BGN 12: Evidence that two loci, Adh1 and Adh2, code for Alcohol dehydrogenase, the second being inducible by anaerobiosis BARLEY GENETICS NEWSLETTER, VOL. 10, II. RESEARCH NOTES
Harberd and Edwards, pp. 26-30

II. 11. Evidence that two loci, Adh1 and Adh2, code for Alcohol dehydrogenase, the second being inducible by anaerobiosis.

N. Harberd and K.J.R. Edwards, Department of Genetics, University of Cambridge, Downing Street, Cambridge CB2 3EH, United Kingdom. "R"

Alcohol dehydrogenase (ADH) activity is high in the scutellum and embryo, and a little lower in the endosperm of the barley grain. Upon germination, activity in the germinal tissue (scutellum and embryo) falls, and very little activity is to be found in two day old roots or coleoptiles. This is true of the older plant with the exception that ADH activity can be induced to fairly high levels in the roots by flooding them (Brown et al., 1978). The pollen also contains detectable levels of ADH activity.

Starch gel electrophoresis (Brown et al., 1978) was used to investigate the effects of anaerobiosis on the Alcohol dehydrogenases in the germinal tissues and endosperm. Anaerobic conditions were imposed by submerging the grains in deionized water at 20°C. In order to maintain constant conditions between experiments, grains were always submerged in pairs in 5 mls of deionized water in a glass vial. Water lost from the vials by evaporation was restored by topping up to volume each day.

In the dry seed, and during germination under aerobic conditions (on moistened filter-paper in a petri-dish) the ADH activity in extracts from both germinal tissue and endosperm can be resolved as a single anodally migrating band in starch gel electrophoresis. Recent investigations of electrophoretic variation in samples from wild populations of Hordeum spontaneum (closely related to, and widely regarded as the wild progenitor of the cultivated Hordeum vulgare) have revealed three electrophoretic variants of this band, designated F, M, and S (for fast, medium and slow migrating bands, respectively) (Brown et al., 1978). We have screened some thirty cultivars of Hordeum vulgare originating from various parts of the world. All were found to contain an M variant of this band.

The effect of submergence on ADH activity in extracts from germinal tissue and endosperm was investigated (Fig. 1). As before, extracts from dry seeds give a single ADH band. After only one day of submergence, however, the germinal tissue extract contains two new ADH bands. They migrate anodally and do so faster than the main band. After three days submergence these anaerobically induced bands are almost as intense as the main band. Interestingly, anaerobiosis does not induce any new bands in the endosperm, whereas extracts from flooded roots again give three distinct ADH bands in the same position as those induced by anaerobiosis in the seed germinal

Figure 1. Photograph of gel demonstrating the effect of anaerobiosis. Sample origin is underneath photograph. Samples: A-germinal tissue extract from dry seed, B-endosperm extract dry seed, C-germinal tissue extract from seed submerged 1 day, D-endosperm extract one day, E-germinal tissue extract 2 days, F-endosperm extract 2 days, G-germinal tissue extract 3 days, H-endosperm extract 3 days.

We have isolated three Alcohol dehydrogenase (Adh1) null mutants (Adh1-M9, Adh1-M140, Adh1-M146) from M2 seeds from Proctor barley mutagenized with azide (Kleinhofs et al., 1978), using selective and screening techniques to be described elsewhere. The response of seeds of these mutants to anaerobiosis maintained by submergence has also been investigated (Fig. 2). After three days submergence, and with prolonged gel staining, activity can be detected at the position of the fastest migrating band, but not in the other two positions. The intensity of this band is much lower than it is in extracts from wild-type grains submerged for three days.

Figure 2. Photograph of gel demonstrating the effect of anaerobiosis on Adh1 null mutants. Sample origin is underneath photograph. Samples: A-germinal tissue extract from wild-type (Adh1+) seed submerged 2 days, B-germinal tissue extract from Adh1-M9 submerged 2 days, C-germinal tissue extract from Adh1-M140 submerged 2 days, D-germinal tissue extract from Adh1-M146 submerged 2 days.

These data support the proposal of a two-gene model for the Alcohol dehydrogenases in barley analogous to that proposed by Schwartz (Schwartz, 1966) to explain similar findings in maize (Fig. 3). The 'major' ADH band found in scutellum, embryo and endosperm under normal conditions is specified by the Adh1 locus. This band is designated the Set I band. Under anaerobic conditions two new bands, the Set II and Set III bands appear. These can be explained by the anaerobic induction of the product of another locus, Adh2. ADH is known to be a dimeric enzyme in maize (Fischer and Schwartz, 1973), wheat (Hart, 1969) and pearl millet (Banuett-Bourrillon and Hague, 1979). In barley its existence as a dimer has been inferred (Hart, 1975). The existence of the Set II band can thus be explained by postulating a dimerization of the Adh1 and Adh2 gene products producing a heterodimer. The Set I and Set III bands represent homodimers of the Adh1 and Adh2 gene products, respectively.

Figure 3. Diagram of the 'two-gene' model for Alcohol dehydrogenases in barley, analogous to that proposed for maize (Schwartz, 1966).

It should be noted that the locus numbering system (Adh1, Adh2) adopted here is a deliberate reversal of that employed by Brown and coworkers (Brown et al., 1978). This has been done in order to allow the designation of the locus coding the 'major' ADH isozyme as Adh1, and to bring the model for ADH in barley into line with that explaining the very similar situation in maize.

The data presented here support this model. When the Adh1 null mutants are placed under anaerobic conditions only the Set III band is observed. The conclusion must be that both the Set I and Set II bands are affected by a lesion at a single locus (Adh1), as would be expected on the basis of the above model. It is possible to envisage a mutational lesion at Adh1 resulting in the production of an enzymatically inactive but normally dimerizing product. Anaerobiosis might then be expected to result in a visible (though low activity) Set II band as well as the Set III band. No such mutant has yet been isolated.

The model predicts that the Set II band should be equidistant from the Set I and Set III bands. That this is indeed the case can be shown by measuring the distances concerned on the photograph (Fig. 1). Brown and co-workers (Brown et al., 1978) have found electrophoretic variants of both Adh1 and Adh2 in Hordeum spontaneum. Their data indicate that the variants of Adh1 and Adh2 are all freely dimerizing with one another, and the patterns they obtain are those expected on the basis of a two-gene model.

As stated previously, the Set III band induced in Adh1 null grains by submergence is of much lower intensity than that obtained by anaerobic induction of wild-type grains. On the basis of the two-gene model the Set III band would, providing no other factors are involved, be expected to be more intense in Adh1 null grains than in wild-type grains since none of the Adh2 gene product is involved in the formation of a Set II band. The cause of this discrepancy is possibly explicable in terms of the effect of submergence on the overall seed physiology. After two days of submergence in the conditions noted above wild-type barley grains are clearly germinating with extended radicles. None of the Adh1 null mutants show any signs of germination under these conditions. A similar effect has already been described for Adh1 null maize kernels (Schwartz, 1969). It is possible that inhibition of Set III induction and inhibition of germination of Adh1 null barley grains during submergence are related phenomena.

Another interesting feature of these data is that submergence of wild-type grains results in induction of Set II and Set III bands in the germinal tissue but not in the endosperm. Clearly, induction of Adh2 is tissue specific.

Further investigation of this system, particularly into the effects of plant hormones on the Adh1/Adh2 balance is envisaged.

Banuett-Bourillon, F. and D. K. Hague. 1979. Genetic analysis of Alcohol dehydrogenase isozymes in pearl millet (Pennisetum typhoides). Biochem. Genet. 17:537-552.

Brown, A.H.D., E. Nevo, D. Zohary, and 0. Dagan. 1978. Genetic variation in natural populations of wild barley (Hordeum spontaneum). Genetica 49:97-108.

Fischer, M. and D. Schwartz. 1973. Dissociation and reassociation of maize Alcohol dehydrogenases: Allelic differences in requirement for zinc. Molec gen. Genet 127:33-38.

Hart, G. E. 1969. Genetic control of Alcohol dehydrogenase isozymes in Triticum dicoccum. Biochem. Genet. 3:617-625.

Hart, G. E. 1975. In vitro hybridization of Alcohol dehydrogenase subunits of Triticum and Hordeum. Isozyme Bulletin 7:41.

Kleinhofs, A., R. L. Warner, F. J. Muehlbauer, and R. A. Nilan. 1978. Induction and selection of specific gene mutations in Hordeum and Pisum. Mut. Res. 51:29-35.

Schwartz, D. 1966. The genetic control of Alcohol dehydrogenase in maize: Gene duplication and repression. Proc. Natl. Acad. Sci. 56:1431-1436.

Schwartz, D. 1969. An example of gene fixation resulting from selective advantage in suboptimal conditions. Am. Nat. 103:479-481.


NH acknowledges the support of a Science Research Council CASE award.

BGN 10 toc
BGN Main Index