Direct Appraisal of the Potato Tuber ADP-glucose Pyrophosphorylase Large Subunit in Enzyme Function by Study of a Novel Mutant Form
2008; Elsevier BV; Volume: 283; Issue: 11 Linguagem: Inglês
10.1074/jbc.m707447200
ISSN1083-351X
AutoresSeon‐Kap Hwang, Yasuko Nagai, Dongwook Kim, Thomas W. Okita,
Tópico(s)Plant nutrient uptake and metabolism
ResumoThe higher plant ADP-glucose pyrophosphorylase is a heterotetramer consisting of two subunit types, which have evolved at different rates from a common ancestral gene. The potato tuber small subunit (SS) displays both catalytic and regulatory properties, whereas the exact role of the large subunit (LS), which contains substrate and effector binding sites, remains unresolved. We identified a mutation, S302N, which increased the solubility of the recombinant potato tuber LS and, in turn, enabling it to form a homotetrameric structure. The LS302N homotetramer possesses very little enzyme activity at a level 100-fold less than that seen for the unactivated SS homotetramer. Unlike the SS enzyme, however, the LS302N homotetramer enzyme is neither activated by the effector 3-phosphoglycerate nor inhibited by Pi. When combined with the catalytically silenced SS, SD143N, however, the LS302N-containing enzyme shows significantly enhanced catalytic activity and restored 3-PGA activation. This unmasking of catalytic and regulatory potential of the LS is conspicuously evident when the activities of the resurrected LK41R·T51K·S302N homotetramer are compared with its heterotetrameric form assembled with SD143N. Overall, these results indicate that the LS possesses catalytic and regulatory properties only when assembled with SS and that the net properties of the heterotetrameric enzyme is a product of subunit synergy. The higher plant ADP-glucose pyrophosphorylase is a heterotetramer consisting of two subunit types, which have evolved at different rates from a common ancestral gene. The potato tuber small subunit (SS) displays both catalytic and regulatory properties, whereas the exact role of the large subunit (LS), which contains substrate and effector binding sites, remains unresolved. We identified a mutation, S302N, which increased the solubility of the recombinant potato tuber LS and, in turn, enabling it to form a homotetrameric structure. The LS302N homotetramer possesses very little enzyme activity at a level 100-fold less than that seen for the unactivated SS homotetramer. Unlike the SS enzyme, however, the LS302N homotetramer enzyme is neither activated by the effector 3-phosphoglycerate nor inhibited by Pi. When combined with the catalytically silenced SS, SD143N, however, the LS302N-containing enzyme shows significantly enhanced catalytic activity and restored 3-PGA activation. This unmasking of catalytic and regulatory potential of the LS is conspicuously evident when the activities of the resurrected LK41R·T51K·S302N homotetramer are compared with its heterotetrameric form assembled with SD143N. Overall, these results indicate that the LS possesses catalytic and regulatory properties only when assembled with SS and that the net properties of the heterotetrameric enzyme is a product of subunit synergy. ADP-glucose pyrophosphorylase (AGPase, EC 2.7.7.27) 2The abbreviations used are:AGPaseADP-glucose pyrophosphorylaseDTTdithiothreitolGlc 1-Pglucose 1-phosphateL or LSlarge subunit3-PGA3-phosphoglycerateS or SSsmall subunit.2The abbreviations used are:AGPaseADP-glucose pyrophosphorylaseDTTdithiothreitolGlc 1-Pglucose 1-phosphateL or LSlarge subunit3-PGA3-phosphoglycerateS or SSsmall subunit. is a key enzyme in the control of starch synthesis in higher plants. The higher plant enzyme has a tetrameric structure consisting of two distinct subunit types, large subunit (LS) and small subunit (SS) (1Okita T.W. Nakata P.A. Anderson J.M. Sowokinos J. Morell M. Preiss J. Plant. Physiol. 1990; 93: 785-790Crossref PubMed Scopus (150) Google Scholar, 2Iglesias A.A. Barry G.F. Meyer C. Bloksberg L. Nakata P.A. Greene T. Laughlin M.J. Okita T.W. Kishore G.M. Preiss J. J. Biol. Chem. 1993; 268: 1081-1086Abstract Full Text PDF PubMed Google Scholar, 3Slattery C.J. Kavakli I.H. Okita T.W. Trends Plant Sci. 2000; 5: 291-298Abstract Full Text Full Text PDF PubMed Scopus (142) Google Scholar, 4Ballicora M.A. Iglesias A.A. Preiss J. Photosynth. Res. 2004; 79: 1-24Crossref PubMed Scopus (234) Google Scholar). These subunit types share considerable sequence identity (∼53%) and similarity (∼73%) indicating that both subunits originated from a common gene ancestor which has diverged at different rates over time (5Georgelis N. Braun E.L. Shaw J.R. Hannah L.C. Plant Cell. 2007; 19: 1458-1472Crossref PubMed Scopus (52) Google Scholar, 6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar, 7Ballicora M.A. Dubay J.R. Devillers C.H. Preiss J. J. Biol. Chem. 2005; 280: 10189-10195Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). When expressed in bacteria, the SS (SWT) of the potato tuber AGPase is capable of forming a catalytically active homotetrameric enzyme (8Salamone P.R. Greene T.W. Kavakli I.H. Okita T.W. FEBS Lett. 2000; 482: 113-118Crossref PubMed Scopus (32) Google Scholar). Compared with the wild-type AGPase heterotetramer, the potato SWT homotetramer requires nearly 24-fold greater levels of 3-PGA for maximal enzyme activity and is more sensitive to the inhibitor inorganic phosphate (Pi) (8Salamone P.R. Greene T.W. Kavakli I.H. Okita T.W. FEBS Lett. 2000; 482: 113-118Crossref PubMed Scopus (32) Google Scholar, 9Hwang S.K. Salamone P.R. Okita T.W. FEBS Lett. 2005; 579: 983-990Crossref PubMed Scopus (40) Google Scholar). When fully activated by high concentrations of 3-PGA, the kinetic behavior of the SWT homotetramer is similar to the wild-type heterotetramer and cyanobacterial AGPases (10Gomez-Casati D.F. Preiss J. Iglesias A.A. Arch. Biochem. Biophys. 2000; 384: 319-326Crossref PubMed Scopus (7) Google Scholar). These observations indicate that the LS is essential for optimal allosteric regulatory properties of the heterotetrameric enzyme. It is also noteworthy that introduction of only one or two amino acid changes generates a SS homotetramer with regulatory properties that are comparable to or more sensitive to 3-PGA than the wild-type heterotetrameric enzymes (8Salamone P.R. Greene T.W. Kavakli I.H. Okita T.W. FEBS Lett. 2000; 482: 113-118Crossref PubMed Scopus (32) Google Scholar, 11Salamone P.R. Kavakli I.H. Slattery C.J. Okita T.W. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 1070-1075Crossref PubMed Scopus (44) Google Scholar).Earlier studies (12Ballicora M.A. Frueauf J.B. Fu Y. Schurmann P. Preiss J. J. Biol. Chem. 2000; 275: 1315-1320Abstract Full Text Full Text PDF PubMed Scopus (134) Google Scholar, 13Fu Y. Ballicora M.A. Leykam J.F. Preiss J. J. Biol. Chem. 1998; 273: 25045-25052Abstract Full Text Full Text PDF PubMed Scopus (123) Google Scholar) suggested that the inter-subunit disulfide bond between two SSs of the potato tuber AGPase heterotetramer also plays a significant role in controlling enzyme activity and stability. When the bond was cleaved by reduction of cysteine residues by DTT or thioredoxins, the AGPase became more sensitive to activation by low levels of 3-PGA and less stable at elevating temperature. This redox modification of the plant AGPase was more evident in planta and significantly affected AGPase activities and starch synthesis (14Geigenberger P. Kolbe A. Tiessen A. J. Exp. Bot. 2005; 56: 1469-1479Crossref PubMed Scopus (168) Google Scholar, 15Kolbe A. Tiessen A. Schluepmann H. Paul M. Ulrich S. Geigenberger P. Proc. Natl. Acad. Sci. U. S. A. 2005; 102: 11118-11123Crossref PubMed Scopus (296) Google Scholar, 16Tiessen A. Hendriks J.H. Stitt M. Branscheid A. Gibon Y. Farre E.M. Geigenberger P. Plant Cell. 2002; 14: 2191-2213Crossref PubMed Scopus (339) Google Scholar, 17Hendriks J.H. Kolbe A. Gibon Y. Stitt M. Geigenberger P. Plant Physiol. 2003; 133: 838-849Crossref PubMed Scopus (312) Google Scholar).By contrast, less information is available on the role of the LS subunit in enzyme function. Unlike the SS, the potato tuber AGPase LS is unable to self-assemble into an active tetramer when expressed in bacteria and readily forms inclusion bodies. Hence, all available evidence on the role of LS for the potato tuber AGPase activity has been obtained by studying its operability with the SS. Site-directed mutagenesis studies on conserved residues predicted to bind to effectors and substrates in the potato tuber subunits, suggest that the SS is catalytic and regulatory, whereas the LS lacked these properties and simply modulates the regulatory activity of the SS by protein-to-protein interactions (18Fu Y. Ballicora M.A. Preiss J. Plant Physiol. 1998; 117: 989-996Crossref PubMed Scopus (47) Google Scholar, 19Frueauf J.B. Ballicora M.A. Preiss J. Plant J. 2003; 33: 503-511Crossref PubMed Scopus (39) Google Scholar). An alternative view on subunit roles in enzyme function was obtained by the study of potato heterotetrameric enzymes formed by different combinations of wild-type and mutant LSs and SSs (9Hwang S.K. Salamone P.R. Okita T.W. FEBS Lett. 2005; 579: 983-990Crossref PubMed Scopus (40) Google Scholar). Such results showed that the net allosteric properties of the heterotetrameric enzyme is contributed by both subunits and is a product of synergy between LS and SS interactions. A similar conclusion for a regulatory role for the LS was also made for the maize endosperm AGPase (20Cross J.M. Clancy M. Shaw J.R. Greene T.W. Schmidt R.R. Okita T.W. Hannah L.C. Plant Physiol. 2004; 135: 137-144Crossref PubMed Scopus (81) Google Scholar).The LS of the potato tuber AGPase is also likely to significantly participate in catalysis by affecting the capacity of the heterotetrameric enzyme to bind substrates and effectors (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar). Substantial evidence for a catalytic role, albeit indirect, for the LS was obtained by identification of a substrate binding site for ATP by photoaffinity labeling studies with 8-N3-ATP (21Hwang S.K. Hamada S. Okita T.W. FEBS Lett. 2006; 580: 6741-6748Crossref PubMed Scopus (23) Google Scholar). Mutations of selected residues located within ATP binding site of the LS significantly altered the catalytic properties of the AGPase heterotetramer.Based on homology studies of other structurally related sugar nucleotide pyrophosphorylases (22Brown K. Pompeo F. Dixon S. Mengin-Lecreulx D. Cambillau C. Bourne Y. EMBO J. 1999; 18: 4096-4107Crossref PubMed Scopus (167) Google Scholar, 23Blankenfeldt W. Asuncion M. Lam J.S. Naismith J.H. EMBO J. 2000; 19: 6652-6663Crossref PubMed Scopus (160) Google Scholar, 24Frueauf J.B. Ballicora M.A. Preiss J. J. Biol. Chem. 2001; 276: 46319-46325Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar), the metal-binding Asp143 (originally assigned as Asp145) of the potato AGPase SS was suggested to be essential for catalytic activity, a prediction verified by site-directed mutagenesis of this residue, which lowered catalytic rate more than four orders of magnitude compared with the wild-type enzyme. A similar mutation in the conserved Asp158 of the LS, however, reduced catalysis only 1.5- to 2.6-fold (19Frueauf J.B. Ballicora M.A. Preiss J. Plant J. 2003; 33: 503-511Crossref PubMed Scopus (39) Google Scholar). Our recent result, however, showed that the replacement of Asp158 by Leu lowered catalytic rates by 11-fold and catalytic efficiencies by 23- to 26-fold depending on the substrates of the enzyme (21Hwang S.K. Hamada S. Okita T.W. FEBS Lett. 2006; 580: 6741-6748Crossref PubMed Scopus (23) Google Scholar), indicating the Asp158 residue in the LS is important for enzyme catalysis. Interestingly, introduction of a single residue replacement, T51K, or double mutations, K41R and T51K, into the wild-type LS resurrected partially (5% of wild-type activity) the catalytic activity of the heterotetrameric enzyme containing the catalytic silenced SD143N (7Ballicora M.A. Dubay J.R. Devillers C.H. Preiss J. J. Biol. Chem. 2005; 280: 10189-10195Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). Overall, these results, together with those obtained from earlier studies on substrate and effector binding, indicate that the LS is catalytically inefficient but has many of the necessary elements for this function.Although available evidence indicates that the LS, which is capable of binding substrates (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar, 18Fu Y. Ballicora M.A. Preiss J. Plant Physiol. 1998; 117: 989-996Crossref PubMed Scopus (47) Google Scholar) and effectors (25Ball K. Preiss J. J. Biol. Chem. 1994; 269: 24706-24711Abstract Full Text PDF PubMed Google Scholar), is catalytically defective, direct evidence for this property has yet to be obtained as all available evidence in support of this view is deduced from studies of the heterotetrameric enzyme forms. To further enhance our understanding of the role of LS in enzyme function, we identified a mutant LS containing a S302N replacement, which significantly elevated solubility of the LS and enabling assembly and formation of LS homotetramer. Results from biochemical and kinetic analysis of LS homotetramer and heterotetrameric forms with catalytic-silenced SD143N showed that the LS in the absence of SS displays very low catalytic activity and is allosteric-insensitive. When operating with the SS, however, catalysis by the LS is stimulated, and allosteric regulatory properties of the LS are unmasked.EXPERIMENTAL PROCEDURESMaterials—[14C]Glucose 1-phosphate and [32P]pyrophosphate were purchased from ICN Pharmaceuticals and PerkinElmer Life Sciences, respectively. Radioactive 8-azidoadenosine 5′-[α-32P]triphosphate ([α-32P]8-N3-ATP) and non-radioactive 8-N3-ATP were purchased from Affinity Labeling Technologies, Inc. Reagents including ATP, Glc 1-P, and ADP-glucose were obtained from the Sigma-Aldrich and were of analytical grade or higher.Expression and Purification of the AGPase Proteins—The wild-type and various LS mutants were expressed in the absence or presence of wild-type SS (SWT) or catalytic-silenced D143N SS (SSi) in the Escherichia coli EA345 host strain, which contains a null mutation in the AGPase gene glgC (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar). The plasmids pSH274 and pSH208 (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar) were used for expression of the His6-tagged LS and SS of the potato tuber AGPase, respectively. For high throughput expression of AGPase subunits E. coli EA3457 cell was made by transforming the pRARE plasmid isolated from Rosseta™ cell (Novagen) into EA345 cell. For purification of the SS homotetramer the His6-tagged SWT protein (26Hwang S.K. Salamone P.R. Kavakli H. Slattery C.J. Okita T.W. Protein Expression Purif. 2004; 38: 99-107Crossref PubMed Scopus (17) Google Scholar) was expressed in EA3457 cells. AGPases were purified as described previously using a Bio-Logic DuoFlow Chromatography system (Bio-Rad) (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar, 21Hwang S.K. Hamada S. Okita T.W. FEBS Lett. 2006; 580: 6741-6748Crossref PubMed Scopus (23) Google Scholar) with minor modifications. Briefly, the active AGPase fractions obtained from a DEAE-Sepharose FF (Amersham Biosciences) chromatography and TALON™-IMAC (Clontech Lab) were subjected to POROS 20 HQ (PerSeptive Biosystems) column chromatography. Aliquots of the purified enzyme preparation were stored frozen in liquid nitrogen and shelved at –80 °C until used for analysis.AGPase Assay and Kinetics—AGPase activities were assayed in the pyrophosphorylase direction (Assay A) during enzyme purification and in the ADP-glucose synthesis direction (Assay B) for kinetic characterization (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar). Unless stated, saturating amount of substrates was used for both methods. One enzyme unit is defined as 1 micromole of ATP (Assay A) or ADP-glucose (Assay B) formed for 1 min at 37 °C. KaleidaGraph 3.5 (Synergy software) was used to fit the experimental data to the modified Hill equations (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar). The kinetic values (S0.5, A0.5, kcat, and kcat/S0.5) were determined as described previously (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar).Chemical and Site-directed Mutagenesis—200 μg of plasmid DNA (pSH345 or pSH274 containing LK41R·T51K) were subjected to chemical mutagenesis by incubating at 37 °C for 20 h in 4 ml of 0.1 m sodium phosphate (pH 6.0), 0.8 m hydroxylamine-HCl, and 1 mm EDTA (27Greene T.W. Chantler S.E. Kahn M.L. Barry G.F. Preiss J. Okita T.W. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 1509-1513Crossref PubMed Scopus (49) Google Scholar). After neutralization with 400 μl of 1.5 m Tris-HCl (pH 8.8) the plasmid DNAs were then precipitated in 80% (v/v) ethanol. The plasmid DNAs were then used to transform E. coli EA345 cells expressing SSi. Cells were grown on NZCYM media supplemented with 100 mg/liter ampicillin and 50 mg/liter kanamycin with 0.4% (v/v) glycerol as the carbon source, which readily allows the efficient induced expression of both subunits from the lac promoters (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar). After overnight culture at 37 °C, the bacterial colonies were screened for glycogen production by exposure to iodine vapor. Plasmid DNAs expressing the LS mutants were purified and were subjected to DNA sequencing to identify changes in nucleotide sequence.Site-directed mutagenesis of the AGPase subunit sequence was accomplished using the QuikChange site-directed mutagenesis kit (Stratagene) as described previously (21Hwang S.K. Hamada S. Okita T.W. FEBS Lett. 2006; 580: 6741-6748Crossref PubMed Scopus (23) Google Scholar): LS302N-F, 5′-AAATCGTTTTATAATGCTaacTTGGCACTCACACAAGAG-3′ and LS302N-R, 5′-CTCTTGTGTGAGTGCCAAgttAGCATTATAAAACGATTT-3′ (target sequences are in lowercase).Labeling of AGPases with 8-Azido-adenosine 5′-Triphosphate—The purified AGPases were reduced, desalted, and labeled as described previously (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar, 21Hwang S.K. Hamada S. Okita T.W. FEBS Lett. 2006; 580: 6741-6748Crossref PubMed Scopus (23) Google Scholar).Modeling of AGPase Structures—DeepView, the Swiss-Pdb-Viewer was used for modeling of the three-dimensional structure of the LS based on the crystal structure of the potato SS homotetramer (28Jin X. Ballicora M.A. Preiss J. Geiger J.H. EMBO J. 2005; 24: 694-704Crossref PubMed Scopus (125) Google Scholar) as the template scaffold. Coordinates for the AGPase structures (1yp3) were retrieved from RCSB Protein Data Bank. Pov-Ray was used for rendering the structure.Protein Analysis—Protein concentration was measured using the Advanced Protein Assay Reagent from Cytoskeleton (Denver, CO) with bovine serum albumin (fraction V) as the standard. SDS-PAGE was performed as described in a previous study (26Hwang S.K. Salamone P.R. Kavakli H. Slattery C.J. Okita T.W. Protein Expression Purif. 2004; 38: 99-107Crossref PubMed Scopus (17) Google Scholar). Immunoblot analysis was performed using anti-potato AGPase LS or anti-potato AGPase SS as described before (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar).RESULTSSubstitution of Ser302 with Asn in the LS Significantly Enhanced Glycogen Production in Bacterial Host Cells—To provide further insights on the role of the potato tuber LS in AGPase function, the expression plasmid DNA containing His6-tagged LK41R·T51K (abbreviated LRK) was subjected to chemical mutagenesis (supplemental Fig. S1A) and co-expressed with the catalytic-silenced SD143N (abbreviated SSi) mutant (7Ballicora M.A. Dubay J.R. Devillers C.H. Preiss J. J. Biol. Chem. 2005; 280: 10189-10195Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar, 21Hwang S.K. Hamada S. Okita T.W. FEBS Lett. 2006; 580: 6741-6748Crossref PubMed Scopus (23) Google Scholar) in bacterial cells lacking AGPase activity. These cells were then analyzed for glycogen production by exposure to iodine vapor. Under these conditions, cell expressing LWTSSi are devoid of glycogen and do not stain, whereas cells expressing LRKSSi enzyme stain lightly as they accumulate small but readily detectable levels of glycogen due to the low catalytic activity of this enzyme (supplemental Fig. S1B). More than 6 × 104 bacterial colonies derived from the mutagenized potato tuber AGPase LS were examined by iodine staining resulting in the identification of 26 colonies, which exhibited significantly enhanced glycogen accumulation (supplemental Fig. S1B). Interestingly, sequence analysis showed that all 26 LS sequences obtained from these excess glycogen accumulating cells contained a common third mutation, S302N, in addition to the pre-existing K41R and T51K mutations (for brevity, LS containing these three mutations will be denoted as LRKN). Six plasmids contained silent mutations in addition to S302N, whereas a single plasmid contained a fourth mutation V159I in LS. Thus, addition of S302N mutation in the LS enhances net AGPase activity which, in turn, increases glycogen production.The S302N Mutation Dramatically Increases LS Solubility—Heterotetrameric forms of the LRKSSi and mutant LRKNSSi enzymes were purified to near homogeneity (>95%) by multiple chromatography steps. During the final clean-up of the enzyme activity using a strong anion-exchange chromatography step, an anomalous protein elution profile was observed for the LRKNSSi enzyme. Typically, the wild-type AGPase elutes from the anion-exchange column as a single major protein peak at 0.3 m NaCl with very little protein eluting at lower salt concentrations (21Hwang S.K. Hamada S. Okita T.W. FEBS Lett. 2006; 580: 6741-6748Crossref PubMed Scopus (23) Google Scholar). Similar elution profile was also observed for the LRKSSi enzyme (Fig. 1A). Although much of the LRKNSSi enzyme also eluted at 0.3 m NaCl, a prominent peak was also observed eluting at 0.15 m NaCl (peak a of Fig. 1A, lower panel). Results from SDS-PAGE and immunoblot analysis showed that the 0.15 m NaCl fraction contained only the LS (Fig. 1B), indicating that a significant amount of LRKN was unassembled with the SS and remained soluble throughout the enzyme purification steps. Analysis of the 0.15 m NaCl protein fraction by Superdex-200 gel filtration chromatography indicated that the LRKN eluted with a molecular size of 223 kDa (supplemental Fig. S2). Hence, the soluble LRKN readily self-assembles into a homotetrameric form. Further studies showed that the soluble LN also readily self-assembles into a 220 kDa oligomer (supplemental Fig. S2).S302N Mutation Is Responsible for the Enhanced Solubility and Formation of the Potato LS Homotetramer—To determine whether the Asn302 residue was responsible for the increase in solubility of the potato tuber LS we compared the total and soluble expression of various LSs containing or lacking this mutation. When expressed in E. coli cells alone without the SS counterpart, all of the various LS types were expressed at nearly equal levels when total cell extracts were examined by SDS-PAGE (supplemental Fig. S3A). When only the soluble fractions were examined, however, only the LS forms containing S302N were found at substantial levels while the same LS forms lacking S302N were present at very low levels indicating that the bulk of the expressed LS was insoluble (supplemental Fig. S3, B and C). Quantitative analysis of the soluble forms of the homotetramers after enzyme purification showed that LRKN and LN were 376- and 77-fold more soluble than LKR and wild-type L (LWT), respectively (supplemental Table S1). These observations indicate that introduction of an Asn residue at position 302 is responsible for the enhanced solubility of the potato tuber LS.The LS Homotetramer Is Not Affected by Allosteric Effectors—Examination of the enzymatic activity of purified LN homotetrameric protein showed that it had very low but measurable activity (∼0.4 unit/g). This activity was 93-fold less than that measured for the homotetramer SWT (37 units/g) when assayed in the absence of the activator 3-PGA (Table 1). AGPase activity was elevated >800-fold by the introduction of K41R and T51K substitutions into LN to yield LRKN (330 units/g).TABLE 1Regulatory properties of the AGPasesSubunit3-PGA, A0.5Specific activityLargeSmallNo 3-PGA3-PGAaSpecific activity was obtained at saturated concentrations of 3-PGA except that 0.5 mm 3-PGA was used to measure activity of LRKN and LN.ActivationbFold activation = maximal specific activity in the presence of saturated concentration of 3-PGA/specific activity in the absence of 3-PGA.μmnHunits/g-foldLRKSSi25 ± 1(0.8)63 ± 103,200 ± 4551LRKNSSi6 ± 1(0.9)48 ± 53,400 ± 16071LNSSi78 ± 17(0.7)0.4 ± 0.112 ± 130LRKN—n/ccn/c: non-calculable due to lack of activation by 3-PGA (LRKN) or very low enzyme activity (LN and SSi).(n/c)330 ± 12320 ± 31LN—n/c(n/c)0.4 ± 0.10.4 ± 0.11—SSin/c(n/c)<0.2 1 mm (Fig. 2). Likewise, LRKN homotetramer activity was not affected by the metabolic inhibitor, Pi, in the presence of 0.5 mm 3-PGA (Fig. 3). Hence, these results indicate that the LRKN homotetramer is not allosterically regulated by 3-PGA or Pi. Other metabolites (0.5 mm each of 2-PGA, phosphoenol pyruvate, fructose 6-phosphate, fructose 1,6-diphosphate, and AMP) tested showed no discernible effects on enzyme activity of the LS homotetramer (supplemental Fig. S4).FIGURE 23-PGA activation profiles of the AGPase wild-type and mutants. Reaction was performed in the forward (ADP-glucose synthesis) direction at various 3-PGA concentrations under saturating substrate conditions (A–F) (see legend of Table 1). The experimental data were analyzed using modified Hill equation (6Hwang S.K. Hamada S. Okita T.W. Phytochemistry. 2007; 68: 464-477Crossref PubMed Scopus (36) Google Scholar) with the curve fitting software, KaleidaGraph 3.5. A, LWTSWT; B, LRKNSWT; C, LNSWT; D, SWT; E, LRKN; and F, LN.View Large Image Figure ViewerDownload Hi-res image Download (PPT)FIGURE 3Phosphate inhibition profiles of the AGPases. Reactions were performed in the forward (ADP-glucose synthesis) direction under saturating substrate conditions and with 0.5 mm 3-PGA (see legend of Table 1). The smooth fit feature of KaleidaGraph 3.5 was used for curve fitting. LRKN (▵), SWT (□), and LRKNSWT (○).View Large Image Figure ViewerDownload Hi-res image Download (PPT)In addition to the homotetrameric enzymes, we also evaluated the potential allosteric regulatory properties of these LS types when assembled with SSi. Interestingly, unlike the LN and LRKN homotetramers, which are not subject to allosteric activation by 3-PGA, LNSSi, LRKSSi, and LRKNSSi readily respond to this effector. Activity of LNSSi increased 30-fold in the presence of 3-PGA. LRKSSi and LRKNSSi showed a different activity pattern in the absence and presence of 3-PGA. Without 3-PGA, the activities of the heterotetramers LRKSSi and LRKNSSi were significantly lower (5- and 7-fold, respectively) than the LRKN homotetramer, whereas the presence of 3-PGA mediated a 51- to 71-fold increases in enzyme activities of these heterotetramers.The LS Mutant Homotetramer Is Catalytically Very Inefficient—The LRKN homotetramer showed significantly low affinity toward ATP and Glc 1-P compared with the heterotetrameric enzyme forms (Table 2). The S0.5 values for ATP were 1927 μm (nH = 1.9) in the presence of 0.5 mm 3-PGA and 1880 μm (nH = 1.6) in the absence of 3-PGA. The S0.5 values for Glc 1-P were 4307 or 4175 μm (nH = 1.3 or 1.4) in the presence or absence of 0.5 mm 3-PGA. When kinetic study was done in the pyrophosphorylase direction, similar trends in S0.5 values were obtained for ADP-glucose and PPi with the LRKN homotetramer having S0.5 values of 3- to 5-fold and 2- to 3-fold higher than LRKNSWT, respectively (Table 3). These results also indicate that the catalytic properties of the LS mutant are not affected by 3-PGA, and the LS mutant is less efficient at substrate binding than the enzyme forms containing SS.TABLE 2Catalytic properties of the various AGPases in the ADP-gluco
Referência(s)