Peroxidase isozymes revisited
-- Emil E. Khavkin and M. V. Zabrodina

Numerous temporal and spatial changes in peroxidase expression have been documented previously by Brewbaker, J et al. (J. Hered. 76:159, 1985) who employed flat PAGE with lithium-borate/citrate buffer systems. We found that vertical PAGE, with alkaline Davis, acidic Reisfeld and especially with neutral Taber and Sherman buffer systems, provided better resolution of peroxidase electromorphs. Besides, several previously unsurveyed elite inbreds and additional tissues (especially those of seedlings) were included in our study. Peroxidase loci are designated below according to the nomenclature by Brewbaker et al.

Anodal peroxidases. The slowest electromorphs were produced by the px7 and px8 loci, and we could not always discriminate these isozymes properly. Among investigated inbreds, intermediate px7 electromorphs prevailed, though more than one band could be found in several genotypes. px8 was expressed in most tissues investigated, with the exception of endosperm and anther, and its activity in pollen sometimes was rather low. The highest px8 expression was found in young leaves. A two-banded pattern of this zone was observed in most genotypes. Inbreds C103 and Va35 exhibited faster electromorphs than most other genotypes, and both fast and slow allelomorphs were present in ICA L223 and WF9.

px3 was the most universally expressed locus, yet in anther its activity was sometimes too low to be detected. There were two or three allelic variants of this isozyme: fast in A239, B14, B37, B73, C103, Oh43, Va35 and W64A, apparently intermediate in Tx601 and slow in A188 and ICA L223, and this evidence did not always comply with the scores by Brewbaker et al. A double-band pattern was found in most inbreds. We also found several distinct minor bands of peroxidase activity between px8 and px3 zones with wider tissue specificity than the root-specific product of px11 described earlier in this region of electropherograms.

The weakly stained px6 zone was easily discerned in embryo axis, leaf and root, and traced in scutellum. More mobile zone in this region could be px10 product, though tissue specificity of this zone and px10 did not completely coincide.

A double-banded px2 pattern was found exclusively in already mature pollen; we were not able to trace this isozyme even in the pre-shedding anthers. Two allelic forms were found among screened inbreds: the fast one in A188, A344, A619, A632, A654, A682, B89, Sd41 and 517/3, and the slow one in A677, A681, B91, Pa91 and W23, while both allelomorphs were present in several A632 and A654 plants.

Mobile bands expressed by px9 were present in all embryo tissues and were especially active in the root as compared to the leaf. px12 product was found only in the root. px12 exhibited allelic polymorphism (apparently a faster band in Oh43 and Va35, and a slower band in A188, B37, C103 and ICA L223).

Cathodal peroxidases. The two most spectacular features of the cathodic pattern were the complete absence of peroxidase staining in mature pollen and a characteristic band in the endosperm and scutellum that apparently was not described previously. It was more mobile than the px1 product. The expression of this fast electromorph in the scutellum rapidly increased from the 1st to the 6th day of germination and slowly declined later, while in endosperm the decrease was already evident by the 3rd day of germination.

With the exception of pollen, px1 was the second most common peroxidase locus producing heavily stained double bands in seedling tissues. In addition to data by Brewbaker et al., active px1 bands were found in the endosperm and scutellum, and weaker but quite distinct expression was characteristic of the anther. In the scutellum this band manifested rapid growth of activity in the course of germination, while in the endosperm and leaf it degraded as seedling growth progressed. A fast px1 allele prevailed (A188, A239, B14, B73, ICA L223, Oh43, Tx601 and apparently WF9) over intermediate (Va35) and slow (C103, W64A and apparently B73) alleles. Upon tissue vs. genetic variability in the px1 pattern, we suggest that this zone comprises the products of at least two loci producing different combinations of fast and slow electromorphs.

The next zone towards the cathode, with two bands well-resolved in acidic gel, was apparently produced by px5. This isozyme was monomorphic in our sample of inbreds, very active in the axial tissues of the seedlings, including leaf, as well as in different tassel tissues, but it stained weakly in endosperm. Contrary to published evidence, we did not observe px5 expression in mature pollen.

The px4 pattern of most tissues consisted of two bands, sometimes poorly resolved in neutral gel. This locus was extremely active in shoot and root tissues of the seedling, less prominent in scutellum and tassel tissues, and practically silent in endosperm. We could not confirm the data on allelic polymorphism reported earlier.

In acidic gel, axial tissues of the seedlings displayed one more cathodal peroxidase band of little or zero mobility, and we are inclined to relate this band to px7. The band was absent from endosperm and scutellum as well as from tassel tissues. In alkaline and neutral buffer systems this band moved slowly towards the anode (see above).

Brewbaker et al. had already described tandem (dimorphic) bands in px3, px6 and px8 patterns. Two or more bands of monomeric peroxidases in most other loci investigated (px1, px4, px5 and especially evident in px2 and px9) suggest that this phenomenon can be more common, presumably resulting from gene duplication. Another frequent phenomenon, two allomorphs present in the same inbred individual, presents a more difficult problem. 

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage