Published online 23 February 2005
Published in Crop Sci 45:645-652 (2005)
© 2005 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
Analysis of Stalk Rot Resistance and Genetic Diversity among Drought Tolerant Sorghum Genotypes
Tesfaye T. Tessoa,
Larry E. Claflinb and
Mitchell R. Tuinstraa,*
a Dep. of Agronomy, Kansas State Univ., Manhattan, KS 66506
b Dep. of Plant Pathology, Kansas State Univ., Manhattan, KS 66506
* Corresponding author (drmitch{at}ksu.edu)
 |
ABSTRACT
|
|---|
Forty-five drought tolerant sorghum [Sorghum bicolor (L.) Moench] genotypes were screened for stalk rot resistance by artificial inoculation with Fusarium proliferatum in field trials near Manhattan, KS, in 2000. Four lines, SC134, SC1154, SC1039, and SC564, were identified with lesions that were smaller than the standard resistant check, SC599. These lines and four other checks were intercrossed in a Design II mating scheme with ‘AWheatland’ and ‘ARedlan’ to produce hybrids for further testing. The inbred parent lines and corresponding hybrids were evaluated for resistance to four major stalk rot pathogens, F. proliferatum, F. thapsinum, F. andiyazi, and Macrophomina phaseolina, using randomized complete block designs in field trials near Manhattan and Hesston, KS, in 2001. Significant differences in lesion length were detected among the inbred parent lines and their corresponding hybrids. Comparisons among entries indicated that crosses of SC701 and SC564 were susceptible and produced very large lesions following inoculation with each of the stalk rot pathogens evaluated in this study. Only small lesions were produced in SC599 and its hybrids following inoculation with either Fusarium species or M. phaseolina. SC134 also exhibited high levels of resistance to Fusarium species and SC35 to M. phaseolina. Comparisons among pathogens showed that inoculations with M. phaseolina produced larger lesions than the Fusarium species in most sorghum genotypes. The most resistant accessions in these studies were all from East Africa; therefore, an analysis of genetic diversity was conducted to evaluate the pattern of genetic relationships among the full set of drought tolerant lines. DNA fingerprinting followed by cluster analyses grouped the entries into five distinct clusters, mainly according to geographical origin. Three of the resistant lines grouped together in a cluster that contained mostly Ethiopian landraces while SC599 grouped in another cluster.
Abbreviations: COI, co-occurrence index • MCA, multiple correspondence analysis • PBS, phosphate buffered saline • PCR, polymerase chain reaction • PDA, potato dextrose agar
|
INTRODUCTION
|
|---|
STALK ROT DISEASES in grain sorghum are caused by several fungi, and the incidence is more pronounced under drought and heat stress conditions (Seetharama et al., 1987). The disease reduces yield and quality of the grain by exposing the plants to excessive lodging; the magnitude of the impact being dependent on the severity of the disease (Mughogho and Pande, 1984). Charcoal rot [caused by Macrophomina phaseolina (Tassi) Goidanich] incidence is much higher when the plants are exposed to prolonged drought and high temperature stress during grain development (Edmunds, 1964). Fusarium stalk rot (caused by Fusarium spp.) is typically more severe when drought and high temperature stress occurs during grain development followed by wet, cool conditions near physiological maturity (Zummo, 1980). These observations suggest that postflowering drought tolerant genotypes not affected by late-season drought may have reduced susceptibility to stalk rot diseases. This hypothesis is bolstered by the fact that many lodging resistant sorghum genotypes have also been found to possess one or more postflowering drought tolerance characteristics such as the staygreen trait (Tenkouano et al., 1993).
Although the predisposing factors generally are similar, the reaction of genotypes to stalk rot diseases may vary depending on the causal organism. Most genetic sources of stalk rot resistance were initially identified in areas where charcoal rot is the predominant stalk rot disease. Consequently, one or more of these genotypes may fail to prove resistant to Fusarium stalk rot. In preliminary studies conducted at multiple environments in Kansas, charcoal rot resistant lines and their hybrids did not express high levels of resistance when inoculated with F. proliferatum. The Kansas Agricultural Experiment Station and Cooperative Extension Service reported similar observations. These results highlight the need to identify germplasm with superior Fusarium stalk rot resistance for specific uses in locations such as Kansas where Fusarium species are the predominant causal organisms of stalk rot (Jardine and Leslie, 1992). The objectives of this study were to (i) characterize selected drought tolerant lines for Fusarium stalk rot resistance by artificial inoculation, (ii) evaluate the performance of superior lines against multiple stalk rot pathogens and investigate the mode of inheritance; and (iii) analyze the extent and pattern of genetic relationships among sorghum genotypes with contrasting stalk rot reactions.
|
MATERIALS AND METHODS
|
|---|
Evaluation of Sorghum Genotypes for Stalk Rot Resistance
Forty drought tolerant sorghum genotypes representing major botanical races of sorghum and geographical centers of origin and five improved lines (three females and two standard pollinators) were selected to evaluate differences in stalk rot resistance and relationships with patterns of genetic diversity. The origin and race classification of the entries and previously reported drought tolerance characteristics are presented in Table 1. In the summer of 2000, these entries were evaluated for their resistance to stalk rot caused by F. proliferatum in a replicated field trial at Manhattan, KS. The trial was conducted on Smolan Silty Clay Loam soils using a randomized complete block design with four replications. The seeds were planted in single-row plots (6.5 x 0.76 m) at a population of 129000 plants ha–1. Seeds were treated with N-trichloromethylthio-4-cyclohexene-1,2-dicarboximide (Captan, Bonide Products, Inc., 6301 Sutliff Road, Oriskany, NY) at the label rate. Cultural practices for field experiments included application of 41 kg N ha–1 before planting and 0.24 L quinclorac ha–1 plus 0.68 kg atrazine ha–1 applied postemergence for weed control.
View this table:
[in this window]
[in a new window]
|
Table 1. Mean lesion lengths of sorghum genotypes having diverse origin, race, and drought tolerance characteristics following basal stalk inoculation with Fusarium proliferatum at Manhattan, KS.
|
|
Plants were injected with F. proliferatum (strain KSLM), a strain initially cultured from diseased sorghum stalks collected in a grain production field in KS. The fungus was grown on potato dextrose agar (PDA) at 24°C. After 3 to 4 d, small sections (2–3 mm2) of the fungal mat were placed in 500-mL Erlenmeyer flasks containing potato dextrose broth (DIFCO, Detroit, MI). Inoculum was grown on a shaker (60 rpm) for approximately 2 d at 24°C to produce conidia. Conidia were separated from the mycelial mass by straining the culture through four layers of cheesecloth. Conidial concentrations were ascertained with a hemacytometer. Spore concentrations were adjusted to 5 x 104 conidia mL–1 using 10 mM (pH 7.2) phosphate buffered saline (PBS). The adjusted suspension was maintained on ice until inoculation. At flowering, three plants in each plot were randomly selected and tagged for inoculation. Two weeks after flowering, tagged plants were inoculated with approximately 1 mL of the conidial suspension in the pith of the basal stalk using an Idico filler-plug gun (Forestry Suppliers, Inc., Jackson, MS) equipped with a stainless steel needle similar to that described by Toman and White (1993). Twenty-eight days after inoculation, plants were harvested and longitudinally split to measure the length of the discolored lesion in the pith of the stalk as described by Tesso et al. (2004). Differences in disease reaction were evaluated using the PROC MIXED procedure of SAS (SAS Institute, 1989).
Four lines that showed acceptable levels of resistance (SC1039, SC134, SC1154, SC564) in 2000 were selected and crossed to AWheatland and ARedlan using a Design-II mating scheme (Hallauer and Miranda, 1988). SC599 (resistant to Fusarium stalk rot), SC35 (resistant to charcoal rot), SC701 (susceptible to stalk rot), and the elite U.S. pollinator lines Tx436 and Tx2737 were also crossed to AWheatland and ARedlan to produce hybrid checks for comparison. The 11 inbred parent lines and 18 crosses were evaluated for stalk rot resistance in Manhattan and Hesston, KS, in 2001. The trials in Hesston were conducted on Ladysmith (fine, smectitic, mesic Udertic Argiustolls) silty clay loam soils. The experiments were conducted in RCB designs with four replications using cultural management practices similar to those described for experiments in 2000. The 29 genotypes evaluated in each experiment were inoculated with conidia of F. proliferatum (D00720), F. thapsinum (F04092), and F. andiyazi (X11152) (Courtesy of Dr. J.F. Leslie, Kansas State University). The M. phaseolina strain was isolated from diseased sorghum stalks collected in Corpus Christi, TX, by Dr. Gary Odvody, Texas A&M University. Liquid inocula for Fusarium isolates were prepared following the procedure described for experiments conducted in 2000. The M. phaseolina inoculum was grown on PDA. Macrophomina phaseolina fails to produce large numbers of conidia or microconidia in culture; therefore, hyphal fragments were used to inoculate plants. The mycelial suspension was fragmented for 10 min with an Osterizer blender. The suspension was strained through four layers of cheesecloth and then collected in a sterile beaker. The number of fragments capable of inciting infection (infective propagules) was determined by plating 200 µL of serial dilution suspensions on PDA. The number of colony forming units was determined by counting colonies after 2 d of incubation at 30°C. These studies indicated that most hyphal fragments were viable and suspensions could be estimated using a hemacytometer. Inoculum concentrations were adjusted to 5 x 104 hyphal fragments mL–1 in PBS. The adjusted suspension was maintained on ice until inoculation.
Ten plants of uniform maturity were tagged in each plot at flowering using five different colored flagging tapes (two plants for each color in each plot). Two weeks after flowering, the tagged plants were inoculated with liquid suspensions of the respective pathogens. Two plants were inoculated with each of the four pathogens and the remaining two plants were inoculated with sterile, distilled water as a control. Inoculation was performed by injecting approximately 1 mL of the pathogen suspension into the pith of the basal stalk 10 cm above the soil surface using the Idico filler-plug gun as described above. Twenty-eight days after inoculation, plants were harvested and longitudinally split to measure the length of the discolored lesion in the pith of the stalk.
Analyses of variance and combining ability were performed per established methods (Hallauer and Miranda, 1988) using PROC GLM and PROC MIXED procedures of SAS (SAS Institute, 1989). In the combined analysis, locations, blocks, and location interactions were considered random effects while the remaining variables were considered fixed effects. Tests of significance for entry, inbred, inbred vs. hybrid, hybrid, male parent, female parent, and male x female effects for all traits were made by comparison with their respective environmental interaction.
DNA Fingerprinting and Analysis of Genetic Diversity
Seeds of the 45 lines listed in Table 1 were germinated on moistened paper towels and coleoptile tissues were harvested after 5 d. Samples were immediately frozen in liquid nitrogen and stored at –20°C until DNA extraction. DNA was extracted from samples as described by Saghai-Maroof et al. (1984). DNA concentrations were adjusted to 20 µg mL–1 for polymerase chain reactions (PCRs). Seventeen SSR markers described by Brown et al. (1996), 11 by Taramino et al. (1997), and four by Bhattramakki et al. (2000) were used to fingerprint the 45 lines. The PCR reactions were performed using a total reaction volume of 40 µL with the reaction mixture containing 30 ng genomic DNA, 5 pmole of each primer, 0.2 nM of each deoxynucleotide triphosphate, 0.75 units Taq polymerase, 2.5 mM MgCl2, and 1 mM Cresol red, with reaction conditions as described by Taramino and Tingey (1996). The PCR products were differentiated by electrophoresis using 3% Metaphor agarose (O,O-dimethyl O-p-nitrophenyl phosphorothioate, FMC Corporation, Philadelphia, PA) at 5 V/cm in TBE (1,1,2,2-tetrabromoethane) buffer for 6 h. Gels were stained with 0.015% ethidium bromide for 1 h and the bands were visualized using a Gel Doc 2000 (Bio-Rad Laboratories, Milan, Italy) documentation system under UV light. Quantity One software (Bio-Rad Laboratories, 1998) was used for scanning gel images and scoring the bands.
Markers were evaluated for polymorphism, and nonpolymorphic markers were excluded from the analysis. Bands with more than 20 base pair differences were scored as different alleles. For statistical analyses, alleles for each marker were assigned independent numerical values with the smallest fragment being given a value of 1 and the next smallest fragment assigned a value 2 until each allele was similarly defined. Genetic diversity among entries was computed as [1-GS] using SAS, where GS is genetic similarity between lines. Genetic similarity was calculated using Jaccard's similarity coefficient based on Rohlf (1993). Multiple correspondence analysis (MCA) was conducted using WARD's method on the distance matrix generated using Jaccard's method. The MCA plot was produced using the JMP option in SAS to reveal the first three principle components defining genetic distance among the genotypes. Resampling analyses with 1000 iterations were conducted to calculate a co-occurrence index (COI) to determine how often each pair of genotypes within the original clusters was grouped together during the resampling process.
|
RESULTS
|
|---|
Germplasm Characterization
The initial characterization of drought tolerant genotypes indicated significant differences among entries for lesion length (Table 1). The most susceptible genotypes had disease lesions that were 10 to 15 fold longer than the most resistant lines. The highest infection was recorded in SU439 with 63-cm mean lesion length. Limited infections were noted in SC1154 with a mean lesion length of 4 cm. SC599, an established source of genetic resistance, had a mean lesion length of 7 cm. SC1154, SC1039, and SC134 from Ethiopia and SC564 from Uganda had shorter lesion lengths than SC599 (Table 1).
Reaction of Selected Genotypes to Various Stalk Rot Pathogens
SC1154, SC1039, and SC134 from Ethiopia and SC564 from Uganda, along with five checks and their crosses with AWheatland and ARedlan, were tested for resistance to several different stalk rot pathogens. In a combined ANOVA, mean squares for location, pathogen, and location x pathogen interactions were highly significant (Table 2). A comparison among locations indicated that the mean lesion lengths at Manhattan were approximately twice those at Hesston (Table 3). The mean lesion lengths produced by M. phaseolina was greater than those produced by the Fusarium species tested in this study. No significant differences in lesion length were detected among the Fusarium species inoculated on different sorghum lines and hybrids (Tables 3). Lesions also were noted in the control treatment; however, these lesions were significantly shorter than those resulting from the pathogen treatments (Table 3). These lesions probably were caused by infections resulting from entry of soil- or air-borne fungal or bacterial pathogens through the wound site. Some of these small lesions could also reflect the discoloration resulting from the normal sorghum wound response with reddening of the wounded tissue.
View this table:
[in this window]
[in a new window]
|
Table 2. Mean squares from the combined analysis of variance for reaction of sorghum entries to stalk rot pathogens at Manhattan and Hesston, KS, in 2001.
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Mean lesion length produced by different pathogens averaged across sorghum genotypes at Manhattan and Hesston, KS, in 2001.
|
|
Significant differences among sorghum entries were also observed in the combined analysis (Table 2). Partitioning of the entry effect into its inbred and hybrid components indicated significant differences in mean lesion length among the inbreds and hybrids (Tables 2). Among inbreds, the largest lesions were noted in SC701 and Redlan, and the smallest lesions were noted in SC1154 and SC134 (Table 4). Among hybrids, crosses of SC599 (resistant check) consistently had the shortest lesions across locations and pathogen groups while crosses of SC701 and SC564 generally had the longest lesions (Table 4). Crosses of SC1154 and SC134 also produced short lesions for all pathogen groups. Crosses of Tx2737 had short to intermediate lesion lengths following inoculation with Fusarium spp., but the lesions were much longer following inoculation with M. phaseolina. SC35 had lesion lengths similar to Tx2737 when inoculated with Fusarium spp., but the lesions were smaller when inoculated with M. phaseolina (Table 4).
View this table:
[in this window]
[in a new window]
|
Table 4. Mean lesion lengths for sorghum lines and hybrids inoculated with various stalk rot pathogens at Manhattan and Hesston, KS, in 2001.
|
|
Further partitioning of the hybrid effect into male and female components revealed that the largest proportion of variability among hybrids was contributed by the male parent (Table 2). Crosses of SC599 consistently had the smallest lesions for all pathogens, followed by SC134 (Table 5). SC35 produced hybrids that exhibited comparatively short lesions following inoculation with M. phaseolina, but produced longer lesions when inoculated with some Fusarium species (Table 5). Some lines that expressed high levels of resistance as inbreds, such as SC564, were found to be highly susceptible in hybrid combinations and developed long lesions following inoculation with different pathogen species (Table 5).
View this table:
[in this window]
[in a new window]
|
Table 5. Mean lesion lengths for hybrids inoculated with Fusarium spp. and Macrophomena phaseolina stalk rot pathogens at Manhattan and Hesston, KS, in 2001.
|
|
Although entry, location, and pathogen interaction effects were highly significant (Table 2), genotype rankings were similar across environments (Table 4). Most of the interaction effects appeared to have resulted from change of scale as compared with change of rank effects. Correlation analyses comparing entry performance in Manhattan and Hesston indicated r = 0.89 for F. proliferatum, r = 0.91 for F. thapsinum, r = 0.84 for F. andiyazi, and r = 0.68 for M. phaseolina. The pattern of disease reactions for entries was generally consistent across pathogen species except M. phaseolina, where some entries showed distinct reactions. The lines SC599 and SC35 had shorter mean lesion lengths for M. phaseolina than the Fusarium species, while the rest of the entries were generally more susceptible to M. phaseolina than Fusarium species (Table 4).
Analysis of Genetic Diversity
A total of 32 SSR markers were evaluated to study the pattern of genetic relationships among the 45 sorghum genotypes. The nine primer pairs that failed to produce visually detectable polymorphisms across genotypes were excluded from the analysis. Profiles for the remaining 21 SSR loci were used to discriminate and uniquely classify all pairs of genotypes studied. A total of 77 polymorphic alleles were produced from the 21 marker pairs. The number of alleles per locus ranged from two for markers Sb4-15, Sb4-22, Sb4-32, and Sb5-236 to five for Sb1-10, Sb4-32, Sb6-325, SbAGE01, SbAGF08, and SbAGH04. The mean number of alleles per locus was 3.7, with allelic sizes in close agreement with reports in the literature (Brown et al., 1996; Taramino et al., 1997; Bhattramakki et al., 2000).
Genetic diversity analyses performed on the raw data matrix revealed a mean overall distance of 0.56 among all genotypes. The mean distance of each genotype from all other genotypes ranged from 0.49 for SU629 to 0.61 for Tx435. The distance between pairs of genotypes ranged from 0.11 for the two closest Sudanese lines, SU629 and P898012, to 0.86 between Tx436 and SC56. Principal component analysis sorted the distance data matrix into 21 principal components, where the first three principal components accounted for 44% of the total variability. Multiple correspondence analyses grouped the genotypes into five distinct clusters (Table 6). Resampling analyses (1000 iterations with replacement) indicated a COI of 0.50 to 0.99, meaning that every pair of genotypes within the original cluster grouped together in 50 to 99% of the time throughout the resampling iterations. Further characterization of the clusters using the TRIM option set to 10% removed five genotypes including SC56 and Tx436, the genotypes that were most distant from each other. However, this result neither changed the number of clusters nor caused reshuffling to members of the clusters. Within cluster distance for all groups was smaller than any pair of between cluster distances and overall mean distance.
Fourteen, ten, six, ten, and five genotypes were grouped into Clusters 1, 2, 3, 4, and 5, respectively (Table 6). The mean genetic distance within clusters ranged from 0.41 to 0.47 and between clusters from 0.55 to 0.64. The lowest within cluster distance was found in Cluster 2 that only included Ethiopian lines and highest in Cluster 3 that comprised genotypes from three different geographical regions. Similarly, the smallest between cluster distance was between Clusters 1 and 4 and 1 and 5, and the highest distance between Clusters 3 and 5. These relationships are graphically presented in Fig. 1
.
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 1. Multiple correspondence analysis indicating the first three principal components in the analysis of genetic distance between sorghum genotypes represented within five distinct clusters.
|
|
|
DISCUSSION
|
|---|
Analysis of Stalk Rot Resistance
Significant differences in mean lesion lengths among pathogen groups may indicate differences in their ability to cause disease. Inoculation with M. phaseolina resulted in the highest overall mean lesion lengths at both test locations for almost all entries. At Manhattan, certain susceptible entries died within 28 d after inoculation with M. phaseolina, while others were nearly dead when data was recorded. However, there were no significant differences among Fusarium species, and none caused complete death to any of the entries. This generally agrees with reports that M. phaseolina is the most aggressive stalk rot pathogen (Anahosur and Patil, 1982; Rosenow, 1984; Mughogho and Pande, 1984). Among entries tested, only SC35 and SC599 were found to have shorter mean lesion lengths after inoculation with M. phaseolina compared with inoculations by Fusarium species. Both in this and previous studies, SC35 was listed with average to moderately susceptible entries for Fusarium stalk rot (Tesso et al., 2004). However, it had lesions as small as the established resistance source SC599 when inoculated with M. phaseolina. These results are consistent with previous reports indicating that stalk resistance in SC599 and B35 (or SC35) are different and cannot be equated (Rosenow, 1984; Tenkouano et al., 1993). The simplest explanation for these differences is that different genes are involved in regulating stalk rot resistance in these two genetic backgrounds.
Significant differences in stalk rot resistance were noted among entries and hybrids. This was not unexpected, as they included genotypes with contrasting reactions to the disease. The variability among entries for mean lesion length was so diverse that the most susceptible entries had lesion lengths that were 10 to 15 times longer than the most resistant entries. Some forms of resistance were recessive and others dominant or incompletely dominant. The resistance gene(s) derived from SC564 were recessive, with testcross hybrids producing very large lesions following inoculation with each of the pathogens. Resistance derived from SC1039 and SC1154 was more additive in nature. The hybrids exhibiting the smallest lesions were produced using SC599 and SC134. These responses were generally consistent for all pathogen groups except M. phaseolina, where SC35 and SC599 produced the smallest lesions. The fusarium and charcoal rot resistance traits in SC599, SC134, and SC35 should provide useful breeding materials for resistance characteristics of commercial sorghum hybrids.
Analysis of Genetic Diversity
Because of the unique mechanism responsible for generating SSR allelic diversity by replication slippage (Tautz, 1989), microsatellite markers have become the marker of choice for analyzing genetic diversity in many species, including sorghum (Smith et al., 2000; Dje et al., 1999) and corn (Taramino and Tingey, 1996; Smith et al., 1997). Because of their ability to distinguish genotypes at the DNA level, molecular markers are considered the most reliable tools for genetic studies. They are widely used for genetic diversity analyses to predict hybrid performance. Application of molecular markers in the field of agriculture is increasing for use in detecting and mapping quantitative traits that are otherwise difficult to identify and manipulate through conventional genetic strategies.
In this study, 21 polymorphic markers were used for determining genetic distances among the genotypes. Previous mapping studies indicated that 15 of the marker loci used in this study mapped to eight different sorghum linkage groups indicating a broad sampling of the genome in this analysis (Brown et al., 1996; Taramino et al., 1997). Multiple correspondence analysis conducted on the genetic distance data classified the entries into five distinct clusters. Cluster analysis generally differentiated the genotypes according to their geographical origin with little or no consideration to botanical races. This was shown by significant correlation (r = 0.34) between clusters and geographic origins. There was no correlation between races and clusters (data not shown). Failure of the clusters to reflect botanical diversity among entries probably is attributable to our selection of markers. The markers used in this study were not known to be linked with genes for panicle morphology. Eight of the nine Sudanese lines were grouped in Clusters 1 and 5. SC279 and SC694 from Nigeria, SC265 and SC284 from Burkina Faso, and one line from Mali and South Africa were also grouped in Cluster 1 along with the Sudan lines. This might be because of a relatively higher genetic flow between these countries at some time, perhaps through trade. Closer scrutiny of the relationships between genotypes within the same cluster reveals that the pattern of relationship between them was largely based on geographical hierarchy. For example, the mean distance among the Sudan lines within Cluster 1 was 0.38 and among the West African sources was 0.39, while the mean distance between the two regions was 0.49. This suggests geographical proximity played a significant role in determining the extent of genetic diversity. All 10 lines grouped in Cluster 2 were of Ethiopian origin and this constitutes 77% of the Ethiopian lines included in the study. The remaining three lines, SC1158, SC1014, and SC146, were grouped with Clusters 1, 3, and 5, respectively. Similarly, all of the Indian lines and a majority of the Ugandan lines were grouped together in Cluster 4.
Cluster analysis categorized the resistant sources into two separate clusters. Of the three lines that expressed comparable levels of resistance with SC599 based on hybrid performance, three (SC134, SC1154 and SC35) are of Ethiopian origin and were grouped together in Cluster 2, while SC599 was grouped in Cluster 4 along with the Indian and Ugandan lines. When compared with results from the phenotypic characterization, these results clearly highlight the importance of East African germplasm in breeding for resistance to stalk rots in sorghum. The genetic variability among sources of resistance can be further classified by differences in mode of inheritance for resistance and patterns of response to various pathogen groups. Some landraces exhibited resistance to both charcoal and Fusarium stalk rots while others, such as SC35, appear to be more resistant to M. phaseolina. If future studies identify multiple genes for stalk rot resistance, crop improvement efforts to combine these genes into a common genetic background may provide a powerful tool for controlling stalk rots in sorghum.
|
CONCLUSIONS
|
|---|
This study confirmed the existence of significant variation among sorghum accessions for resistance to stalk rot diseases that could be used in breeding programs. There also were significant differences in disease severity caused by the different pathogen groups when inoculated on the same lines and hybrids indicating differences in virulence among pathogen isolates and species. Macrophomina phaseolina produced significantly larger lesions in almost all entries; however, no significant differences were detected between the Fusarium species.
When compared with results from the phenotypic characterization, these results clearly highlight the importance of East African germplasm in breeding for resistance to stalk rots in sorghum. The genetic variability among sources of resistance can be further classified by differences in mode of inheritance for resistance and patterns of response to various pathogen groups. Some landraces exhibited resistance to both charcoal and Fusarium stalk rots, while others such as SC35 appear to be more resistant to M. phaseolina.
Analysis of the genetic relationships among these lines indicated that differences among entries were primarily associated with geographical origin with little or no consideration of botanical race. Stalk rot resistant sources were grouped into two clusters; however, the assignment of genotypes to these clusters did not necessarily predict stalk rot resistant characteristics of all entries within the cluster.
|
NOTES
|
|---|
Contribution No. 04-051-J from the Kansas Agric. Exp. Stn.
Received for publication November 18, 2003.
|
REFERENCES
|
|---|
- Anahosur, K.H., and S.H. Patil. 1982. Some promising sources of resistance to charcoal rot of sorghum. Sorghum Newsl. 25:118.
- Bhattramakki, D.J., A.K. Chhabra, and G.E. Hart. 2000. An integrated SSR and RFLP linkage map of Sorghum bicolor (L.) Moench. Genome 43:988–1002.[Medline]
- Bio-Rad. 1998. Quantity One users guide. v. 4. Bio-Rad Laboratories, Hercules, CA.
- Brown, S.M., S.M. Hopkins, S.E. Mitchell, M.L. Senior, T.U. Wang, R.R. Duncan, F. Gonzalez-Candelas, and S. Kresovich. 1996. [Sorghum bicolor (L.) Moench] Multiple methods for the identification of simple sequence repeats (SSRs) in sorghum. Theor. Appl. Genet. 93:190–198.[ISI]
- Dje, Y., D. Forcioli, and M. Ater. 1999. Assessing population genetic structure of sorghum landraces from northwestern Morocco using allozyme and microsatellite markers. Theor. Appl. Genet. 99:157–163.[ISI]
- Edmunds, L.K. 1964. Combined relation of plant maturity, temperature and soil moisture to charcoal stalk rot development in grain sorghum. Phytopathology 54:514517.
- Hallauer, A.R., and J.B. Miranda. 1988. Quantitative genetics in maize breeding. 2nd ed. Iowa State Univ. Press, Ames, IA.
- Jardine, D.J., and J.F. Leslie. 1992. Aggressiveness of Gibberella fujikuroi (Fusarium moniliforme) isolates to grain sorghum under greenhouse conditions. Plant Dis. 76:897–900.
- Mughogho, L.K., and S. Pande. 1984. Charcoal rot of sorghum. p. 11–24. In L.K. Mughogho (ed.) Sorghum root and stalk rots, a critical review. Proc. of the Consultative Group Discussion on Research Needs and Strategies for Control of Sorghum Root and Stalk Rot Diseases, Bellagio, Italy. 27 Nov.–2 Dec. 1983. ICRISAT, Patancheru, A.P. India.
- Rohlf, F.J. 1993. NTSYS-pc, Numerical Taxonomy and multivariate analysis system. v. 2.02h. Applied BioStatistical, New York.
- Rosenow, D.T. 1984. Breeding for resistance to root and stalk rots in Texas. p. 209–217. In L.K. Mughogho (ed.) Sorghum root and stalk rots, a critical review. Proc. of the Consultative Group Discussion on Research Needs and Strategies for Control of Sorghum Root and Stalk Rot Diseases, Bellagio, Italy. 27 Nov.–2 Dec. 1983. ICRISAT, Patancheru, A.P. India.
- Roseneow, D.T., C.A. Woodfin, L.E. Clark, and J.W. Sij. 1999. Drought resistance in exotic sorghums. p. 166. In 1999 Annual Meetings Abstracts. ASA, CSSA, and SSSA, Madison, WI.
- Saghai-Maroof, M.A., K.M. Soliman, R.A. Jorgensen, and R.W. Allard. 1984. Ribosomal DNA spacer-length polymorphisms in barley: Mendelian inheritance, chromosomal location, and population dynamics. Proc. Natl. Acad. Sci. USA 81:8014–8018.[Abstract/Free Full Text]
- SAS Institute. 1989. SAS/STAT user's guide. v. 6, 4th ed. SAS Inst., Cary, NC.
- Seetharama, N., F.R. Bidinger, K.N. Rao, K.S. Gill, and M. Mulgund. 1987. Effect of pattern and severity of moisture deficit stress on stalk rot incidence in sorghum. I. Use of line source irrigation technique, and the effect of time of inoculation. Field Crops Res. 15:289–308.
- Smith, J.S.C., E.C.L. Chin, H. Shu, O.S. Smith, S.J. Wall, M.L. Senior, S.E. Mitchel, S. Kresovich, and J. Jiegle. 1997. An evaluation of utility of SSR loci as molecular markers in maize (Zea mays L.): Comparisons with data from RFLP and pedigree. Theor. Appl. Genet. 95:163–173.[ISI]
- Smith, J.S.C., S. Kresovich, M.S. Hopkins, S.E. Mitchel, R.E. Dean, W.L. Woodman, M. Lee, and A. Porter. 2000. Genetic diversity among elite sorghum inbred lines assessed with simple sequence repeats. Crop Sci. 40:226–232.[Abstract/Free Full Text]
- Taramino, G., R. Tarchini, S. Ferrario, M. Lee, and M.E. Pe. 1997. Characterization and mapping of simple sequence repeats (SSRs) in Sorghum bicolor. Theor. Appl. Genet. 95:66–72.[ISI]
- Taramino, G., and S. Tingey. 1996. Simple sequence repeats for germplasm analysis and mapping in maize. Genome 39:277–287.[Medline]
- Tautz, D. 1989. Hypervariability of simple sequences as a general source for polymorphic DNA markers. Nucleic Acids Res. 17:6463–6467.[Abstract/Free Full Text]
- Tesso, T., L.E. Claflin, and M.R. Tuinstra. 2004. Estimation of combining ability for resistance to Fusarium stalk rot in grain sorghum. Crop Sci. 44:1195–1199.[Abstract/Free Full Text]
- Tenkouano, A., F.R. Miller, R.A. Frederiksen, and D.T. Rosenow. 1993. Genetics of nonsenescence and charcoal rot resistance in sorghum. Theor. Appl. Genet. 85:644–648.[ISI]
- Toman, J., Jr., and D.G. White. 1993. Inheritance of resistance to anthracnose stalk rot of corn. Phytopathology 83:981–986.[ISI]
- Zummo, N. 1980. Fusarium disease complex of sorghum in West Africa. p. 297–299. In G.D. Bengtson (ed.) Sorghum diseases: A world review. Proc. of Int. Workshop of Sorghum Diseases, Hyderabad, India. 11–15 Dec. 1978. ICRISAT, Patancheru, Andhra Pradesh, India.
TOP
ABSTRACT
INTRODUCTION
