We have been screening the maize genome for genes which express dosage effects on the content and composition of the oil of the kernel, and we have reported the identification of such factors (Plewa and Weber, Can. J. Genet. Cytol. 15:313; Genetics 81:211; Shadley and Weber, Can. J. Genet. Cytol. 22:11; MNL 55:126). Nearly all of the oil of the kernel is located in the embryo. The percent oil of the kernel can depend upon the amount of oil present in the embryo, the size of the embryo and the percent embryo of the kernel. Because of the negative correlation between percent oil of the kernel and the levels of linoleic acid of the oil (Alexander, Corn and Corn Improvement, 1977) these factors may also affect the composition of the oil. Therefore, we were interested in determining if dosage of regions of the maize genome known to alter kernel oil also affected kernel and embryo traits.
Kernels from tester ears segregating for TB-5La, TB-10L(19), or one of three compound TB-5La-3L translocations (Shadley and Weber, MNL 54:89; 55:124) were analyzed. In each case kernels from an ear were weighed and hydrated for a 24 hr period. The hydration was necessary for measuring kernel and embryo volume using the procedure of Plewa and Weber (Can. J. Genet. Cytol. 15:313). A dissecting microscope equipped with an ocular micrometer and set at 10X total magnification was used. The length and width of each kernel and embryo were measured. A longitudinal section was then made parallel and adjacent to the embryonic axis. The portion of the kernel containing the embryonic axis was used to measure the kernel and embryo depth. All measurements were made at the greatest linear distance. The products of the three measurements were used as an estimate of embryonic and kernel volumes, and will be designated "embryo and kernel volume." The ratio of the embryo volume to the kernel volume was used to calculate the percent embryo. The measured portion of the kernel containing the embryonic axis was germinated. Chromosome counts were obtained from primary root tip meristematic cells using the standard Feulgen squash technique. All measurements made on kernels from a single ear were done during a single sitting. Students' t-tests were used for statistical analysis. One ear of each translocation type was analyzed. Tables 1-5 present the observations. For all B-A translocations analyzed, kernels with hypoploid endosperm and hyperploid embryos had significantly lower kernel weights, kernel volumes and embryo volumes than sibling kernels with hyperploid endosperm and hypoploid embryos (p = 0.001 in all cases). Where measured, kernels with hypoploid endosperms and hyperploid embryos also had significantly lower kernel weights, kernel volume, and embryo volumes than those with euploid endosperms and euploid embryos. However, when kernels with hyperploid endosperms and hypoploid embryos were compared to those with euploid endosperms and euploid embryos, no significant differences were found. For all B-A translocations analyzed, the values for percent embryo were not significantly different between any of the kernel types.
The lower weights of the hypoploid endosperm kernels are consistent with the findings reported by others (Birchler, Genet. Res. 36:111; Lin, Genetics 100:475; Beckett, Can. J. Genet. Cytol. 25:346) and are apparently due to a disturbance of the 2:1 maternal to paternal ratio for the chromosome segment whose dosage is being altered. The embryo volume in all cases was reduced in proportion to the kernel volume. The disturbance in the endosperm has an effect on the embryo through the reduction in kernel size. As a result the percent embryo of the kernel remains the same for all kernel types. The data in Table 2 show that there are no significant differences in the traits measured between kernels heterozygous for the translocation (10 10B B10) and those not bearing the translocation (10 10). Therefore, the translocation itself does not affect the traits measured, nor does the presence of the B chromosome.
Plewa (Ph.D. dissertation, 1975, Illinois State University) found no significant difference in percent oil of the kernel between kernel types segregating for TB-5La. Given this and the data in Table 1, the smaller hyperploid embryos must have a lower amount of oil than the other embryo types. The amount of oil they synthesize is proportional to their embryo size. Therefore, there are no gene(s) on this segment of 5L that express dosage effects upon oil concentration.
Plewa also reported in his dissertation that kernels with embryos hyperploid for TB-10L(19) had a significantly higher percent oil than their hypoploid embryo sibs. However, the milligram oil content in these two kernel types was the same. We found that the hyperploid embryos are smaller than the hypoploid embryos, but both make up the same percent of the kernel. The hyperploid embryos do not synthesize oil in an amount proportional to their size. A gene(s) on this segment of 10L which expresses dosage effects on oil concentration may be responsible for this. The hypoploid embryo kernels may not have a lower content of oil compared to the euploid or diploid due to their equivalent embryo size, which may compensate for the missing copy of the gene(s). Using monosomics, Plewa and Weber (Can. J. Genet. Cytol. 15:313) reported that kernels with embryos monosomic for chromosome 10 had significantly lower percent oil than their diploid sibs.
Altered dosages of the regions of the maize genome reported here have an effect upon kernel size. The effect of this upon the percent oil of the kernel and the oil content depends upon the presence of genes in these regions of the genome which express dosage effects upon oil of the kernel. In the case of TB-5La the small kernel yields a small embryo, but, because there are no genes expressing dosage effects on oil in this region of 5L, the size of the embryo determines the amount of oil present. Oil content is affected but not percent oil. In the case of TB-10L(19) a gene(s) expressing dosage effects upon oil overrides the effect of the small embryo. Percent oil is affected but not oil content. The small kernel effect does not prevent the detection of genes acting on oil of the kernel as long as one compares percent oil values.
J. D. Shadley and D. F. Weber
Return to the MNL 58 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage
