Dual sources of abscisic acid in maize kernels

Abscisic acid (ABA) is the plant hormone most commonly implicated in dormancy induction events in plants, and the premature sprouting associated with the viviparous mutants of maize has long been suspected to result from an ABA deficiency in the embryos. That this is true was first reported by Brenner, et al. (Plant Physiol. 59:S76, 1977), who reported that 24 days after pollination (DAP) ABA concentrations in mutant embryos were only 10-50% as high as in normal embryos. In our experiments with 20 DAP embryos of vp2, vp5, vp7, vp9 and w3, ABA levels in homozygous mutant embryos were 20-37% as high as we determined for normal homozygous and heterozygous embryos segregating on the same ears, although older embryos had levels as high as 60% of wild type levels.

At least three alternative explanations would appear to exist for the reduced, but significant, levels of ABA found in viviparous embryos. The first of these relates to the biosynthetic pathway for ABA. Two pathways have been proposed, but neither has been critically established. The direct pathway suggests that ABA is derived from a 15-carbon precursor, farnesyl pyrophosphate. The indirect pathway suggests that ABA is derived from a 40-carbon carotenoid, probably violaxanthin. Since the viviparous mutants mentioned previously are all carotenoid-deficient, the observations concerning reduced ABA content of viviparous embryos could be explained if both pathways were operative in maize. This hypothesis presumes that ABA synthesis via the carotenoid pathway is blocked in the mutants, and that the ABA observed in the mutant embryos is synthesized via the farnesyl pyrophosphate pathway.

A second possible explanation is that ABA is actually synthesized via the carotenoid pathway in maize embryos, but the mutations are leaky. However, vp5 blocks the 11-12 C desaturation and accumulates phytoene, w3 blocks 11'-12'C desaturation and accumulates phytofluene while vp7 blocks the second cyclization reaction and accumulates g-carotene and d-carotene. Since we have not identified a-carotene, b-carotene, lutein or zeaxanthin, which are distal in the pathway to the intermediates that accumulate in the mutants and are normally present as relatively large pools, these mutants do not appear to be leaky with respect to carotenogenesis. Under these circumstances, the mutants could not be leaky for ABA synthesis if ABA is actually a terminal carotenoid derivative.

A third possible explanation is that part of the ABA in the embryo, "maternal ABA", is synthesized in the plant and moves into the developing caryopsis through the phloem transport system, and the rest of the ABA, "in situ ABA", is synthesized within the kernel. The large day to day and plant to plant variation in embryo ABA content which we have observed has caused us to suspect a maternal contribution for some time. We determined that 3H-ABA (18 Ci/mM) injected into the shank of the ear could be recovered from the kernels. While this showed that ABA could be translocated into the kernel, it doesn't show that such transport actually operates in the plant. However, when Karssen, et al. (Planta 157:158-165, 1983) reported dual sources of ABA in seed of Arabidopsis thaliana, we were convinced that a similar system exists for maize.

It finally occurred to us that a remarkably simple experiment, based upon the fact that ABA is rapidly metabolized, could be used to demonstrate dual origins for ABA in maize kernels. If viviparous mutant kernels or embryos are rescued, the albino seedlings can be cultured for a period sufficiently long to allow maternal ABA that was initially present in the embryos to be metabolized. Thus, such seedlings should not contain ABA if in situ ABA synthesis is completely blocked and all of the ABA in the mutant embryos is of maternal origin.

Table 1 summarizes the results of several experiments. Although the failure to detect ABA in homozygous mutant seedlings does not prove that no ABA is present, it argues very strongly that the ABA found in the mutant embryos actually is of maternal origin. The corollary of this argument is that the difference between ABA contents of wild type and mutant embryos from the same ear should provide a rough estimate of in situ ABA in wild type embryos.

In addition to establishing that the ABA in maize embryos has dual origins, our failure to detect ABA in homozygous, carotenoid-deficient seedlings supports the hypothesis that ABA is synthesized via the carotenoid pathway in maize.

Table 1.

J. D. Smith, F. Fong, C. W. Magill, B. G. Cobb and D. J. Hole

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

Return to the MNL 60 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page