Altered Maize Endosperm Adp-glucose Pyrophosphory- Lases from Revertants

Four phenotypically wild-type seeds were obtained from separate Actiuatorinduced events in the Dissociation-inhibited allele sh2-ml (shrunken-2, mutable1 ) . Endosperm adenosine diphosphoglucose pyrophosphorylase, the enzyme controlled by sh2, was extracted and partially purified from the four revertants and was compared to enzyme produced by the progenitor Sh2 allele and the sh2-m allele. The revertants contained 50 to 140% of the activity conditioned by the progenitor allele. Each of the revertants appears to be unique as judged by differences in K m (glucose-1-PO,), 3-phosphoglycerate (3-PGA) activation, and phosphate-inhibition. In one case the reversion event apparently increased the sensitivity of ADP-glucose pyrophosphorylate to 3-PGA activation. AIZE transposab'le elements are components of the maize genome that are M. recognized by their ability to move throughout the genome and inhibit gene function. In the Actiuator-Dissociation ( Arc-Ds) system, Ds inhibits the function of the gene at which it resides, while Ac acts remotely, promoting movement of Ds. Upon removal of Ds, gene function may be restored in the cell and its progeny. Thus, reversion events can often be seen as patches of wild type cells in a tissue generally lacking the gene product (MCCLINTOCK 1956). Several investigators have studied the effect of Ds on the protein affected by Ds inhibition of a locus. SCHWARTZ (1960) showed that the effect of association of Ds with the Shrunken-l ( S h l ) locus resembJed the effect of most spontaneous shl mutants in that the endosperm lacked the Sh l protein (later identified as SUcrose synthetase by CHOUREY and NELSON 1976). Association of DS with the sh2 locus conditions an ADP-glucose pyrophosphorylase with a K, (glucose-I -PO,) two to three times higher than that of the enzyme from. normal endosperm (HANNAH and NELSON 1976). DOONER and NELSON (1977) studied the effect of assmiation of Ds with the Bronze ( B z ) locus on the enzyme UDP-glucose: flavonoid 1 This research was supported by the Science and Education Administration of the U.S. Department of Agriculture under Grant No. 5901-0410-8-0054-0 from the Competitive Grants Office. Florida Agricultural Experiment Station Journal Series No. 3104. Parts of this work were taken from a thesis submitted by D . M. TUSCHALL in partial fulfillment of the Master of Science degree, University of Florida. 2 To whom correspondence should be addressed: Vegetable Crops Dept., I255 HS/PP Bldg., University of Florida, Gainesville, Florida 32611. Genetics 100: 105-111 January, 1980. 106 D. M. TUSCHALL A N D L. C. H A N N A H glucosyltransferase (UFGT) . They found that some Bz-Ds alleles conditioned UFGT which was more thermolabile than enzyme from wild-type allele. One allele exhibited an altered developmental profile for UFGT, and other alleles conditioned a complete absence of UFGT. The investigators concluded that DF could alter the coding sequence of the structural components of the Bz gene. Reversion of controlling element-inhibited genes has been studied at the biochemical level in only one instance. DOONER and NELSON (1979) showed that Ac-mediated reversion events of the Ds-inhibited Bz allele, bz-m2 (DZ) . gave rise to a t least two classes of UFGT molecules. Whereas enzymes from 5 of 15 revertants could not be distinguished from enzyme found in the wild-type endosperm, IJFGT from the remaining 10 revertants had reduced heai-stability and had much lcwer activity in mature seed. These data indicate that the reversion events mediated by Ac-Ds are not always identical since heterogeneity exists among alleles which condition ideniical (wild-type) phenotypes. In these studies we asked whether such Ac-Ds mediated heterogeneity might be found among revertants at another locus, Sh2. Endosperm ADP-glucose pyrophosphorylase, which catalyze; the synthesis of the starch intermediate ADPglucose from ATP and glucose-l-PO,, is under the control of Sh2 (chromosome 3 ) as well as Brittle-2 (Bt2) on chromosome 4 (TSAI and NELSON 1966; DICKINSON and PREISS 1969b). Several lines of evidence suggest that Sh2 is a structural gene for ADP-glucose pyrophosphorylase: ( 1 ) enzyme preparations from spontaneous, recessive sh2 alleles show allele-specific K, (glucose-1 -PO4) values and differ in urea stability (HANNAH and NELSON, 1976), (2) enzyme activity shows sh2 dosage-dependence (HANNAH and NELSON 1975) and (3) deletion of Sh2 abolishes all endosperm ADP-glucose pyrophosphorylase (HANNAH, TUSCHALL and MANS 1980). Apparently, endosperm ADP-glucose pyrophosphorylase is essential for viability since this deletion is a zygotic lethal, as are several bt2 and sh2 alleles which are known to condition the lowest detectable level of ADPglucose pyrophosphorylase so far observed ( HANNAH, TUSCHALL and MANS 1980). Multiple forms of endosperm ADP-glucose pyrophosphorylase have been detected by electrophoresis and sucrose gradient centrifugation ( HANNAH and NELSON 1975), kinetic characterizatim (HANNAH and NELSON 1976). and thermolability studies (HANNAH. TUSCHALL and MANS 1980). While molecular weight differences may be artifactual (HANNAH and NELSON 1975), characterization of two differentially heat-stable forms has been informative (HANNAH, TUSCHALL and MANS 1980). In wild type, 95% of the activity exists in a form which is labile at 57", and is destroyed by electrophoretic conditions. The heatlabile and heat-stable activities may be interconvertible, and have similar if not identical purification properties. Although deletion of Sh2 abolishes both fcrms. mutations of this locus primarily reduce the heat-labile form. Characterization of 11 spontaneous sh2, 10 bt2 alleles and two double mutants showed that in every mutant the ratio of heat-stable/total activity was much greater than that found in wild type (HANNAH, MANS and TUSCHALI, 1980). Thus. while both forms of enzyne are under the control of sh2 and bt2, the heat-labile form is reduced t3 a greater extent than the heat-stable form in all spontaneous mutants so far examALTERED ADP-GLUCOSE PYROPHOSPHORYLASES 107 ined. As noted previously (HANNAH, MANS and TUSCHALL 1980), these observations may suggest a nonstructural role for Sh2 and/or Bt2. For instance, Sh2 could control a series of similar, if not identical, post-translational steps. If this were so, we would predict that the heat-stable enzyme would (1 ) be an intermediate in the path for the synthesis of the heat-labile enzyme, (2) have a K, (glucose1-PO,) higher than that of heat-labile enzyme and (3) contain at least one site modified by Sh2 and Bt2. This hypothesis predicts that in reversion events which restore wild-type activity at sh2 or bt2 the resulting enzyme molecules should also be wild-type. I n the studies described below we examined wild-type revertants induced by the Ac-Ds system. Some revertants contain wild-type enzyme activity levels, yet the enzyme molecules are kinetically altered. These observations strengthen the argument that sh2 is a structural gene for ADP-glucose pyrophosphorylase. Further, the data are in agreement with the conclusion of DOONER and NELSON (1979) that reversion events mediated by Ac-Ds are not identical and thus these transposable elements may be used to create heterogeneity within a locus. MATERIALS A N D M E T H O D S Gemtic Stocks: The allele designated sh2-ml was generated by HANNAH and NELSON (1976) from a stock containing Ds in the a locus (a-m3) and Ac somewhere in the genome. The method used to obtain the phenotypically wild-type revertants was described (HANNAH, TUSCHALL and MANS 1980). Because the original, wild-type kernel phenotype was restored, these alleles were termed revertants. These revertants, designated Sh2‘-2 through Sh2’-5, were detected in crosses of the Ac-containing stock a-m3 sh2-ml with either a sh2 inbred or a sh2 hybrid. Sh2‘-5 was from a cross involving the sh2 sweet corn ‘Florida Stay Sweet’ while Sh2‘-2, Sh2‘-3 and Sh2‘-4 came from crosses involving an inbred parent of this cultivar. The progenitor allele(Sh2-Prog) refers to the Sh2 allele before Ds was incorporated into the locus. It was supplied generously by 0. E. NELSON. The progenitor allele was used for comparison in order to minimize differences due to factors other than the passage of Ds through the locus. Endosperms analyzed enzymatically were homozygous for the specified allele except for Sh2’-5. In this case, endosperms resulting from self-pollination of ShZf-5/sh2 plants were studied. The recessive allele is that carried in ‘Florida Stay Sweet’ and is most likely sh2-R. Enzyme Preparation and Assay: Enzyme was prepared as described previously (HANNAH and NELSON 1975,1976; HANNAH, TUSCHALL and MANS 1980). Crude enzyme preparations were used to measure total, as well as heat-stable and heat-labile enzyme levels. Partially purified enzyme was used for K, (glucose-I-PO,), phosphate inhibition and 3-phosphoglycerate (3-PGA) activation determinations. Enzyme was assayed using the method of DICKINSON and PREISS (1969a) as modified by HANNAH and NELSON (1975, 1976). Heat treatment was previously described (HANNAH, TUSCHALL and MANS 1980). An enzyme unit is one micromole of ADP-glucose