privatc·sector efforts are now under way to sequence the rice genome.
Much plant biotechnology research is currently directed at the improvement of speci.fic 'quaJity' traits in modern varieties (Mazur et al, 1999). It is likely that sorne landraces, both locaily and widely adapted ones, can also be improved in this way. Paradoxically, deciding not to take this course may in the longer term only hasten the displacement of landraces by other crops or improved varieties that can provide such quality traits.
Increastng fanners' access ta tratts from wtld relatives
As we have seen, farmers' varieties may lack genes for traits useful to fanners or other ehd users. The wild relatives of crops have already contributed many useful traits to crop production (Stalker, 1980;
Prescott-Allen, 1988; Lenné and Wood, 1991). While the use of genes from wild species has so far been confined mainly to major cereaJ and cash crops. it is likely that almost aH crops can benefit from the addition of agronomically desirable traits from this source, although these traits may not necessarily be easily accessible (e.g., Muehlbauer et al, 1994; Grimanelli et al, 1995; Singh and Ocampo, 1997).
Biotechnology-Assisted PPB: Comp(ement or Contrad iction?
There are examples of geneflow from wild relatives to domcsticates (e.g., Oka and Chang, 1961; de Wet and Harlan, 1975; Longley, 1999), but farmers on their own seldom systematically access useful genes from wild relatives a nd related species. There are major barriers to such access, such as reproduc tive isolation, embryo breakdown, hybrid sterility, and limited genetic recombination (Spillane and Gepts, 2000).
The disincentives faced by formal plant breeders in using wild relative s are felt even more acutely by farmers, who typically must seU or cat what they breed or select.
Nevertheless, access to useful genes from wild relatives can benefit resource-poor farmers. Baudoin e t al (1997) demonstrated the usefulness of embryo reseue in tissue c ulture to achieve the wide -cross transfer of uscful traits from wild s trruns of common bean (Phaseolus vulgaris) into the Andean cultivatcd genepool. Through on-farm trials and farmer participation, the best enhanced germplasm was then rapidly selected by farmers for incorporation into their existing beanmaize multiple cropping systems. Without the use of wide-cross embryo techniques it is highly unlikely that these Andean highland farmers would have had access to wild bean germplasm.
Conventional plant breeding has had major successes in transferring useful genes into cultivated varieties using either bridging crosses or wide crosses. For example, bridging crosses have often been used to access alien genetic variation in potato breeding (Iwanaga et al, 1991; Ortiz, 1998), while wide crosses have made significant contributions to wheat improvement (Jiang et al, L994). Biotcchnologies such as embryo reseue have also increased the opportunities for transfer (Sharma, 1995). One of the few examples of the farmer participatory dissemination of biotechnology products has occurred through the work of the West Africa Riee Development Association (WARDA), where progeny from an in vitro-facilitated inter-species cross between the indigenous African and Asian rice species have been entered into PVS trials (WARDA, 1999).
Wide crossing. especially of the less commercial crops, is considered by sorne to be a n egleeted area for research (Duvick, 1989).
Yet advances in wide -crossing teehniques su eh as hybrid embryo culture (Sharma et al, 1996) and the use of crossing strategies such as bridge erosses are making the wild relatives of many crops ever more aecessiblc (Stalker, 1980; Muehlbauer et al, 1994). The sueeess rate of gene transfer in wide crosses can be increased by knowledge of chromosome pruring mechanisms and their genetic control. This knowledge is essentiaJ to promote recombination between heterologous or homologous chromosomes if the size of the introgrcsscd chromosome segment(s) needs to be either minimized or maximized (e.g.• Luo et al, 1996). Continuing advanees in structural genomies (e.g., comparative mapping) and genetic engineering (e.g., crossability tran sgenes) are likely to facilitate wide crossing s till further in the coming years.
Biotechnology as a Set ofTools for Fonnal and lrlformal Plan! Breeding
Althaugh erop wild relatives are valued as a unique souree af genetic variatian, they have rarely beeo used to improve quantitative traits. It is aclmowledged that exotic germplasm of this kind is infrequently used by breeders (Duvick, 1996; Spillane and Gepts, 2000). Achieving a wide cross is, oC course. only the first step in successful gene transfer from wild to domesticated species. The problem oí 1inkage drag' of undesirable genes with the desirable gene can only be solved by long cycles of repetitive backcrossing to break the linkage. Studies have shown that, even after 20 or more years of conventional breeding, a single gene transferred from a wild species can still be linked with enough chromosomal DNA to contain more than 100 other potentially undesrrable genes (Young and Tanksley, 1989).
One example of how undesirable linkages limit aecess to useful traits is the low protein quality oC cultivated maize kernels (Or et al, 1993). Storage proteins (zeins) eontaining high levels of the essenlÍal amino acids methionine and lysine have been identified in unseleeted wild germplasm, but not in domesticated germplasm. It is thought that undesirable genetie linkages betwecn the zein loei and other loei have, since domestication, prevented both farmers and formal plant breeders from selecting for this trait using conventional breeding techniques (Swarup et al, 1995). MAS or genetic engineering may yet help to break this linkage.
New opportunities have been opened up by the recent development of a molecular marker-based technique that enables the transfer of QTLs conferring complex traits su eh as yield and organ size (Paterson, 1995; Tanksley and McCouch, 1997). This technique has now been demonstrated for rice (Xiao et al, 1998) and tornata (deVicente and Tanksley, 1993). Once its applicability to other crop/wild relative combinations is demonstrated, the technique may prove useful in developing trait-enriched germplasrn populations for both conventional and PPB projeets. One way forward may be the deliberate choice of diverse genotypes from erop eore colleetions (collections of lmes known to contain maxirnum levels of gene tic diversity and to be adapted lo difTerent agro-environments) for inclusion in QTL analysis studies (van Hintum, 1999).
Prouldtng useft" traits through transgenesis
Transgenie approaches to providing the genetic variation needed to solve a plant breeding problem are usually tried only ir suitable conventional approaehes are laeking or do not work-for example, if germplasm conferring resistance to an important pest or rosease has not becn found or is very diffieult to aeeess in the genepool of a major eornmercial crop. Many erop genepools are poor in agronomically useful traits, sueh as protein quality or abiotic stress tolerance. that are available in the genepools of other crops or species. In sorne cases Biotechnology-Assisted PPB: Compfement or ContradiChon?
transgenic approaches may be the only way of obtaining resistant or improved varieties (J. Tohme, pers. comm.).
A number of serious pest5 and diseases are already being tackled in thi s way. One example is 50ft rot or blackleg (Enuinia carotovora) in potato, which causes crop losses estimated at US$1 00 million per year worldwide (Perombelon and Kelman, 1980). Resistance is lacking in the potato genepool but has been identified in the wild species Solanum breuidens, which cannot be easily crossed with S. tuberosum (Austin et al, 1988; Wi11iams et al, 1993). A transgenic route is thus the only possible one.
Sorne other examples of pests or diseases for which conventional resistance options are lacking include;
Insects in cowpea (lITA, 1992) Leaf ro11 virus (PLRV) in potato (Corsini et al, 1994) Rice hoja blanca virus (Madriz et al, 1998) Rice grassy stunt virus (Swaminathan, 1982) Black sigatoka disease in banana (Swennen and Vuylsteke, 1991) CoITee seed weevil (CENICAFE, 1997) Sean golden mosalc virus (Hidalgo and Seebe, 1997) African cassava mosaic virus (Cours et al, 1997; Obm-Nape et al, 1997) Viruses in papaya (Gonsalves, 1998; Prasartsee et al, 1998) Insects in cotton (Estruch et al, 1997).
Similarly, crops contaln no known genes for resistance against viroids (the smallest infectious agents of plants). At present, the only practical way of protecting crops from viroid epidemics is to diagnose infected plants and then to eliminate them from cultivation. Two genetic engineering strategies using antisense genes (Yang et al, 1997) or a yeast ribonuclease (Sano et al, 1997) have been developed to provide new sourees of genetic resistance against specific viroids.
Although there are stin problems in developing efficient transformation systems in many erops, a crop's accessible germplasm already extends in principie to many other organisffis and could even inc1ude synthetic genes (e.g., Rotino et al, 1997). In particular, pest and disease resistance provides a multitude of examples in which transgenes have been obtained from diverse species and organisms.
A range of other agronomically useful genes have now been isolated and suceessfully transfcrred to crops. Many single plant genes are also now being transferred between sexually incompatible crop plant species (e.g., Whitham et al, 1996; Molvig et al, 1997; Wilkinson et al, 1997). For ¡nstanee, pathogen resistance genes can be transferred from one plant species to another (e.g., tobacco to tomato, and vice -versa) and rernain functional (Rommens et al, 1995).
Biotechnology as Se! of Tools for Formal and Informal Plant Breeding Q While the majority of agronomic traits are quantitative and hence difficult to improve using existing transgenic technology, many monogenes are also known to con(er majar agronomic benefits (Table 1). In addition, monogene mutations are ofmajor importan ce in breeding programs. Examples inelude 'opaque -2', which improves the nutritional value of maize kernels, 'nor', which increases the shelf life of toma toes, and 'Rht1' and 'Rht2', which reduce the height ofwheat plants (e.g., Lohmer et al, 1991). lndeed, the Rht-BljRht-DI and dwarfd8) genes that were largely responsible for the Green Revolution have recently been shown to be mutant genes that are insensitive to certain growth hormones (Peng et al, 1999). The identification, isolation, and transfer of such monogenes between crop species or varieties may offer new opportunities to bring about genetic gain rapidly, in landraces as well as modero varieties.
Transferring desirable monogenic traits from exotic to adapted cultivated germplasm through conventional plant breeding can be highly time-consuming (Ronald, 1997). Transgenic technology is often equated with transferring genes between species, but it can equally well be used to transfer genes within a crop. For instance, if a desirable resistance gene homolog is available in a particular accession but not in the variety of choice, transgenic techniques can be uscd to move it. In sorne crops, once a resistance (or other) gene has been eloned (e.g., Kilian et al, 1997), transgenic cultivars can be generated within 2 years, compared with 5 -7 or 10 years using a c1assical backcross approach (Ronald, 1997; C. Qualset, pers. comm.). Where PPB programs require access to specific monogenic traits, transgenic approaches can definitely help deliver them quickly.
Transgenic technology can be used to enhance landraces. For example, cassava farmers in Tanzania like both MulundijS, which is a selection from an on-station variety trial, and their local variety.