Some insights into the genetics of flowering time

Heterochromatic knobs in maize exhibit a highly nonrandom geographical and racial distribution in spite of the apparent polymorphism for their number and size. The frequency of knobs, and even the size of a given knob in a particular race, is inversely related to the changes in latitude and altitude. Furthermore, the knob number is positively correlated with the time to reach maturity. Generally, low knob lines flower before high knob lines.

The maturity and flowering time in maize have long been considered quantitatively inherited characters. However, a large number of studies have failed to resolve the major controversy regarding the number of genes that differentiate parental lines with respect to flowering time; the estimates range from as few as two genes to as many as twenty-nine (see references in Bonaparte, Can. J. Genet. Cytol. 19:251-258, 1977).

For about 4 years, we have gained insights into the genetics of flowering time in maize. The focal point has been to elucidate the role of knob heterochromatin in determining the rate of floral development. Undoubtedly, the genetics of flowering time is not simple. We have been able to partition variously involved factors. Certain factors have proved critical in the analysis, especially inbreds and varieties with various genetic backgrounds that are essentially day-length neutral. The knobless genotypes with their F1's and F2's have been most important since they provided a simple genetic system.

Since the rate of development is related to the number of knobs, our first assumption was to consider the degree of delay in pollen shedding in relation to knob number. One might expect that the higher the knob DNA content, the longer the cell cycle would be and thus the slower the development. This simplest model proved to be incorrect in crosses using Zapalote chico, a high-knob Mexican race; the F1 hybrids flowered at more or less the same time as the knobless or low-knob parent. Similarly, the F1 hybrid between Zapalote grande (another high-knob and late maturing Mexican race) and the knobless Tama flint flowered at precisely the flowering time of its knobless parent during the 1st week of January, while Zapalote grande did not flower until the last week of February in our winter 1984-85 nursery at Molokai, Hawaii.

In our summer 1985 planting at Urbana, the knobless flints (Tama, Parker's and Wilbur's) were extensively crossed to a large number of inbreds of diverse genetic backgrounds. The parents and the F1 progeny from a multitude of crosses were grown out in our New Zealand and Molokai winter 1985-86 nurseries. With very minor exceptions, the F1 hybrid corresponded in flowering, within a day or two, with the knobless parent regardless of the knob composition, maturity group or the genetic background of the other inbred parent. Simply stated, the knobless genotypes are completely dominant over the genotypes with knobs and the knobless parent determines the rate of development in these hybrids.

Further analyses regarding the segregation of knobs in the F2 generation enabled us to understand the genetics. Initially, the segregation of knobs was followed in F2 progeny of crosses of Wilbur's knobless flint to four inbreds, KYS, A634, Oh43 and A619, each with 5 knobs. Not unexpectedly, "polygenic" inheritance-type data were obtained. The data from crosses of the knobless flints to low-knob inbreds made the picture clearer. The F2 progeny from such crosses did segregate in a discontinuous fashion, rather than continuous. The F2 progeny gave a bimodal distribution pattern when segregating for a single knob. The large F2 populations from selfed F1's involving the genetically related inbreds with known knob compositions (i.e., B14 x B37, B14 x A635, B14 x B73, and Mo17 X C103) were analyzed in 1986 plantings at Urbana. These are segregating for only one or two knobs and thus served as the model system. These and various other crosses exhibited a discontinuous and simple Mendelian inheritance pattern. From these data we conclude that each homozygous knob combination (e.g. 9S/9S) delays plant development by three to four days. A heterozygous knob has no delay effect. Thus, the delay in development and the later time of flowering is positively correlated with the number of homozygous knobs. In simple terms, it is not just the number of knobs nor the amount of knob DNA, but the number of homozygous knobs causing delays. The observation that homozygous knobs delay development is consistent with the association between knob number and the time to reach maturity for the races of maize. Our best data are for 4L, 7L and 9S knobs. Interestingly, these are the most frequent knobs in U.S. corn belt inbreds. Other knobs are being analyzed.

Since the amount of knob DNA is not related to the rate of development and thus its effect in delaying development is not via lengthening the cell cycle, our alternate hypothesis is that knob DNA affects development via controlling the expression of the bracketing genes by a cis-acting position effect. The examples of cis-acting position effect variegation showing spreading effect that might involve DNA binding proteins are well known in Drosophila and mammalian X-chromosome inactivation.

We propose the presence of a protein that specifically binds to 180-bp repeat of knob DNA, brings about a conformational change and is probably involved in condensation of the knob DNA which could shut the neighboring genes off by a spreading effect.

The proposed hypothesis could explain the complete dominance of the knobless genotype, the knob-mediated developmental delay in homozygous and its absence in heterozygous knob condition. Thus, the F1 hybrids between knobless and knobbed inbreds involve the co-existence of "two genomes" which differ temporally in their RNA and DNA synthetic activity. Since replication and transcription are temporally correlated, the knobless genotype should replicate early in the S phase, and so most of these newly replicated genes will be expressed. On the other hand, those genes closely linked to knobs will not replicate until the end of S when the knob DNA replicates (Pryor et al., Proc. Natl. Acad. Sci. 77:6705-6709, 1980).

If our working hypothesis on the role of knob DNA in gene expression is correct, then it might provide clues to understanding the dogma of hybrid vigor and combining ability. For example, in crosses between knobless (or low knob) and high-knobbed inbreds (e.g., Mo17 X B73 with good combining ability), the early replication and expression of knobless genome coupled with the late expression of knob-controlled genes will provide overlapping and extended periods of gene activity.

Studies are being actively pursued to test our various predictions, especially the role of knob heterochromatin in maize development.

Sajjad R. Chughtai and Dale M. Steffensen

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

Return to the MNL 61 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page