Previous PageTable Of ContentsNext Page

Load of deleterious genes in maize estimated by inbreeding depression, yield potential per plant and CV

Ioannis S. Tokatlidis1 and Leonidas Giakalis2

1 Department of Agricultural Development, Democritus University of Thrace, Pantazidou 193, Orestiada 682 00 Greece. www.duth.gr Email itokatl@agro.duth.gr
2
Technological Education Institute of W. Macedonia, 53100 Florina, Greece. www.teikoz.gr Email leon3@in.gr

Abstract

Inbreeding depression is usually used to measure the load of deleterious genes. In this work inbreeding depression was related with reduced yield potential per plant and increased coefficient of variation (CV) of individual plant yield, as they had been estimated in F1 and F2 of eight maize (Zea mays L.) hybrids under very low density (0.74 plants/m2). Hybrids developed through selfing and on the basis of line performance per se, were found to have improved yield potential per plant in both F1 and F2, lower CV values in F2, and lower inbreeding depression. Results were indicative of effective exploitation of the additive genetic effects, in the aim to reduce the load of deleterious genes and to produce less heterozygous hybrids characterized by improved yield potential per plant, an essential trait for high and more stable productivity.

Media summary

Lower load of deleterious genes in maize is reflected by improved yield potential per plant and low CV, two determinant factors for high and stable productivity.

Key words

Low density, Honeycomb selection, Heterosis, Combining ability

Introduction

In the cross-pollinated maize, the performance of homozygous genotypes is lower than that of the hybrid cultivars. Homozygosity permits the expression of recessive alleles that may have been masked by a dominant allele under heterozygosity. When the recessive alleles are less favorable than dominant ones, a reduction in individual performance results, referred to as inbreeding depression. Inbreeding has been used to reduce the frequency of deleterious recessive genes. One selfed generation may permit the expression and subsequent elimination of deleterious alleles (Fehr 1987). The load of deleterious genes is usually estimated by measuring inbreeding depression (ID): ID(%)=100(F1-F2)/F1. The larger the difference between F1 and F2, the larger the load of deleterious genes. Tokatlidis et al. (1999) stated that higher load of deleterious genes is associated with lower yield potential per plant and higher CV values in both hybrids and their parents. The importance of yield potential per plant as a component of the crop yield, being a determinant parameter for greater stability and dependability, was emphasized in a resent review study (Tokatlidis and Koutroubas 2004). In an attempt to improve the yield potential per plant Tokatlidis et al. (1998) developed 40 recycled S5xS5 hybrids, through honeycomb selection in the F2 of the widely adapted commercial hybrid PR 3183, under very low density (0.74 plants/m2), and on the basis of individual plant yield and line performance per se, in order to reduce the load of deleterious genes by exploiting the additive genetic effects. These hybrids exhibited on average 67% higher yield potential per plant, compared with their original hybrid. The objective of this work was to estimate the load of deleterious genes of the recycled hybrids by measuring the inbreeding depression, as well as the possible association of inbreeding depression with yield potential per plant and CV.

Methods

Honeycomb selection in the F2 of the hybrid PR 3183, under very low density (0.74 plants/m2), based on individual plant yield and line performance per se, resulted in 40 recycled single-cross hybrids (Tokatlidis et al 1998). The original hybrid PR 3183 (PR) was the leading hybrid in Greece for more than 10 years (e.g., 1980-90). Six S7xS7 of those recycled hybrids (RC1-6), chosen because they combined higher yield per plant and lower CV (Tokatlidis et al. 2001), along with their original hybrid and the check hybrid B73xMo17 (CH) were used in the study. F1 and F2 seed of these hybrids were sown in a honeycomb R16 experimental design (Fasoulas and Fasoula 1995), at the Technological Education Institute farm of Florina, Greece, and during two successive growing seasons (1999-2000). Plants were spaced 125 cm apart in a triangular grid (0.74 plants/m2), arranged systematically so that every F1 plant adjoined an F2 plant of the same hybrid, and plants of a given entry were located uniformly and surrounded by plants of the remaining entries (Fig. 1). This systematic arrangement ensured effective soil heterogeneity control and comparable conditions between entries. Each entry was replicated about 50 times in both experiments. Mean grain yield per plant (YP), coefficient of variation (CV) of individual plant yields, and inbreeding depression (ID), were calculated.

Figure 1. The way of experimentation (R16 honeycomb design). Every F1 plant adjoined an F2 plant of the same hybrid, and plants of each entry were located uniformly surrounded by plants of the remaining entries (e.g., RC5F1)

Results

All the recycled hybrids (RC) exhibited significantly greater yield potential per plant, compared with both their original hybrid (PR) and the check hybrid (CH), either in F1 or F2 (Table 1). The superiority of the RC hybrids over PR ranged from 36 (RC5) to 54% (RC2) in F1 and from 47 (RC5) to 75% (RC2) in F2. On the other hand the check hybrid yielded lower by 23% in F1 and by 32% in F2 than the PR hybrid. It is noteworthy that two F2 of RC hybrids did not differ significantly from the F1 of PR hybrid, and all F2 of RC hybrids yielded higher than the F1 of CH hybrid (three with significant superiority). The RC hybrids had CV values ranging from 28.3 to 32.1%, and on average (30.6%) being similar to those of PR (29.4%) and CH (30.7%) (Fig. 2). However, in F2 generation RC hybrids had lower CV values (on average 40.2%) than PR (45.8%) and CH (47.9%). Inbreeding depression of RC hybrids ranged from 38.5 to 40.3%, while ID of PR was 45.7% and that of CH was 52.4%. According to ID values the highest load of deleterious genes was observed in CH hybrid, and the lowest in RC hybrids. It seems that improved yield potential per plant reflects lower deleterious genes (Fig. 2). Indicative was the negative relationship of ID with yield per plant in both F1 and F2 as r values were -0.94 and -0.97, respectively (Table 2). Although the six RC hybrids were of the most promising (Tokatlidis et al. 1998; 2001), their yield potential per plant may have been underestimated in this work. This because in a previous work (Tokatlidis et al. 1998), although the original hybrid yielded the same (668 g/plant), the average superiority of the forty recycled hybrids over PR was 67%. Poorer performance of RC hybrids in this work could be attributed to two reasons. First recycling had been conducted in a single region and different than that of present experimentation, so genes for homeostasis may have been lost. Second CV, which reflects stability of performance, was not used as a selection criterion from the first generations of selfing. The last may explain also why improved yield potential per plant was not followed by reduced CV values, and the lack of relationship between yield per plant and CV among hybrids (r=-0.11) (Table 2). The smaller the CV in the absence of competition, the more favorable are the gene interactions within entries and the more increased is the stability of performance and productivity (Fasoula and Fasoula 1997a). However, a significant negative relationship between yield potential per plant and CV was found when r was computed among the recycled hybrids (r=-0.96), being consistent with previous work (Tokatlidis et al. 1999). Additionally this relationship was apparent in F2 generation (r=-0.82, and -0.95) (Table 2). It seems also that a positive relationship exists between ID and CV, according the data obtained from F2s (Fig. 2, Table 2). A similar positive relationship (r=0.95) reported by Fasoulas (1993). The lack of such relationship in F1s (Table 2) may be attributed to the same reasons mentioned above.

Table 1. Yield potential per plant (YP), coefficient of variation for individual plant yield (CV) and inbreeding depression (ID) of the recycled hybrids (RC), their original hybrid (PR) and the check hybrid (CH).

Hybrid

F1

F2

 

YP (g) *

CV%

YP (g) *

CV%

ID(%)

RC1

887 c

32.1

530 fg

43.5

40.3

RC2

1026 a

28.3

631 de

35.6

38.5

RC3

999 ab

29.3

614 de

35.4

38.5

RC4

984 ab

30.5

580 ef

38.0

41.1

RC5

902 c

32.2

532 fg

42.9

41.0

RC6

940 bc

31.4

535 fg

45.8

43.1

PR

665 d

29.4

361 h

45.8

45.7

CH

515 g

30.7

245 i

47.9

52.4

* values followed by a common letter do not differ significantly at the 5% level according to z-test for independent samples and different standard deviations (n≈100).

Figure 2. Data of the six recycled hybrids on average (RC1-6), and the check hybrid (CH) expressed % of the respective values of the original hybrid (PR). Higher load of deleterious genes, reflected by higher inbreeding depression (ID), is followed by lower yield potential per plant (YP) and higher CV.

Table 2. The simple correlation coefficients between yield potential per plant (YP), coefficient of variation (CV) and inbreeding depression (ID) in F1 and F2 of the eight hybrids.

 

F1

F2

YP

ID

YP

ID

CV

-0.11

0.14

-0.82 (P<2%)

0.79 (P<2%)

CV

-0.96 (P<3‰)

0.70

-0.95 (P<4‰)

0.84 (P<4%)

ID

-0.94 (P<1‰)

 

-0.97 (P<1‰)

 

ID

-0.54

 

-0.78 (P<7%)

 

† among the eight hybrids ‡ among the six recycled hybrids

Lower ID values of the recycled hybrids constitute good evidence that the procedure used to develop them, (i.e., improving yield potential per plant through selfing to exploit additive genetic effects), did reduce the load of deleterious genes. Since the combining ability is tightly related to heterosis, which is connected with the presence of deleterious genes, selection on the basis of combining ability implies retention of deleterious genes (Fasoulas 1993). Close inbreeding may convert repulsion into coupling phase linkages, while selection for line performance per se eliminates the deleterious genes and exploits the favorable additive genetic effects, resulting in improved yield potential per plant (Fasoula and Fasoula 1997b). Given that yield potential per plant is affected by the load of deleterious genes, the significant superiority of the RC hybrids over PR and CH is explained by the differences in the load of deleterious genes, as determined by the CV values in F2s and ID values. The more the deleterious genes are replaced by favorable additive alleles, the less the heterosis in F1 (Koutsika et al. 1990; Fasoulas 1993). The negative correlation between the yield potential per plant and ID suggests that selection in maize must be based on productivity through selfing in order to exploit the favorable additive and eliminate the deleterious gene effects. According to Hallauer and Miranda (1981), the accumulated evidence in maize suggests that additive genetic variance with partial to complete dominance of favorable genes is the predominant type of gene action in the expression of yield. According to data from Meghji et al. (1984) the rate of yield improvement was similar for hybrids and inbreds during the past three eras (1930, 1950, 1970), indicating that the improvement of inbred line performance per se and not heterosis played the predominant role in the rate of improvement of hybrid performance. The bulk of high performance of maize hybrids is due to additive and dominance effects (Crow 2000).

Conclusion

The positive relationship of ID with CV, as well as the negative relationship of yield potential per plant with ID and CV, show that either yield potential per plant or CV determine the load of deleterious genes. Consequently, the improvement of yield potential per plant, by selection based on line performance per se (high mean yield combined with low CV), may contribute to reduced load of deleterious genes. Selection for combining ability leads to the preservation of deleterious genes because of evaluation under heterozygosity. Hybrids that combine improved yield potential per plant and tolerance to environmental effects, reflected by higher grain yield and lower CV values under low densities, are expected to have wider optimum density (density under which maximum crop yield is obtained), a key factor for high and stable productivity (Tokatlidis et al. 2001; Tokatlidis and Koutroubas 2004).

References

Crow JF (2000). The rise and fall of overdominance. In Plant Breeding Reviews (Ed. J Janick) 17, 225-257.

Fasoula DA and Fasoula VA (1997a). Competitive ability and plant breeding. In Plant Breeding Reviews (Ed. J Janick) 14, 89-138.

Fasoula DA and Fasoula VA (1997b). Gene action and plant breeding. In Plant Breeding Reviews (Ed. J Janick) 15, 315-374.

Fasoulas AC (1993). Principles of crop breeding. A.C. Fasoulas, P.O. Box 1555, GR-54006, Thessaloniki 17, Greece, pp. 127.

Fasoulas AC and Fasoula VA (1995). Honeycomb selection designs. In Plant Breeding Reviews (Ed. J Janick) 13, 87-139.

Fehr WR (1987). Principles of cultivar development. Vol. 1. Macmillan Publ. Co. New York. pp. 536.

Hallauer AR and Miranda JB (1981). Quantitative genetics in maize breeding. Iowa State University Press/Ames.

Koutsika-Sotiriou M and Bos I (1996). Heterosis after several numbers of cycles of mass honeycomb selection in maize. Journal of Genetics & Breeding 50, 7-13.

Meghji MR, Dudley JW, Lambert RJ and Sprague GF (1984). Inbreeding depression, inbred and hybrid grain yields, and other traits of maize representing three eras. Crop Science 24, 545-549.

Tokatlidis IS and Koutroubas SD (2004). A review of maize hybrids’ dependence on high plant populations and its implications for crop yield stability. Field Crops Research, in press.

Tokatlidis IS, Koutsika-Sotiriou M and Fasoulas AC (1999). Load of deleterious genes and stability of performance in maize. Maydica 44, 127-132.

Tokatlidis IS, Koutsika-Sotiriou M and Fasoulas AC (2001). The development of density-independent hybrids in maize. Maydica 46, 21-25.

Tokatlidis IS, Koutsika-Sotiriou M, Fasoulas AC and Tsaftaris AS (1998). Improving maize hybrids for potential yield per plant. Maydica 43, 123-129.

Previous PageTop Of PageNext Page