As part of a maize breeding program with the purpose of obtaining waxy maize commercial hybrids, a selection process by high and low endosperm protein content in and among inbreds was carried out.

This program was initiated in 1990 at the IFSC with the collaboration of CITIM. Four waxy maize foundational populations (SCV1, SCV2, WEM and FW) were used and characterized. S1, S2 and S3 inbreds were obtained from these populations using the ear-to-row method developed by Shull, and applying a divergent selection procedure. This way, ears with high and low endosperm protein content are identified in each generation, similarly to Illinois Selections. In our case, each generation of inbreds is obtained by selfings and their endosperm protein content is measured at the laboratory using the microKjeldahl procedure (A.O.A.C., 1981). Values over 10.1% are considered high protein content in endosperm defatted flours, whilst minor values are considered low.

On the other hand, crosses of low endosperm protein content inbreds x high and the reciprocal crosses were done. A similar process, but using heterozygotes, was also done. The F1's obtained, as well as both parents of the crosses, were measured for their endosperm protein content. The results obtained let us confirm our previous hypothesis about the inheritance mechanism for the trait under study and set down an elucidative genetic model (see MNL 69, 1995).

In Table 1, for each population studied, as well as for the Sn inbred lines derived from them, the average endosperm protein content, range of variation, standard deviation and variance are shown. In each early inbred generation there are low and high endosperm protein content genotypes. This confirms the heterozygous condition for the trait of the original populations. Then, and according to the genetic model opportunely presented, the populations SCV2 and WEM will denote in their endosperm the genotype a/a/A, as they show low protein content in the endosperm, but the S1 inbreds derived show variation for the trait. This way, the polar cells of the original populations could be aa or AA whilst the second antherozoid genotype will be a or A. The possible genotypes of the S1 generation obtained by selfing plants of the original populations could be: a/a/a; a/a/A; a/A/A or A/A/A. This fact would explain the variation found for the trait in S1 inbred lines.

The populations SCV1 and FW show high protein content. Thus, their endosperm genotype could only be a/A/A since in the S1 to S3 inbreds obtained by selfings it is possible to find individuals with high or low endosperm protein content.

It must be pointed out that in all the inbred generations derived from each original waxy maize population, individuals exist with a higher endosperm protein content than usual in commercial maize hybrids grown in Argentina. On the other hand, whether inbreeding normally causes depression for yields and plant traits, in the case of endosperm protein content it does not seem to do so. This is surely related to the inheritance mechanism of the trait and the divergent selection procedure used. Through consecutive selfing generations and selecting in and among the inbreds it is possible to increase the protein content in the endosperm.

Table 1. Descriptive statistics for endosperm protein content in the populations (So) and the inbreds derived from them.
 

Material	Gn	Average	S.D.	Variance	Range of Variation
SCV1	So	10.80
	S1	8.76	0.94	0.88	8.11-9.83
SCV2	So	8.14
	S1	8.12	1.46	2.14	6.07-10.16
	S2	10.22	1.09	1.19	8.25-11.65
WEM	So	8.73
	S1	9.21	4.60	21.19	5.96-12.47
	S2	11.70	0.70	0.49	11.21-12.20
FW	So	11.38
	S1	10.74	2.50	6.27	8.28-14.79
	S2	9.58	1.39	1.92	8.43-11.21
	S3	11.37	2.43	5.92	9.65-13.09

Return to the MNL 71 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage