Protein profiles in selected dosage series

In a previous report (Birchler, 1977, MGCNL 51:13) it was noted that glucose-6-phosphate, 6-phosphogluconate and isocitrate dehydrogenase activities were negatively correlated with the number of 1L chromosome arms present in a dosage series produced by TB-1La. These enzymes were increased in monosomics and decreased in trisomics relative to the disomic. Other enzyme activities were constant through the dosage series.

To test whether a similar phenomenon occurred on the protein level, aneuploids of several regions of the maize genome and their respective euploids were analyzed on SDS polyacrylamide gels to compare protein profiles. Electrophoresis and staining were by the method of Laemmli (Nature 227:680-685, 1970). Gels were 10% acrylamide. One scutellum was extracted per one ml of sample buffer. Fifteen microliters of extract were applied to each channel.

The regions studied were 1L, 3L, 4S, 5L, 7L and 9S. The dosage series for each region was generated as follows. For 1L, plants homozygous for an Adh-F (alcohol dehydrogenase) allele were crossed as females by males hyperploid for TB-1La (1 B1 B1 1B) and homozygous for an Adh-C allele. Monosomic kernels have only the Adh-F allele and originate from fertilization of the egg by the deficient sperm which is formed as a result of nondisjunction of the B1 chromosome at the second microspore division in B1 1B male gametophytes. The disomic scutella have Adh-F and Adh-C alleles present which form a three banded isozyme pattern in which the CC:CF:FF band ratio is approximately 1:2:1. These zygotes originate from fertilization of the egg by either a sperm with the normal chromosome one or a sperm from a 1B B1 pollen grain in which disjunction of the B centromere occurred. The trisomic scutella have one Adh-F and two Adh-C alleles present. The CC:CF:FF isozyme patterns of these scutella are in an approximate 4:4:1 ratio. These arise from fertilization of the egg by sperm carrying two B1 chromosomes as a result of nondisjunction at the second microspore division. Hyperploid (1 B1) pollen tubes are successful in competition at a frequency of less than 0.0002 (Birchler, unpublished) and would not contribute to the last class. Kernels were classified by subjecting an extract of a sliver of each scutellum to starch gel electrophoresis.

The dosage series for 3L was generated by crossing an a-m-1 A2 C C2 R-scm-2 female by a hyperploid male (3 B3 B3 3B) which carried the recessive a allele in the normal chromosome three. The B3 chromosomes carried the dominant A allele. This female line allows the expression of anthocyanin in scutellar tissue when crossed by another line possessing a dominant A allele. Since it is known that there is little recombination between the normal chromosome and the B3 translocated chromosomes and that hyperploid pollen duplicated for such long regions cannot successfully compete against haploid pollen, the following phenotypes are believed to correspond to the respective dosage of 3L: colorless embryo, purple aleurone = 1 dose embryo; purple embryo and scutellum = 2 dose embryo; purple embryo and colorless aleurone = 3 dose embryo; and colorless embryo and colorless aleurone = 2 dose embryo.

Aneuploids for 4S were generated by the following protocol. Females homozygous for sugary, su, were crossed by a hyperploid 4 B4 B4 4B plant homozygous for the starchy allele, Su. The sugary kernels in the progeny are segmental trisomics in the scutellum; the starchy kernels have either disomic or monosomic scutella. As a control on the effects of sugary endosperm on the scutellum, a backcross of Su/su X su/su was used as a source of segregating starchy vs. sugary kernels.

Aneuploids for 5L were generated in the following manner. Females of A A2 C C2 R-scm-122 pr were crossed by hyperploid 5 B5 B5 5B males which had pr on the normal chromosome 5 and Pr on the two B5 chromosomes. Since the pr locus is uncovered by TB-5La, kernels with pr scutellum and Pr aleurone are considered to be monosomic for 5L. Kernels with Pr aleurone and scutellum are considered to be disomics, resulting from union of a haploid egg with a balanced sperm from a pollen grain in which disjunction of the B centromere occurred. We note that if 5 B5 pollen tubes could compete with the euploid tubes, the resulting phenotype would be the same, but heavy pollinations were made to minimize this possibility. Kernels with Pr scutellum and pr endosperm are segmentally trisomic for 5L. This translocation was kindly supplied to me by Dr. J. B. Beckett.

Aneuploids for 7L were generated by crossing Esterase-1 N/N females by a hyperploid male 7 B7 B7 7B with E-F on the normal chromosome seven and E-S on the B7 chromosomes. The E locus is uncovered by TB-7Lb (D. Schwartz, personal communication). Monosomic kernels have only the E-N allele present. Disomic kernels are E-N/S and trisomic kernels are E-N/S/S. The latter two types of scutella can be distinguished from each other on the basis of isozyme band ratios. Moreover, the N/S/S class has an associated endosperm which is reduced in size.

Aneuploids for 9S were generated as follows. An A A2 C C2 R-scm-2 stock was used as a female for homozygous TB-9Sb also homozygous for C-I. This stock was kindly supplied to me by Dr. D. S. Robertson. In this case the colored scutellum, colorless aleurone class is considered monosomic; the completely colorless class disomic; and the colorless scutellum, colored aleurone class segmentally trisomic.

Scutella of the dosage series of these regions were examined on SDS polyacrylamide gels for the pattern of proteins in each aneuploid and the diploid. The number of discrete protein bands was > 60. Each dosage series produced a unique set of protein intensities. The most common effects were specific protein increases in monosomics, decreases in trisomics or a complete negative correlation with chromosome dosage. A positive correlation between protein intensity and chromosome dosage was observed for certain proteins in certain dosage series.

Of particular note from this study is the response of Protein-1, one of the major extractable scutellar proteins. Variants in molecular weight have been found for this protein by Drew Schwartz (personal communication). He has mapped the locus which encodes this protein to between bz2 and Adh on chromosome 1. In this study it was demonstrated that the gene for this protein was uncovered by TB-1La, by crossing a female with one molecular weight variant by a TB-1La stock homozygous for a second variant. The monosomics, as determined by the Adh genotype, have only the maternal variant. The intensity of this variant decreased through the 1, 2, 3 dosage series, with the monosomic intensity being the greatest. The paternally contributed variant increased slightly in the trisomic compared to the disomic. In the 1L dosage series in which Protein-1 did not differ in molecular weight between the maternal and TB-1La lines, there was little or no change in intensity through the dosage series. Thus, it is concluded that even though the locus for Protein-1 is uncovered by TB-1La and the number of structural genes is varied in a 1, 2, 3 relationship in the 1L series, there is a compensation such that the total amount of protein is similar in the monosomic, disomic and trisomic. The intensity of several other proteins is negatively correlated with 1L dosage but most of the protein species have similar intensities through the 1L series.

In the various other dosage series different protein species are affected. For some chromosome arms there are several proteins with altered intensities, while other dosage series only affect a few proteins. Two of the most striking cases are the aneuploids of 4S and 5L. The intensity of Protein-1 differs greatly between the trisomic scutella and the 1 or 2 dose scutella for 4S. The intensity of Protein-1 is much less in the trisomic scutella. Examination of starchy vs. sugary kernels from a segregating ear showed that this effect was not due to the sugary endosperm phenotype associated with the trisomic scutella.

Protein-1 is also negatively affected by the dosage of 5L. In this case the intensity of this protein was increased in the monosomic scutella and decreased in the trisomics relative to the euploid class. Dosage series of 3L, 7L and 9S had little or no effect upon the intensity of Protein-1.

Thus, there are at least two regions in the maize genome that produce a negative effect upon the level of Protein-1. In addition, the level of this protein is compensated in aneuploids of 1L. The basis of this compensation is not known but these results indicate that it occurs at least on the protein level. Whether the compensation of Protein-1 and ADH (see previous report) have a similar basis is not known. The compensation could be due to feedback control, a limiting factor produced elsewhere in the genome, a cancellation of a gene dosage effect by an inverse effect or some other basis.

As pointed out previously (1977, MGNL 51:18), multiple inverse effect regions can affect a single locus in any one tissue. In this case 4S and 5L produce a negative correlation between the intensity of Protein-1 and chromosome dosage. Furthermore, this study shows that monosomic increases are usually associated with trisomic decreases. This is an important consideration. It has been previously noted that there exist in the literature data indicating specific enzyme level reductions in trisomics of Drosophila, Datura and Hordeum. With only disomic and trisomic data it is always necessary to consider that such effects are due to the reduced vigor of trisomics or to a slight developmental or anatomical asynchrony. However, when the corresponding monosomic shows an increase for the same enzyme or protein that the trisomic decreases, such an interpretation is no longer possible, since monosomics are also almost always less vigorous than the disomic. Since there is no correlation with vigor or developmental stage, an explanation of the phenomenon must take into account the fact that there is a negative correlation of enzyme/protein level with chromosomal dosage. The only segmental monosomy data available are from maize.

James A. Birchler

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

Return to the MNL 52 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page