Endosperm protein variation in maize "triads"
--Kuanhung R. Lin, A. J. Bockholt, Brian H. Taylor and J. D. Smith

The nutritional advantage conferred by the o2 allele has been well documented, but deleterious agronomic characteristics associated with opaque grain have inhibited its general acceptance by producers. Most of these characteristics appear to result from the soft, floury endosperm typical of opaque kernels. Consequently, several groups in the U.S., Mexico and South Africa have taken different approaches toward solving this problem by selecting for hard endosperm forms of o2/o2 lines.

Brian Larkins, B and his associates have reported that QPM populations developed by CIMMYT have high levels of (-zein (27kD), and their investigations suggest that (-zein levels may be more important than total endosperm protein in determining endosperm hardness. By itself, though, an increase in zeins is counter productive, since it reduces the percentage of lysine in endosperm protein.

We have analyzed a series of inbred "triads" to investigate the effects selection for hard endosperm may have on expression of the various solubility fractions of endosperm proteins. A triad represents 3 forms of an inbred line: normal, o2 and QPM. The o2 forms were derived by backcrossing W64A O2 (non-recurrent parent) to the standard inbreds at least 4 times. QPM forms were derived by selecting inbreds from CIMMYT QPM o2 populations previously selected for hard endosperm, and backcrossing the QPM inbreds (non-recurrent parent) to the o2 inbreds of each triad 2-4 times with selection for hard endosperm. To the best of our knowledge, the original CIMMYT lines also carried the W64A o2 allele. Although the triads are far from being isogenic, genetic variation among the forms of a triad should be significantly reduced for genes that are not associated with o2 or regions related to hard endosperm expression.

Table 1 presents data related to total endosperm protein, protein extracted for albumin, globulin, glutelin and zein solubility fractions, and lysine levels for the B73, Mo17, T224 and T232 triads.

Although variation in total protein and solubility fraction levels was observed both within and among triads, the low levels of albumin and globulin fractions suggest that they have relatively small effects, if any, on endosperm hardness and lysine content. Thus, if non-zein proteins actually contribute to endosperm hardness, it seems probable that this contribution stems from enhanced expression of glutelin proteins.

Table 1. Endosperm protein distribution and lysine levels in inbred triads. Total protein levels are presented as mg prot/g dry weight of endosperm. Solubility fractions are given as % total protein. Lysine levels are given as mole % of total protein.
Triad Total protein Albumin Globulin Glutelin Zein Lysine
B73 85.2 2.5 2.2 37.1 59.6 1.53
B73 o2 72.7 8.7 4.0 46.5 46.5 3.20
B73 QPM 81.7 3.3 3.1 44.9 48.7 2.69
Mo17 96.5 1.9 1.7 35.3 61.0 1.29
Mo17 o2 85.4 2.9 3.3 45.3 48.6 2.56
Mo17 QPM 98.2 2.9 2.7 48.0 46.4 3.04
T224 84.7 3.2 2.5 29.2 65.2 1.87
T224 o2 82.6 4.6 4.2 41.6 49.5 3.86
T224 QPM 94.8 3.0 2.5 39.3 55.2 2.17
T232 82.4 2.5 2.3 35.3 60.0 1.07
T232 o2 73.0 3.0 2.6 47.7 46.6 2.27
T232 QPM 81.7 3.0 3.2 46.3 47.5 2.19

Samples of protein for each solubility class were run on 8-25% SDS-polyacrylamide minigels using a Phast System gel electrophoresis unit (Pharmacia LKB) and staining with Coomassie blue. Band intensity, area and molecular weight were measured with a high resolution laser densitometer using Gelscan software (Pharmacia LKB). Relative intensities of selected zein and glutelin bands determined by this method are shown in Table 2.

Table 2. Relative intensities of selected zein and glutelin bands. Minor glutelin bands are not shown. The indicated intensities were measured by laser densitometer scans of Coomassie blue stained protein gels.
Mo17 0.37 0.35 0.26 0.14 0.33
Mo17 o2 0.19 0.29 0.45 0.20 0.35
Mo17 QPM 0.06 0.43 0.48 0.45 0.40
T224 0.39 0.31 0.28 0.20 0.18
T224 o2 0.09 0.21 0.45 0.33 0.32
T224 QPM 0.24 0.45 0.44 0.33 0.52
B73 0.25 0.23 0.30 0.15 0.25
B73 o2 0.11 0.18 0.46 0.16 0.47
B73 QPM 0.14 0.25 0.55 0.21 0.48
T232 0.35 0.38 0.32 0.18 0.19
T232 o2 0.18 0.25 0.44 0.26 0.30
T232 QPM 0.20 0.30 0.45 0.28 0.32

The intensity of the 22kD zein band in each line reflects the amount of a-zein present, while that of the 27kD bands reflects the amount of g-zein. No 27kD band was detected in the glutelin fraction, indicating that the extraction method used efficiently removed all the g-zein from the glutelin fraction. Both the Mo17 and T224 QPM lines showed substantially increased levels of g-zein, while the B73 and T232 QPM lines had only a slight increase, suggesting that in these lines the hard endosperm phenotype does not result from an increase in g-zein. All of the triads showed higher levels of 37kD, 45kD and 75kD glutelin bands in the o2 and QPM forms relative to the standard inbred. However, B73 and Mo17 QPM lines had sharply higher intensities of the 37kD and 45kD bands, respectively. These elevated bands may reflect the presence of specific QPM modifier genes in these lines, and might be useful as phenotypic markers for RFLP tagging. 

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page