| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
a Dep. of Agronomy, 915 West St., Purdue Univ., West Lafayette, IN 47907-2054
* Corresponding author (gejeta{at}purdue.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONREFERENCES |
|---|
Abbreviations: AGA, agar gel assay • DMBQ, 2,6-dimethoxy-1,4-benzoquinone • EAGA, extended agar gel assay • HR, hypersensitive response • IR, incompatible response • LGS, low production of germination stimulant • LHF, low production of the haustorial initiation factor • PRA, paper roll assay • QTL, quantitative trait loci • SSR, simple sequence repeat
|
ACKNOWLEDGMENTS |
|---|
Received for publication April 4, 2007.
a Dep. of Agronomy, 915 West St., Purdue Univ., West Lafayette, IN 47907-2054
* Corresponding author (gejeta{at}purdue.edu).
The parasitic weed Striga Lour. has been one of the more intractable agricultural problems seriously limiting productivity of cereal and legume crops in sub-Saharan Africa. The development of crop plants with resistance to Striga has been limited because of the complexity of interactions between host, parasite, and the physical environment. We formulated a novel approach based on developing and exploiting a thorough understanding of the biology of this intricate association. We employed this knowledge-based approach to develop powerful selection assays, to conduct genetic analyses, and to characterize the range of mechanism involved in host plant resistance to Striga. Information thus generated has been used to identify unique sources of resistance to Striga, introgress these genes into selected cultivars, and deploy sorghum [Sorghum bicolor (L.) Moench] cultivars with known sources of Striga resistance either independently or in tandem as genotypes with multiple mechanisms of resistance. High yielding sorghum cultivars with Striga resistance and evident grain quality characteristics have been developed and deployed in a number of African countries to be used as cultivars per se or as a central component of an integrated Striga management program.
Abbreviations: AGA, agar gel assay • DMBQ, 2,6-dimethoxy-1,4-benzoquinone • EAGA, extended agar gel assay • HR, hypersensitive response • IR, incompatible response • LGS, low production of germination stimulant • LHF, low production of the haustorial initiation factor • PRA, paper roll assay • QTL, quantitative trait loci • SSR, simple sequence repeat
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONREFERENCES |
|---|
Although a number of Striga control measures have been suggested (Parker, 1991), methods developed through conventional approaches have been of limited value to subsistence farmers in combating this menace in the semiarid tropics (Ogborn, 1987). The low efficacy of control measures such as chemical fertilizers, herbicides, or new cultural practices is due in part to lack of access to an already overpriced market of production input (Eplee and Norris, 1987). The limited understanding of the host–parasite biology that curtailed the potential impact of these treatments has also been suggested as another major impediment (Ejeta and Butler, 1993). Hand weeding and cultivation, the most prevalent control practices, are practiced after the parasite emerges above ground and has already inflicted significant damage to the crop. Farmers with crop fields severely infested with Striga resort to abandoning their fields contributing to an already severe pressure on availability of farm lands. Therefore, measures that minimize impact on crop losses, deplete the Striga seed bank in the soil, reduce further Striga seed production, and diminish the spread of Striga to uninfested fields are needed. Host plant resistance, when effectively deployed, offers many of these benefits with an insignificant increase in cost, as the technology is embedded in the genetics of the seed of the crop cultivars to be planted (Ejeta, 2005). Genetic control thus offers a practical and economically feasible measure (Parker and Riches, 1993; Ejeta et al., 1997) either independently or as a central component of a more comprehensive and integrated mix of Striga control approaches.
Understanding Host–Parasite Biology
Striga control measures that are based on a better understanding of the biology of the parasite and its interactions with host plants and the physical environments are likely to be more effective. Host plant resistance to Striga involves physiological and genetic mechanisms and requires a thorough understanding of the biophysical processes governing parasitism as shown in the Striga life cycle (Fig. 1
). In the last 50 to 60 yr, considerable effort has been aimed at unraveling the intricate biology of Striga and its association with its diverse host species. Insightful observations have been made resulting in a greater understanding of key biological processes by which these parasites grow and develop as well as how they form an intricate network of coordination with their hosts.
|
The eventual survival of the parasitic seedling requires successful attachment to a host plant via a special organ called the haustorium. Production of this organ, however, requires that the parasitic seedling is in close proximity to the host so that another host-derived signal may trigger this developmental transition. Formation of haustoria is the start of the parasitic process in which the parasite begins to tap water and nutrients from the host, a crucial step to the eventual development of the parasite. Unlike the initial key signal required for germination of Striga seeds, host-produced compounds that are involved in haustorial formation have not been identified. Yet it is known that the chemistry of haustorial induction is distinct from that of germination stimulants. Kinetin, simple phenolic compounds, and quinines like 2,6-dimethoxy-1,4-benzoquinone (DMBQ) have been found to be active haustorial initiators (Riopel and Timko, 1995).
Parasitic attachment to host root surface takes place immediately on contact and facilitated by a secretion of a hemicellulose-based adhesive substance that fixes the parasite to the host root (Baird and Riopel, 1985). The binding that takes place is strong, but not very specific as haustorial produced by Striga are known to attach to host or nonhost roots (Hood et al., 1998). Attachment is a prerequisite of the transition to the penetration phase of haustorial development that may involve additional chemical or tactile signals from the host root. Penetration of host root to tap nutrients occurs rapidly since seed reserves are small and mostly expended during germination and haustorial differentiation. Eventual connection to the vascular core of a host root appears to be aided by enzymatic activity that breaks down wall components of host cortical cells. After penetration of the xylem, haustorial cells loose their protoplasts transforming them into water-conducting elements that are continuous with host xylem (Dörr, 1997). However, Striga does not appear to possess capacity to establish direct connections with host phloem (Rogers and Nelson, 1962).
Soon after attachment to host tissue, the parasite seedling develops a tubercle to assist with accumulation of nutrients. The cotyledenous Striga leaves emerge from the seed coat within a day after vascular connections are established with the host (Hood et al., 1998). Eventually, the parasite matures forming flowers within 6 wk after aboveground emergence, later forming seeds. Some Striga species are self-pollinating while others are obligate outcrosses. In either case, the Striga fruit capsules form mature seeds in as little as 2 wk after pollination. Throughout the process, the parasite is regarded as part of the host with cambial activities coordinated with that of the host resulting in formation of continuous elements bridging between them (Joel, 2000). The survival of Striga as a parasite and its successful development as a plant depends on its interactions with the host plant. Both metabolic and developmental processes are needed in bridging connection of the parasite with its host as well as in its eventual survival (Joel et al., 2007).
Successful parasitism is therefore a series of interactive process between the host and parasite conditioned by a large number of genetic and physiological events each possibly influenced by additional array of environmental factors. Host plant resistance based on observation of emergence above ground of parasitic seedlings and level of infestation, therefore, is a complex, quantitatively inherited trait that is difficult to select for using conventional approaches of plant breeding. In our sorghum breeding program, we made the characterization and dissection of Striga resistance into specific mechanisms based on a series of host–parasite signal exchanges the central focus and premise for our research approach and effort aimed at developing Striga resistant cultivars (Ejeta et al., 1991).
Development of Bioassays
Adequate genetic variation and availability of effective selection tools are essential requisites for successful plant breeding efforts. Unfortunately, selection methods that work well for improving other desirable crop traits have not operated at the same efficiency for Striga resistance (Ejeta and Butler, 1993). Field selection for Striga resistance has not been successful because of the difficulty in clearly identifying resistant variants in segregating germplasm populations, lack of clarity of information on the genetic control of field resistance, and the difficulty of establishing uniform infestation of the parasite population under varying environmental conditions. Lack of rapid and efficient screening techniques has been a major constraint, thereby slowing the progress of Striga resistance breeding. Laboratory methods that permitted observation of the early events in the developmental association between the host and parasite were needed.
As a result, development of bioassays received primary focus and consideration in our Striga resistance breeding. Targeting the initial processes around Striga germination, we first developed an in-vitro laboratory procedure, the agar gel assay (AGA), that separated sorghum genotypes on their capacity to produce the exudates required for Striga seed germination (Hess et al., 1992). Using this assay, we established that sorghum genotypes varied in the amount and type of the germination signal they produced (Weerasuriya et al., 1993). In field tests, sorghum genotypes that produced very low levels of the germination stimulants were found to be resistant to Striga (Ramaiah, 1987; Hess et al., 1992). Subsequent efforts led to the development and refinement of two new in-vitro bioassays namely, the extended agar gel assay (EAGA) and the paper roll assay (PRA) (Ejeta et al., 2000).
The EAGA, though not a quantitative assay for the production of host-derived signal for haustorial formation in host roots, distinguishes host genotypes qualitatively on the basis of their ability to induce haustorial formation. The EAGA is conducted by treating petri plates of agar containing sorghum seedlings grown in the presence of conditioned Striga seed with ethylene or another growth hormone, GR-24, to overcome host genotypic differences in germination stimulant production. Exposure to ethylene or GR-24 offers a more even germination of nearly all conditioned Striga seeds over the plate. With this assay, haustoria are only observed if the sorghum root produces a signal that induces its development. Presence of haustoria can be detected around the growing host root at 48 h after ethylene or GR-24 treatment using a stereomicroscope.
The PRA was developed to permit observation of the early stages of Striga attachment following germination and development of the haustoria as the organ of attachment to host roots. The procedure involves growing sorghum seedlings with their roots between rolled layers of germination paper. When seedlings are 1 wk old, papers are unrolled and filter-paper strips containing artificially germinated Striga seed are placed on sorghum roots. Papers are then rolled and placed in an enclosed glass container which allows light to reach growing sorghum shoots. After an interval of 2 to 3 wk, papers are unrolled to reveal progressive invasion of the parasite on host roots. By this method we typically see several attachment events. The PRA works best if Striga seeds are pregerminated with ethylene or GR-24. Treatment with these germination stimulants overcomes differences in host genotypes for germination stimulant production. Moisture is also critical for encouraging invasion. Germination papers that are saturated or too dry reduce attachment events in the assay. The PRA can be an effective tool for identifying early postinfection resistance mechanisms. We have been using the assay, in combination with the quicker agar assay for germination stimulant production, to screen germplasm and breeding populations of sorghum for resistance to Striga.
Use of these assays has enhanced our ability to more systematically evaluate and exploit sorghum germplasm as sources of Striga resistance by focusing on each stage of the parasitic process individually. The bioassays have also provided further insights to the interactive biological processes between Striga and the roots of host plants during the early stages of the infection process giving us an increased understanding of the specific mechanisms of resistance associated with each source of host genotype (Cai et al., 1993). Furthermore, defense responses triggered in response to infection could also be monitored and exploited via these assays. Hence, disrupting these interactions offers unique opportunities for controlling Striga through identification of genetic variants with single or multiple interventions at key critical stages throughout the life cycle. Our working hypothesis has been that host plant resistance derived as a result of disruption in any one of these critical stages is likely to be simply inherited, therefore easy to select for and to introgress through conventional breeding or manipulate via other biotechnological approaches for the development of Striga resistant crop cultivars.
Characterization of Mechanisms of Striga Resistance
Little has been known about the actual host defenses that discourage parasitic growth and establishment. We felt that development of simple reproducible in vitro assays to examine parasitic associations will not only assist our breeding effort but may elucidate the specific mechanisms involved. We embarked on the use of the above-described assays to classify established Striga resistant cultivars based on their defined reactions to Striga invasion at key stages of parasitic development. Our bioassays allowed us to describe and classify four mechanisms of Striga resistance (Ejeta et al., 2000). We screened a large collection of cultivated sorghums as well as a smaller group of wild sorghum accessions in our breeding program and identified unique genetic variants including those that could only be found among the wild sorghum accessions (Rich et al., 2004). The specific Striga resistance mechanisms we have described to date include resistance associated with low germination stimulant (LGS) production, low production of the haustorial initiation factor (LHF), hypersensitive response (HR), and the incompatible response (IR) to parasitic invasion of host genotypes (Ejeta et al., 2000).
LGS genotypes of sorghum produce insufficient amounts of the exudates required for germination of conditioned Striga seeds. A chemical stimulus produced by host roots elicits parasitic seed germination, but as described above an additional metabolic process of conditioning needs to take place before the Striga seed can respond to this external stimulus with germination. Sorghum genotypes that produce very low levels of the germination stimulants have been found to be resistant to Striga in field tests (Ramaiah, 1987; Hess et al., 1992), but not all Striga resistant sorghum genotypes are low producers of the stimulant exudate. All highly susceptible sorghum genotypes appear to be high producers of the germination stimulant. We have also identified and characterized host root–produced chemical signals of two different types (Netzly et al., 1988; Siame et al., 1993). Though several classes of chemicals have been shown to elicit Striga seed germination, the sorgolactones appear to be the most common and important in terms of controlling Striga germination in the field (Hauck et al., 1992).
In resistance based on LHF, germinated Striga near the roots of sorghum lines possessing this trait normally do not form haustoria and therefore die from their inability to attach to their potential host. Although the need for chemical signals exuded by host and nonhost plants to elicit Striga germination has been known for many years, evidence for the requirement of an additional host signal to encourage production of the haustorium to facilitate attachment to host roots has only emerged recently (Riopel and Timko, 1995). Striga seeds germinated using artificially synthesized germination stimulants in vitro do not develop beyond the formation of a radicle unless these radicles are placed close to a developing root of a host or some nonhosts Unlike the signals required for germination of Striga seeds, host-produced compounds that are involved in haustorial formation have not been identified. Yet, it is known that the chemistry of haustorial induction is distinct from germination stimulants. A large number of phenolic compounds have been shown to function as haustoria initiators in Striga. A simple quinone, DMBQ, though not found in root exudates, has been shown to act as a strong haustorial initiating factor (Lynn and Chang, 1990).
Resistance based on the HR involves localized necrosis of host tissues surrounding the site of attachment, presumably coupled with a release of phytoalexins that kill the attached Striga. We began the characterization of the HR with an adaptation of laboratory assays to measure the trait and identify sources of HR among cultivated and wild sorghums. A description of HR and its association with failed Striga parasitism has been reported (Mohamed et al., 2003). In a screening of a collection of sorghum cultivars, wild accessions, and breeding lines, all susceptible lines showed no necrosis. Whereas absence of necrosis and unobstructed parasite development characterized susceptible genotypes, some resistant sorghum cultivars (e.g., Framida and Dobbs) and a wild accession, P47121, showed necrosis in nearly 70% of attached Striga. In each of these lines, attached Striga did not penetrate host tissue or develop further. The wild sorghum accession, P47121 showed the highest level of necrosis (90%) and discouraged haustoria penetration (83%) resulting in eventual withering and death of the parasites. Framida and Dobbs also were effective with approximately 70% necrosis and 50% failed penetration. HR expression has been extensively studied in a number of host–parasite systems, where it is generally characterized by the appearance of a necrotic zone around the site of attempted infection (Agrios, 1988). Host cell death results in unsuccessful establishment of the parasite and leads to its ultimate demise.
Sorghum genotypes with the IR mechanism of resistance do not exhibit an apparent necrosis in host root tissue surrounding the parasite attachment site. Otherwise, the host response is similar to the HR in that it serves to discourage development of Striga beyond attachment (Grenier et al., 2001). In host genotypes whose Striga resistance is based on IR, Striga seedlings that succeed in penetration of host tissue may not develop beyond first emergence of leaves. Some Striga appear to develop normally at first but show signs of stunted growth. This is similar to that observed when Striga unsuccessfully infests nonhost plants, thus the use of the term IR. We have identified several sources of IR among cultivated and wild sorghums. Striga resistant sorghum genotypes exhibiting IR as a defense response have been observed among cultivated sorghums as well as among wild sorghum genotypes evaluated. Histological observations (C. Grenier et al., unpublished data, 2007) suggest that IR reactions develop as a result of failure to establish adequate vascular connections, perhaps caused by deficiency of vital factors, or because of the production of toxic factors that disrupt sustained growth and development of the parasite.
Evidence for mechanical barriers to penetration has been reported in certain host–parasite associations by increased lignification (Maiti et al., 1984), deposition of cellulose layers (Olivier et al., 1991), and encapsulation (Labrousse et al., 2001). Hypersensitive responses against attaching parasites have been reported in Striga resistant cowpeas (Lane et al., 1994) and sorghum (Mohamed et al., 2003), as well as vetch (Vicia spp.) resistant to Orobanche L. (Goldwasser et al., 2000). The release of toxic substances that kill or hinder development of haustorial cells were noted in sorghum–Striga association (Arnaud et al., 1999; Neumann et al., 1999). An incompatible relationship was described with S. hermonthica (Del.) Benth. on sorghum (Grenier et al., 2001) and on Tripsacum dactyloides (L.) L. (Gurney et al., 2003) a wild relative of maize, whereby growth and development of attached parasites was retarded. A similar Striga resistant reaction was earlier described in var. Framida (Arnaud et al., 1999), although our own observation shows that this line possesses a combination of low germination stimulant production and hypersensitive response resistance mechanisms (Mohamed et al., 2003). Incompatible relationships with resistant hosts have also been reported for O. cumana Wallr. growing on sunflower (Helianthus annuus L.) (Labrousse et al., 2001) and O. crenata Forsk. on a variety of legumes (Pérez-de-Luque et al., 2005). In addition to having low germination stimulant activity to Striga, the resistant sorghum SRN39 possesses an apparent incompatible response to attached Striga that essentially halts parasite development at an early stage, even if vascular connections are made (C. Grenier et al., unpublished data, 2007). The overall effect of the IR reaction is that a small proportion of attached parasites proceed to slowly establish on resistant host, but most of the attached show diminished growth and delayed emergence above ground.
Genetic Analysis of Striga Resistance
The development of simple bioassays has facilitated our research in the genetic analysis of the Striga resistance trait both in our effort to determine genetic factors that condition the specific responses to Striga invasion as well as in molecular mapping and identification of quantitative trait loci (QTL) associated with each of these response reactions. The genetics of low germination stimulant was studied in populations of sorghum derived from the resistant cultivar SRN39 (Vogler et al., 1996). The AGA was employed to determine the inheritance of low stimulant production in progenies of SRN-39 and three susceptible lines, Shanqui Red, P954063, and IS4225. Segregation ratios suggested that this trait was inherited as a single, nuclear, recessive gene with largely additive gene action. The gene symbol LGS was proposed. The same approach was employed to study the inheritance of two additional mechanisms of Striga resistance, the low production of the haustorial factor and the hypersensitive response but using the EAGA (Mohamed, 2003). Analysis of progenies derived from a cross of Striga susceptible lines and a wild sorghum accession P78 with LHF, suggests inheritance of the trait through a dominant allele of a single gene. Analysis of F2:3 progenies from crosses between HR expressers CK32 and KP33 and susceptible lines TX430 and TX2737 resulted in a segregation of progenies for presence or absence of necrosis at the point of attachment in a ratio that reflected the presence of one dominant allele from either of two genes. The mode of inheritance of the IR mechanism of Striga resistance has not been clearly established. However, we have established that IR is independently inherited from low germination stimulant production (Grenier et al., 2001).
To map QTL associated with Striga resistance, we generated several genetic populations including recombinant inbred lines, large F2– and F2–derived F3 populations, and advanced backcross populations. Using both our laboratory assays and Striga-infested fields in Africa, we conducted phenotypic evaluation of these populations to identify putative markers for potential use in marker-assisted selection for Striga resistance. Some of these populations were created utilizing field-tested Striga resistant germplasm sources while others were based on lines selected in laboratory screening using bioassays during the characterization of specific mechanisms of resistance described above. The overall approach has been to identify putative QTL by genotyping a large initial population, followed by confirmatory studies using advanced backcross populations and near isogenic lines so that QTL that can be used in marker-assisted selection efforts could be undertaken with confidence. Different levels of progress have been reached for each of the mechanisms of Striga resistance we have studied. Our sorghum linkage map has an estimated map size of 1628 cM with an average interval of 9.5 cM between adjacent loci. The LGS gene mapped between two flanking markers, ISSR617 g and the restriction fragment length polymorphism marker PIO20025BamH1, at 7.9 cM and 5.7 cM, respectively. To map the LHF gene, a subset of 122 F2:3 families from a cross involving P78 and Shanqui Red was used for phenotypic and genotyping characterization. Simple sequence repeat (SSR) marker TXP358 was identified to associate with the LHF locus at a map distance of 7.5 cM. A large set of advanced backcross progenies from a cross involving P47121, the hypersensitive response parental wild sorghum line with a susceptible sorghum cultivar were used to map the HR factor. The HR locus was found to be at 7.5 cM and from SSR markers TXP96 and 12.5 cM from marker SBKAFGK1 (Ejeta, 2005).
Development of Striga Resistant Sorghum Cultivars
The approach in breeding for resistance to obligate root parasites based on understanding of the biology of host–parasite interaction has been effectively applied to develop and release a number of Striga resistant sorghum cultivars to several African countries (Ejeta et al., 2000; Ejeta, 2005). New sources of resistance to these pests have been discovered and to varying degrees exploited in breeding programs. The best characterized resistance phenotype against Striga is low germination stimulant production. Cultivar differences in sorghum to stimulate Striga germination are well correlated to field resistance (Hess et al., 1992). Low Striga germination stimulant production in sorghum is controlled by recessive alleles at a single locus (Vogler et al., 1996). A bioassay for this character has been exploited in developing Striga resistant sorghum cultivars (Hess et al., 1992). The nature of induction of these genes is now known, although the relationship between the activity of these genes and the formation of germination stimulants has not yet been clearly established (Bouwmeester et al., 2003).
Beyond low germination stimulant production by host plants, several other resistant phenotypes are being discovered and to some degree exploited in breeding programs. A laboratory method was used to screen wild and cultivated sorghums for the ability to cause haustorial initiation of germinated S. asiatica (L.) Kuntze, and wild accessions of sorghum were found that showed reduced haustorial formation (Rich et al., 2004). Exudation of phytotoxins by the host that kill unattached parasites has been reported in sunflowers resistant to O. cumana (Serghini et al., 2001).
As crop-breeding programs adopt use of laboratory procedures, identify or create appropriate genetic populations, and employ deliberate selection for resistance to root parasites, host plants with high levels of resistance to parasitic weeds would likely emerge. It is interesting to note that in many of the above examples, potential resistance phenotypes are derived from wild relatives of crop plants. Molecular markers may facilitate the transfer of resistance genes into crop cultivars and facilitate pyramiding of multiple resistance genes into agronomically desirable ones. In sorghum, mapping populations have been developed that are polymorphic for genes associated with low germination stimulant production, low haustorial initiation, mechanical barriers, hypersensitive response, and incompatible response mechanisms of Striga resistance (Ejeta et al., 2000; Haussmann et al., 2004) Continued improvements in laboratory and field evaluation methods would likely continue to drive gains in resistance to parasitic weeds in all crop plants.
Selection for Specific Resistance Mechanisms
Targeting the early stages of parasitic development and using the AGA described above, we demonstrated that sorghum genotypes can be separated on their capacity to produce exudates required for Striga germination (Hess et al., 1992). Sorghum genotypes vary significantly in the amount and type of the germination stimulants they produce (Netzly et al., 1988; Weerasuriya et al., 1993; Rich et al., 2004). Sorghum genotypes that produce little or no germination stimulants have been shown to be resistant to Striga in field tests (Ramaiah, 1987; Hess et al., 1992). Not all Striga resistant sorghum genotypes are low stimulant producers, however, genotypes susceptible to Striga appear to be generally high stimulant producers and the trait is highly heritable (Vogler et al., 1996). The low germination stimulant (LGS) gene has been successfully introduced into high yielding and broadly adapted sorghum cultivars that have been deployed into several African countries (Ejeta et al., 1997).
A quantitative assay for production host-derived signals for haustorial formation in sorghum has not been available. However, using a modified AGA, host genotypes have been qualitatively separated on the basis of their ability to induce haustoria formation (Rich et al., 2004). Through this assay, haustoria is only observed if the sorghum root produces a signal that induces its development, and no haustoria was developed in genotypes that were LHF producers. Among the collection of sorghums studied, Rich et al. (2004) found sorghum variants that did not stimulate haustorial development in a small collection of wild sorghums, but no such variant was found among cultivated sorghums. We have also transferred the lhf gene found in wild sorghum lines into improved sorghum cultivars and pyramided them with other resistant sources. As described earlier, sorghum genotypes with specific mechanisms of resistance at early developmental stages before attachment have been developed and released. In addition, resistant cultivars with post-attachment mechanisms of resistance have also been developed (SRN39, PSL85061, P9401) and new germplasm identified among a collection of wild sorghums (P52, P101, IS47121).
Pyramiding Genes for Multiple Mechanisms of Striga Resistance
Our approach to exploiting specific traits in Striga resistance breeding has broadened since the development of sorghum germplasm possessing the LGS trait. As we discovered new sources of genes that render heritable traits associated with Striga resistance, we introgressed them into selected genotypes. The development of locally adapted African landraces, with Striga resistance genes newly added, is important in areas where the improved cultivars either do not perform well or have traits not acceptable to the local population. We have germplasm at various stages of introgression that have pyramided Striga resistance traits of LGS, LHF, HR, and IR. Stacking genes for Striga resistance was done into both improved modern cultivars with high yield as well as African landraces that possess unique adaptation and fit in specific niches of local environments.
Pyramiding Genes into Improved Sorghum Lines
A collection of sorghum lines with known genes for Striga resistance is maintained in our program. Crosses are made with these lines as donor sources and elite lines selected for their agronomic superiority with the intention of pyramiding genes for Striga resistance into these cultivars with an array of desired traits. Cultivars with broad adaptation and high yield are selected as recurrent parents to which the traits are transferred. Two different approaches were used. Phenotypic selection for Striga resistance was practiced on true breeding progenies derived from breeding populations in which selection has been undertaken at early generations for agronomic attributes of yield, grain quality, and maturity. Selection for Striga resistance was deferred to later generations after pure lines were extracted. In the second approach, crosses were also made between donor sources and selected cultivars in a direct two-way or four-way crosses with backcrosses to the recurrent parent. Advanced backcross populations of various genomic proportions are generated providing a select choice of populations for a marker-assisted selection program. These populations were used to both generate QTL markers associated with resistance as well as confirm previously detected markers as a confirmatory procedure so that marker-assisted selection is practiced only on those markers found consistently across populations and across physical environments.
Agronomically superior cultivars in which genes for two or more Striga resistance mechanisms have been added were developed. Many such progenies are under extensive phenotypic and genotypic evaluation in both field and the laboratory conditions. One such set was regionally tested as part of our International Striga Nursery that has been distributed to collaborators annually and on a request basis. A superior sorghum cultivar with strong Striga resistance and high grain yield was found to be very well adapted in the Amhara region of Ethiopia. This cultivar, designated as PSL85061 in our breeding programs was officially released in 2001 under the name ‘Brhan’ in Ethiopia. This cultivar is high yielding and contains genes for three different mechanisms of Striga resistance. A large-scale seed production was initially undertaken at Purdue University in 2002 and 3 Mg of seed of this cultivar was shipped to the Amhara region for wide distribution. This nuclear seed was also used for an in-country large-scale seed production and over 15 Mg of seed was produced in 2003 for wide distribution in northern Ethiopia during the ensuing crop season. Brhan is also currently being tested in Kenya and Tanzania both at experiment stations and on farmers' fields for its regional adaptation.
Introgression of Genes into African Landraces of Sorghum
Selected African landrace sorghums obtained from collaborators in eastern Africa (Ethiopia, Sudan, and Tanzania) and western Africa (Mali and Niger) were used to initiate crosses. African landraces with photoperiod sensitivity do not flower readily without artificial manipulation by decreasing daylight hours, to which these plants are exposed in the early stages of plant development when the floral initiation mechanism is triggered. Normally, making deliberate crosses between sorghum lines is a routine procedure. However, because of the photoperiod sensitivity associated with the landraces and differences in maturity in donor sources, it required an undue amount of time and attention becoming a significant and major project in the program. Nevertheless, with organized manipulation of day-length using greenhouse, growth chamber, as well as an off-season nursery in Puerto Rico, we have been able to generate a large number of crosses and backcrosses between these African landraces and the source cultivars and accessions available to us from our program. Two-way, three-way, and four-way crosses and backcrosses were made to the cultivar of choice. We now have advanced backcross germplasm at various stages of introgression that contain pyramided Striga resistance traits of LGS, LHF, HR, and IR. Proper evaluation of these populations requires testing in large replicated plot trials with data collected accordingly. This is an expensive undertaking. As a result, only two of these introgressed populations involving the African landrace El-Mota from Niger have been evaluated. Field evaluation for Striga resistance was undertaken at two locations in Niger during the 2002 and 2003 crop seasons. Based on data on field Striga resistance, ranking of progenies was made from which bulk segregates of the best and worst 20% were selected for genotyping. Markers associated with Striga resistance were identified to initiate marker-assisted selection in collaboration with the national research program in Niger. Progenies with high proportion of the genome of the local African landrace El Mota and Striga resistance were selected. These recombinants will be used in marker-assisted selection as well as for subsequent field test to check their usefulness in Niger even before additional backcrossing. Evaluation of the remaining introgressed populations need to be undertaken in each of the collaborating countries, Ethiopia, Tanzania, Mali, and Sudan.
Deployment and Adoption of Striga Resistant Cultivars
A number of Striga resistant sorghum cultivars, developed and released by the Sorghum Research Program at Purdue University, have been widely distributed for use in Striga endemic regions of a number of African countries. The first official release of cultivar SRN 39 took place in Niger and Sudan in 1991. This release was a result of breeding through conventional approach of selection in field tests from natural source populations. Subsequently, a set of eight Striga resistant sorghum cultivars were developed using our new paradigm and based on bioassay-mediated selection. These cultivars were carefully tested for field resistance to Striga in collaboration with plant breeders in several countries. Based on the results of these field tests, eight food-grade sorghum cultivars that also possessed excellent drought tolerance were officially released for wide cultivation in 1995. To accelerate the process of wide testing and diffusion, 8 Mg of seed was centrally produced and distributed to 12 African countries in collaboration with World Vision and with financial support from the Office of Foreign Disaster Assistance of the U.S. Agency for International Development.
Extensive testing has been conducted in each of these countries, but success in generating data required for official release was variable depending on the strength of national research efforts. As a result, two sorghum cultivars were released in Tanzania (2002), one cultivar in Ethiopia (2002), and two cultivars in Eritrea in 2005. With earlier releases already made in Ethiopia (2), Sudan (2), and Niger (1), the list of Striga resistant sorghum cultivars from our breeding program in commercial cultivation has grown to 11 cultivars (Ejeta, 2005). These sorghum cultivars represent a significant research advance in food-grade sorghums and to date, they are the only research-derived sources of high yielding Striga resistant cereal cultivars available for large-scale cultivation or use by subsistence farmers in Striga endemic areas.
These cultivars have performed well in each of these countries since their release. They have become popular among farm communities that have tried the effectiveness of these cultivars under heavy Striga infestation. Farmers who have grown the Striga resistant sorghum cultivars have appreciated the added merits of food quality and drought resistance characteristics, in addition to the continued stability of resistance to Striga. As a result, demand for seed of these cultivars exceeds available seed supply. Seed distribution has primarily been through shared farmer exchanges. However, a much greater diffusion and adoption of these cultivars would have been possible if it was not for lack of a functional seed multiplication programs in many of these countries.
Integrated Striga Management
Striga resistant sorghum cultivars have not been available until recently as the complex nature of the host–parasite relationship had hampered progress from selection in field-based breeding. Increased understanding of the basic biology of the parasite and discovery of novel screening procedure opened the way for improving Striga resistances described. This has made significant progress in a set of the arsenal available to fight the Striga menace. No single solution is likely to offer long-lasting solution to the serious problem of parasitic weeds. In recognition of this, we undertook a study to evaluate the synergistic effect of combining the use of Striga resistant sorghum cultivars with the cultural control options of soil fertility management and moisture conservation. The results showed that integration of resistant cultivars with chemical fertilizer and soil moisture conservation using tied ridges significantly increased grain yield and reduced infestation by Striga (Ejeta, 2005). Comparison of individual control options showed that Striga resistant cultivar played a major role in reducing Striga pressure, both in terms of Striga count and vigor, than fertilizer and soil moisture conservation. Both fertilizer and soil moisture conservation had no effect on Striga count and vigor in a resistant cultivar. However, in the susceptible cultivar, chemical fertilizers and soil moisture conservation significantly reduced Striga count and vigor. Alternatively, in both resistant and susceptible cultivars, soil fertility management and moisture conservation factors significantly contributed to increase in yield indicating that application of optimum production practices, use of the recommended rate of fertilizer, and soil moisture conservation would increase productivity.
Financial support for research that led to the results documented herein has been primarily obtained through a series of grants by the United States Agency for International Development (through INTSORMIL) and by the Rockefeller Foundation. Basic research was undertaken as graduate research dissertation of several M.Sc. and Ph.D. students at Purdue University. The more applied component of varietal testing and deployment was accomplished jointly with collaborating scientists from the national agricultural research systems in several countries.
Received for publication April 4, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
