| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Crop Sciences, University of Illinois at Urbana-Champaign, Urbana, IL 61820
* Corresponding author (trochefo{at}uiuc.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Abbreviations: GCA, general combining ability • SCA, specific combining ability
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
A number of sources of relative resistance to Aspergillus ear and kernel rot and inhibition of aflatoxin production have been identified by screening and selection (Campbell and White, 1994, 1995a, 1995b; Scott and Zummo, 1988, 1990, 1992). The breeding strategies developed to introgress resistance from these sources and potential new sources into elite germplasm will depend in part on the level and genetic basis of resistance. Generally, resistance to Aspergillus ear rot and aflatoxin accumulation is considered quantitatively inherited, with additive gene action more important than dominance gene action (Campbell and White, 1995b; Darrah et al., 1987; Gardner et al., 1987; Hamblin and White, 2000; Zuber et al., 1988). However, generation means analysis for some sources of resistance showed that additive genetic effects were important, whereas dominance effects were significant for other sources (Campbell, 1995b). Comparison of two diallel studies for resistance to aflatoxin accumulation, with one using a subset of parental lines of the other study and performed in different years, indicated additive effects were more important in one study and dominance effects more important in the other (Darrah et al., 1987; Gardner et al., 1987). Large genotype x environment interactions for aflatoxin accumulation has been reported in various studies (Hamblin and White, 2000; Widstrom et al., 1984; Zuber et al., 1983). The large variation that exists between years has been considered an impediment to breeding for resistance to aflatoxin production and Aspergillus ear rot (Darrah et al., 1987) and also limits the precision of genetic studies. These observations indicate the need and merit of evaluating potential, promising sources of resistance to aflatoxin accumulation in different genetic studies and collectively in several environments before general, reliable conclusions are made on a particular source of resistance.
New sources of resistance to Aspergillus ear rot and aflatoxin accumulation are continually being identified (Campbell and White, 1994, 1995a, 1995b; Scott and Zummo, 1988, 1990, 1992). On the basis of what is known about the genetic basis for resistance to A. flavus and aflatoxin production from previous studies, it is advisable to closely examine genetic relationships and interactions among the more promising new resistance sources in the general environment in which the materials will be grown. Several inbreds that had reduced ear rot severity and/or aflatoxin accumulation when evaluated in F1 hybrid combination with B73 and/or Mo17 were identified in a large study conducted for two years (Campbell and White, 1995a). However, since more than 900 inbreds were screened, evaluation of each individual hybrid was limited, with only two replications per year. Furthermore there were hybrid x environment interactions and in some cases missing data points because of technical difficulties. Therefore six of the more promising inbreds in this initial evaluation were selected for further and more detailed study: CI2, 75-R001, LB31, Oh513, Oh516, and Tex6. The objectives of this study were to (i) determine if any of these six inbreds had a complementary genetic basis of resistance such that their F1 hybrids would result in higher levels of resistance than hybrid combinations involving Mo17 and/or B73; (ii) investigate the genetic basis of resistance for these six individual inbreds through diallel analysis; and (iii) identify which of these six inbreds are the most promising sources of resistance for conventional and molecular marker assisted breeding programs.
The purpose of making resistant x resistant crosses was to examine if higher levels of resistance could be obtained than previously reported, rather than limiting evaluation just to crossing promising resistant inbreds to Mo17 or B73. This strategy will provide information on the potential levels of resistance that might be obtained, regardless of current maize breeding heterotic patterns. The mating structure in this study also serves to examine how a resistance source performs when combined with a number of other inbreds, not just Mo17 and B73. The number of good sources of resistance to aflatoxin are very limited and thus a good resistance source might be used in breeding programs in regions of the United States and the world that do not use Mo17 or B73 related materials. Therefore information on how resistance sources combine with a set of diverse inbreds is useful. Another purpose of this diallel analysis was to estimate general combining ability (GCA) and specific combining ability (SCA) effects for specific inbreds that in an initial study were tentatively considered resistant when crossed just with Mo17 and/or B73. The GCA and SCA estimates provide genetic information on the specific inbreds beyond the simple mean performance of hybrids produced from these inbreds. The historically important inbreds Mo17 and B73 were included in this study to provide a common reference to earlier studies, and also to provide more information on the performance of potential resistance sources in combination with Mo17 and B73.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The set of 28 F1 crosses were grown at the University of Illinois Agronomy/Plant Pathology South Farms, Urbana, IL in 1997 and 1998. The plots were grown on a Flanagan silt loam (fine montmorillonitic, mesic Aquic Argiudolls). The experimental design was a randomized complete block design with four replications. F1 hybrids were grown in one-row plots, 5.34 m long spaced 0.76 m apart with 15 plants per plot after thinning.
Approximately 20 to 24 d after midsilk, 14–15 ears were inoculated through the husks with a spore suspension (1 x 106 conidia mL-1) of four isolates of Aspergillus flavus (NRRL Isolates 6536, 6539, 6540, and an isolate from Illinois collected in 1988), using the pinboard inoculation technique previously described (Campbell and White, 1994). The four isolates were previously found to be highly toxigenic, and a group of four was used as a safeguard in case any one isolate suddenly became non-toxigenic after lab culture. Approximately 50 to 60 d after inoculation, the ears were husked and the percent of fungal colonization in the inoculated area was visually estimated. The ear rot ratings for each plot were averaged. Harvested ears were air dried before they were shelled. Kernels from inoculated ears within a family were bulked and ground with a Romer mill (Model 2A) to sizes below 1 mm. Ground kernels were subsampled and 1 to 2 g quantities were removed from the 250 to 500 g bulked samples and assayed for aflatoxin levels by an indirect competitive ELISA assay (Campbell and White, 1995a). All samples were run in triplicate. The lower detection limit by this procedure is 2 ng g-1 aflatoxin B1, so samples with non-detectable aflatoxin were recorded as 2ng g-1.
Ear rot and aflatoxin data were analyzed as a randomized complete block design by the SAS general linear model procedure (SAS Institute Inc., Cary, NC). Hybrids and years were declared fixed effects. Natural logarithmic transformations of aflatoxin concentration (ng g-1) were used to stabilize variances. Homogeneity of error mean squares was tested with HOVTEST option of SAS. Means separation was performed by Waller-Duncan test with the WALLER option of SAS.
Diallel Analysis
The data from the complete set of 28 possible F1 hybrids from the inbreds CI2, 75-R001, LB31, Oh513, Oh516, Tex6, B73, and Mo17, grown in 1997 and 1998, were used for diallel analysis. Parents were not included because the inoculation method used is best suited to the larger F1 hybrid ears. Reciprocal crosses were not made and the resultant half diallel ignores possible reciprocal effects, which have not been reported for resistance to Aspergillus ear rot and aflatoxin accumulation. Griffing's (1956) Method 4 of diallel analysis was employed as only F1 hybrids are included. The experiment was performed and analyzed according to Griffing's (1956) Model 1 of assumptions which indicates the variety and block effects are constants. Since the inbreds used in this experiment were a subset selected for possible resistance, the hybrids produced from these inbreds are considered a fixed effect. Since the response of maize hybrids to inoculation by Aspergillus flavus in terms of fungal growth and particularly aflatoxin production can vary considerably in different parts of the United States and in different years (Payne, 1992), environments were considered a fixed effect. Evaluation in only two years in central Illinois does not provide an adequate sampling of different environments to be able to confidently make more general statements about resistance beyond these two environments in central Illinois and for different parts of the United States. There have been a number of instances in which certain inbreds and hybrids considered relatively resistant after evaluation in just two environments were subsequently determined not to be relatively resistant upon evaluation in additional years and locations.
Data were analyzed with a SAS program (DIALLEL-SAS) developed by Zhang and Kang (1997), based on the linear contrast matrix for each genetic component in the diallel analysis, described by Griffing (1956) and Cockerham (1963). Sample matrices for each of Griffing's methods, generated by the array statements in the program, are given in Kang (1994). The applicable general linear model for the analysis was yijklc = m + al +bkl + vij + (av)ijl + eijkl, where yijklc is the observed trait (either ear rot or natural logarithmic transformations of aflatoxin concentration) of the (i,j)th hybrid in year l and the kth replicate; m = the overall mean; al = year - effect; bkl = block or replication within year effect; vij = F1 hybrid effect = gi + gj + sij [where gi = general combining ability (GCA) effect for the ith parent, gj = GCA effect for the jth parent, sij = specific combining ability (SCA) effect for the ijth F1 hybrid], (av)ijl = interaction between F1 hybrids and year = al (gi + gj + sij); and eijkl = random error term associated with the (ij)th cross in the kth replication in year l. To test the significance of mean squares for F1 hybrids, GCA and SCA effects, and the interaction between environments and the corresponding component, the residual error term is used in the fixed model. Standard errors of GCA and SCA estimates were calculated according to formulas provided by Griffing (1956): var(gi) = p - 1/p(p - 2) 
2; var(sij) = (p - 3)/(p - 1) 2 (i 
j); where p is the number of parents and 2 is the expectation of mean squares for error term.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
|
|
The mean of all hybrids with Mo17 as a parent rated 5.06 ng g-1 for loge aflatoxin and 42.3% for ear rot, similar to hybrids with B73 as a parent having values of 5.3 ng g-1 for loge aflatoxin and 41.2% for ear rot. (Mo17xB73 was not included in these means since it was common to each group). However, comparisons of hybrid means demonstrated that certain resistant x resistant combinations resulted in higher levels of resistance to aflatoxin than related hybrids involving B73 whereas other combinations involving Mo17 did not. The hybrid Tex6xOh516 had significantly lower levels of aflatoxin accumulation than Tex6xB73. The hybrid Tex6xOh513 was significantly more resistant to aflatoxin accumulation than the hybrids Oh513xB73 or Tex6xB73 (Table 2). In contrast, Tex6xOh516 was not significantly different than Tex6xMo17 or Mo17xOh516.
For ear rot, there were two comparisons in which a resistant x resistant cross had a significantly lower level of ear rot than a related hybrid involving B73. These were Oh516xOh513 and Tex6xOh513 in comparison to B73xOh513. In contrast, the Mo17xOh513 hybrid was not significantly different than Oh516xOh513 and Tex6xOh513. Furthermore Oh513x75-R001 and LB31x 75-R001 were significantly worse for ear rot than the corresponding cross of Mo17xOh513 and LB31xMo17, but were not significantly worse than the corresponding cross of B73xOh513 and LB31xB73. Thus Mo17 in some cases appears to combine better than B73 with resistance sources to provide levels of resistance for ear rot as well as aflatoxin comparable to resistant x resistant hybrids. This suggests that introgressing resistance alleles from some of the better aflatoxin resistance sources into B73 type lines as opposed to Mo17 type lines may be more beneficial to improve a B73xMo17 type hybrid.
Diallel Analysis
Analyses of variance for ear rot and aflatoxin data (Table 3) obtained from Griffing's Method 4, Model 1 diallel analysis showed that GCA for both ear rot rating and aflatoxin concentration was highly significant. SCA was not significant for both ear rot rating and aflatoxin concentration. This suggests that SCA effects overall are less important than GCA effects for this set of lines. The environment (E) x hybrids (H), E x GCA, and E x SCA interactions were highly significant for aflatoxin. In contrast, none of the interactions with environment were significant for ear rot. These findings are consistent with the higher levels of genotype x environment interactions commonly observed for aflatoxin in comparison to ear rot.
|
Highly significant positive GCA effects for both ear rot and aflatoxin were obtained with 75-R001. This shows that relative to the other inbreds in this experiment, 75-R001 is associated with a significant increase in ear rot and aflatoxin levels in hybrids. The 75-R001 inbred was selected for further evaluation in this study based on performance of the 75-R001xB73 hybrid for ear rot and aflatoxin levels in the Campbell and White (1995a) study. The results in our study clearly indicate that 75-R001 is not a good, reliable source of resistance, contrary to initial evaluations.
Although Mo17 and B73 were included for their historical importance and to assess how resistance sources combine with these lines, the diallel provided useful results for these two lines. B73 was associated with significant GCA effects for higher levels of aflatoxin concentration and Mo17 was not (Table 2). This finding is consistent with results that showed B73 in some cases did not combine as well as Mo17 with resistant inbreds to provide higher levels of resistance (Table 2).
This study has confirmed other studies (Campbell and White, 1995b; Hamblin and White, 2000) that have shown that Tex6 is one of the better sources of resistance to aflatoxin accumulation. Oh516 was not included in these studies because the initial results on Oh516 were limited. The finding that the magnitude of GCA effects of Tex6 and Oh516 are similar for both ear rot and aflatoxin production in this set of materials is very encouraging. This suggests that Oh516 may be comparable with the now well-proven Tex6 line in terms of level of resistance. Oh516 provides another promising, potential source of resistance that is complementary to Tex6 in hybrid combination as the Tex6xOh516 hybrid has the lowest values for ear rot and aflatoxin in this study (Table 2). In contrast, diallel analysis indicated that CI2 has only limited value as a resistance source for reducing both ear rot and aflatoxin production. These results are consistent with a recent study that indicated CI2 and progenies derived from it were inconsistent when evaluated as a source of resistance to aflatoxin in two different years (Rozzi and White, 2001).
The comparison of results in this study to other studies indicates the value of performing diallel studies with the most promising new sources of resistance to Aspergillus ear rot and aflatoxin production. Diallel analysis should enable useful assessment of promising resistance sources before using them in expensive molecular marker mapping studies and in breeding programs. Our study also shows the merit of evaluating sources of resistance in multiple environments as some initially promising sources of resistance were determined less desirable or not resistant. Evaluation in just two environments does not appear to be adequate to make general predictive statements on performance of promising sources of resistance for Aspergillus ear rot and aflatoxin production in other environments.
The genetic information obtained from this study is useful for identifying and assessing the best sources of resistance, and also for making more informed decisions in choice of parents for genetic studies to identify quantitative trait loci (QTL) associated with lowered ear rot incidence and aflatoxin accumulation levels in maize. We have ongoing QTL studies with Tex6 and CI2. However, based on our results and those of other studies, future emphasis will be placed on Tex6 and Oh516, whereas CI2 will be dropped from QTL study. The highly significant GCA for lower ear rot and aflatoxin accumulation associated with Oh516 suggests that this inbred may be useful for incorporating resistance into diverse germplasm sources and therefore may be of use to other research programs. We developed a BC1S1 population derived from a backcross of a B73 type inbred to Oh516 which will be inoculated and evaluated for segregation for resistance to ear rot and aflatoxin production in multiple environments. If the results are promising, efforts will be made to map QTL for resistance to ear rot and aflatoxin production in this population. If favorable QTL from Oh516 are identified on different chromosome regions than for Tex6, we would introgress favorable QTL from both Tex6 and Oh516 into elite lines with the goal of increasing levels and stability of resistance over environments.
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 17, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Zhang, M. S. Kang, and K. R. Lamkey DIALLEL-SAS05: A Comprehensive Program for Griffing's and Gardner-Eberhart Analyses Agron. J., June 17, 2005; 97(4): 1097 - 1106. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Plant Registrations | Soil Science Society of America Journal | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
and for loge transformed concentrations of aflatoxin
B1 produced in F1 crosses among eight inbred lines. Mean values for ear rot (shown below diagonal) and loge aflatoxin B1 concentration (shown above diagonal) obtained from the 28 F1 hybrids evaluated in 1997 and 1998 in Urbana, IL.