Plant Physiol. (1998) 117: 425-435

Characterization of SU1 Isoamylase, a Determinant of Storage Starch Structure in Maize¹

Afroza Rahman, Kit-sum Wong, Jay-lin Jane, Alan M. Myers, and Martha G. James^*

Department of Biochemistry and Biophysics (A.R., A.M.M., M.G.J.), and Department of Food Sciences and Human Nutrition (K.-s.W., J.-I.J.), Iowa State University, Ames, Iowa 50011

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

Function of the maize (Zea mays) gene sugary1 (su1) is required for normal starch biosynthesis in endosperm. Homozygous su1- mutant endosperms accumulate a highly branched polysaccharide, phytoglycogen, at the expense of the normal branched component of starch, amylopectin. These data suggest that both branched polysaccharides share a common precursor, and that the product of the su1 gene, designated SU1, participates in kernel starch biosynthesis. SU1 is similar in sequence to -(16) glucan hydrolases (starch-debranching enzymes [DBEs]). Specific antibodies were produced and used to demonstrate that SU1 is a 79-kD protein that accumulates in endosperm coincident with the time of starch biosynthesis. Nearly full-length SU1 was expressed in Escherichia coli and purified to apparent homogeneity. Two biochemical assays confirmed that SU1 hydrolyzes -(16) linkages in branched polysaccharides. Determination of the specific activity of SU1 toward various substrates enabled its classification as an isoamylase. Previous studies had shown, however, that su1- mutant endosperms are deficient in a different type of DBE, a pullulanase (or R enzyme). Immunoblot analyses revealed that both SU1 and a protein detected by antibodies specific for the rice (Oryza sativa) R enzyme are missing from su1- mutant kernels. These data support the hypothesis that DBEs are directly involved in starch biosynthesis.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

Starch is a reserve carbohydrate that accumulates in the storage organs of many higher plants. This storage compound consists of a mixture of two Glc homopolymers, amylopectin and amylose, in which linear chains are formed via -(14) glucosyl linkages and branches are introduced by -(16) glucosyl linkages. Starch synthesis in maize (Zea mays) occurs within the amyloplasts of endosperm cells during kernel development via the concerted actions of ADP-Glc pyrophosphorylase, starch synthases, and starch-branching enzymes (for reviews, see Preiss, 1991; Hannah et al., 1993; Martin and Smith, 1995; Nelson and Pan, 1995; Preiss and Sivak, 1996; Smith et al., 1996). In addition, selective removal of branch linkages by DBEs is proposed to play an essential role in the final determination of amylopectin structure (Ball et al., 1996).

Physical and chemical analyses of granular starch have led to a widely accepted model for amylopectin structure called the "cluster model," in which amorphous and crystalline regions alternate with a defined periodicity (for reviews, see French, 1984; Manners, 1989; Jenkins et al., 1993). Within amylopectin the crystalline component is composed of parallel arrays of linear chains packed tightly in double helices. Branch linkages, which account for approximately 5% of the glucosyl linkages in amylopectin, are located at the root of each cluster in the amorphous region. This periodic clustering of branches allows for the alignment of the intervening linear chains and their dense packing into crystalline regions, thus providing an efficient mechanism for nutrient storage. Undoubtedly, the enzymatic processes required to achieve this ordered spatial positioning of branch linkages and extension of linear glucans must be highly regulated and coordinated.

Mutations that alter or eliminate the cluster organization within amylopectin can provide clues to the molecular mechanisms that give rise to its structure. Such is the case with mutations of the maize su1 gene. Phytoglycogen, which accumulates in su1- mutant kernels, has twice the frequency of branch linkages as amylopectin, a shorter average chain length (average degree of polymerization is approximately 10 versus an average of 20-25 for amylopectin), and a significantly different chain-length distribution (Yun and Matheson, 1993). Thus, phytoglycogen is multiply branched and lacks the packed crystalline helices of amylopectin (Gunja-Smith et al., 1970; Alonso et al., 1995). These structural alterations cause the molecule to be water soluble, whereas amylopectin in endosperm cells is insoluble. Biochemical analysis has revealed that su1- mutants are deficient in the activity of a specific DBE (Pan and Nelson, 1984). This fact, correlated with the accumulation of phytoglycogen in su1- mutant kernels, suggests that the DBE participates in the organization of regularly spaced clusters within amylopectin. Similar evidence is available from sugary mutants of rice (Oryza sativa) and the STA-7 and STA-8 mutants of Chlamydomonas reinhardtii, all of which accumulate phytoglycogen and also are deficient in the activity of a DBE (Mouille et al., 1996a, 1996b; Nakamura et al., 1996a, 1996b).

The two types of DBEs that have been identified in plants are classified as pullulanases (also referred to as R enzymes or limit dextrinases) and isoamylases (Lee and Whelan, 1971; Manners, 1971; Doehlert and Knutson, 1991). Both types of enzyme directly hydrolyze -(16) branch linkages, but differ in their activities toward specific polysaccharides. Plant pullulanases hydrolyze both pullulan, a polymer of -(16)-linked maltotriosyl units, and -limit dextrins at much higher rates than amylopectin, but they have little or no hydrolytic activity toward glycogen. In contrast, isoamylases readily hydrolyze the -(16) branch linkages of amylopectin and glycogen, but do not act on pullulan. The DBE shown to be missing in su1- mutants of maize and rice is of the pullulanase class (Nakamura et al., 1996b; Pan and Nelson, 1984).

Both isoamylases and pullulanases are present in developing maize endosperm during the time of starch biosynthesis (Doehlert and Knutson, 1991), consistent with the genetic evidence indicating DBE involvement in the determination of amylopectin structure. The participation of a specific pullulanase or isoamylase in the biogenesis of kernel starch, however, has yet to be demonstrated directly. In addition to having potential biosynthetic functions, both types of DBE are believed to be involved in the degradation of endosperm starch after seed germination (Manners and Rowe, 1969; Toguri, 1991).

Molecular cloning of the maize gene su1 and the Su1 cDNA revealed that its polypeptide product, SU1, is similar in amino acid sequence to members of the -amylase family of starch-hydrolytic enzymes (Jesperson et al., 1993; James et al., 1995; Beatty et al., 1997). SU1 is significantly similar to Ps. isoamylase, with 32% identical residues among 695 aligned amino acids, although its relation to known plant or bacterial pullulanases is significantly less (James et al., 1995). This result is an apparent discrepancy with the finding that the particular DBE deficient in su1- mutant endosperm is of the pullulanase type (Pan and Nelson, 1984).

To resolve this discrepancy and gain a better understanding of the role Su1 plays in starch biogenesis, this study made use of two recombinant forms of the SU1 protein. Antibodies specific for SU1 were produced and used to characterize its native size, aspects of its subcellular location, and its expression pattern in developing endosperm. In addition, a nearly native-size recombinant form of SU1 was expressed in Escherichia coli, purified, and characterized in terms of its enzymatic properties. The results clearly demonstrate that SU1 is an enzyme of the isoamylase class and indicate that at least two distinct DBEs are deficient in su1- mutants as a result of a primary deficiency of SU1 isoamylase. Furthermore, SU1 is expressed in maize kernels during the period of starch production, consistent with the proposed biosynthetic role for this DBE.

	MATERIALS AND METHODS

Top Abstract Introduction Methods Results Discussion References

Vector Construction

View larger version (15K): [in this window] [in a new window]   Figure 1. Su1 cDNA expression constructs. A, Restriction map of the coding region of the Su1 cDNA. The region denoted "N-terminal extension" indicates the first 67 amino acids of the polypeptide, which is not represented in Ps. isoamylase in its alignment with SU1. B, Plasmid pAR2, comprising the 580-bp EcoRI-HindIII fragment of the Su1 cDNA cloned downstream of the GST coding sequence. The fusion gene is under the direction of the tac promoter. C, Plasmid pAR4, comprising the 2327-bp EcoRI-XhoI fragment of the Su1 cDNA fused to the S-tag sequence in pET-29(b)+. The fusion gene is under the direction of the T7 promoter. Restriction sites are indicated as follows: N, NcoI; H, HindIII; X, XhoI; E, native EcoRI; and E*, EcoRI introduced during cDNA construction.

Plasmid pAR4 was constructed to express biochemically active SU1 in E. coli. pAR4 contains the nearly full-length Su1 cDNA (nucleotides 202-2529) as an in-frame fusion with the 3 end of the S-tag sequence (a 15-amino acid peptide derived from RNase A) in the expression vector pET-29b(+) (Novagen, Madison, WI) (Fig. 1C). pAR4 was constructed by first replacing the 1.3-kb KpnI-XhoI fragment of the Su1 cDNA in pAR1 with the longer, 1.8-kb KpnI-XhoI cDNA fragment from pMJ99 (James et al., 1995).

The resultant plasmid was linearized with XhoI and partially digested with EcoRI to excise a 2327-bp fragment comprising the nearly full-length Su1 cDNA. This fragment was cloned into pET-29b(+) to create pAR4.

Protein Expression and Purification

The recombinant protein SU1r (for SU1 recombinant protein) was expressed from pAR4 in E. coli strain BL21(DE3)pLysS (Novagen) and purified as follows. A pure clone was grown overnight at 30°C in Luria-Bertani medium containing 30 µg/mL kanamycin and 34 µg/mL chloramphenicol, diluted 1:20 with fresh medium, and grown at 30°C until the A₆₀₀ reached 0.8 to 1.2. Cultures were cooled to room temperature, and expression was induced by the addition of 1 mM isopropylthio--galactoside and growth at room temperature for 16 to 20 h. Cells from 100 mL of induced culture were harvested by centrifugation, suspended in 10 mL of 20 mM Tris-hydrochloride, pH 7.5, 150 mM sodium chloride, 5 mM DTT, 0.1% Triton X-100, 1 mM PMSF (GIBCO-BRL), and 0.01 mM E64 (Sigma), and lysed by treatment with lysozyme (100 µg/mL) and sonication. Lysates were centrifuged at 39,000g for 20 min. SU1r was affinity purified according to the manufacturer's protocol (Novagen) by incubating the total soluble extracts with S-protein agarose beads, followed by cleavage between the S-tag sequence and SU1r with biotinylated thrombin. Residual biotinylated thrombin was removed from the purified protein by treatment with streptavidin-agarose.

Preparation of Polyclonal Antibodies

Nomenclature and Maize Stocks

Wild-type plants were in the W64A or Oh43 genetic background, as were plants homozygous for the reference mutation of the su1 locus su1-Ref (Correns, 1901). Plants homozygous for su1- R4582::Mu1 were of a mixed genetic background (James et al., 1995).

Isolation of Proteins from Maize Kernels

SDS-PAGE and Immunoblot Analysis

Enzyme Assays

Quantitative determination of reducing-end formation was performed according to the method of Fox and Robyt (1991). Assay conditions were the same as for blue-value assays, except that all substrate concentrations were increased to 5 mg/mL. Pullulan (Sigma) was included as an additional potential substrate. SU1r was present at 0.5 µg/mL for assays of its activity toward amylopectin and -limit dextrin, and at 1 µg/mL for other substrates. Products were assayed over a time course of 1 to 6 h; aliquots of 100 µL were removed after each hour, followed by inactivation of the enzyme with 50 µL of 1 M sodium carbonate. A baseline reducing value obtained for each substrate before the addition of the enzyme was subtracted from the reducing value measured at each time point. Maltose was used as the standard for reducing value. In preliminary experiments the rate of product formation obtained under these reaction conditions was in all instances shown to be approximately linearly proportional to the enzyme concentration (data not shown). Thus, the reactions were performed under conditions in which the enzyme was saturated with substrate.

Enzyme activity as a function of pH was determined from the reducing activity measured after incubating 5 mg of amylopectin with 0.5 µg of SU1r in 50 mM sodium citrate, 100 mM sodium phosphate buffers varying from pH 3.0 to 7.0; in 100 mM sodium phosphate, pH 8.0; or in 50 mM Gly-sodium hydroxide buffers at pH 9.0 or 10.0. Formation of reducing ends was assessed hourly during a 6-h time course, and specific activities were calculated. In a similar time course, the thermal optimum for SU1r activity toward amylopectin was measured at reaction temperatures varying from 15 to 60°C at pH 6.0. The thermal stability of SU1r was determined by preincubating the enzyme at temperatures from 30 to 60°C for 10 min, then assessing enzymatic activity at pH 6.0 at 30°C. The effect of divalent cations on SU1r activity toward amylopectin was determined by adding 10 mM calcium chloride, 10 mM magnesium chloride, or 10 mM EDTA to the pH 6.0 reaction buffer, and then measuring activity according to the reducing-sugar assay.

Analysis of SU1r Reaction Products

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Immunological Identification of Native SU1

View larger version (57K): [in this window] [in a new window]   Figure 2. Immunoblot analyses with anti-SU1. A, E. coli extracts. Anti-SU1 detects GST-SU1 in soluble extracts of E. coli cells bearing pAR2 only when GST-SU1 expression is induced from the tac promoter. B, Total kernel extracts. Anti-SU1 identified a polypeptide of approximately 79 kD in wild-type kernels of inbred W64A collected 20 DAP. This polypeptide was not detected in the indicated homozygous su1- mutant kernels. C, Soluble and granule-associated proteins. Anti-SU1 detected the 79-kD polypeptide in the soluble fraction of proteins from wild-type maize kernels, but not in the fraction containing granule-associated proteins (pellet).

Anti-SU1 identified a protein of approximately 79 kD in immunoblot analysis of total soluble extracts from maize kernels collected 20 DAP (Fig. 2B). This protein was not detected in extracts of kernels homozygous for either of two independent su1- mutations, the reference allele su1-Ref (Correns, 1901) or the transposon insertion allele su1-R4582::Mu1 (James et al., 1995). These data, together with previous characterizations of the cloned su1 gene (James et al., 1995), indicate that the 79-kD protein is native SU1. The 79-kD size of the protein is approximately that predicted for native SU1 based on the length of the Su1 cDNA, considering the likely removal of an amyloplast-targeting peptide.

Investigation of the physical association of SU1 with starch granules revealed that it is exclusively a soluble endosperm protein. Proteins from maize endosperm cells collected 20 DAP were separated into those bound to starch granules and those in the postgranule fraction. Immunoblot analysis indicated that SU1 is present solely in the postgranule fraction (Fig. 2C). These results suggest that SU1 acts at the surface of the granule or within the amyloplast stroma, rather than within the granule interior or tightly bound to a substrate that is integral to the granule.

The expression pattern of Su1 during the course of endosperm development was determined by immunoblot analysis using anti-SU1 and also by RNA gel-blot analysis (Fig. 3). Anti-SU1 detected the 79-kD protein in extracts from kernels harvested 10 to 30 DAP; this protein was present in nearly constant abundance throughout this time period (Fig. 3, bottom). However, the 79-kD protein was not detected in extracts from 6-DAP kernels. These results were supported by an analysis of Su1 transcript accumulation in kernels harvested at various times after pollination. Gel blots of RNAs isolated from kernels 8 to 20 DAP showed that the Su1 transcript was present at a relatively constant level during this period (Fig. 3, top and data not shown). Thus, SU1 is present in the endosperm coincident with the initiation of starch biosynthesis, and persists at a high level throughout the time period of starch accumulation.

View larger version (68K): [in this window] [in a new window]   Figure 3. Temporal expression of Su1 mRNA and SU1 protein. Top, RNA gel-blot analysis using a Su1 cDNA probe. Su1 mRNA was detected in total RNA isolated from wild-type kernels of maize inbred W64A harvested 8, 10, 12, and 14 DAP. RNA loading was standardized based on ethidium-bromide staining of rRNAs. Bottom, Immunoblot analysis with anti-SU1. SU1 was detected in wild-type kernels of inbred Oh43 harvested 10 DAP and later.

Expression of Recombinant SU1 and Purification to Apparent Homogeneity

Inducible expression of a fusion protein of approximately 75 kD was detected in both the soluble and insoluble fractions of crude E. coli cell extracts by SDS-PAGE, and was identified specifically as SU1r by immunoblot analysis using anti-SU1 (data not shown). Equivalent protein fractions from uninduced E. coli cells and also from induced E. coli cells harboring the empty plasmid pET-29(b)+ served as negative controls. SU1r was purified from the total soluble extracts by means of the affinity of the fused S-tag region for S-protein derived from RNase A (Novagen), and was released from the affinity matrix by cleavage with thrombin. Analysis of the purified material by SDS-PAGE revealed a single polypeptide of approximately 75 kD on a silver-stained gel, and immunoblot analysis with anti-SU1 also identified a 75-kD protein (Fig. 4). The expressed protein is approximately the size predicted by the Su1 cDNA sequence, and at this sensitive level of detection appears to be free of contaminating proteins.

View larger version (46K): [in this window] [in a new window]   Figure 4. Purification of SU1r. A, Silver stain of an SDS-PAGE gel. The gel contains approximately 0.5 µg of affinity-purified protein collected from E. coli cells bearing pAR4 grown in inducing conditions, and shows a single polypeptide of approximately 75 kD. B, Immunoblot analysis with anti-SU1. The gel analyzed was a duplicate of that shown in A. The purified 75-kD protein is recognized by anti-SU1.

SU1r Cleaves -(16) Branch Linkages

View larger version (13K): [in this window] [in a new window]   Figure 5. Absorption spectra of glucan-iodine complexes. The indicated substrates were incubated with SU1r, and aliquots of the reaction mixtures were combined with iodine/potassium iodide stain. Spectra were recorded before the addition of SU1r (dashed lines) and after 1 h of incubation (solid lines).

Substrates of SU1r were also identified by a quantitative colorimetric assay that measures increases in the reducing value of the reaction products relative to the substrate (Fig. 6). The formation of new reducing ends was assessed hourly during the course of a 6-h incubation period of various substrates with SU1r. Again, SU1r was shown to have significant activity toward amylopectin, with the number of reducing ends formed increasing linearly for the entire incubation period (Fig. 6). Hydrolysis of -limit dextrin proceeded in a similar manner but at a slower rate under these reaction conditions (Fig. 6; Table I).

View larger version (21K): [in this window] [in a new window]   Figure 6. Hydrolytic activity of SU1r. The indicated polysaccharides were incubated with SU1r, and the concentration of reducing ends present in the reaction mixtures was determined in terms of micrograms of maltose equivalents per milliliter. The amount of SU1r present in the reactions using glycogen and phytoglycogen was twice that present in the reactions using amylopectin and -limit dextrin.

Phytoglycogen and glycogen are also substrates of SU1r, although the rate of formation of new reducing ends leveled off after 5 and 3 h, respectively (Fig. 6). Significantly, there were no new reducing ends formed during the incubation of pullulan with SU1r, indicating that SU1r is unable to hydrolyze -(16) glycoside bonds in this substrate. These results were supported by TLC analysis of the products of the pullulan/SU1r reaction, which indicated that no maltotriose was released (data not shown). The specific activities of SU1r toward each of the substrates tested under these reaction conditions is provided in Table I, as is a comparison of the activities relative to that for amylopectin. The finding that SU1r has DBE activity toward each of the substrates tested except pullulan classifies it as an isoamylase rather than a pullulanase type of DBE. The Su1 product, therefore, is referred to as SU1 isoamylase.

SU1r activity toward branched cyclodextrins containing either glucosyl or maltosyl side chains was also tested. Cycloheptaose (also called -Schardinger dextrin) to which a glucosyl or a maltosyl branch was attached by an -(16) linkage (called Glc-cGlc7 and Glc2-cGlc7, respectively), was incubated with SU1r, and the reducing value of the product was measured over time. No increase was seen in the number of reducing ends after the incubation of either Glc-cGlc7 or Glc2-cGlc7 with SU1r, indicating that SU1r is unable to hydrolyze the -(16) linkage between the glucosyl or maltosyl side chain and the cyclodextrin (data not shown). These results suggest that SU1 requires a chain length of greater than two Glc units as a minimal substrate for its presumed hydrolytic activity. This interpretation is supported by TLC analysis of the products of the -limit dextrin/SU1r reaction, in which no maltose was released (data not shown). Maltotriose was observed in this TLC analysis, however, indicating that a minimum chain length of three Glc residues is needed for debranching by SU1r.

Properties of SU1r Isoamylase

View larger version (14K): [in this window] [in a new window]   Figure 7. pH and temperature optima for SU1r activity and stability. A, Activity of SU1r relative to pH. SU1r was incubated with amylopectin in buffers adjusted to pH values from 3.0 to 10.0, and activity was determined at 30°C as illustrated in Figure 6. B, Activity of SU1r as a function of temperature. SU1r was incubated with amylopectin at pH 6.0 at temperatures ranging from 15 to 60°C. C, Thermal stability of SU1r. SU1r was preincubated for 10 min at temperatures ranging from 30 to 60°C before its incubation with amylopectin at 30°C and pH 6.0.

Neither divalent cations nor sulfhydryl agents were required for the activity of SU1r. Although SU1r enzymatic reactions were conducted routinely in buffers devoid of divalent cations, the addition of 10 mM calcium or magnesium ions to the buffer did not alter SU1r activity (data not shown). SU1r activity was not dependent on residual calcium ions, because the addition of 10 mM EDTA to the reaction buffer did not alter the reaction rate. Maintenance of enzyme stability did require the addition of 5 mM DTT to the storage buffer.

Products of SU1r Hydrolysis

View larger version (39K): [in this window] [in a new window]   Figure 8. Chain-length distribution produced by SU1r and Ps. isoamylase. Amylopectin from waxy maize was debranched by SU1r or Ps. isoamylase for 1 or 24 h, respectively. Chain-length distribution was determined quantitatively by HPAEC-ENZ-PAD and normalized to the most abundant glucan chain.

Potential Regulation of a Second Maize DBE by Su1

View larger version (25K): [in this window] [in a new window]   Figure 9. Immunoblot analysis with rice R-enzyme antibody. Total protein extracts from kernels in the Oh43 inbred background were analyzed. The polypeptide of approximately 100 kD present in wild-type kernels was deficient in the homozygous su1-Ref mutant kernels.

Further support for the conclusion that a rice R-enzyme homolog is present in maize, provided by characterization of a nearly full-length cDNA sequence isolated from an endosperm library, will be described in detail in a paper in preparation (M. Beatty, A. Rahman, A.M. Myers, and M.G. James, unpublished data). This cDNA codes for a protein of approximately 100 kD that has high identity to the rice R enzyme (Nakamura et al., 1996a). Genetic mapping by restriction fragment-length polymorphism linkage analysis showed that the gene that codes for this pullulanase homolog is located on chromosome 2 (data not shown). This map location is distinct from that of su1, which is located on chromosome 4S (Neuffer et al., 1997).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

This report demonstrates that the product of the maize gene su1 is a polypeptide with an apparent molecular mass of 79 kD that is located exclusively in the soluble fraction of endosperm cells and is present during the time of starch biosynthesis. Furthermore, SU1 was specifically identified as an -(16) glucan hydrolase of the isoamylase class. Together with previous characterizations of the cloned su1 gene (James et al., 1995), the fact that the polypeptide identified by anti-SU1 antibodies is missing in various su1- mutant endosperms is confirmation that the cloned cDNA was correctly identified as a copy of the Su1 transcript. Thus, the phenotypic effects of su1- mutations must be explained as a result of the primary defect in the DBE characterized here as SU1 isoamylase.

An increase in the absorbance of a glucan-iodine complex after incubation with an enzyme is the standard means of demonstrating debranching activity; this measurement, the blue value, is known to increase with the linearity of the polysaccharide (Banks et al., 1971). In addition, a shift in the _max is known to reflect an increase in the number of linear chains. Increases in both the blue value and the _max were observed after incubation of amylopectin with SU1r, defining SU1 as a DBE. This conclusion was confirmed by analysis of the glucan chains produced after a relatively brief digestion of amylopectin with SU1r. The resultant chain-length distribution indicated complete debranching of amylopectin, and closely matched the distribution obtained by digestion with Ps. isoamylase.

Quantitation of hydrolysis rates revealed that the specific activity of SU1r toward amylopectin was significantly greater under these reaction conditions than that toward the -limit dextrin of amylopectin, phytoglycogen, or oyster glycogen. For -limit dextrin this effect may be attributable to the short lengths of the exterior branch stubs that remain in the dextrin after treatment of amylopectin with -amylase. Maltosyl stubs were shown directly to be poor substrates by the lack of activity of SU1r toward Glc2-cGlc7. This is also true for Ps. isoamylase, which cleaves maltosyl branches at a lower rate than maltotriosyl branches, and in general is more active toward longer glucan chains (Kainuma et al., 1978). Reduced activity of SU1r toward phytoglycogen and glycogen compared with amylopectin is also likely to result from the substrate structure, given that these molecules have shorter average chain lengths than amylopectin and relatively uniform distributions of branch-linkage positions. Overall, these results suggest that SU1r does not require the ordered branching structure of amylopectin to recognize a substrate, and that chains with as few as three Glc residues can be released by the enzyme. There was no observable activity of SU1r toward pullulan. Taken together, the data clearly indicate that SU1r is a DBE of the isoamylase class.

The recombinant form of SU1 analyzed in vitro exhibits a slightly smaller apparent molecular mass than the native protein (75 as opposed to 79 kD). The amino acid sequence of the mature N terminus of native SU1 has not yet been determined, and it is likely that the native protein and SU1r have distinct N termini. The possibility exists, therefore, that the enzymatic activity of SU1 in cells could differ from the properties determined in vitro. That SU1 functions as an isoamylase in vivo is most likely, however, considering the in vitro characterization and the fact that SU1 is more closely related to isoamylases than to pullulanases (James et al., 1995).

SU1 is distinct from bacterial isoamylases with respect to substrate preference and optimal pH and temperature conditions. Ps. isoamylase, a monomeric enzyme of 81 kD, is optimally active in the acidic pH range of 3.0 to 4.0 and at a temperature of 52°C (Yokobayashi et al., 1970). This isoamylase completely hydrolyzes the branch linkages of amylopectin and glycogen, releasing maltotriose and larger maltosaccharides, but its activity is relatively equal toward both substrates (Yokobayashi et al., 1973; Kainuma et al., 1978). In contrast, SU1r debranches amylopectin more readily than glycogen, and is optimally active within the neutral range of pH 6.0 to 7.0 and at temperatures between 25 and 37°C. A plausible explanation for the different conditional and substrate requirements for such similar-acting enzymes is that each has developed a functional adaptive response to its own unique environmental conditions while still using the same structural motifs for substrate binding and catalysis.

Based on its enzymatic characteristics, SU1 may be the same enzyme that was defined biochemically as isoamylase II (Doehlert and Knutson, 1991). Like SU1r, isoamylase II is present in extracts of developing maize kernels, is active at a neutral pH, hydrolyzes amylopectin more rapidly than phytoglycogen, and requires reducing agents in the buffer for maintenance of activity. Isoamylase II has an estimated molecular mass of 141 kD as determined by gel filtration, which is significantly larger than SU1. However, this observation by itself does not exclude the possibility that SU1 and isoamylase II are the same, because the size determination was made on relatively crude fractions and the possibility exists that SU1 is oligomeric in vivo. Another maize isoamylase was identified in extracts from mature su1- mutant sweet corn kernels (Manners and Rowe, 1969). The fact that SU1 is not present in su1- mutant kernels rules out identity with this sweet-corn isoamylase, which most likely participates in endosperm-starch degradation after seed germination.

The finding that su1- mutants are deficient in a DBE was first demonstrated by Pan and Nelson (1984). In that study, the assay used to detect hydrolysis of -(16) linkages employed pullulan as a substrate, so the missing enzyme must be classified as a pullulanase. These results do not agree with the characterization of SU1 as an isoamylase that has no activity toward pullulan. Further analysis of su1- mutant endosperms in the present study, however, revealed deficiencies of two proteins, SU1 isoamylase and a protein immunologically related to a known R enzyme. These data suggest that the su1 locus controls the accumulation of two different DBEs.

An alternative possibility, that a deficiency of the pullulanase occurs as the result of a second mutation present in the su1- stock, can be ruled out by a previous characterization of multiple, independent su1- mutations (Pan and Nelson, 1984). In that study, pullulanase activity was assayed in six distinct su1- mutant lines and found to be deficient in each. If a mutation other than su1- were responsible for the pullulanase deficiency, it would have had to occur simultaneously with the su1- mutation in each instance. The likelihood of this happening is remote. We conclude, therefore, that su1 function is necessary for accumulation of a pullulanase type of DBE in addition to the isoamylase for which it codes. Such a pleiotropic effect would reconcile the results of the initial study of Pan and Nelson (1984) with the current characterization of SU1. A similar situation exists in rice, in which the R enzyme found to be deficient in sugary- mutants is not coded for by the sugary gene (Nakamura et al., 1996b).

Two hypotheses are offered that might explain why the primary deficiency in SU1 isoamylase secondarily results in a decrease of the putative pullulanase. First, the two DBEs may associate in vivo in a multisubunit complex. Deficiency of SU1 could disrupt the complex, which could result in decreased accumulation of the pullulanase. The second suggestion is that su1- mutations indirectly result in an altered expression of the gene coding for the pullulanase, possibly as a result of the increased mono- and disaccharide concentrations in the mutant endosperm.

Mutations of su1 have been studied for nearly a century because of their unique effects on kernel starch. Here we show that these mutations cause a primary deficiency in an isoamylase normally present in the endosperm at the time that starch is produced. These observations suggest that SU1 isoamylase participates in starch biosynthesis. Furthermore, the fact that the combined deficiencies of SU1 isoamylase and a pullulanase result in phytoglycogen production at the expense of amylopectin is most simply explained by direct participation of DBEs in amylopectin synthesis. A mechanism by which this might occur was recently proposed in a model for the synthesis of storage starch, which suggests that DBEs act to selectively trim newly introduced branches to restore spatial order to the growing amylopectin molecule (Ball et al., 1996). Further analysis of the nature of the conditions and substrates required by both SU1 isoamylase and pullulanase, as well as characterization of the regulatory mechanisms that govern these enzymes, is expected to elucidate the specific roles played by DBEs in starch biosynthesis.

	FOOTNOTES

   Received November 13, 1997; accepted February 27, 1998.
   The accession number for the Su1 cDNA sequence reported in this article is U18908.

	ABBREVIATIONS

Abbreviations: DAP, days after pollination. DBE, starch-debranching enzyme. ENZ, amyloglucosidase enzyme reactor. Glc-cGlc7, glucosyl-cycloheptaamylose. Glc2-cGlc7, maltosyl-cycloheptaamylose. GST, glutathione-S-transferase. HPAEC, high-performance anion-exchange chromatography. _max, wavelength at which the complex exhibited maximal absorbance. PAD, pulsed-amperometric detectionPs.. isoamylase, isoamylase from Pseudomonas amyloderamosa.

	ACKNOWLEDGMENTS

We thank Y. Nakamura for providing antiserum against the rice R enzyme and J. Robyt for providing the Glc-cGlc7 and Glc2-cGlc7 substrates.

	LITERATURE  CITED

Top Abstract Introduction Methods Results Discussion References

Alonso MD, Lomako J, Lomako WM, Whelan WJ (1995) A new look at the biogenesis of glycogen. FASEB J 9: 1126-1137 [Abstract]

Ball S, Guan H-P, James M, Myers A, Keeling P, Mouille G, Buleon A, Colonna P, Preiss J (1996) From glycogen to amylopectin: a model for the biogenesis of the plant starch granule. Cell 86: 349-352 [CrossRef][ISI][Medline]

Banks W, Greenwood CT, Khan KM (1971) The interaction of linear, amylose oligomers with iodine. Carbohydr Res 17: 25-33

Beatty MK, Myers AM, James MG (1997) Genomic nucleotide sequence of a full-length wild type allele of the maize sugary1 (su1) gene (accession no. AF030882) (PGR 97-187). Plant Physiol 115: 1731

Beavis W, Berlyn M, Burr B, Chandler V, Coe E, Fauron C, Nelson O, Polacco M, Rodermel S, Sachs M, and others (1995) A standard for maize genetics nomenclature. Maize Genet Coop Newsl 69: 182-184

Correns C (1901) Bastarde zwischen maisrassen, mit besonder Berucksichtung der Xenien. Bibliotheca Bot 53: 1-161

Doehlert DC, Knutson CA (1991) Two classes of starch debranching enzymes from developing maize kernels. J Plant Physiol 138: 566-572

Fox JD, Robyt JF (1991) Miniaturization of three carbohydrate analyses using a microsample plate reader. Anal Biochem 195: 93-96 [CrossRef][ISI][Medline]

French D (1984) Organization of the starch granule. In RL Whistler, JM BeMiller, EF Paschall, eds, Starch: Chemistry and Technology. Academic Press, Orlando, FL, pp 183-248

Gunja-Smith Z, Marshall JJ, Mercier C, Smith EE, Whelan WJ (1970) A revision of the Meyer-Bernfeld model of glycogen and amylopectin. FEBS Lett 12: 101-104 [CrossRef][ISI][Medline]

Hannah LC, Giroux M, Boyer C (1993) Biotechnological modification of carbohydrates for sweet corn and maize improvement. Sci Hortic 55: 177-197

Harlow E, Lane D (1988) Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

James MG, Robertson DS, Myers AM (1995) Characterization of the maize gene sugary1, a determinant of starch composition in kernels. Plant Cell 7: 417-429 [Abstract]

Jenkins PJ, Cameron RE, Donald AM (1993) A universal feature in the starch granules from different botanical sources. Staerke 45: 417-420

Jesperson HM, MacGregor EA, Henrissat B, Sierks MR, Svensson B (1993) Starch- and glycogen-debranching and branching enzymes: prediction of structural features of the catalytic (/)8-barrel domain and evolutionary relationship to other amylolytic enzymes. J Protein Chem 12: 791-805 [CrossRef][ISI][Medline]

Kainuma K, Kobayashi S, Harada T (1978) Action of Pseudomonas isoamylase on various branched oligo- and polysaccharides. Carbohydr Res 61: 345-357 [Medline]

Koerner TJ, Hill JE, Myers AM, Tzagoloff A (1990) High-expression vectors with multiple cloning sites for construction of trpE-fusion genes: pATH vectors. Methods Enzymol 194: 477-490

Lee EYC, Whelan WJ (1971) Glycogen and starch debranching enzymes. In P Boyer, ed, The Enzymes, Vol 5. Academic Press, New York, pp 191-234

Manners DJ (1971) Specificity of debranching enzymes. Nature New Biol 234: 150-151 [Medline]

Manners DJ (1989) Recent developments in our understanding of amylopectin structure. Carbohydr Polym 11: 87-112 [CrossRef]

Manners DJ, Rowe KL (1969) Studies on carbohydrate-metabolizing enzymes. XX. Sweet corn debranching enzyme. Carbohydr Res 9: 107-121

Martin C, Smith AM (1995) Starch biosynthesis. Plant Cell 7: 971-985 [CrossRef][ISI][Medline]

Mouille G, Colleoni C, Maddelein M, Libessart N, Decq A, Delrue B, Ball S (1996a) Glucan trimming: a novel mechanism that explains the asymmetric distribution of -1,6 branches in amylopectin. In Y Nakamura, ed, 5th NIAR/COE International Symposium, Tsukuba, Japan. National Institute of Agrobiological Resources, Tsukuba, Japan, pp 27-32

Mouille G, Maddelein M-L, Libessart N, Talaga P, Decq A, Delrue B, Ball S (1996b) Preamylopectin processing: a mandatory step for starch biosynthesis in plants. Plant Cell 8: 1353-1366 [Abstract]

Mu-Forster C, Huang R, Powers JR, Harriman RW, Knight M, Singletary GW, Keeling PL, Wasserman BP (1996) Physical association of starch biosynthetic enzymes with starch granules of maize endosperm. Plant Physiol 111: 821-829 [Abstract]

Nakamura Y, Umemoto T, Ogata N, Kuboki Y, Yano M, Sasaki T (1996a) Starch debranching enzyme (R-enzyme or pullulanase) from developing rice endosperm: purification, cDNA and chromosomal location of the gene. Planta 199: 209-218 [ISI][Medline]

Nakamura Y, Umemoto T, Takahata Y, Komae K, Amano E, Satoh H (1996b) Changes in structure of starch and enzyme activities affected by sugary mutations in developing rice endosperm: possible role of starch debranching enzyme (R-enzyme) in amylopectin biosynthesis. Physiol Plant 97: 491-498 [CrossRef]

Nelson O, Pan D (1995) Starch synthesis in maize endosperms. Annu Rev Plant Physiol Plant Mol Biol 46: 475-496 [CrossRef][ISI]

Neuffer MG, Coe EH, Wessler SR (1997) Mutants of Maize. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Ou-Lee T, Setter TL (1985) Effect of increased temperature in apical regions of maize ears on starch-synthesis enzymes and accumulation of sugars and starch. Plant Physiol 79: 852-855 [Abstract/Free Full Text]

Pan D, Nelson OE (1984) A debranching enzyme deficiency in endosperms of the sugary-1 mutants of maize. Plant Physiol 74: 324-328 [Abstract/Free Full Text]

Preiss J (1991) Biology and molecular biology of starch synthesis and its regulation. Oxf Surv Plant Mol Cell Biol 7: 59-114

Preiss J, Sivak MN (1996) Starch synthesis in sinks and sources. In E Samski, AA Schaffer, eds, Photoassimilate Distribution in Plants and Crops. Marcel Dekker, New York, pp 63-96

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Smith AM, Denyer K, Martin C (1996) The synthesis of the starch granule. Annu Rev Plant Physiol 48: 67-87 [CrossRef]

Sumner JB, Somers GF (1944) The water soluble polysaccharides of sweet corn. Arch Biochem 4: 4-7

Toguri T (1991) Changes of a rice debranching enzyme during seed formation and germination. J Plant Physiol 137: 541-546

Wong K-S, Jane J (1995) Effects of pushing agents on the separation and detection of branched amylopectin by high performance anion exchange chromatography with pulsed amperometric detection. J Liquid Chromatogr 18: 63-80

Wong K-S, Jane J-L (1997) Quantitative analysis of debranched amylopectin by HPAEC-PAD with postcolumn enzyme reactor. J Liquid Chromatogr 20: 297-310

Yokobayashi I, Akai H, Sugimoto T, Hirao K, Sugimoto K, Harada T (1973) Comparison of the kinetic parameters of Pseudomonas isoamylase and Aerobacter pullulanase. Biochim Biophys Acta 293: 197-202 [Medline]

Yokobayashi K, Misaka A, Harada T (1970) Purification and properties of Pseudomonas isoamylase. Biochim Biophys Acta 212: 458-469 [Medline]

Yun S, Matheson NK (1993) Structures of the amylopectins of waxy, normal, amylose-extender, and wx:ae genotypes and of the phytoglycogen of maize. Carbohydr Res 243: 307-321 [Medline]

Copyright Clearance Center:   0032-0889/98/117/0425/11
 
© 1998 American Society of Plant Physiologists