Journal of Heredity 96(5):614-617 (2005)
A Genetic Test of Bioactive Gibberellins as Regulators of Heterosis in Maize
D. L. Auger, E. M. Peters, and J. A. Birchler

Heterosis or hybrid vigor refers to the superior performance of hybrid plants relative to the better of the two parents. Traditionally the issue of heterosis has been posed in genetic terms such as dominance and overdominance (Birchler et al. 2003). Ultimately it is necessary to understand the molecular basis of heterosis.

It was proposed that gibberellin levels are responsible for the vigorous plant growth associated with heterosis (Paleg 1965; Rood et al. 1988; Sarkissian et al. 1964). This proposition was supported by the observations that inbred maize plants were more responsive to the application of exogenous gibberellin A3 (GA3), a synthetic analogue of gibberellin A1 (GA1), than are hybrids (Nickerson 1959; Rood et al. 1983, 1990). These results are consistent with the idea that inbreeding depression is due to an insufficiency of GA, while hybrids possess GA at near saturation. It was demonstrated that levels of GA1 in inbreds are much lower than in hybrids (Rood et al. 1988). At the time of these experiments it was believed that GA1 was the only version of GA that was bioactive in the regulation of maize shoot elongation (Phinney 1985). Subsequent evidence indicated that GA3, as well as another gibberellin, GA5, are endogenous to and bioactive in maize (Spray et al. 1996).

Any maize mutant that is deficient for producing or unresponsive to the presence of bioactive GAs is known as a dwarf. The dwarf 1 (d1) gene encodes a factor that is responsible not only for the conversion of GA20 to GA1 (Spray et al. 1984), but also for the conversion of GA20 to GA5 and GA5 to GA3 (Spray et al. 1996). The homozygous recessive d1 mutants are incapable of synthesizing any bioactive version of GA and exhibit a phenotype indicative of a reduced level of bioactive GAs (Spray et al. 1996). The d1 mutants are short, compact plants with shortened internodes, short wide leaves, and short erect tassels that exhibit difficulty in anther extrusion (Neuffer et al. 1997). In this study we tested the hypothesis that genetic modulation of GA levels affects heterosis. We developed near-isogenic lines that segregated for d1 and then produced hybrids that also segregated for d1. If modulation of the GA level is the underlying basis of heterosis, then the d1/d1 hybrids should experience little or no heterotic response relative to d1/d1 inbreds.


The effects of heterosis among the normal plants were apparent (Table 1). The only comparisons between an inbred and a hybrid that were not significant (P < .05) were for total nodes and ear node location between the B73 inbred and the B73/Mo17 hybrid and for the number of tassel branches between the B73 inbred and either of the hybrids. The effects of heterosis are also apparent among the dwarves. Like their normal siblings, the B73 dwarves did not show significant differences for the total nodes and ear node location with the B73/Mo17 hybrid. The B73 dwarves also failed to show significant differences for silk emergence with either hybrid. All other differences between the dwarves of either of the inbreds compared to the dwarves of either of the hybrids were significant (P < .05).

The heterotic effects in dwarf genotypes appear to be no less than in the normal plants. Figure 1 shows the average measurements of the hybrids expressed as a ratio relative to the average of the inbreds. The average measurements for the hybrids were obtained by pooling data from all of the hybrids (B73/Mo17 and Mo17/B73) and the averages for the inbreds was obtained by pooling data from all of the inbreds (B73 and Mo17). For each trait, the differences between the means of the pooled hybrids and the inbreds were highly significant (P < .01 per t test). In no case do dwarves appear to suffer a decrease in heterotic effect. If anything, there seems to have been a slight increase in the relative advantage of hybrid dwarves over inbred dwarves. This was not expected since the hybrid dwarves grew in the same rows as their normal siblings that crowded and towered over them more than the normal siblings of the dwarf inbreds.


The heterotic response seen among the hybrid dwarves indicates that heterosis occurs without any limitation due to reduced amounts of bioactive GA. This finding is not consistent with the possibility that modulation of GA is a controlling factor in the heterotic response in maize. If modulation of bioactive GAs was the basis of heterosis, then the dwarves would have been severely restricted in the heterotic response.

Another recent study indicated that GAs do not underlie interspecific heterosis in poplars (Pearce et al. 2004). Although Populus trichocarpa with faster elongating shoots has higher GA concentrations than Populus deltoides, the hybrids possessed GA levels similar to the slower growing species, even though they exhibited heterosis for shoot elongation. Interestingly, among the F2 descendents there was a negative correlation between internode length and GA level.

We interpret the previous findings in maize (Paleg 1965; Rood et al. 1988, 1990; Sarkissian et al. 1964) to indicate that GAs may be a target of the heterotic response, if not an underlying mechanism. Considering heterosis at the cellular level, the growth potential, both in terms of cell size and rate of cell division, is controlled by rate-limiting factors. Clearly availability of resources such as energy, water, and essential nutrients are the ultimate limitations, but next in importance is the ability to obtain and metabolize these resources in an efficient manner. Inasmuch as hormones, such as GAs, can stimulate more effective metabolism of these resources, they are potential targets of heterosis. If all of the physiological mechanisms are working near maximum potential, then the application of exogenous GAs should have little stimulatory effect. If the metabolism of GA is reduced to a level where it becomes rate limiting for the processes it regulates, then the application of exogenous GA will have a stimulating effect on those processes. The dramatic effect of the application of exogenous GA to maize inbreds compared to maize hybrids indicates that it was a rate-limiting factor in the inbreds, but not in the hybrids. This interpretation could explain the results of Pearce et al. (2004). The observation that hybrid poplars had increased growth with lower GA concentrations than the faster growing parental genotype indicates that factors other than GA were rate limiting and were acted upon by heterosis. It is possible that, among the hybrid poplars, the levels of GA may have been down-regulated in order to achieve the optimal stimulatory signal. The present experiments indicate that GA modulation is not the major underlying basis of heterosis. Further work is needed to address the basis of hybrid vigor.