Inheritance of Seedling Hydrocyanic Acid Potential and Seed Weight in Sorghum - Sudangrass Crosses 1

The hydrocyanic acid potential (HCN-p) of sorghum [Sorghum bicolor (L.) Moench] plants is recognized as a heritable trait, but previous tudies on the mode of inheritance of HCN-p have produced inconsistent results. The objective of this study was to investigate the inheritance patterns of seedling HCN-p and also of seed weight in reciprocal crosses of sorghum and sudangrass [formerly S. sudanense (Piper) Stapf]. Both traits were found to be inherited quantitatively. Generation means analysis indicated that additive genetic effects were most important for both seed weight (53% of variation) and seedling HCN-p (74% of variation). A maternal effect was found for both traits in the F~ and backcross generations. No evidence of this reciprocal effect was found in the F2, suggesting that cytoplasmic inheritance was not involved. A highly positive correlation between seed weight and HCN-p was found across all entries, across both parents, and in the F~ and backcross generations. However, correlation coefficients between seed weight and seedling HCN-p for individual entries, the pooled F2’S, or within types of seed parents in the F~ or backcross generations were generally nonsignificant. It was concluded that seed weight per se does not have a large effect on seedling HCN-p. Additional index words: Dhurrin, Prussic acid, Sorghum blcolor (L.) Moench, S. sudanense (Piper) Stapf. p REVIOUS studies of the inheritance of the cyanogenic glucoside, dhurrin [p-hydroxy-(S)-mandelonitrile-#-D-glucoside], in sorghum [Sorghum bicolor (L.) Moench] have not yielded consistent results. Nass (17) in 1972 reviewed published reports on the inhero itance of cyanogenesis in sorghum, various Lotus species, and white clover (Trifolium repens L.), and concluded that the situation in sorghum was more complex than that in other species. The studies agreed that the dhurrin content or hydrocyanic acid potential (HCNp) of sorghum leaves was a heritable trait, but disagreed on the presence or absence of dominance, dominance of high or low HCN-p, and the number of genes involved. Krauss (14), in a study not included in Nass’s review, concluded that HCN-p was conditioned by four genes with additive effects without dominance. Other studies conducted since Nass’s review reported different conclusions on the importance of additive and nonadditive effects (2, 4, 9). Recently, Gorz et al. (10) found that seedlings of sorghum lines KS8 and N32 were high in HCN-p, but the HCN-p of flag leaves from field-grown plants of KS8 was only about 1/10 as high as that of N32 flag leaves. A single gene pair was found to be primarily responsible for the large difference in mature leaves, and neither low nor high HCN-p was completely dominant. The lack of agree~ Contribution from the USDA-ARS and the Nebraska Agric. Res. Div., Lincoln, NE 68583. Published as Paper no. 8119, Journal Series, Nebraska Agaric. Res. Div. Research was conducted under Project 12-114. Recetved 7 Aug. 1986. -~ Former graduate research assistant, George Holmes professor of agronomy, and supervisory research geneticists, USDA-ARS, respectively. Published in Crop Sci. 27:522-525 (1987). ment among these studies was no doubt due in part to differences in lines or cultivars used, conditions of growth, types of tissue assayed, and the analytical procedures used. The primary objective of the present study was to investigate the inheritance of seedling HCN-p in crosses of grain sorghum and sudangrass [formerly S. sudanense (Piper) Stapf] lines. The sudangrass lines had been selected for low HCN-p; crosses of these lines to the high-HCN-p sorghum lines provided a wider range in HCN-p than was possible in most of the previous studies of HCN-p inheritance. Also, this study utilized the spectrophotometric assay described by Gorz et al. (8). This assay is independent of the activity of catabolic enzymes in the plant. MATERIALS AND METHODS Four inbred parental lines were used in this study. Two were lines from sudangrass populations, and two were grain sorghum cultivars. The two sudangrass lines, 81-1901-7 and 81-1904-74, hereafter referred to as 1901 and 1904, respectively, were selected primarily for low HCN-p. Both lines carried the ms3 gene for genetic male sterility (gms); this recessive gene caused 25 to 30% of the plants in each line to be sterile. Both sudangrass lines also restored fertility when crossed with cytoplasmic-male-sterile (cms) sorghum lines having A1 cytoplasm. The population from which line 1901 was selected has since been released and registered as NP25 low-dhurrin sudangrass germplasm (12). Line 1904 was selected from a population with most of its background from ’Piper’. Both 1901 and 1904 are very low in seedling HCNp compared to commercially available sudangrasses, and they are considered to be homozygous for low HCN-p. The two grain sorghums used were ’Redlan’ and ’Combine Kafir 60’ (CK60). A-lines (A 1 cytoplasm) and B-lines of these cultivars have been maintained for at least 10 yr by making paired crosses annually to produce A-line seed and also selfing the B-line plants used in the A-line crosses. The lines are considered to be homozygous for high seedling HCN-p. All possible crosses, including reciprocals, were made among the four parental lines. Crosses were made using gms plants from the sudangrass lines and cms plants from the sorghum cultivars as females. Use of these male-sterile plants allowed for production of a large number of FI seeds with minimal risk of contamination during crossing. The FI plants were selfed to produce F2 seed, and were also used as pollen sources in backcrosses to gms sudangrass and cms sorghum parental plants. To get backcross seed from the sorghum × sorghum F~’s, the cms hybrids were used as females, and pollen was taken from the parental sorghum B-lines. All crosses were made at the University of Nebraska Agronomy Farm, Lincoln, NE, in the summers of 1982 and 1983 or in the greenhouse in early 1984. All seedlings analyzed in the laboratory were grown from seed produced either in the field in 1983 or in the greenhouse in early 1984. In total, 52 entries were assayed in the laboratory, including 12 F~’s, 10 F2’s, 24 backcrosses, and 6 parental lines (1901, 1904, ARedlan, BRedlan, ACK60, and BCK60). Twenty-five seed packets were prepared for each of the 52