Multiple molecular forms of the gibberellin-induced α-amylase from the aleurone layers of barley seeds

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 224, No. 1, July 1, pp. 224-234, 1983

Multiple Molecular Forms of the Gibberellin-induced a-Amylase from the Aleurone Layers of Barley Seeds’ J. CALLIS

AND

TUAN-HUA

Department of Botany, University of Illinois, Received November

DAVID

HO3

Urbana, I&unk

9, 1982, and in revised form February

61801 14, 1983

A class of plant growth regulators, gibberellins, induce the synthesis of a-amylase (1,4-a-D-glucan glucanohydrolase, EC 3.2.1.1) in the aleurone layers of barley (Hordeurn vulgare L. var. Himalaya) seeds. The purified a-amylase is composed of multiple isozymic forms with indistinguishable molecular weights, but different net charges. These cY-amylase isozymes separate on isoelectric focusing gels into two groups, each containing multiple species. One group has an apparent isoelectric point (~1) of approximately 5.8 (the high pI group). The other group’s p1 values are around 4.5 (the low pI group). On some gels a small amount of protein focuses between the high and low pI isozymes. These proteins comigrate with the low pI isozymes upon reelectrophoresis. The synthesis of these two groups is temporally regulated. The high pI group is the dominant set of isozymes secreted from embryoless half seeds during the first two days of gibberellin administration. After four days, however, the major isozymes are those of the low pI group. This shift in isozyme pattern is due to a shift in their relative rates of synthesis. Peptide analysis of these two groups of isozymes with Staphylococcus aureus V3 protease and cyanogen bromide shows amino acid sequence differences. However, members within the same group have similar peptide patterns. Both groups of isozymes are synthesized in vitro in a wheat germ extract primed with poly(A)+ RNA isolated from gibberellin-treated aleurone layers. This indicates that the synthesis of the two groups of a-amylase isozymes is probably directed by two or more different populations of mature mRNA. A model that explains these observations and the available genetic information is that barley aleurone a-amylase isozymes are encoded by at least two sets of structural genes.

Gibberellins play an important integrative role during the germination of cereal grains such as barley and wheat. These hormones diffuse from their site of synthesis in the embryo to the aleurone layers, the living cells surrounding the starchy endosperm. There gibberellins induce the synthesis and secretion of a number of hydrolases as well as the secretion of preex-

isting enzymes (1, 2). These hydrolases mobilize the starch and protein stored in the endosperm for use by the growing embryo. Gibberellin control of starch hydrolysis occurs via the induction of cu-amylase (1,4a-D-glucan glucanohydrolase, EC 3.2.1.1). Little a-amylase activity exists in the dry mature seed. However, after 3 to 4 h of incubation with gibberellic acid (GAB, one of the gibberellins),4 isolated aleurone lay-

’ This investigation was supported by National Science Foundation Grants PCM 80-21632 and 82-40868. ’ Current address: Department of Biological Sciences, Stanford University, Stanford, Calif. 94305. a To whom correspondence should be addressed. 0003-9861/83 $3.00 Copyright All rights

0 1933 hy Academic Press, Inc. of reproduction in any form reserved.

“Abbreviations used: GAs, gibberellic acid; SDS, sodium dodecyl sulfate; IEF, isoelectric focusing; CNBr, cyanogen bromide; ~1, isoelectric point;

224

GIBBERELLIN-INDUCED

a-AMYLASE

ers start to synthesize cu-amylase de novo (3,4). By 24 h cu-amylase is the most abundant protein synthesized and secreted from the aleurone layer. This treatment with GA3 leads to the accumulation of total poly(A)+ RNA (5) and an increase in the level of translatable message for ar-amylase (6,7). Aleurone layers incubated with buffer alone synthesize little a-amylase and do not contain a significant amount of translatable cY-amylase message. Although aleurone cY-amylases cannot be separated on the basis of molecular weight, they have different net charges. Previous investigators, using native gel analyses, find multiple forms of cr-amylase, but disagree on the number of isozymes present (8-13). These multiple electrophoretic forms also differ in their biochemical properties (8, 9). Furthermore, Jacobsen and Knox (14) and Bog-Hansen and Daussant (15) have demonstrated the presence of at least two immunologically distinct a-amylase activities. Despite the above observations, the molecular basis for this heterogeneity is not fully understood. The present study was undertaken to further characterize and determine the source of heterogeneity of barley aleur’one cY-amylase. MATERIALS

AND

METHODS

Treatment oj‘ plant material. Barley seeds (Hordeum vulgare L,. var. Himalaya) were obtained from Washington State University, Pullman, Washington. Seeds from a single harvest year, 19’74, were used throughout the study. With few modifications, embryoless half seeds (aleurone layers with starchy endosperm) and isolated aleurone layers were prepared, treated with GA3 (Sigma), and homogenized as described by Chrispeels and Varner (16). Labeling conditions. Typical experiments used 10 isolated aleurone layers with 2 ml 20 mM sodium succinate buffer, pH 5.0, containing 20 mM CaClz in a 25ml sterile flask. They were incubated aseptically at 25°C in a reciprocal shaker at 120 rpm. Prior to pulse labeling with [“Slmethionine, buffer was removed aseptically and replaced with 1 ml of fresh buffer.

TEMED, tetramethylethylenediamine; hydroxyethyl)-1-piperazineethanesulfonic EGTA, ethylene glycol bis(fi-aminoethyl tetraacetic acid.

Hepes, 4-(2acid; ether) N,N’-

OF BARLEY

ALEURONE

LAYERS

225

Typical labeling conditions were 1 h with 50 &i/ml [SsS]methionine (sp act > 1000 Ci/mmol, New England Nuclear). After labeling, the buffer was removed and the layers rinsed with 1 mM methionine before being homogenized in 0.5 ml of the incubation buffer minus GA,. The homogenate was spun in an Eppendorf microfuge for 2 min and the supernatant (crude extract) was used. Pulse-labeled half seeds were treated the same way as isolated aleurone layers except that the starchy endosperm was removed aseptically immediately before the addition of label to facilitate uptake. The incorporation of methionine into protein was assayed by 5% trichloroacetic acid precipitation as described by Mans and Novelli (17). Purification of a-amylase. cY-Amylase was isolated from the incubation medium of GAa-treated half seeds. Under aseptic conditions, typical experiments used two 250-ml flasks with 250 half seeds each. The half seeds were shaken at 25”C, 200 rpm in the buffer described above. At designated intervals, the medium was decanted and the half seeds washed with a small volume of buffer. The half seeds in fresh buffer were then put back on the shaker. a-Amylase was purified essentially according to Rodaway (18). a-Amylase activity was assayed as described by Varner and Mense (19). Protein content was quantified according to Lowry et al (20). Sodium dodecyl sulfate (SDS) gel electrophoresis. SDS gels were prepared and run according to the method of Laemmli (21). Proteins were boiled for 2 min prior to loading. Native Laemmli gels were run like the SDS gels described above except that the SDS and boiling were omitted. Isoelectric focusing (IEF) gels. Slab isoelectric focusing gels of 1.5-mm thickness were used to analyze multiple samples simultaneously. The gel solution, 5% acrylamide/0.3% Tris/G M urea (Schwarz/Mann/ 2% Ampholine (LKB), after degassing was polymerized with the addition of TEMED and ammonium persulfate between two glass plates separated along the edges with a 1.5-mm thick rubber spacer as described by Righetti and Drysdale (22). Samples were adsorbed onto 4 X 8-mm pieces of Whatman 3MM paper. The paper was then placed on top of the gel surface 2 cm from the edge on the cathode side. Two strips of 7-mm wide 3MM paper, one wetted with 1 M NaOH and the other with 1 M H2S01 were placed along the entire length of the gel at its edge. The loaded gel was then transferred to a 4°C cooling plate that was covered with a thin layer of water to assure good gel-plate and cooling-plate contact. The gels were usually run for 4 h at a constant power of lo-20 W. The amount of power depended on the length of the gel. Before fixing, the pH gradient of the IEF gel was determined at l-cm intervals with a flat-bottom pH electrode (Beckman 39507 combination electrode). The electrode was calibrated with standard solution at 4°C while the gel was kept at 4°C. Ui (23) determined

226

CALLIS

that gels containing urea have measured pH values higher by 0.42 units than the actual values; so the p1 determinations for a-amylase were adjusted accordingly. For labeled samples, the proteins were acetone precipitated and resuspended in 9.2 M urea before loading. Nonradioactive samples were either loaded directly or treated as described for radioactive samples. When the same sample was run both on a SDS gel and an IEF gel, samples were processed essentially as described by O’Farrell (24). Fluorography. After fixing, SDS and IEF gels were prepared for fluorography. They were either treated with Enhance (New England Nuclear) or with 2,5diphenyloxazole (PPO) in dimethyl sulfoxide (DMSO) (25,26). They were then dried and exposed at -70°C to either Kodak SB-5 or SR-5 X-ray film. Protein cleavages. Partial proteolysis with Staphylococous aureUS V8 protease (Miles) was performed essentially as described by Cleveland et aL (27). Gel slices containing protein (5 X 8 mm in size) were cut from IEF gels. Bands of interest were localized by briefly staining adjacent lanes that contained the same protein sample. Further treatment was as described by Cleveland et al (27). Partial proteolysis of in vitro synthesized products was done similarly except that the gel slice was cut out of an SDS gel instead of an IEF gel. Chemical cleavage with cyanogen bromide (CNBr) was done according to Nikodem and Fresco (28). RNA isolation. RNA was isolated from 24-h GAStreated aleurone layers as modified from Higgins et al (6). A typical preparation used 200 aleurone layers in a 250-ml Erlenmeyer flask which was incubated as described previously. The aleurone layers were frozen in liquid Nz. ground to a dry powder, and then mixed with 0.1 M sodium glycinate, pH 9.510.01 M EDTA/O.l M NaCl/l% SDS/l% insoluble polyvinylpyrrolidone. Phenol extraction and ethanol precipitation of RNA was as described by Higgins et al. (6). RNA was resuspended after ethanol precipitation in 10 mM Hepes, pH 7.5/l mM EDTA10.1 mM EGTA/ 0.1% SDS (eluting buffer of oligo(dT) column). RNA samples were boiled for 3 min and cooled to room temperature in an ice-water bath before l/lOth vol of 5 M LiCl was added. The RNA was slowly passed through an oligo(dT) cellulose column that had been prepared as follows. Oligo(dT) cellulose (Collaborative Research) (50 mg) was soaked in 5 ml of 0.1 N NaOH for 10 min. It was then loaded onto a small column and washed with 25 mM Hepes, pH 7.5110 mM EDTA/O.l mM EGTA/l% SDS/O.5 M LiCl until the buffer’s ODzso was zero. After the RNA solution was passed through twice, the column was again rinsed with the same buffer until the OD, was zero. Poly(A)+ mRNA was eluted from the column with 4 ml of eluting buffer (see above). This RNA was ethanol pre-

AND

HO

cipitated twice, lyophilized, and redissolved in sterile water at a concentration of 0.1 rg/pl for in vitro translations. In vitro translation. Wheat germ S-30 extract was prepared essentially as described by Roberts and Paterson (29). The wheat germ reaction components were as described by Kemper et al. (30) except that the K+ concentration was 50 mM. [?l]Methionine (7-12 PCi; sp act >lOOO Ci/mmol) was used in a 25-pl reaction. The incorporation into protein was assayed by hot 10% trichloroacetic acid precipitation. Soluble proteins were analyzed on SDS and IEF gels and fluorographed as described above.

RESULTS

Effect of GAS on the synthesis of a-amylase. The GA3 treatment of aleurone layers results in a substantial enhancement in the rate of cy-amylase synthesis as shown in Fig. 1. cY-Amylase is very stable; halflife estimates are on the order of 15 h (31). Therefore, [35S]methionine incorporation into a-amylase protein is probably an accurate reflection of its rate of synthesis. Isolated aleurone layers were pulse labeled for 1 h with [35S]methionine after 24 h of incubation in buffer (lane 1) or buffer with 1 PM GA3 (lane 2). The two lanes contain equal amounts of radioactivity. A comparison of the amounts of newly synthesized proteins indicates that GA3 enhances the rate of synthesis of a 45 kDa protein that comigrates with the secreted purified cY-amylase (Fig. 1, lane 3) and is immunoprecipitated by anti-a-amylase antibodies (data not shown). Heterogeneity of p&Ted wamylase. Purified cY-amylase, though a single band on SDS gels, separates into several bands when electrophoresed on denaturing IEF gels containing 6 M urea (Fig. 2 insets). Denaturing IEF gels are used in order to reduce the possibility of heterogeneity due to conformational isomers and/or varying degrees of denaturation and to get an estimate of the number of Lu-amylase isozymes present. Two pH gradients were used, a linear pH 3.5 to 10 gradient (Fig. 2B) and a gradient where the pH range from 4 to 6 was expanded (Fig. 2A). In most of the experiments the expanded pH 4 to 6 gradient was used since it gave greater separation of the isozymes. As the

GIBBERELLIN-INDUCED

1

2

wAMYLASE

3

FIG. 1. The SDS gel profile of newly synthesized proteins in barley aleurone layers. Both lanes were labeled for 1 b after 24 h in succinate buffer only (lane 1) or succinate buffer plus 1 PM GA3 (lane 2). Lanes 1 and 2 are the autoradiogram of an SDS gel containing equal amounts of radioactivity in each lane. Lane 3, Coomassie blue stained SDS gel of cY-amylase purified from the incubation medium of aleurone layers incubated with 1 pM GAS for 48 h.

graphs in Fig. 2 show, cu-amylase isozymes focus into two groups. One group, designated as the high pI group, consistently contains two major bands, labeled 1 and 2, and several minor bands. The two major high pI bands focus with apparent PI’S of 5.75 and 5.88. The low pI group consists of bands 3,4, and 5 with apparent PI’S of 4.48, 4.53, and 4.68, respectively. The PI’S of the isozymes and the ratio of the low pI to the high pI group are independent of major changes in preincubation conditions, including lo-fold increase in salt concentration, variations in pH from 3 to 12, and inclusion of a number of different denaturants. However, cu-amylase loaded near the anode does not focus, probably due to leaching of sulfuric acid from the anode

OF BARLEY

ALEURONE

LAYERS

227

(Fig. 4, lane 10). We, therefore, always load samples 2 cm from the cathode. The arrows in the insets to Figs. 2A and B correspond to a protein that copurifies with cY-amylase as described by Rodaway (18). It is a dimer of a 20-21 kDa monomer and has a pI of approximately 7. Neither the reason it copurifies with cu-amylase nor its physiological role are known. This protein is absent from the medium of 48 to 72 h GAS-treated aleurone layers. Thus, when necessary, cY-amylase free of the copurifying protein was obtained from the medium of half seeds incubated with GA3 for more than 48 h. Since the presence of this protein does not interfere with the isoelectric focusing of cy-amylase and since it can be separated from cY-amylase electrophoretically, the properties of this proetin were not investigated further. As seen in Fig. 2 insets, a small amount of Coomassie blue staining material is present between the two groups of isozymes. This is not due to incomplete focusing, since extended electrophoresis does not alter the pattern. The intermediate species are not the result of detectable proteolytic cleavage or degradation since they comigrate with the high and low pI groups on the SDS dimension of a twodimensional gel system (data not shown). To determine the relationship between the two groups of isozymes and the protein which focuses between the two groups, a lane of purified cy-amylase was electrophoresed on one IEF gel, cut out, rotated 90”, and reelectrophoresed on a second IEF gel (Fig. 3). Since the focusing pattern of the low pI group and the high pI group fall on the diagonal, they refocus at their original PI’S in the second gradient. However, the diffuse intermediate pI protein species migrates both on the diagonal and also to the position of the low pI group. This indicates that the intermediate pI species can be converted to members of the low pI group. The ratio between the intermediate and low pI groups can be altered by some specific sample treatments without affecting the isoelectric points of the high pI isozymes. Incubation of purified a-amylase at pH 2 (Fig. 4, lanes 8 and 9) or acetone pre-

228

CALLIS

,o _

Mostly

2

AND

HO

4-6

4

8

6

Distance

10~

0

Linear

“““““““2

(cm)

3.5-10

4

8

6

Distance

12

14

(cm?

FIG. 2. pH gradients and isozymie patterns of purified a-amylase on IEF gels. (A) Mostly pH 4 to 6 gradient; inset, Coomassie blue stained lane of purified a-amylase. (B) Linear pH 3.5 to 10 gradient; inset, Coomassie blue stained lane of purified a-amylase. The arrow denotes a contaminating protein that copurifies with cu-amylase.

cipitation in the presence of thiols (lane 4) or 100 mM CaClz (lane 6) result in a loss of the low pI group with a concomitant increase in the amount of the intermediate pI species. We interpret this to mean that the low pI group can be altered so that members focus at intermediate pH values. To eliminate the possibility that the het-

erogeneity of a-amylase is generated during its purification and to see if the isozymes secreted from the aleurone layer are found intracellularly, we have analyzed newly synthesized cu-amylase extracted from aleurone layers pulse labeled with [35S]methionine after a 24-h incubation with GAB. Figure 6B, lane 1, is a fluoro-

GIBBERELLIN-INDUCED

(x-AMYLASE

OF BARLEY

ALEURONE

LAYERS

229

FIG. 3. Coomassie blue stained two-dimensional IEF gel. Bottom horizontal lane, 26 fig of purified a-amylase onfirst dimension of IEF. It was cut out, rotated 99”, and run on a second IEF gel. The arrois indicate two faint spots in the low pH region.

gram of an IEF gel with such a crude extract. The prominently labeled bands comigrate with the purified, secreted isozymes with the exception of band 4 which

was only slightly labeled. The isozymes are labeled in pulses as short as 20 min; the amount of label in band 4 never increases even after extended labeling times (12 h).

FIG. 4. Effect of various treatments on the IEF pattern of purified a-amylase. Twenty fig of oamylase was treated as follows and loaded on a mostly pH 4-6 IEF gel. Lane 1, no treatment; lane 2, acetone precipitated and resuspended in either 9.2 M urea or H&; lane 3, boiled 2 min; lane 4, acetone precipitated and resuspended in 9.2 Y urea/5% @-mercaptoethanol; lane 5, 196 mru CaCla treatment; lane 6, acetone precifiitation with I0S mM CaCl,, resuspended in 9.2 M urea; lane 7, acetone precipitation with 100 mru CM&, resuspended in HaO; lane 8, supernatant of pH 2 treatment; lane 9, pellet of pH 2 treatment resuspended in 9.2 M urea; lane 19, sample loaded 2 cm from anode . (on the left).

230

CALLIS

AND

HO

B c

4

3

2

1

I.

FIG. 5. The IEF analysis of cY-amylase secreted from barley endosperm half seeds treated with 1 pM GAS for different lengths of time. (A) cY-Amylase synthesized and secreted between: lane 1, 0 to 24 h of GAS treatment; lane 2,24 to 48 h of GA3 treatment; lane 3,48 to 72 h of GA3 treatment; lane 4, ‘72 to 96 h of GA3 treatment. (B) Native Laemmli gel of purified ol-amylase. The label of lanes is the same as in A. (C) IEF gel of proteins extracted from the four bands on the native Laemmli gel as in B. Lane 1, total a-amylase; lane 2, top band of B; lane 3, second band of B, lane 4, third hand of B, lane 5, bottom band of B.

Developmental expression of a-amylase isoxymes. The relative proportion among the different a-amylase isozyme species depends on the length of incubation with GAS. With dry half seeds simultaneously hydrated and incubated in GAS, virtually all of the a-amylase purified from the medium after 24 h is the high pI group (Fig. 5A, lane 1). The low p1 proteins then increase, as a group, so that by 72 h the two groups are present in approximately equimolar amounts (Fig. 5A, lane 3); by the fourth day the amount of high pl isozymes present declines (Fig. 5A, lane 4). The low pI proteins can be detected for 2-3 more days until the half seed stop secreting LYamylase (data not shown). The same samples run on native Laemmli gels (Fig. 5B) also show a developmental shift from lesser anionic proteins to more anionic proteins. Individual protein bands were eluted from the gel and rerun on an IEF gel (Fig. 5C). The lesser anionic proteins correspond to the high pI isozymes and the more anionic proteins to the low p1 isozymes. There is a small amount of cross-contamination since it is difficult to cut out only one band. However,

this experiment demonstrates that, with the two gel systems, the same temporal shift in isozyme pattern occurs; from the presence of predominantly the high pl group early during GA3 incubation to predominantly the low pI group after longer GA3 treatment. Aleurone layers are pulse labeled with [35S]methionine for 1 h at various times during GA3 incubation to distinguish whether changes in the secreted a-amylase population are due to differential expression or to selective secretion and/or release of individual species from the aleurone layer cell walls. Representative time courses are shown in Fig. 6 for both half seeds (Fig. 6A) and isolated aleurone layers (Fig. 6B). Arrows correspond to labeled proteins in the crude extracts which comigrate with purified a-amylase. Isolated aleurone layers do not incorporate labeled amino acids into protein after 48 h of GA3 incubation; so identical time points are not possible. For half seeds, the time course of newly synthesized cr-amylase roughly parallels the time course of secreted cY-amylase isozymes (compare Fig. 5A with Fig. 6A). At

GIBBERELLIN-INDUCED

a-AMYLASE

OF BARLEY

ALEURONE

LAYERS

231

FIG. 6. The IEF analysis of newly synthesized proteins in barley half seeds or aleurone layers treated with 1 PM GA, for different lengths of time. Each lane contains equal amounts of radioactivity. The arrows indicate the a-amylase isozymes. (A) Fluorogram of mostly pH 4 to 6 IEF gel with crude extract of [%]methionine-labeled half seed after: lane 1, 24 h of GAa treatment; lane 2,48 h of GA3 treatment; lane 3,72 h of GA, treatment; lane 4,96 h of GA3 treatment. (B) Same as in A except with pulse-labeled isolated aleurone layers after: lane 1,6 h of GAa treatment; lane 2, 18 h of GA3 treatment: lane 3,24 h of GAa treatment; lane 4, 48 h of GA* treatment.

24 and 48 h after GA3 treatment, both groups of isozymes are labeled (Fig. 6, lanes 1 and 2). However, by ‘72 h and extending through 96 h, only the low pI isozymes become labeled (Fig. 6, lanes 3 and 4). In contrast, isolated aleurone layers shift to the synthesis of only the low pI group by 48 h of GA3 incubation, 24 h earlier than half seeds (Fig. 6B, lane 4). The difference seems to be whether the aleurone layers are rehydrated then treated with GA3 (isolated aleurone layers), or hydrated and incubated with GA3 simultaneously (half seeds). Molecular nature of the heterogeneity of a-amylase isozymes. Because all the cy-amylase isozymes have nearly identical molecular weights, the sources of the charge differences observed are limited to either amino acid composition differences or posttranslation modifications. To detect protein sequence differences, the individual isozymes were cut from an IEF gel and subjected to electrophoresis in an SDS gel by the method of Cleveland et al (27). Lanes 1 and 2 are the peptide pattern from low pI proteins; lanes 3 and 4 represent members of the high pIgroup. The two isozyme groups have distinct proteolytic patterns, while members within a group have nearly identical patterns. The peptides obtained are relatively insensitive to the isozyme concentration, since gel slices with less protein give the same pattern (Fig. 7A, lanes 1 and 3). Increasing the protease concentration does not change the characteristic fragmentation patterns of either group, but does lead to an increase in the quantity of the lower-molecular-weight

peptides at the expense of the high-molecular-weight ones. Figure ‘7B is the same V8 protease digestion as in Fig. 7A, but run on a higher percentage acrylamide gel to improve resolution of the low-molecular-weight peptides. To verify the differences observed with V8 protease digestion, the individual isozymes cut from IEF gels as before were cleaved with cyanogen bromide. As shown in Fig. 7, each group of isozymes generates a unique fragmentation pattern while members of the same group produce nearly identical patterns. To test whether both groups of isozymes are synthesized in vitro, poly(A)+ mRNA from 24-h GAB-treated aleurone layers was isolated and translated in vitro in a wheat germ translation mixture (Fig. 8A). Poly(A)+ mRNA from GAS-treated aleurone layers directs the synthesis of a prominently labeled protein approximately 1.5 kDa larger than authentic aamylase (6) (Fig. 8A). This protein has been shown to be the precursor of a-amylase since it can be immunoprecipitated by anticY-amylaseantibodies ((6); data not shown) and its peptide pattern is characteristic of a-amylase (Fig. 8D). To analyze in vitro synthesized a-amylase, total in vitro translation products were separated by size and charge according to O’Farrell (24). Figure 8D shows the two-dimensional gel distribution of in vivo synthesized a-amylase which is used to compare with the in vitro synthesized proteins (Fig. 8C). Proteins corresponding to both the low and high pI a-amylase groups are synthesized in the in vitro translation

CALLIS

AND

HO

12

34567

FIG. 7. Peptide analysis of individual a-amylase isozymes. (A) The 12.5% acrylamide-SDS gel analysis of Staphylococcus aureus V8 protease digested individual a-amylase isozymes. Lanes 1 and 2, pattern of two low pI proteins; lanes 3 and 4, pattern of two high pI proteins. (B) Same as A except 15% gel. Lane 1, low pI protein pattern; lane 2, high pI protein pattern. (C) CNBr cleavage patterns on 15% acrylamide-SDS gel. Lanes 1 and ‘7, control; lane 2, hand 5 (see Fig. 2 for notation); lane 3, band 4; lane 4, band 3; lane 5, band 2; lane 6, band 1. Lanes 2 through 4 represent the low pl isozymes, lanes 5 and 6 the high pI group.

low p1 group so that it focuses at the intermediate pH values. Finally, the isoelectric focusing pattern one observes for secreted a-amylase is identical, with the exception of one isozyme, to the IEF pattern observed for aleurone layers pulse labeled with [?S]methionine. For these reasons, we feel that the separation of Ly-amylase into at least two groups of isozymes is an accurate reflection of in viva heterogeneity. The number of species within each of the groups is somewhat variable. The bands specified in Figs. 2A and B are always observed; others, especially the minor bands in the high pI group, appear irregularly. The reason for this is not understood. Rodaway (18) determined the carbohyDISCUSSION drate content of a mixture of a-amylase In this study, we have investigated the isozymes to be approximately 0.5 moleheterogeneity of GAS-induced cu-amylase cules each of mannose, glucose, and N-acefrom barley aleurone layers. Purified a- tylglucosamine per molecule of a-amylase. amylase can be classified into three groups Since the percentage of carbohydrate is so based on their ~1’s. The intermediate group low, it is unlikely that glycosylation is rewe believe to be an altered form of the low sponsible for the proteolytic and chemical pI group. This conclusion is based on the cleavage pattern differences. The unique observations that the intermediate pro- cleavage patterns suggest amino acid seteins comigrate with the low pIgroup when quence differences and therefore the presreelectrophoresed and the ability of cer- ence of at least two unique proteins. In tain preincubation conditions to alter the support of this are the -in vitro translation system primed with mRNA, isolated from GAs-treated aleurone layers. This same conclusion was reached when in vitro synthesized cy-amylase is digested with V8 protease (Fig. 8D). In the lanes containing in vitro synthesized cw-amylase, cold carrier a-amylase is added to increase the chemical quantity present for protease digestion. The cold a-amylase digestion pattern is a composite of the low and high p1 groups (compare Fig. 8D, lane 5 with lanes 2 and 3). The fluorogram of V8cleaved in vitro a-amylase is nearly identical with its cold carrier; therefore it contains peptides found in both the high and low p1 groups.

GIBBERELLIN-INDUCED

a-AMYLASE

OF BARLEY

ALEURONE

LAYERS

233

peptides from various secretory proteins show little sequence homology, but all contain a high percentage of hydrophobic amino acids (34). Thus, the exact amino acid composition for the leader sequence of cY-amylase cannot be predicted from the sequence of known leader sequences. We have observed, however, that the difference in charge between in vivo and in vitro synthesized cu-amylases is small. The GAS-induced expression of cY-amy2 3 1 lase in barley aleurone layers involves the presence of at least two sets of poly(A)’ mature mRNA species. The two most likely mechanisms are: products of two different loci or the result of differential RNA processing of initial transcripts from the same DNA sequence. An example of the latter mechanism is the concomitant expression of IgM and IgD from the same B cell (35). Initial transcripts are processed into two 3456 12 different mature messages to yield the two distinct classes of immunoglobulins. It has FIG. 8. Gel analysis of in vitro synthesized o-amalso been reported that a single cu-amylase ylase. (A) Fluorogram of in vitro synthesized prodgene in mouse can specify two different ucts of SDS gel Lane 1, purified labeled cY-amylase; mature mRNAs, each found in separate lane 2, products from wheat germ in vitro translation primed with poly(A)+ RNA from aleurone layers tissues (36). treated with 1 pM GA3 for 24 h; lane 3, same as lane However, we favor the first mecha2 with no added RNA. (B) Two-dimensional gel analnism-the presence of two loci. Previous ysis of purified a-amylase. The gel was stained with isozyme analyses in barley (37) indicate Coomassie blue and only the part of the gel containthat the segregation of one group of isoing cu-amylase is shown. The acid end of the pH grazymes follows single Mendelian inheridient is on the right and the alkaline end is on the tance. However, their data also indicate left. (C) The fluorogram of a two-dimensional gel the presence of another isozyme for which analysis of in vitro synthesized o-amylase. (D) Pepno variants are observed. This is compattide analysis with V8 protease of in vitro synthesized ible with the model that cY-amylase isoa-amylase. Lanes 1 and 4, fluorogram of V%proteasedigested in vitro synthesized a-amylase; lane 2, V8 zymes in barley are coded by two indeprotease pattern characteristic of the low pl isopendent loci. zymes; lane 3, V8 protease pattern characteristic of We do not understand the physiological the high p1 isozymes; lane 5, Coomassie blue stained basis for multiple aleurone cu-amylases. lane 1 showing peptide pattern of cold carrier used; Wheat has also been found to contain mullane 6, undigested a-amylase. tiple a-amylase activities whose appearance is temporally separated (33). The first wheat cr-amylase produced after GAB inexperiments. The wheat germ cell-free duction shows greater activity toward intranslation system does not possess the soluble polysaccharides. The other cu-amylcapacity to glycosylate proteins. The synase appears later. It is inactive against inthesis of both groups of isozymes in vitro soluble polysaccharides, but active against reduces the probability that post-transsmaller, soluble oligosaccharides. Since lational glycosylation is a source of the differences between the two groups of iso- starch molecules in the dry seed are mostly large insoluble ones, it is conceivable that zymes. It is not known how the presence of a the temporal expression of two types of CYleader sequence on in vitro synthesized LY- amylase would allow an efficient digestion -. amylase affects its isoelectric point. Leader of starch. Perhaps there is an analogous

234

CALLIS

phenomenon occurring in barley; however, whether barley cu-amylases vary in the substrate specificities remains to be determined. GA3 induction of a-amylase synthesis in barley aleurone layers during seed germination is a convenient system for the study of hormone-induced and developmentally regulated gene expression in higher plants. Models developed to explain GAS’s mechanism of action must be consistent with the existence of multiple, coordinately regulated genes for a-amylase. Further studies on a-amylase in barley aleurone layers promise to increase our knowledge of plant growth and development. ACKNOWLEDGMENTS We would like to thank Judy Willis for advice on running IEF gels, Julie Gross for preparing the wheat germ extract, and Charles Gasser for reviewing the manuscript. We are also grateful to Sheila Hunt for preparing this manuscript in a professional fashion. Note added in prooJ Since this manuscript was submitted, Jacobsen et al. ((1982) Plant PhysioL 70,16471653) have published a similar analysis on barley aamylase isozymes. REFERENCES 1. YOMO, H., AND VARNER, J. E. (1971) in Current Topics in Developmental Biology (Moscona, A. A., and Monroz, A., eds.), Vol. 6, pp. lll114, Academic Press, New York. 2. HO, T. D. (1979) in Molecular Biology of Plants (Rubenstein, I., Phillips, R. L., Green, C. E., and Gegenbach, B. G., eds.), pp. 21’7-240, Academic Press, New York. 3. FILNER, P., AND VARNER, J. E. (1967) Proc Nat. Acad Sci USA 58, 1520-1526. 4. Ho, D. T., AND VARNER, J. E. (1978) Arch Biachem Biophys. 187,441-446. 5. Ho, D. T., AND VARNER, J. E. (1974) Proc. Nat. Acad. Sci. USA 71, 4783-4786. 6. HIGGINS, T. J., ZWAR, J., AND JACOBSEN,J. V. (1976) Nature (London) 260. 166-169. 7. MUTHUKRISHNAN, S., CHANDRA, G. R., AND MAXWELL, E. (1979) Proc. Nat. Acad Sci USA 76, 6181-6185. 8. FRYDENBERG, O., AND NIELSEN, G. (1965) Hereditas 54, 123-139. 9. TANAKA, Y., AND AKAZAWA, T. (1970) Plant Physiol 46, 586-591. 10. JACOBSEN, J. V., SCANDALIOS, J., AND VARNER, J. E. (1970) Plant Physid 45, 367-371. 11. BILDERBACK, D. E. (1974) Plant Physiol 53, 480484.

AND

HO

12. MOMOTANI, Y., AND KATO, J. (1966) Plant PhysioL 41,1395-1396. 13. MACGREGOR, A. W. (1977) J. Inst. Brew., London 83,100-103. 14. JACOBSEN, J. V., AND KNOX, R. B. (1973) Plunta 112,213-224. 15. BOG-HANSEN, T. C., AND DAUSSANT, J. (1974) Anal Biochem 61,522-527. 16. CHRISPEELS, M. J., AND VARNER, J. E. (1967) Plant Physid 42, 398-406. 17. MANS, R. J., AND NOVELLI, G. D. (1960) Biochem. Biqvhys. Res. Cwnmun. 3, 540-548. 18. RODAWAY, S. J. (1978) Phytochemistry 17, 385389. 19. VARNER, J. E., AND MENSE, R. M. (1972) Plant Physid 49,187-189. 20. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. BioL Chem 193, 265-275. 21. LAEMMLI, U. K. (1970) Nature @v&on) 227,680685. 22. RIGHET~I, P. G., AND DRYSDALE, J. W. (1976) in Laboratory Techniques in Biochemistry and Molecular Biology (Work, T. S., and Work, E., eds.), Vol. 5, pp. 440-469, Elsevier, Amsterdam. 23. UI, N. (1971) B&him Biophys. Acta 229,567-581. 24. O’FARRELL, P. H. (1975) J. Bid Chem 250,40074021. 25. BONNER, W. M., AND LASKEY, R. A. (1974) Eur. J. Biochxm. 46,83-88. 26. LASKEY, R. A., AND MILLS, A. D. (1975) Eur. J. B&hem. 56, 336-341. 27. CLEVELAND, P. W., FISCHER, S. G., KIRSCHNER, M. W., AND LAEMMLI, U. K. (1977) J. BioL Chem 252. 1102-1106. 28. NIKODEM, V., AND FRESCO, J. R. (1979) Anal. B&hem. 97,382-386. 29. ROBERTS, B. E., AND PATERSON, B. M. (1973) Prop. Nat. Aead Sci USA 70,2330-2334. 30. KEMPER, B., HABENER, J., MULLIGAN, R., Porrs, J., AND RICH, A. (1974) Proc Nat Acad Sci USA 71,3731-3735. 31. HO, T. D. (1976) On the Mechanism of Hormone Controlled Enzyme Formation in Barley Aleurone Layers, Ph.D. Dissertation, Michigan State University, East Lansing. 32. MACGREGOR, A. W., AND DAUSSANT, J. (1981) J. Inst. Brew. Landon 87,155-157. 33. SARGEANT, J. G. (1979) in Recent Advances in the Biochemistry of Cereals (Laidman, D. L., and Wyn Jones, R. G., eds.), pp. 339-343, Academic Press, New York. 34. KREIL. G. (1981) Annu. Rev. B&&em. 50,317~348. 35. EARLY, P., ROGERS, J., DAVID, J., DALAYE, K., BOND, M., WALL, R., AND HOOD, L. (1980) CeU 20, 313-319. 36. YOUNG, R. A., HAGENBUCHLE. D., AND SHIBLER, U. (1981) Cell 23,451-458. 37. FRYDENBERG, O., NIELSEN, G., AND SANDFAER. J. (1969) 2. Pjunzenzuecht. 61,201-215.