| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Crop and Soil Sci., Oregon State Univ., Corvallis, OR 97331-3002, USA
b Pioneer Hi-Bred Int., Avda. Reino Unido s/n, Edificio Aditec 2a Planta, 41012 Sevilla, Spain
c Pioneer Hi-Bred Int., Belediye Binasi Kat 3, Ahmetbey-Kirklareli, Turkey
d Pioneer Hi-Bred Int., 18285 County Road 96, Woodland, CA 95695-9340, USA
* Corresponding author (steven.j.knapp{at}orst.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: AFLP, amplified fragment length polymorphism • BSA, bulked segregant analysis • cM, centimorgan • CMS, cytoplasmic-genic male sterility • LOD, likelihood odds • MAS, marker-assisted selection • PCR, polymerase chain reaction • RAPD, random amplified polymorphic DNA • RIL, recombinant inbred line • RFLP, restriction fragment length polymorphism • SCAR, sequence characterized amplified region • SSR, simple sequence repeat
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
While various genetically simple and complex sources of Orobanche resistance have been described in sunflower (Pustovoit, 1976; Russell, 1981; Krokhin, 1983; Kirichenko et al., 1987; Ramaiah, 1987; Saaverdra del Rio et al., 1994b; Dominguez, 1996a), the most important and widely used are single dominant genes (Burlov and Kostyuk, 1976; Pogorletsky and Geshele, 1976; Vranceanu et al., 1980; Burlov and Artemenko, 1983; Ish-Shalom-Gordon et al., 1993; Sukno et al., 1998, 1999; Lu et al., 2000). The development of Orobanche-resistant inbred lines is complicated by the weedy and noxious characteristics of the parasite, the need for geographic containment, susceptible escapes and other screening variability, the genetic complexity of physiological races of the pathogen, genetic background effects, and genotype x environment interactions; hence, Orobanche resistance is an ideal target for molecular breeding. Despite the complexities underlying Orobanche resistance breeding in sunflower, race-specific dominant genes seem to protect the crop and are ideal sources of resistance for single-cross hybrid breeding because they only need be incorporated into one parent or the other. Moreover, allelic and nonallelic resistance genes can be pyramided by working opposite sides of a hybrid pedigree.
The first Orobanche-resistant sunflowers were developed by introgressing resistance genes from Jerusalem artichoke (H. tuberosus L.) to cultivated sunflower (Vranceanu et al., 1980). The first (Race A) resistant cultivars (Kruglik A-41 and Saratovsky 169) were developed by 1916 (Pustovoit, 1976; Parker and Riches, 1993). Resistance to Race A was overcome by the emergence of the more virulent Race B by 1928 (Pustovoit, 1976; Parker and Riches, 1993). By 1935, open-pollinated cultivars (e.g., Jdanovsky 8281 and 8885) resistant to Race B had been developed. Resistance to Race B was ultimately transferred to Peredovik and VNIIMK1646 (Melero-Vara et al., 1989; Fernandez-Martinez et al., 2000). By the early 1960s, Race A and B resistance genes were defeated by the emergence of Race M, purportedly, a complex of 17 to 22 highly virulent subraces (Petrov, 1968; Melero-Vara et al., 1989).
Vranceanu et al. (1980)(1986), in a classic study, identified five physiological races (A to E) of Orobanche by using five dominant genes (Or1, Or2, Or3, Or4, and Or5) resistant to Race A, A + B, A + B + C, A + B + C + D, and A + B + C + D + E, respectively. The five races were subsequently identified and substantiated in other analyses and geographic regions (Melero-Vara et al., 1989; Bulbul et al., 1991; Saaverdra del Rio et al., 1994a; Shindrova, 1994). Vranceanu et al. (1980)(1986) identified homozygous differentials for Or1, Or2, Or3, Or4, and Or5 for discriminating between the races and resistance genes (Kruglik A-41 for Race A, Jdanovsky 8281 for Race B, Record for Race C, S-1358 for Race D, and P-1380 for Race E). Similar to many of the downy mildew [Plasmopara halstedii (Farl.) Berl. & de Toni in Sacc.] resistance genes described in sunflower (Mouzeyar et al., 1995; Roeckel-Drevet et al., 1996, Vear et al., 1997), new Orobanche resistance genes often confer resistance to earlier races. Several analyses in segregating populations have shown that Or1 to Or5 are either allelic or tightly linked (Vranceanu et al., 1980, Ish-Shalom-Gordon et al., 1993; Sukno et al., 1998, 1999; Fernandez-Martinez et al., 2000).
Because physiological races of broomrape seem to rapidly evolve or are exposed by selection pressure stemming from the broad deployment of individual resistance (R) genes, the search is continually on for new resistant gene specificities (Sackston, 1992; Dominguez et al., 1996; Ruso et al., 1996; Sukno et al., 1998, 1999; Fernandez-Martinez et al., 2000). During the last several years, resistance to Race E (conferred by Or5) has been defeated by the emergence of Race F in Spain. Genes for resistance to the new race have been identified in cultivated and wild sunflowers and seem to confer resistance to earlier races (A through E), as has been the historical pattern (Vranceanu et al., 1980, 1986; Melero-Vara et al., 1989; Dominguez, 1996a; Ruso et al., 1996; Gagne et al., 1998; Sukno et al., 1998, 1999; Fernandez-Martinez et al., 2000).
The development of inbred lines resistant to Orobanche can be accelerated by marker-assisted selection (MAS), for example, by identifying and culling heterozygotes and susceptible escapes (phenotyping errors) and performing genotypic selection on seedlings and other preanthesis growth stages. Or5 has not been placed on the SSR map of sunflower, and few DNA markers are presently found in the region surrounding Or5. Lu et al. (2000) used bulked segregant analysis (BSA) (Michelmore et al., 1991) to identify a random amplified polymorphic DNA (RAPD) marker (UBC120_660) and five DNA sequence characterized amplified region (SCAR) markers (RTS05, RTS28, RTS40, RTS29, and RTS41) linked to Or5. The RAPD and SCAR markers and Or5 were mapped in two segregating populations (Lu et al., 2000) and the SCAR markers were placed on LG17 of the CARTISOL restriction fragment length polymorphism (RFLP) map (Lu et al., 1999). The closest SCAR marker mapped 5.6 cM distal to Or5, and no DNA markers other than the RAPD (22.5 cM upstream) flanked Or5.
The goal of the present study was to identify SSR markers tightly linked to Or5 and, in the process, to place the Or5 locus on the public molecular genetic linkage map of sunflower (Tang et al., 2002; Yu et al., 2003). Secondarily, we tested the utility and sensitivity of fluorescent PCR-multiplex SSR genotyping (Tang et al., 2003) as a tool for screening bulked-segregant DNA samples (Michelmore et al., 1991) in sunflower.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Orobanche Resistance Phenotyping
The parents and RILs (F5 seedlings) were screened for resistance to broomrape Race E in a growth chamber at Pioneer Hi-Bred Agroservicios Spain S.L., Sevilla, Spain, in 2000 using seeds of the Race E broomrape population collected from Ecija, Sevilla, Spain, in the summer of 1998. Broomrape seeds were collected from plants infesting the hybrid Florasol (resistant to Race D, but susceptible to Race E) in a nursery where several hundred rows of Race E-resistant hybrids were grown and none were infected. The broomrape seeds were homogeneously mixed with a 1:1 mixture of sand and peat at the rate of 250 mg per kg. PHC, PHD, two susceptible controls (Coronil and Florasol), and the RILs were screened for resistance to Orobanche. Five seeds of each entry were planted in 6- x 10- x 10-cm plastic pots filled with the infested soil mixture. The plants were grown under a 14-h photoperiod using 25°C day and 18°C night temperatures and constant 
60% humidity. Two-month-old plants were carefully removed from the pots to phenotype for the presence or absence of emerged or underground broomrape stalks (nodules). RILs with no infected plants were scored as resistant (R), RILs with 100% infected plants were scored as susceptible (S), and RILs with a mixture of infected and uninfected plants were scored as segregating (H).
Bulked Segregant Analysis and SSR Genotyping
Bulked segregant analysis (Michelmore et al., 1991) was performed by screening 78 SSR marker loci amplified by the PCR-multiplexes (13 six-plexes) described by Tang et al. (2003). Resistant (R) and susceptible (S) bulks were produced by pooling equal quantities of DNA from 10 putatively homozygous resistant and 10 putatively homozygous susceptible RILs, respectively. Two independent R and S bulks were produced and screened. The 78 SSR markers were screened for polymorphisms between PHC and PHD and between the replicate bulked DNA samples using the PCR-multiplexes and genotyping methods described by Tang et al. (2003). The forward primers in each six-plex were labeled with different combinations of fluorophores (6FAM, HEX, TET, or NED) to facilitate multiplex genotyping. PHC, PHD, and the bulks were subsequently screened for polymorphisms using 11 additional SSR markers from the upper end of LG3 (CRT392, CRT314, ORS1040, ORS1112, ORS683, ORS372, ORS820, CRT197, ORS777, ORS657, and ORS1021). The RILs were genotyped for 13 SSR markers found to be polymorphic between PHC and PHD and between R and S bulks. The genotyping assays were performed using post-PCR multiplexing.
Genetic Analyses and Map Construction
The expected segregation ratio for Or5 among F5 RILs was 0.4375 homozygous-resistant (Or5Or5) to 0.125 segregating (3 Or5__:1 or5or5) to 0.4375 homozygous susceptible (or5or5). The fit of the observed ratio of Orobanche resistance phenotypes to the expected ratio of Or5 genotypes was checked using 
2 statistics. The RIL mapping function of MAPMAKER (Lander et al., 1987) was used to construct a genetic linkage map for LG3 among the PHC x PHD RILs. Loci were grouped using a likelihood odds (LOD) threshold of 10.0 and map distances (cM) were calculated using the Kosambi (1944) mapping function.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2 = 54.826, P < 0.001). We observed a deficiency of susceptible (-60.625) and excesses of segregating (+37.25) and resistant (+23.375) RILs. The segregation distortion undoubtedly arose from misclassifying homozygous susceptible RILs (or5or5) with escapes (resistant plants) as segregating and segregating RILs (3 Or5__: or5or5) with escapes (no susceptible plants) as resistant.
Bulked Segregant Analysis
Two replicate samples of 10 susceptible and 10 resistant RILs were selected for producing and screening bulked-segregant DNA samples. Forty-three out of 78 SSR markers amplified by the PCR-multiplexes (Tang et al., 2003) were polymorphic between PHC and PHD. When screened on replicate R and S bulks, three SSR markers from LG3 (ORS1222 in Set 1, ORS1036 in Set 11, and ORS1114 in Set 13) were polymorphic (Fig. 1), suggesting that Or5 might reside on LG3. None of the SSR markers from other linkage groups were polymorphic between R and S bulks. The other two SSR marker loci on LG3 amplified by the PCR multiplexes (ORS665 in Set 4 and ORS949 in Set 7 of the PCR multiplexes) were monomorphic between PHC and PHD.
|
|
|
The locus orders for SSR markers on the PHC x PHD RIL and reference genetic linkage maps were identical (Fig. 3). Or5 mapped distal to the SSR marker loci and was the upper terminus of LG3 (Fig. 3), as demarcated by SSR marker loci on the genetic linkage map (Burke et al., 2002; Tang et al., 2002; Yu et al., 2003). The closest proprietary SSR marker (CRT392) was 6.2 cM downstream of Or5, while the closest public SSR marker (ORS1036) was 7.5 cM downstream of the Or5 locus.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
700 mapped SSR marker loci (Burke et al., 2002; Tang et al., 2002; Yu et al., 2003) and, through comparisons to other maps, 900 mapped RFLP marker loci (Berry et al., 1995, 1996, 1997; Gentzbittel et al., 1995, 1999; Gedil et al., 2001). Of >1600 RFLP and SSR marker loci mapped in sunflower thus far, only three are within 6 to 10 cM of the Or5 locus. The map distances between Or5 and the DNA markers reported here (Fig. 3) and elsewhere (Lu et al., 1999, 2000) could be upwardly biased by phenotyping errors because susceptible escapes misclassified as resistant introduce spurious recombinants and inflate map distances. Of the two SCAR markers closest to Or5 (Lu et al., 2000), RTS05 failed to amplify bands from PHC and PHD, whereas RTS28 amplified monomorphic bands from PHC and PHD (data not shown). RTS05 was 5.6 and RTS28 was 13.6 cM downstream of Or5 on the map of Lu et al. (2000), while the closest RFLP marker was 7.1 cM downstream of Or5 (Gentzbittel et al., 1995, 1999; Lu et al., 1999). None of the SCAR or RFLP markers mapped upstream of Or5; however, using a LOD threshold of 1.4, Lu et al. (2000) placed a RAPD marker locus (UBC120_660) 22.5 cM upstream of Or5 on LG3. Because of the low LOD threshold (high probability of Type I error), the placement of UCB120_660 is tenuous and must be rechecked. UBC120_660 RAPD primers failed to amplify alleles from PHC and PHD; thus, we could not substantiate or refute the location of UBC120_660 upstream of Or5. If the reported position of the RAPD marker locus is correct (Lu et al., 2000), then the RFLP (Berry et al., 1995, 1996, 1997; Gentzbittel et al., 1995, 1999; Jan et al., 1998; Gedil et al., 2001) and SSR (Burke et al., 2002; Tang et al., 2002; Yu et al., 2003) maps of sunflower are missing 20 or more cM on the upper end of LG3 (Fig. 3), and none of the >100 SSR and RFLP markers mapped to LG3 reside in the interval between Or5 and UBC120_660.
The goal of identifying flanking DNA markers tightly linked to Or5 has not been met, partly because the Or5 locus resides in a telomeric or near-telomeric region of apparently high recombination. Because none of the DNA markers described thus far are tightly linked to or flank Or5, MAS is presently limited to the centromeric side of the Or5 locus, and Or5 genotypes cannot be unequivocally identified from SCAR or SSR marker genotypes. However, less than one in 12 individuals selected for ORS1036 SSR marker genotypes should be recombinant for Or5. The location of Or5 in a region of apparently high recombination is advantageous for breeding and potentially advantageous for map-based cloning (Zhang et al., 1994; Tanksley et al., 1995; Qi and Gill, 2001), and has almost certainly played an important role in the recovery of Or5 recombinants in H. tuberosus (hexaploid) x H. annuus (diploid) and other interspecific segregating populations (Vranceanu et al., 1980; Ruso et al., 1996; Sukno et al., 1999). The scarcity of DNA markers near Or5 could stem from a scarcity of DNA polymorphisms; however, because Or5 and other Orobanche resistance genes have been introgressed from wild sunflowers, a scarcity of DNA polymorphisms in the dragged DNA segments seems improbable. Quite the contrary, DNA sequences flanking Orobanche resistance genes from wild sunflowers should be extraordinarily polymorphic when compared with DNA sequences commonly found in elite inbred lines of cultivated sunflower (Tang and Knapp, 2003).
The SSR markers described here complement the RAPD and SCAR markers developed by Lu et al. (2000) for MAS of broomrape resistance genes in sunflowers. Two proprietary (CRT392 and CRT314) and two public (ORS1036 and ORS1040) SSR markers are located within 6.2 to 11.2 cM of Or5. The heterozygosities of the four SSR markers among elite inbred lines fall in the range of 0.41 (ORS1036) to 0.73 (CRT392) (Tang et al., 2003; unpublished data) and supply much needed molecular marker diversity and polymorphic DNA sequences in the Or5 region. DNA sequences for the SSR markers have been deposited in public databases (Tang et al., 2002; Yu et al., 2002). ORS1036 was mapped for the first time in PHC x PHD, whereas the other SSR markers had been mapped in other populations (Tang et al., 2002; Yu et al., 2003). ORS1036, the closest public SSR marker to Or5, amplified 245- and 255-base-pair-long alleles from PHC and PHD, respectively, an allele length difference long enough for genotyping on agarose.
ORS1036 and CRT314 seem to be universally codominant, whereas CRT392 and ORS1040 are mixed dominant-codominant (Tang et al., 2003; unpublished data). CRT392 and ORS1040 were dominant in PHC x PHD. Similar to many other SSR markers in sunflower (Tang et al., 2002, 2003), CRT392 and ORS1040 produce three or more alleles (one being null) among diverse germplasm accessions and, consequently, are dominant in some crosses and codominant in others. Presumably, the null alleles (PCR amplification failures) were caused by single nucleotide polymorphisms in DNA sequences targeted by the SSR primers (Mogg et al., 2002).
On the basis of the distribution of gene-rich regions and ratios of physical-to-genetic distance in centromeric and telomeric regions in other taxa (Ganal et al., 1989; Segal et al., 1992; Zhang et al., 1994; Tanksley et al., 1995; Umehara et al., 1995; Gill et al., 1996a,b; Künzel et al., 2000; Qi and Gill, 2001), we speculate that the region flanked by Or5 and the next closest DNA markers (RTS05, CRT392, and ORS1036) may be physically shorter than randomly selected and centromeric regions in the sunflower genome. The density of SSR marker loci was lowest in the distalmost and highest in the centermost regions of the genetic linkage map for LG3 (Tang et al., 2002; Yu et al., 2003) (Fig. 3), the classic pattern for a metacentric chromosome (Gill et al., 1996a, b; Künzel et al., 2000). The density of SSR markers was 2.54 cM per locus in the upper 17.8 cM (the region distal to ORS1112-3), 4.78 cM per locus in the lower 33.5 cM (the region distal to ORS149), and 1.05 cM per locus in the centermost 35.8 cM region between ORS1112-3 and ORS149 on LG3 (Fig. 3). DNA markers densities are typically lower in telomeric than centromeric regions, recombination tends to be greater in gene-rich than gene-poor regions, and ratios of physical-to-genetic distance tend to be less in telomeric than centromeric regions because of suppressed recombination in the latter (Ganal et al., 1989; Brown and Sundaresan, 1991; Segal et al., 1992; Zhang et al., 1994; Tanksley et al., 1995; Pedersen et al., 1995; Gill et al., 1996a,b; Schnable et al., 1998; Künzel et al., 2000; Sandhu et al., 2001; Qi and Gill, 2001).
The Utility and Sensitivity of PCR-Multiplexes for Identifying SSR Polymorphisms in Mixed DNA Samples
The present study was our initial test of the utility and sensitivity of the PCR-multiplexed SSR markers (Tang et al., 2003) for bulked-segregant analysis in sunflower. The PCR-multiplexes facilitated a rapid scan of the sunflower genome for SSR markers linked to Or5. The analysis was completed by performing 78 PCRs (six DNA samples x 13 six-plexes), but could have been completed by performing as few as 26 PCRs (screening one set of R and S bulks) (Fig. 1). Historically, bulked segregant analyses have been performed, out of necessity, by screening randomly selected, unmapped RAPD or amplified fragment length polymorphism (AFLP) markers (Mouzeyar et al., 1995; Brahm et al., 2000; Lu et al., 2000). With the emergence of dense SSR genetic linkage maps for sunflower (Tang et al., 2002; Yu et al., 2003), BSA can be systematically performed by screening strategically positioned SSR marker loci, instead of randomly selected RAPD or AFLP markers. Moreover, SSR markers immediately supply DNA sequence-tagged-sites for subsequent analyses and DNA marker development. While the multiplex ratios of individually typed SSR markers are less than RAPD and AFLP markers (Powell et al., 1996), they can be greatly increased by pre- and post-PCR multiplexing (Tang et al., 2003). Typically, the multiplex ratios for multiplexed SSR markers range from six to 14, and are on par with or greater than the multiplex ratios for RAPDs, but less than the multiplex ratios for AFLPs (Powell et al., 1996).
We screened independent replicates of the R and S bulks to check the sensitivity of the PCR- multiplexes for identifying SSR marker polymorphisms in bulked DNA samples. The mapped markers (a mixture of dominant and codominant SSR markers) permitted a retrospective analysis of the sensitivity of fluorescent SSR marker assays for BSA. The codominant SSR markers were sensitive to allele dosage differences. When recombinants were present in bulks screened with codominant SSR markers, differences in DNA template concentrations (allele dosages) between bulks yielded progressively different band intensities as a function of the number of recombinants between Or5 and the SSR marker loci (Fig. 2, 3). Dominant SSR polymorphisms can only be observed between R and S bulks when recombinants are absent in both bulks or when recombinants are absent in null allele bulks only, for example, when the susceptible parent is homozygous for the null allele and the susceptible bulk lacks recombinants or when the resistant parent is homozygous for the null allele and the resistant bulk lacks recombinants. ORS1112-3 and ORS1040-3 fell in this category (Fig. 3). Conversely, when the susceptible parent is homozygous for the dominant allele and recombinants are present in the resistant bulk or when the resistant parent is homozygous for the dominant allele and recombinants are present in the susceptible bulk, then bands of equal intensity are amplified from both bulks and the dominant SSR polymorphism is not observed between R and S bulks. ORS820 fell in this category and, as previously noted, was polymorphic between PHC and PHD but not between R and S bulks.
The number of recombinants between Or5 and the SSR marker loci in bulks was ascertained from the SSR genotypes of CRT392, ORS1036, CRT314, ORS1040, and ORS1222 among the RILs pooled to create the bulks. CRT392, ORS1036, CRT314, and ORS1040 mapped 6.2 to 11.2 cM downstream of Or5, had up to three recombinant alleles out of 20 alleles per bulk, and sharply discriminated between R and S bulks (Fig. 1, 2). ORS1036 and ORS1222 are codominant SSR markers from opposite ends of the region on LG3 mapped in PHC x PHD (Fig. 3). One out of 20 ORS1036 (7.5 cM) alleles were recombinant, while 7 out of 20 ORS1222 (29.5 cM) alleles were recombinant in the bulks. Naturally, the intensity of the ORS1036 polymorphisms were greater than the ORS1222 polymorphisms between R and S bulks, but both were clearly polymorphic. ORS1222 demonstrated that codominant SSR markers 30 cM away could be identified by allele intensity differences. The 0- to 30-cM sensitivity range for SSR markers is similar to that found for RAPD markers (Michelmore et al., 1991). Occasionally, of course, DNA marker loci separated from the target locus by distances greater than 30 cM can be identified by BSA, as was found for ORS1114 (74.3 cM from CRT392) in the present study (Fig. 1 and 3) and by Mouzeyar et al. (1995) for a RAPD marker 43.7 cM downstream of the Pl1 locus in sunflower.
The 78 SSR marker loci amplified by the PCR-multiplexes are highly polymorphic, dispersed throughout the sunflower genome, and estimated to be within 6.4 cM of any locus in the genome, and hence should be routinely powerful for identifying SSR marker loci linked to phenotypic loci. Candidates SSR marker loci can be identified by screening the PCR-multiplexed SSR markers for some phenotypic loci, as was the case for Or5, but not for others, depending on polymorphisms in the segregating population. We have identified another 222 highly polymorphic single-locus SSR markers for BSA and NIL screening for cases where leads are not produced by screening the SSR marker loci amplified by the PCR multiplexes (Tang et al., 2003). The density, distribution, and polymorphisms of the 300 single-locus SSR markers should be sufficient for tracking down polymorphic SSRs linked to virtually any unmapped phenotypic locus in sunflower.
|
ACKNOWLEDGMENTS |
|---|
Received for publication September 12, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
