• affinity gel electrophoresis;
  • carbohydrate-binding module;
  • kinetics;
  • Phaseolus vulgaris L.;
  • starch synthase III


  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Conclusions
  5. Experimental procedures
  6. Acknowledgements
  7. References

In plants and green algae, several starch synthase isozymes are responsible for the elongation of glucan chains in the biosynthesis of amylose and amylopectin. Multiple starch synthase isozymes, which are classified into five major classes (granule-bound starch synthases, SSI, SSII, SSIII, and SSIV) according to their primary sequences, have distinct enzymatic properties. All the starch synthase isozymes consist of a transit peptide, an N-terminal noncatalytic region (N-domain), and a C-terminal catalytic region (C-domain). To elucidate the enzymatic properties of kidney bean (Phaseolus vulgaris L.) SSIII and the function of the N-domain of kidney bean SSIII, three recombinant proteins were constructed: putative mature recombinant SSIII, recombinant kidney bean SSIII N-domain, and recombinant kidney bean SSIII C-domain. Purified recombinant kidney bean SSIII displayed high specific activities for primers as compared to the other starch synthase isozymes from kidney bean. Kinetic analysis showed that the high specific activities of recombinant kidney bean SSIII are attributable to the high kcat values, and that the Km values of recombinant kidney bean SSIII C-domain for primers were much higher than those of recombinant kidney bean recombinant SSIII. Recombinant kidney bean SSIII and recombinant kidney bean SSIII C-domain had similar chain-length specificities for the extension of glucan chains, indicating that the N-domain of kidney bean SSIII does not affect the chain-length specificity. Affinity gel electrophoresis indicated that recombinant kidney bean SSIII and recombinant kidney bean SSIII N-domain have high affinities for amylose and amylopectin. The data presented in this study provide direct evidence for the function of the N-domain of kidney bean SSIII as a carbohydrate-binding module.


affinity gel electrophoresis


carbohydrate-binding module


C-terminal catalytic domain of starch synthases


degree of polymerization


granule-bound starch synthase


internal repeat regions of SSIIIs


N-terminal noncatalytic domain of starch synthases


kidney bean (Phaseolus vulgaris L.) SSIII


recombinant kidney bean (Phaseolus vulgaris L.) SSI


recombinant kidney bean (Phaseolus vulgaris L.) SSII-1


recombinant kidney bean (Phaseolus vulgaris L.) SSIII


recombinant C-terminal catalytic domain of kidney bean (Phaseolus vulgaris L.) SSIII


recombinant N-terminal noncatalytic domain of kidney bean (Phaseolus vulgaris L.) SSIII


starch-binding domain


starch synthase


SSIII-specific domain


transit peptide


variable region of SSIIIs.

Starch, the major carbon reserve accumulated in plastids, consists of two glucose homopolymers, amylose and amylopectin. Amylose is essentially a linear form composed of α-1,4-glucosidic linkages, whereas amylopectin is a highly branched glucan of α-1,4-linked chains with branched α-1,6-linkages. The biosynthesis of these glucans requires the concerted action of at least four enzymatic activities: ADP-glucose pyrophosphorylase, starch synthase (SS), starch-branching enzyme, and starch-debranching enzyme [1–5]. SS (EC and EC catalyzes the transfer of the glucosyl moiety from ADP-glucose to the nonreducing end of an α-1,4-glucan, and therefore is responsible for the elongation of α-1,4-glucan chains in amylose and amylopectin. Multiple SS isozymes have been found in individual plant species, and are divided into five classes [granule-bound SS (GBSS), SSI, SSII, SSIII, and SSIV] on the basis of their primary sequences [4]. Each isozyme appears to have distinct enzymatic properties and to play a specific role in starch biosynthesis [4,6]. One SS isozyme, GBSS, is known to solely contribute to amylose biosynthesis [1,7,8]. In contrast to GBSS members, other SS isozymes play a role in amylopectin synthesis and might not participate in amylose synthesis [1–8].

All SS isozymes have an N-terminal transit peptide, an N-terminal noncatalytic domain (N-domain), and a C-terminal catalytic region (C-domain) (Fig. 1A). The C-domain is highly conserved in all SS members, whereas no significant similarity is found between the N-domains of different classes. The SSIII members have a unique structure, with the longest N-domain among SS isozymes. This domain is further divided into a variable region (VR) that varies in length and homology, and an SSIII-specific domain (SSIII-SD) that contains three internal repeats (IRs) and is highly conserved in plants (Fig. 1A) [9]. Recent computational analysis of SSIII sequences showed that each of the three repeats is predicted to be a starch-binding domain (SBD) [10] and is in the carbohydrate-binding module (CBM) family 21 [11]. CBMs are noncatalytic carbohydrate-recognizing modules of carbohydrate-active enzymes, and promote the association of the enzyme (catalytic domain) with the substrates by increasing the concentration of the substrate on the surface of the enzyme [12,13]. Currently, CBMs are divided into 49 families on the basis of amino acid sequence similarity ( A large number of CBMs have been identified experimentally, but there is no evidence that the predicted SBD of SSIII functions as a CBM.


Figure 1.  Schematic representation of PvSSIII and sequence alignment of the three IRs among dicot SSIIIs. (A) Comparison of the primary sequences of several SSIII members showed that PvSSIII consists of a TP (white bar), an N-domain, and a C-domain (black bar). The N-domain is further divided into a VR (dotted bar) and an SSIII-SD (gray bar) containing three IRs (gray arrows). (B) The following sequences were obtained from the GenBank/EMBL/DDBJ databases: PvSSIII from Ph. vulgaris L. (AB293998); AtSSIII from Arabidopsis thaliana (DQ241810); StSSIII from Solanum tuberosum L. (X95759); and VuSSIII from Vigna unguiculata L. (AJ225088). The sequences of the IRs were aligned by the clustalw program [24]. The secondary structure of the IRs predicted by the pred[25] program is shown above the sequences as a cylinder (helix) and black arrows (β-sheet). The asterisks indicate the residues corresponding to two aromatic residues that are essential for binding activity of the SBD from As. niger glucoamylase.

Download figure to PowerPoint

Although some genetic mutants and/or transgenic plants with absent or reduced SSIII activity have been analyzed to investigate its function, the enzymatic characterization of SSIII isozymes is still incomplete. Analysis of maize dull1 mutants without SSIII activity indicates that the mutation causes changes in amylopectin structure, with increased branching frequency [14–16]. Transgenic potato plants expressing antisense RNA for SSIII have altered amylopectin with an increased proportion of extra-long chains [17–19]. Similar results were obtained for amylopectin from leaf starch of Arabidopsis mutants lacking SSIII activity [20]. These observations suggest that SSIII is essential for starch biosynthesis in these plants, and that the function of SSIII is not completely compensated for by other SS isozymes. However, it is difficult to determine the precise function of SSIII, because of the pleiotropic effects of the absence or reduction of SSIII activity on other starch biosynthetic enzymes [18–21]. Hence, to understand the role of SSIII in starch synthesis, it is also important to elucidate its enzymatic properties.

In this study, we describe the isolation of cDNA for SSIII from kidney bean (Phaseolus vulgaris L.) (PvSSIII) and the enzymatic properties of the recombinant PvSSIII (rPvSSIII) and its N-terminal truncated form (rPvSSIII-C). We also show the glucan-binding abilities of rPvSSIII, rPvSSIII-C and the recombinant N-domain of PvSSIII (rPvSSIII-N) using affinity gel electrophoresis (AGE), and demonstrate that the N-domain of PvSSIII functions as a real CBM.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Conclusions
  5. Experimental procedures
  6. Acknowledgements
  7. References

Isolation of PvSSIII cDNA

To isolate a cDNA clone for SSIII, RT-PCR was performed with the degenerated primers and the single-strand cDNA mixture from kidney bean leaves. The sequence of the amplified fragment (about 2.3 kb) showed high homology (more than 70% identity) to the corresponding regions of the other plant SSIII members. The sequence information on the full-length clone for PvSSIII was obtained by 3′-RACE and 5′-RACE. The PvSSIII cDNA clone was 4135 bp long and contained a 3498 bp ORF (see supplementary Fig. S1). The protein encoded by the PvSSIII cDNA corresponded to 1165 amino acid residues. When the primary sequence of PvSSIII was analyzed for a putative signal sequence with the targetp[22] and chlorop[23] network programs, the protein was predicted to contain a plastid-targeting sequence with a cleavage site between Val100 and Val101. The alignment of primary sequences among SSIII members by the clustalw program [24] showed that the PvSSIII consisted of a putative transit peptide (TP, 100 amino acids), VR (123 amino acids), SSIII-SD (499 amino acids), and C-terminal domain (443 amino acids) (Fig. 1A). The primary sequence of PvSSIII, except for the TP and VR sequences, displayed a significant sequence similarity (80–94%) with those of SSIII members. In addition, the SSIII-SD of PvSSIII contained three internal repeats (IR1–IR3 in Fig. 1A), each of which is predicted to have a helix and seven β-sheets by the secondary structure prediction program (pred) [25] (Fig. 1B). Computational analysis has shown that each IR from Arabidopsis SSIII is structurally similar to the SBD from Aspergillus niger glucoamylase [26]. The structure determined by multidimensional NMR spectroscopy suggested that the SBD has two independent binding sites (sites 1 and 2), and that two Trp residues (Trp543 and Trp590 in As. niger glucoamylase) are essential for binding activity in site 1 [26–28]. The residues corresponding to Trp543 are completely conserved in IRs from all the SSIII members, whereas the residues corresponding to Trp590 are not Trp but are conserved as the other aromatic residues, Tyr and Phe (Fig. 1B). Although these observations suggest that IRs from SSIII members function as an SBD, no experimental evidence on the function of SSIII-SD and on the kinetic properties of SSIII has been reported.

Preparation of rPvSSIII and rPvSSIII-C

We prepared the recombinant proteins, rPvSSIII and rPvSSIII-C, to examine their enzymatic properties and the effects of the N-domain on enzyme activity (Fig. 2A). The N-terminus of rPvSSIII-C was determined to be in the same position as that of the truncated form of pea SSII [29]. Because of the His-tag at their C-termini, both recombinant proteins were purified by two-step column chromatography. Purified rPvSSIII and rPvSSIII-C migrated as a single polypeptide band on an SDS/PAGE gel (Fig. 2B). The molecular masses of rPvSSIII and rPvSSIII-C were estimated to be 122 and 52 kDa, respectively, and were identical to those predicted from their sequences.


Figure 2.  Recombinant PvSSIII proteins constructed in this study. (A) The recombinant proteins, rPvSSIII, rPvSSIII-C, and rPvSSIII-N, are shown schematically. (B) Purified recombinant proteins were subjected to SDS/PAGE on a 7.5% polyacrylamide gel, and then the gel was stained with Coomassie Brilliant Blue. Lane M: standard proteins. Lanes 1–3: 0.75 µg of purified rPvSSIII, rPvSSIII-C, or rPvSSIII-N, respectively.

Download figure to PowerPoint

Enzymatic properties of rPvSSIII

Several enzymatic properties of purified rPvSSIII are summarized in Table 1, together with those of rPvSSI and rPvSSII-1 [30]. When assayed under standard conditions in the presence of 10 mg·mL−1 amylopectin and 1 mm ADP-glucose, the specific activity of rPvSSIII was determined to be 199 s−1, which is 17-fold and 40-fold higher than those of rPvSSI and rPvSSII-1, respectively (Table 1). rPvSSIII exhibited a slightly higher thermostability and a broader range of pH stability than other recombinant kidney bean SSs.

Table 1.   Comparison of some enzymatic properties of SS isozymes from kidney bean.
  1. a  These values were calculated from the SS activities in the presence of 10 mg·mL−1 amylopectin and 1 mm ADP-glucose, and are expressed as turnover number on a molar basis. b These data are cited from our previous report [27]. c These kinetic parameters are the means with SE of measurements obtained using at least three independent experiments.

Thermal stability (°C)< 35< 35< 45< 40
pH stability7.0–9.57.0–9.0   6.5–10.0   7.5–10.0
Optimum pH8.58.0   8.5   7.5
Specific activity (s−1)a12.05.1199  12.7
For amylopectin (at 1 mm ADP-glucose)
 Km (mg·mL−1) ± 0.16119 ± 26
 kcat (s−1)18.05.14235 ± 10162 ± 26
For glycogen (at 1 mm ADP-glucose)
 Km (mg·mL−1)210.218.7 ± 1.8382 ± 88
 kcat (s−1)37.15.14412 ± 49324 ± 67
For ADP-glucose (at 10 mg·mL−1 amylopectin)
 Km (mm) ± 0.010.33 ± 0.04
 kcat (s−1)11.24.8235 ± 2.220.1 ± 0.93
For ADP-glucose (at 10 mg·mL−1 glycogen)
 Km (mm) ± 0.010.34 ± 0.02
 kcat (s−1)11.24.9333 ± 4.48.8 ± 4.4

The kinetic parameters of rPvSSIII were analyzed with primer glucans (amylopectin and glycogen) and ADP-glucose as substrates (Table 1). Lineweaver–Burk plots of the reaction catalyzed by rPvSSIII in the presence of 1 mm ADP-glucose yielded Km and kcat values of 2.4 mg·mL−1 and 235 s−1 for amylopectin and 8.7 mg·mL−1 and 412 s−1 for glycogen, respectively. The high specific activity of rPvSSIII was predominantly attributed to the elevated kcat value. The catalytic efficiencies of rPvSSIII for primers were higher than those of rPvSSI and rPvSSII-1 (Table 1). The Km values of rPvSSIII for ADP-glucose in the presence of 10 mg·mL−1 glucan primers were nearly identical, indicating that the glucan primer species had no effect on the affinity for ADP-glucose.

Enzymatic properties of rPvSSIII-C

To investigate the effect of the N-domain on enzyme activity, some properties of rPvSSIII-C were compared with those of rPvSSIII (Table 1). Despite the absence of the N-domain, rPvSSIII-C displayed a specific SS activity of 12.7 s−1, indicating that the N-domain is not essential for catalysis. rPvSSIII-C showed a slightly lower thermostability and narrower range of pH stability than rPvSSIII. These results suggest that the N-domain of PvSSIII has only a moderate effect on these physicochemical properties, i.e. on the catalytic domain.

The Km and kcat values of rPvSSIII-C were estimated to be 119 mg·mL−1 and 162 s−1 for amylopectin and 382 mg·mL−1 and 324 s−1 for glycogen, respectively (Table 1). These kcat values of rPvSSIII-C were similar to those of rPvSSIII, but the deletion of the N-domain resulted in drastic decreases in the affinity and catalytic efficiency for the glucan primers, as indicated by about 50-fold increases in the Km values and reductions in the kcat/Km values, respectively. In contrast, the Km values of rPvSSIII-C for ADP-glucose in the presence of 10 mg·mL−1 glucan primers were nearly identical to those of rPvSSIII, indicating that the N-domain had no effect on the affinity for ADP-glucose. The low kcat values of rPvSSIII-C for ADP-glucose were thought to be due to nonsaturation of glucan primer because of the remarkably reduced affinity. These results, together with some physicochemical properties, suggest that the N-domain of PvSSIII is structurally independent of the catalytic domain and is involved in the interaction with glucan primers.

As SSIII members, SSI and SSII are also composed of an N-domain and a C-domain. However, the N-domains of SSI from maize and SSIIs from pea and potato appear to have no significant role in enzymatic catalysis, because these N-domain-truncated forms had nearly the same specific activities as their intact forms [29,31,32]. In pea SSII, analysis of the truncated form has suggested the possibility that the N-domain has a physical role, such as determining the subcellular localization of the enzyme [32]. In addition, analysis of the N-domain-truncated form of maize SSI by AGE has shown that the N-domain is not responsible for the properties of the affinity for polyglucans [33]. The primary sequences of the N-domains in SSI or SSII class members are extremely diverse, and no significant similarity is observed [30,34]. Thus, each N-domain in these members is presumed to have a different role. In contrast, the N-domains of SSIII members, in particular the SSIII-SDs containing three IRs, share high homology (more than 74% similarity and 52% identity) with each other. The sequence conservation between the various SSIII-SDs and structural modeling of an IR as an SBD [10] allow us to assume that the N-domains of SSIII members have a common role in the affinity for glucan primers. Our kinetic data on rPvSSIII and rPvSSIII-C support this idea.

Chain-length specificities of rPvSSIII and rPvSSIII-C

We have previously shown that rPvSSI and rPvSSII-1 have different chain-length specificities for the extension of glucan chains; rPvSSI adds glucose to the chains with shorter outer-chain lengths of degree of polymerization (dp) 6–10, whereas rPvSSII-1 preferentially elongates unit chains of dp 17–20 (Fig. 3A,C) [30]. To elucidate the specificities of rPvSSIII and rPvSSIII-C for chains of amylopectin and glycogen, enzyme reactions were performed with 10 mg·mL−1 primers and 1 mm ADP-[U-14C]glucose. After the products were debranched by isoamylase and separated on a Toyopearl HW-50S column, the reducing power and radioactivity in each fraction were determined (Fig. 3). Although, as observed in the products for glycogen, the reactivity for chains with dp < 10 was somewhat lower than chains around dp 15, the profiles of 14C-label incorporated by rPvSSIII into amylopectin and glycogen nearly paralleled the chain-length distributions of both glucans, respectively (Fig. 3B,D). This suggests that the reactivity of rPvSSIII depends on the number of each chain rather than the chain length, which means that rPvSSIII has a broader specificity than rPvSSI and rPvSSII-1. The profiles of 14C-label incorporation by rPvSSIII-C were very close to those for rPvSSIII, indicating that the N-domain of PvSSIII has almost no effect on the specificity for chains of glucans (Fig. 3B,D).


Figure 3.  Analysis of products formed by the recombinant SS isozymes using amylopectin and glycogen as primers. rPvSSs (10 mU each) were incubated with 14C-labeled 1 mm ADP-glucose and 10 mg·mL−1 of amylopectin and glycogen for 1 h at 30 °C. After the reaction, the remaining ADP-glucose was removed, and glucans were debranched by isoamylase. The debranched products were separated on a Toyopearl HW-50S column. The eluate was collected, and the reducing power, radioactivity and dp were determined. The broken and solid lines indicate reducing power (µmol·mL−1) and incorporated glucose (%), respectively. (A,B) Products formed by rPvSSI (circles), rPvSSII-1 (triangles), rPvSSIII (open squares) and rPvSSIII- C (closed squares) with amylopectin. (C,D) Products formed by rPvSSI (circles), rPvSSII-1 (triangles), rPvSSIII (open squares) and rPvSSIII-C (closed squares) with glycogen.

Download figure to PowerPoint

The activities of rPvSSIII and rPvSSIII-C for several malto-oligosaccharides were compared with each other (Fig. 4). Although rPvSSIII-C had only 1/16th the activity of rPvSSIII for amylopectin (Table 1), no major difference in the activities for each malto-oligosaccharide was observed between rPvSSIII and rPvSSIII-C. The activities of both enzymes for maltotriose were half as much as for maltotetraose, maltopentaose, and maltohexaose. These results suggest that the N-domain of PvSSIII has an inadequate effect on the catalytic activities for malto-oligosaccharides, and that the catalytic center is likely to have subsites that specifically bind to malto-oligosaccharides with a dp greater than 4.


Figure 4.  Comparison of catalytic activities of rPvSSIII and rPvSSIII-C for malto-oligosaccharides. Enzyme activity was assayed using 1 mm ADP-glucose and 50 mm maltose (G2), maltotriose (G3), maltotetraose (G4), maltopentaose (G5) or maltohexaose (G6), and expressed as catalysis turnover number on a molar basis.

Download figure to PowerPoint

Binding of N-domain to polyglucans

In addition to rPvSSIII and rPvSSIII-C, rPvSSIII-N was prepared and analyzed by AGE to examine whether the N-domain of PvSSIII directly interacted with polyglucans. The rPvSSIII-N purified on two chromatographs migrated as a single polypeptide band on SDS/PAGE gel (Fig. 2B). Purified rPvSSIII, rPvSSIII-N and rPvSSIII-C proteins, together with BSA, were run on 4–16% polyacrylamide gels containing different concentrations of amylopectin (Fig. 5A). The relative mobility (Rm) of rPvSSIII and rPvSSIII-N with respect to BSA decreased with the increase of amylopectin concentration, whereas that of rPvSSIII-C was almost constant irrespective of amylopectin concentration. These results indicate that the N-domain interacts with amylopectin, whereas the C-domain has very low affinity for amylopectin, despite its catalytic ability. In addition, the observations from AGE analysis were consistent with the kinetic data showing that deletion of the N-domain caused a dramatic increase in Km values for glucan primers (Table 1).


Figure 5.  AGE analysis. (A) rPvSSIII (lane 1), rPvSSIII-N (lane 2), rPvSSIII-C (lane 3) and BSA (lane 4) were loaded on stepwise 4–16% polyacrylamide gels containing 0, 0.5 and 1 mg·mL−1 amylopectin. (B) Plots of the reciprocal of relative mobility (1/Rm) of rPvSSIII (open circles) and rPvSSIII-N (open squares) against the concentration of amylose, amylopectin, glycogen and pullulan were derived from AGE using 7.5% polyacrylamide gels containing various concentrations of polyglucans.

Download figure to PowerPoint

rPvSSIII and rPvSSIII-N proteins were also analyzed on 7.5% polyacrylamide gels containing different concentrations of amylose, amylopectin, glycogen, and pullulan. The Kd values of both enzymes for these polyglucans were estimated from the reciprocal plots of the relative mobility (1/Rm) against each polyglucan (Fig. 5B and Table 2). Both proteins showed similar Kd values for each polyglucan, and the affinity increased with the increase in the average length of α-1,4 chains forming each polyglucan, which is estimated to be amylose > amylopectin >> glycogen > pullulan. These observations suggest that the N-domain is predominantly responsible for the affinity of PvSSIII for these polyglucans and prefers longer α-1,4 chains.

Table 2.   Estimation of Kd values of rPvSSIII and rPvSSIII-N for polyglucans. These values were estimated from the plots of the reciprocal of the relative mobility against the concentration of polyglucans (Fig. 5B). Values are the means with SE of measurements from three independent experiments.
 Amylose, Kd (mg·mL−1)Amylopectin, Kd (mg·mL−1)Glycogen, Kd (mg·mL−1)Pullulan, Kd (mg·mL−1)
rPvSSIII0.30 ± 0.020.53 ± 0.034.04 ± 0.086.86 ± 0.10
rPvSSIII-N0.41 ± 0.060.68 ± 0.043.91 ± 0.056.88 ± 0.08


  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Conclusions
  5. Experimental procedures
  6. Acknowledgements
  7. References

Previous reports on SSI and SSII isozymes have shown that the N-domain has no significant role in catalytic activity and that the C-domain is responsible for both affinity for (Km value) and molecular activity (kcat value) against polyglucans [29,31–33]. In this study, we first provide a detailed understanding of the relationship between the domain structure and function of SSIII. Our results suggest that the N-domain and C-domain of PvSSIII have different functions in enzyme activity; the N-domain contributes to the predominant affinity for polyglucans as a CBM, whereas the C-domain contributes to the molecular activity.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Conclusions
  5. Experimental procedures
  6. Acknowledgements
  7. References


Amylopectin and amylose from potato, glycogen from rabbit liver and ADP-glucose were obtained from Sigma Chemical Co. (St Louis, MO, USA). ADP-[U-14C]glucose was purchased from GE Healthcare Bio-Sciences Corp. (Piscataway, NJ, USA). Malto-oligosaccharides and pullulan were purchased from Wako Pure Chemical Industries Ltd (Osaka, Japan) and Hayashibara Biochemical Laboratories (Okayama, Japan), respectively. Isoamylase from Pseudomonas was obtained from Nacalai tesque (Kyoto, Japan). Kidney bean plants were grown in the field of Hokkaido University, Japan. Intact leaves were harvested in the early morning and stored in liquid nitrogen until use.

Isolation of PvSSIII cDNA

The cDNA clone for PvSSIII was isolated from the total RNA of kidney bean leaves using RT-PCR, 3′-RACE, and 5′-RACE, described in detail in supplementary Table S1 and supplementary Fig. S1. A 3997 bp cDNA fragment, including the whole coding region for PvSSIII, was amplified by the specific primers ss3-up and ss3-dw (supplementary Table S1), and was cloned into pBluescript II SK (+) (Stratagene, La Jolla, CA, USA) to generate plasmid pBS-SS3.

Construction of plasmids for the expression of PvSSIII, PvSSIII-N, and PvSSIII-C

Construction of the expression vectors is described in detail in supplementary Table S1 and supplementary Fig. S2. All the fragments amplified by PCR were sequenced to verify that no error in polymerization had occurred. The pEXP-SSIII, pEXP-SSIIIN and pEXP-SSIIIC plasmids were constructed for the production of the recombinant proteins rPvSSIII, rPvSSIII-N, and PvSSIII-C, respectively. Escherichia coli BL21 (DE3) pLysS (Novagen, Madison, WI, USA) was transformed by each expression vector.

Expression and purification of recombinant proteins

E. coli cells carrying each expression vector were grown in 2 L of LB medium containing 100 µg·mL−1 of ampicillin at 37 °C. When the absorbance at 600 nm reached 0.5, the transformants with pEXP-SSIII and pEXP-SSIIIC were induced with 0.1 mm isopropyl thio-β-d-galactoside (final concentration) at 25 °C for 5 h. The transformants with pEXP-SSIIIN were incubated at 15 °C for 30 min, and then induced with 0.1 mm isopropyl thio-β-d-galactoside at 15 °C overnight. The cells collected by centrifugation at 6000 g for 10 min (CR20F centrifuge with an RPR10-2 rotor; Hitachi Koki, Tokyo, Japan) were suspended in buffer A (20 mm sodium phosphate, pH 7.5, 0.5 m NaCl, 50 mm imidazole, 10% glycerol). The suspended cells were disrupted with a French press (Ohtake, Tokyo, Japan) and centrifuged at 40 000 g for 20 min (CR20F centrifuge with an R19A rotor). The resulting supernatant was applied to a Chelating Sepharose FF column (GE Healthcare Bio-Sciences) previously charged with nickel ions and equilibrated with buffer A. The column was washed with buffer A, and then proteins with a His-tag were eluted with a gradient of 50–500 mm imidazole. The fractions including each recombinant protein were pooled and dialyzed against buffer B (20 mm Tris/HCl buffer, pH 8.0, containing 1 mm EDTA, 1 mm dithiothreitol, and 10% glycerol) and then fractionated on a DEAE-Sepharose CL-6B column (GE Healthcare Bio-Sciences) equilibrated with buffer B. The column was washed with buffer B, and this was followed by elution with a gradient of 0–0.5 m NaCl. The purified protein solution was desalted and concentrated with Vivaspin (VIVASCIENCE, Hanover, Germany), and then stored at − 20 °C after addition of glycerol (final concentration 50%). The protein concentration was determined by the Bradford method, using BSA as a standard [35].

Assay of SS activity

The reaction mixture contained 100 mm Bicine (pH 8.0), 5 mm EDTA, 25 mm potassium acetate, 10 mm dithiothreitol, 0.1 mg·mL−1 BSA, 1 mm ADP-glucose including ADP-[U-14C]glucose (3.3 GBq·mol−1), and either 10 mg·mL−1 amylopectin from potato, glycogen from rabbit liver or 50 mm malto-oligosaccharide as a primer. The reaction was initiated by the addition of ADP-glucose, incubated at 30 °C for 10 min, and then terminated by boiling for 3 min. Incorporation of radioactive label into primer was determined by liquid scintillation counting as described previously [30,36]. All assays were performed under conditions where the amount of glucose incorporated into primer was linearly proportional to the amount of enzyme and to the incubation time. The optimum pH, pH stability, temperature stability and kinetic parameters of purified enzymes were determined as described previously [36].

Product analysis

The SS reaction was performed as described in the assay of SS activity, except that the volume of the reaction mixture was 1 mL. To remove the remaining ADP-glucose, the reaction mixture was mixed with 0.5 mL of 0.5 g·mL−1 Dowex 1-X8 for 5 min, and 1 mL of supernatant was recovered by centrifugation at 10 000 g for 5 min (MX-160 centrifuge with a TMP-21 rotor; Tomy, Tokyo, Japan). After this procedure was repeated once more, the recovered supernatant was incubated with 50 µL of 1 m sodium acetate (pH 3.5) and 100 units of isoamylase from Pseudomonas at 45 °C for 12 h. The reaction was terminated with the addition of 0.3 mL of 5 m NaOH. One milliliter of the sample filtered through a 0.22 µm filter was applied at a flow rate of 0.5 mL·min−1 to a column (1.6 × 70 cm) of Toyopearl HW-50S (Tosoh Bioscience, Tokyo, Japan) equilibrated and eluted with 20 mm NaOH and 0.2% (w/v) NaCl. Fractions of 2 mL each were collected, and the radioactivity in the fraction was measured by liquid scintillation counting. The carbohydrate content and reducing power in the fractions were determined by the phenol/H2SO4 method [37] and the modified Park–Johnson method [38], respectively. The dp was calculated from the values of carbohydrate content and reducing power.


The capacity of binding of rPvSSIII, rPvSSIII-N and rPvSSIII-C to soluble polyglucans was assayed by AGE. AGE was performed on native gels consisting of seven layers of 4–16% polyacrylamide with various concentrations of amylopectin or consisting of continuous 7.5% polyacrylamide with various concentrations of polyglucans (amylose, amylopectin, glycogen, and pullulan) in 25 mm Tris/HCl and 250 mm glycine buffer (pH 8.3). Approximately 5 µg of each recombinant protein and BSA (noninteracting negative control) were loaded on the gels and subjected to electrophoresis at 12 mA per gel at room temperature in running buffer (6.25 mm Tris/HCl, 50 mm glycine, pH 8.3). The migration distances of each protein were measured after staining with Coomassie Brilliant Blue. The relative mobility (Rm) of each protein with respect to BSA was determined. The dependence of 1/Rm on glucan concentration [s] was plotted. The value of the dissociation constant (Kd) was calculated from the intercept of the [s] axis.


  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Conclusions
  5. Experimental procedures
  6. Acknowledgements
  7. References

This work was supported by grants-in-aid for Scientific Research (C) (13660071, 16580069, and 18580086) and the Akiyama Foundation to H. Ito from the Japan Society for the Promotion of Science, and a grant-in-aid for Young Scientists (B) (18780067) to S. Hamada from the Ministry of Education, Science, Sports, and Culture, Japan.


  1. Top of page
  2. Abstract
  3. Results and Discussion
  4. Conclusions
  5. Experimental procedures
  6. Acknowledgements
  7. References
  • 1
    Smith AM (1999) Making starch. Curr Opin Plant Biol 2, 223229.
  • 2
    Kossmann J & Lloyd J (2000) Understanding and influencing starch biochemistry. Crit Rev Plant Sci 19, 171226.
  • 3
    Nakamura Y (2002) Towards a better understanding of the metabolic system for amylopectin biosynthesis in plants: rice endosperm as a model tissue. Plant Cell Physiol 43, 718725.
  • 4
    Ball SG & Morell MK (2003) From bacterial glycogen to starch: understanding the biogenesis of the plant starch granule. Annu Rev Plant Biol 54, 207233.
  • 5
    Tetlow IJ, Morell MK & Emes MJ (2004) Recent developments in understanding the regulation of starch metabolism in higher plants. J Exp Bot 55, 21312145.
  • 6
    Smith AM, Denyer K & Martin C (1997) The synthesis of the starch granule. Annu Rev Plant Physiol Plant Mol Biol 48, 6787.
  • 7
    Ball SG, van de Wal MHBJ & Visser RGF (1998) Progress in understanding the biosynthesis of amylose. Trends Plant Sci 3, 13601385.
  • 8
    Denyer K, Johnson P, Zeeman S & Smith AM (2001) The control of amylose synthesis. J Plant Physiol 158, 479487.
  • 9
    Li Z, Mouille G, Kosar-Hashemi B, Rahman S, Clarke B, Gale KR, Appels R & Morell MK (2000) The structure and expression of the wheat starch synthase III gene. Motifs in the expressed gene define the lineage of the starch synthase III gene family. Plant Physiol 123, 613624.
  • 10
    Palopoli N, Busi MV, Fornasari MS, Gomez-Casati D, Ugalde R & Parisi G (2006) Starch-synthase III family encodes a tandem of three starch-binding domains. Proteins 65, 2731.
  • 11
    Machovic̆ M & JaneC̆ek S (2006) The evolution of putative starch-binding domains. FEBS Lett 580, 63496356.
  • 12
    Bolam DN, Ciruela A, Mcqueen-Mason S, Simpson P, Williamson MP, Rixon JE, Boraston A, Hazlewood GP & Gilbert HJ (1998) Pseudomonas cellulose-binding domains mediate their effects by increasing enzyme substrate proximity. Biochem J 331, 775781.
  • 13
    Boraston AB, Bolam DN, Gilbert HJ & Davies GJ (2004) Carbohydrate-binding modules: fine-tuning polysaccharide recognition. Biochem J 382, 769781.
  • 14
    Shannon JC & Garwood DL (1984) Genetics and physiology of starch development. In Starch: Chemistry and Technology (Whistler RL, BeMiller JN & Paschall EF, eds), pp. 2886. Academic Press, San Diego, CA.
  • 15
    Wang YJ, White P, Pollak L & Jane JL (1993) Characterization of starch structures of 17 maize endosperm mutant genotypes with Oh43 inbred line background. Cereal Chem 70, 171179.
  • 16
    Gao M, Wanat J, Stinard PS, James MG & Myers AM (1998) Characterization of dull1, a maize gene coding for a novel starch synthase. Plant Cell 10, 399412.
  • 17
    Edwards A, Fulton DC, Hylton CM, Jobling SA, Gidley M, Rössner U, Martin C & Smith AM (1999) A combined reduction in activity of starch synthase II and III of potato has novel effects on the starch of tubers. Plant J 17, 251261.
  • 18
    Lloyd JR, Landschütze V & Kossmann J (1999) Simultaneous antisense inhibition of two starch-synthase isoforms in potato tubers leads to accumulation of grossly modified amylopectin. Biochem J 338, 515521.
  • 19
    Fulton DC, Edwards A, Pilling E, Robinson HL, Fahy B, Seale R, Kato L, Donald AM, Geigenberger P, Martin C et al. (2002) Role of granule-bound starch synthase in determination of amylopectin structure and starch granule morphology in potato. J Biol Chem 277, 1083410841.
  • 20
    Zhang X, Myers AM & James MG (2005) Mutations affecting starch synthase III in Arabidopsis alter leaf starch structure and increase the rate of starch synthesis. Plant Physiol 138, 663674.
  • 21
    Singletary GW, Banisadr R & Keeling PL (1997) Influence of gene dosage on carbohydrate synthesis and enzymatic activities in endosperm of starch-deficient mutants of maize. Plant Physiol 113, 293304.
  • 22
    Emanuelsson O, Nielsen H, Brunak S & von Heijne G (2000) Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J Mol Biol 300, 10051016.
  • 23
    Emanuelsson O, Nielsen H & von Heijne G (1999) A neural network-based method for predicting chloroplast transit peptides and their cleavage sites. Protein Sci 8, 978984.
  • 24
    Thompson JD, Higgins DG & Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 46734680.
  • 25
    Jones DT (1999) Protein secondary structure prediction based on position-specific scoring matrices. J Mol Biol 292, 195202.
  • 26
    Sorimachi K, Jacks AJ, Le Gal-Coëffet M, Williamson G, Archer DB & Williamson MP (1996) Solution structure of the granular starch binding domain of glucoamylase from Aspergillus niger by nuclear magnetic resonance spectroscopy. J Mol Biol 259, 970987.
  • 27
    Sorimachi K, Le Gal-Coëffet M, Williamson G, Archer DB & Williamson MP (1997) Solution structure of the granular starch binding domain of Aspergillus niger glucoamylase bound to β-cyclodextrin. Structure 5, 647661.
  • 28
    Giardina T, Gunning AP, Juge N, Faulds CB, Furniss CSM, Svensson B, Morris VJ & Williamson G (2001) Both binding sites of the starch-binding domain of Aspergillus niger glucoamylase are essential for inducing a conformational change in amylose. J Mol Biol 313, 11491159.
  • 29
    Edwards A, Marshall J, Denyer K, Sidebottom C, Visser RGF, Martin C & Smith AM (1996) Evidence that a 77-kilodalton protein from the starch of pea embryos is an isoform of starch synthase that is both soluble and granule bound. Plant Physiol 112, 8997.
  • 30
    Senoura T, Isono N, Yoshikawa M, Asao A, Hamada S, Watanabe K, Ito H & Matsui H (2004) Characterization of starch synthase I and II expressed in early developing seeds of kidney bean (Phaseolus vulgaris L.). Biosci Biotech Biochem 68, 19491960.
  • 31
    Imparl-Radosevich JM, Li P, Zhang L, McKean AL, Keeling PL & Guan H (1998) Purification and characterization of maize starch synthase I and its truncated forms. Arch Biochem Biophys 353, 6472.
  • 32
    Edwards A, Borthakur A, Bornemann S, Venail J, Denyer K, Waite D, Fulton D, Smith A & Martin C (1999) Specificity of starch synthase isoforms from potato. Eur J Biochem 266, 724736.
  • 33
    Commuri PD & Keeling PL (2001) Chain-length specificities of maize starch synthase I enzyme: studies of glucan affinity and catalytic properties. Plant J 25, 475486.
  • 34
    Edwards A, Marshall J, Sidebottom C, Visser RGF, Smith AM & Martin C (1995) Biochemical and molecular characterization of a novel starch synthase from potato tubers. Plant J 8, 283294.
  • 35
    Bradford MM (1976) A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248254.
  • 36
    Isono N, Ito H, Senoura T, Yoshikawa M, Nozaki K & Matsui H (2003) Molecular cloning of a cDNA encoding granule-bound starch synthase I from kidney bean and expression in Escherichia coli. J Appl Glycosci 50, 355360.
  • 37
    Dubois M, Gilles KA, Hamilton JK, Rebers PA & Smith F (1956) Colorimetric method for determination of sugars and related substances. Anal Chem 28, 350356.
  • 38
    Hizukuri S, Takeda Y & Yasuda M (1981) Multi-branched nature of amylose and the action of debranching enzymes. Carbohydr Res 94, 205213.