engleski

Quantitative Genetic Studies on the Integration of Exotic Germplasm into Adapted Maize Breeding Materials

Narrowing of genetic diversity in breeding populations of crops as a consequence of modern agriculture led to worldwide efforts of collecting, characterizing and utilizing plant genetic resources. The main goal of the utilization of this mostly unadapted (exotic) material is increasing genetic variability of current breeding programs through integration of novel and useful germplasm. Exotic germplasm can provide genes for increased yields, pest resistance and may contain quality traits to meet new market demands. However, the lack of adaptation, as the major obstacle for immediate use of exotics, causes problems in generating recombinations and in identifying superior genotypes. Consequently, an optimal breeding procedure for integration of exotics for polygenic inherited (quantitative) traits is not known, so far. Aim of the present study was to ascertain appropriate breeding scheme for integration of diverse unadapted genetic resources in maize using quantitative genetic parameters. Specifically, the objectives were according to testcross performances i) to evaluate 18 initial crosses between adapted (recipient) and exotic (donor) elite inbred lines for agronomic and forage quality traits, ii) to compare the effects of backcrossing to adapted parent and intermating of F2 generation in relation to nonintermated F2 generation after selection for earliness when used as S1 lines developed from the initial crosses, iii) to examine effects of recipient and donor, and their specific interactions in the adapted  exotic crosses, iv) to estimate genetic variation within four initial crosses between adapted and exotic elite inbred lines each in three selected foundation populations, namely in F2, twice intermated F2 (F2-Syn2), and BC1 generations, v) to determine the relative superiority of each generation for breeding purposes according to quantitative genetic parameters (mean, genetic variance, heritability, selection response, breeding usefulness) and to recommend the superior type of foundation population to be developed from the adapted  exotic cross of elite inbred lines. The basic parental material was comprised of four adapted elite inbred lines (D06, D408, D08, F272) and nine exotic lines (A619, H107, B73, Mo17, Va26, H108 and NC258) . The adapted lines are current elite Central European Dent breeding material of early maturity (about FAO 200 to FAO 300 maturity units). The late maturing exotics from the USA are well known public dent lines ranging from FAO 500 to FAO 700. The non-adaptedness of the exotics was primarily defined by the extreme lateness for Central European environments determined by late flowering dates, and subsequently, a low content of dry matter. Among the adapted lines, D06 and D408 were each crossed with five exotics, and D08 and F272 with another four exotic lines, resulting in a total of two sets of ten and eight adapted  exotic initial crosses, respectively. First backcross generations (BC1) were produced from the F1 hybrids by crossing to the respective adapted line. F1 plants were also selfed to obtain F2 populations, and each F2 population was intermated twice under controlled hand pollination to obtain F2-Syn2 populations. 1150 plants were grown from each of the 54 segregating populations , i.e. foundation populations (18 crosses in F2, F2-Syn2 and BC1, respectively). The earliest 50 plants were selected, selfed and designated as S1 lines representing a very high selection intensity with a selected fraction of α = 4.4%. A bulk of the selected 50 S1 lines from each of the 54 foundation populations, the respective F1 crosses, as well as three of the adapted inbred lines (D06, D408 and D08) were crossed with a common flint-tester. Material has been developed in period 1988-1993. Experiment 1 (the main experiment at three locations in 1994 and 1995) was set up to determine genetic differences in mean of generation -derived testcrosses (F1, F2, BC1 and F2-Syn2) developed from 18 adapted×exotic crosses for following traits: dry matter content of the whole plant, dry matter yield, days to pollen shed, days to silk , plant height, crude fiber content in dry matter, acid detergent fiber content in dry matter, in-vitro digestiblity of organic matter, enzymatic solubility of organic matter, starch content in dry matter, crude protein content in dry matter and net energy content in kg of dry matter. Experiment 2, planted in 1996 at two locations for grain production, was designed i) to estimate genetic variances of 50 S1-testcross progenies from four adapted × exotic crosses in three types of foundation population each (a total of 12 populations), which had been chosen from the main experiment and produced by the single seed descent method , and ii) to determine breeding usefulness of the populations. Grain dry matter content, grain yield, days to pollen shed and days to silk were evaluated. Differences between initial crosses were significant or highly significant in both sets and for all 12 traits measured except for dry matter yield in the second set. Levels of significance for the effect of generation in the 18 adapted × exotic crosses were variable across the sets and traits: significant for agronomic traits, except for dry matter yield in the first set, and nonsignificant for quality traits excluding starch content in both sets and crude protein content in the second set. Significant interactions between generation and initial cross effects were observed at least in one set across all traits evaluated. Significance of interactions between genotypic effects (generation, initial cross) and environment varied across the sets and traits. Due to the complex interactions, the effect of generation was elaborated for each initial cross separately for dry matter content and dry matter yield, and separately for each adapted line, pooled across the respective five or four exotic lines for all traits. Dry matter content ranged from 27.88% for the testcross (D408×B73)F1 to 34.84% for adapted testcross with D08. Dry matter yield ranged from 1303 g m-2 for adapted testcross with D08 to 1549 g m-2 for the S1-testcross of (D06×B73)F2 The difference between 25% of exotics in three investigated generations (F1, F2, F2-Syn2) and 12.5% in backcrosses was sufficient, however, to distinguish the means of these generations, although the overall maturity range was relatively small, and ranges within each group were overlapping to some extent. Hawbaker et al. (1997) supposed that the use of an early maturing tester reduced variability for grain dry matter content among the semi-exotics tested in their study. Even more narrow ranges were observed for the quality traits, in which the differences, although significant among initial crosses, were not considerable in most instances. The 18 adapted × exotic initial crosses reacted differently to process of adaptation across the generations for quality traits. Focus of this study was relationship between dry matter content and yield, since all quality traits except crude protein were tightly associated with maturity. Therefore, aggregating breeding value combining yield and maturity defined as “agronomic usefulness” of exotics was of prime interest. Progenies of all types of foundation population of the initial cross D06×B73 were among ten best ranking entries. This implies that a proper choice of germplasm is more important than the choice of the breeding method. The BC1-derived testcrosses were earlier maturing than F2-derived progenies in all instances. However, this difference was significant in ten crosses only. There was no clear distinction between F2 and BC1 for dry matter yield across initial crosses indicating that backcrossing did not result in a notable shifting towards the recurrent parent for yield. Estimated genetic variances of BC1-derived materials across four initial crosses in Experiment 2 were consistently smaller than those of F2-derived materials for grain dry matter content. However, significant effects of backcrossing on the genetic variances for dry matter content were detected only in two initial crosses, and for grain yield in one of two initial crosses in which these variances were estimable. With few exceptions, intermating did not generate considerable changes neither in the means, nor in the genetic variances. Positive, yet minor recombination effects for earliness, were obtained in some adapted×exotic combinations. There were no significantly different means of F2-Syn2- compared to the respective F2-derived testcrosses among the 18 initial crosses for dry matter yield. However, the intermated, F2-Syn2-derived progenies of all four initial crosses in Experiment 2, had, although only partially significant, consistently lower grain yields than the respective progenies of non-intermated foundation populations. The effects of backcrossing as well as intermating on means and genetic variances were germplasm specific. The optimal proportion of unadapted and adapted material in a foundation population additionally depends on the particular goal(s) of integration. While maximizing the proportion of unadapted germplasm is important for integration programs themselves, for yield enhancement efforts the proportion of the unadapted and adapted germplasm can be variable according to their performance relation. Whether to produce intermated foundation populations appears to be also a question of breeding goals. Intermating does not seem to be worthwhile for inbred line development due to its duration and relative small genetic gains obtained compared to non-intermated F2 populations. On the other hand, intermating can be beneficial for continued programs for some material, especially in first selection cycles in order to obtain earlier S1 lines. With exception of D06B73, the results of the Experiment 2 showed that the predicted selection response generally did not change the ranks presumed by the progeny means of the three types of foundation population for grain dry matter content, emphasizing the importance of the mean differences. The (D06B73)BC1 generation had considerable lower selection response than the other two generations for grain dry matter content, as well as for grain yield (Figure 3.6a,b). Similarly, the ranks were not changed by predicted response as per the means of the foundation populations of D408B73 for grain yield. For the other two initial crosses, comparisons of usefulness parameter were not conclusive, due to small and/or unestimable variances. . It indicates that F2, due to completely lack of association between the traits, would be the most appropriate foundation population to initiate selection for earliness without undesirable alternation of yield.

maize; quantitative genetics; exotic germplasm; breeding

nije evidentirano