Identification of the catalytic residues of bifunctional glycogen debranching enzyme

Eukaryotic glycogen debranching enzyme (GDE) possesses two different catalytic activities (oligo-1,4-->1,4-glucantransferase/amylo-1,6-glucosidase) on a single polypeptide chain. To elucidate the structure-function relationship of GDE, the catalytic residues of yeast GDE were determined by site-directed mutagenesis. Asp-535, Glu-564, and Asp-670 on the N-terminal half and Asp-1086 and Asp-1147 on the C-terminal half were chosen by the multiple sequence alignment or the comparison of hydrophobic cluster architectures among related enzymes. The five mutant enzymes, D535N, E564Q, D670N, D1086N, and D1147N were constructed. The mutant enzymes showed the same purification profiles as that of wild-type enzyme on beta-CD-Sepharose-6B affinity chromatography. All the mutant enzymes possessed either transferase activity or glucosidase activity. Three mutants, D535N, E564Q, and D670N, lost transferase activity but retained glucosidase activity. In contrast, D1086N and D1147N lost glucosidase activity but retained transferase activity. Furthermore, the kinetic parameters of each mutant enzyme exhibiting either the glucosidase activity or transferase activity did not vary markedly from the activities exhibited by the wild-type enzyme. These results strongly indicate that the two activities of GDE, transferase and glucosidase, are independent and located at different sites on the polypeptide chain.     

Crystallization of glycogen debranching enzyme

Crystals of glycogen debranching enzyme from rabbit skeletal muscle have been obtained from solutions of polyethylene glycol 8000 (pH 7.3) containing 10 mM-linear oligosaccharides of lengths from three to seven glucose units in alpha-1,4 linkage. Preliminary X-ray precession photographs indicate an orthorhombie unit cell with dimensions of a = 106.4 A, b = 195.7 A and c = 93.0 A. The space group is P212121 with one monomer per asymmetric unit.     

Fluorogenic substrates of glycogen debranching enzyme for assaying debranching activity

Glycogen debranching enzyme (GDE) degrades glycogen in concert with glycogen phosphorylase. GDE has two distinct active sites for maltooligosaccharide transferase and amylo-1,6-glucosidase activities. Phosphorylase limit dextrin from glycogen is debranched by cooperation of the two activities. Fluorogenic branched dextrins were prepared as substrates of GDE from pyridylaminated maltooctaose (PA-maltooctaose) and maltotetraose, taking advantage of the synthetic action of Klebsiella pneumoniae pullulanase. Their structures were as follows: Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4(Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-6)Glcalpha1-4Glcalpha1-4GlcPA (B3), Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4(Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-6)Glcalpha1-4Glcalpha1-4Glcalpha1-4GlcPA (B4), Glcalpha1-4Glcalpha1-4Glcalpha1-4(Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-6)Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4GlcPA (B5), Glcalpha1-4Glcalpha1-4(Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-6)Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4GlcPA (B6), Glcalpha1-4(Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-6)Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4GlcPA (B7), and Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-6Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4GlcPA (B8). These dextrins were incubated with porcine skeletal muscle GDE. No fluorogenic product was found in the digest of B8. The fluorogenic products from B3, B4, and B5 were PA-maltooctaose only. PA-maltooctaose, PA-maltoundecaose, and 6(7)-O-alpha-glucosyl-PA-maltooctaose were from B7. PA-maltooctaose and 6(6)-O-alpha-glucosyl-PA-maltooctaose were from B6. These results indicate that the maltooligosaccharide transferase removed the maltotriosyl residues from the maltotetraosyl branches by hydrolysis or intramolecular transglycosylation to expose 6-O-alpha-glucosyl residues, and then the amylo-1,6-glucosidase hydrolyzed the alpha-1,6-glycosidic linkages of the products rapidly. Probably, 6-O-alpha-glucosyl-PA-maltooctaoses from B7 and B6 were less susceptible to the amylo-1,6-glucosidase than were those from B3, B4, and B5. Taking this into account, B3, B4, and B5 are suitable substrates for GDE assay.     

Glycogen debranching pathway deduced from substrate specificity of glycogen debranching enzyme

Glycoconjugate Journal - Glycogen debranching enzyme (GDE) is bifunctional in that it exhibits both 4-α-glucanotransferase and amylo-α-1,6-glucosidase activity at two distinct catalytic...     

Binding of glycogen, oligosaccharides, and glucose to glycogen debranching enzyme

The binding of glucose and a series of oligosaccharides to glycogen debranching enzyme was determined by the ability of the saccharides to decrease the rate of reaction of sulfhydryl groups with 5,5'-dithiobis(2-nitrobenzoate) (DTNB). At pH 7.2, the strength of binding increases with chain length from glucose to maltotriose to maltopentaose but not to maltohexaose, and the free energies for binding of the oligosaccharides suggest subsites of equivalent affinities for the four glucose units following the initial reducing moiety. The rate of reaction of DTNB with enzyme saturated with saccharide is the same for all compounds, suggesting that all the saccharides, including glucose, induce the same conformational state. The site of binding may be that which binds the alpha-1,6-linked side chain of the natural limit dextrin substrate. At pH 8.0, this site exhibits similar characteristics, but an additional site, which may bind the four terminal glucose units of the main chain of the natural substrate, is manifested and exhibits different characteristics, including a very low affinity for glucose itself. The binding of glycogen to the debranching enzyme was monitored by centrifugal separation from the protein and exhibits a much lower dissociation constant than that for the oligomers, suggesting that branched polymers have more than one set of subsites.     

Donor substrate specificity of 4-alpha-glucanotransferase of porcine liver glycogen debranching enzyme and complementary action to glycogen phosphorylase on debranching

Glycogen debranching enzyme (GDE) has both 4-alpha-glucanotransferase and amylo-alpha-1,6-glucosidase activities. Here, we examined 4-alpha-glucanotransferase action of porcine liver GDE on four 6(4)-O-alpha-maltooligosyl-pyridylamino(PA)-maltooctaoses, in the presence or absence of an acceptor, maltohexaose. HPLC analysis of digested fluorogenic branched dextrins revealed that in the presence or absence of acceptor, 6(4)-O-alpha-glucosyl-PA-maltooctaose (B4/81) was liberated from 6(4)-O-alpha-maltopentaosyl-PA-maltooctaose (B4/85), 6(4)-O-alpha-maltotetraosyl-PA-maltooctaose (B4/84) and 6(4)-O-alpha-maltotriosyl-PA-maltooctaose (B4/83), whereas 6(4)-O-alpha-maltosyl-PA-maltooctaose (B4/82) was resistant to the enzyme. The fluorogenic product was further hydrolyzed by amylo-alpha-1,6-glucosidase to PA-maltooctaose (G8PA) and glucose. The ratio of the rates of 4-alpha-glucanotransferase actions on B4/85, B4/84 and B4/83 in the absence of the acceptor was 0.15, 0.42 and 1.00, respectively. The rates increased with increasing amounts of acceptor, changing the ratio of the rates to 0.09, 1.00 and 0.60 (with 0.5 mM maltohexaose) and 0.10, 1.00 and 0.58 (with 1.0 mM maltohexaose), respectively. Donor substrate specificity of GDE 4-alpha-glucanotransferase suggests complementary action of GDE and glycogen phosphorylase on glycogen degradation in the porcine liver. Glycogen phosphorylase degrades the maltooligosaccharide branches of glycogen by phosphorolysis to form maltotetraosyl branches, and phosphorolysis does not proceed further. GDE 4-alpha-glucanotransferase removes a maltotriosyl residue from the maltotetraosyl branch such that the alpha-1,6-linked glucosyl residue is retained.     

Association of bi-functional activity in the N-terminal domain of glycogen debranching enzyme

Glycogen debranching enzyme (GDE) in mammals and yeast exhibits α-1,4-transferase and α-1,6-glucosidase activities within a single polypeptide chain and facilitates the breakdown of glycogen by a bi-functional mechanism. Each enzymatic activity of GDE is suggested to be associated with distinct domains; α-1,4-glycosyltransferase activity with the N-terminal domain and α-1,6-glucosidase activity with the C-terminal domain. Here, we present the biochemical features of the GDE from Saccharomyces cerevisiae using the substrate glucose(n)-β-cyclodextrin (Gn-β-CD). The bacterially expressed and purified GDE N-terminal domain (aa 1–644) showed α-1,4-transferase activity on maltotetraose (G4) and G4-β-CD, yielding various lengths of (G)_n. Surprisingly, the N-terminal domain also exhibited α-1,6-glucosidase activity against G1-β-CD and G4-β-CD, producing G1 and β-CD. Mutational analysis showed that residues D535 and E564 in the N-terminal domain are essential for the transferase activity but not for the glucosidase activity. These results indicate that the N-terminal domain (1–644) alone has both α-1,4-transferase and the α-1,6-glucosidase activities and suggest that the bi-functional activity in the N-domain may occur via one active site, as observed in some archaeal debranching enzymes.     

Purification of glycogen debranching enzyme from porcine brain: evidence for glycogen catabolism in the brain

Amylo-1,6-glucosidase from porcine brain was purified to homogeneity by ammonium sulfate fractionation, followed by sequential steps of liquid chromatography on DEAE-Sephacel, Sephacryl S-300, and Super Q. The purified enzyme had both maltooligosaccharide transferase and amylo-1,6-glucosidase activities within a single polypeptide chain, and the combination of these two activities removed the branches of phosphorylase limit dextrin. Based on these results, the purified enzyme was identified as a glycogen debranching enzyme (GDE). The molecular weight of the brain GDE was 170,000 by gel-filtration and 165,000 by reducing SDS-PAGE. The pH profile of maltooligosaccharide transferase activity coincided with that of the amylo-1,6-glucosidase activity (pH optimum at 6.0). The existence of GDE as well as glycogen phosphorylase in the brain explains brain glycogenolysis fully and supports the hypothesis that glycogen is a significant source of energy in this organ.     

Cloning and characterization of the glycogen branching enzyme gene existing in tandem with the glycogen debranching enzyme from Pectobacterium chrysanthemi PY35

The glycogen branching enzyme gene (glgB) from Pectobacterium chrysanthemi PY35 was cloned, sequenced, and expressed in Escherichia coli. The glgB gene consisted of an open reading frame of 2196bp encoding a protein of 731 amino acids (calculated molecular weight of 83,859Da). The glgB gene is upstream of glgX and the ORF starts the ATG initiation codon and ends with the TGA stop codon at 2bp upstream of glgX. The enzyme was 43-69% sequence identical with other glycogen branching enzymes. The enzyme is the most similar to GlgB of E. coli and contained the four regions conserved among the alpha-amylase family. The glycogen branching enzyme (GlgB) was purified and the molecular weight of the enzyme was estimated to be 84kDa by SDS-PAGE. The glycogen branching enzyme was optimally active at pH 7 and 30 degrees C.     

Structural rationale for the short branched substrate specificity of the glycogen debranching enzyme GlgX

Glycogen serves as major energy storage in most living organisms. GlgX, with its gene in the glycogen degradation operon, functions in glycogen catabolism by selectively catalyzing the debranching of polysaccharide outer chains in bacterial glycosynthesis. GlgX hydrolyzes α‐1,6‐glycosidic linkages of phosphorylase‐limit dextrin containing only three or four glucose subunits produced by glycogen phosphorylase. To understand its mechanism and unique substrate specificity toward short branched α‐polyglucans, we determined the structure of GlgX from Escherichia Coli K12 at 2.25 Å resolution. The structure reveals a monomer consisting of three major domains with high structural similarity to the subunit of TreX, the oligomeric bifunctional glycogen debranching enzyme (GDE) from Sulfolobus. In the overlapping substrate binding groove, conserved residues Leu270, Asp271, and Pro208 block the cleft, yielding a shorter narrow GlgX cleft compared to that of TreX. Residues 207–213 form a unique helical conformation that is observed in both GlgX and TreX, possibly distinguishing GDEs from isoamylases and pullulanases. The structural feature observed at the substrate binding groove provides a molecular explanation for the unique substrate specificity of GlgX for G4 phosphorylase‐limit dextrin and the discriminative activity of TreX and GlgX toward substrates of varying lengths. Proteins 2010. © 2010 Wiley‐Liss, Inc.     

Purification of glycogen debranching enzyme from rabbit muscle using omega-aminoalkyl agarose chromatography

Rabbit muscle glycogen debranching enzyme binds to all of a homologous series of ω-aminoalkyl agaroses. The debrancher can be eluted from ω-aminoethyl and ω-aminobutyl agarose with 0.5 m NaCl, and it desorbs more readily and elutes sooner from ω-aminoethyl agarose than from ω-aminobutyl agarose. No activity is eluted from ω-aminohexyl, octyl, or decyl agaroses. An improved purification procedure has been developed which includes chromatography on ω-aminoethyl agarose. This procedure enables the isolation of over 90% yield of the debranching enzyme from muscle within 3 days.     

Activation of 4-alpha-glucanotransferase activity of porcine liver glycogen debranching enzyme with cyclodextrins

Glycogen debranching enzyme (GDE) is a single polypeptide chain containing distinct active sites for 4-alpha-glucanotransferase and amylo-alpha-1,6-glucosidase activities. Debranching of phosphorylase limit dextrin from glycogen is carried out by cooperation of the two activities. We examined the effects of cyclodextrins (CDs) on debranching activity of porcine liver GDE using a fluorogenic branched dextrin, Glcalpha1-4Glcalpha1-4Glcalpha1-4(Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-6)Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4GlcPA (B5/84), as a substrate. B5/84 was hydrolyzed by the hydrolytic action of 4-alpha-glucanotransferase to B5/81 and maltotriose. The fluorogenic product was further hydrolyzed by the amylo-alpha-1,6-glucosidase activity to the debranched product, Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4GlcPA (G8PA), and glucose. alpha-, beta- and gamma-CDs accelerated the liberation of B5/81 from B5/84, indicating that the 4-alpha-glucanotransferase activity was activated by CDs to remove the maltotriosyl residue from the maltotetraosyl branch. This led to acceleration of B5/84 debranching. The extent of 4-alpha-glucanotransferase activation increased with CD concentration before reaching a constant value. This suggests that there is an activator binding site and that the binding of CDs stimulates 4-alpha-glucanotransferase activity. In the porcine liver, glycogen degradation may be partially stimulated by the binding of a glycogen branch to this activator binding site.     

Molecular cloning and nucleotide sequence of cDNA encoding human muscle glycogen debranching enzyme.

cDNA comprising the entire length of the human muscle glycogen debranching enzyme was cloned and its nucleotide sequence determined. The debrancher mRNA includes a 4545-base pair coding region and a 2371-base pair 3'-nontranslated region. The calculated molecular mass of the debrancher protein derived from cDNA sequence is 172,614 daltons, consistent with the estimated size of purified protein (Mr 165,000 +/- 500). A partial amino acid sequence (13 internal tryptic peptides with a total of 213 residues) determined on peptides derived from purified porcine muscle debrancher protein confirmed the identity of the cDNA clone. Comparison of the amino acid sequence predicted from the human glycogen debrancher cDNA with the partial protein sequence of the porcine debrancher revealed a high degree (88%) of interspecies sequence identity. RNA blot analysis showed that debrancher mRNA in human muscle, lymphoblastoid cells, and in porcine muscle are all similar in size (approximately 7 kilobases). Two patients with inherited debrancher deficiency had a reduced level of debrancher mRNA, whereas two other patients had no detectable abnormality in RNA blots. The isolation of the debrancher cDNA and determination of its primary structure is an important step toward defining the structure-function relationship of this multifunctional enzyme and in understanding the molecular basis of the type III glycogen storage disease.     

Active site mapping of amylo-alpha-1,6-glucosidase in porcine liver glycogen debranching enzyme using fluorogenic 6-O-alpha-glucosyl-maltooligosaccharides

Glycogen debranching enzyme (GDE) has two enzymatic activities, 4-alpha-glucanotransferase and amylo-alpha-1,6-glucosidase. Products with 6-O-alpha-glucosyl structures formed from phosphorylase limit dextrin by the 4-alpha-glucanotransferase activity are hydrolyzed to glucose by the amylo-alpha-1,6-glucosidase activity. Here, we probed the active site of amylo-alpha-1,6-glucosidase in porcine liver GDE using various 6-O-alpha-glucosyl-pyridylamino (PA)-maltooligosaccharides, with structures (Glcalpha1-4)(m)(Glcalpha1-6)Glcalpha1-4(Glcalpha1-4)(n)GlcPA (GlcPA, 1-deoxy-1-[(2-pyridyl)amino]-D-glucitol residue). Fluorogenic dextrins were prepared from 6-O-alpha-glucosyl-alpha-, beta-, or gamma-cyclodextrin through partial acid hydrolysis, followed by fluorescent tagging of the reducing-end residues of the hydrolysates and separation by gel filtration and reversed-phase HPLC. Porcine liver GDE hydrolyzed dextrins with the structure Glcalpha1-4(Glcalpha1-6)Glcalpha1-4Glc to glucose and the corresponding PA-maltooligosaccharides, whereas other dextrins were not hydrolyzed. Thus, substrates must have two glucosyl residues sandwiching the isomaltosyl moiety to be hydrolyzed. The rate of hydrolysis increased as m increased and reached maximum at m = 4. The rates were the highest when n = 1 but did not vary much with changes in n. Of the dextrins examined, Glcalpha1-4Glcalpha1-4Glcalpha1-4Glcalpha1-4(Glcalpha1-6)Glcalpha1-4Glcalpha1-4GlcPA (6(3)-O-alpha-glucosyl-PA-maltoheptaose) was hydrolyzed most rapidly, suggesting that it fits the best in the amylo-alpha-1,6-glucosidase active site. It is likely that the active site accommodates 6(2)-O-alpha-glucosyl-maltohexaose and that the interactions of seven glucosyl residues with the active site allow the most rapid hydrolysis of the alpha-1,6-glucosidic linkage of the isomaltosyl moiety.     

Glycogen debranching enzyme in bovine brain

Glycogen debranching enzyme was partially purified from bovine brain using a substrate for measuring the amylo-1,6-glucosidase activity. Bovine cerebrum was homogenized, followed by cell-fractionation of the resulting homogenate. The enzyme activity was found mainly in the cytosolic fraction. The enzyme was purified 5,000-fold by ammonium sulfate precipitation, anion-exchange chromatography, gel-filtration, anion-exchange HPLC, and gel-permeation HPLC. The enzyme preparation had no alpha-glucosidase or alpha-amylase activities and degraded phosphorylase limit dextrin of glycogen with phosphorylase. The molecular weight of the enzyme was 190,000 and the optimal pH was 6.0. The brain enzyme differed from glycogen debranching enzyme of liver or muscle in its mode of action on dextrins with an alpha-1,6-glucosyl branch, indicating an amino acid sequence different from those of the latter two enzymes. It is likely that the enzyme is involved in the breakdown of brain glycogen in concert with phosphorylase as in the cases of liver and muscle, but that this proceeds in a somewhat different manner. The enzyme activity decreased in the presence of ATP, suggesting that the degradation of brain glycogen is controlled by the modification of the debranching enzyme activity as well as the phosphorylase.     

Acceptor specificity of 4-alpha-glucanotransferases of mammalian glycogen debranching enzymes

Glycogen debranching enzyme (GDE) has two distinct active sites for its 4-alpha-glucanotransferase and amylo-alpha-1,6-glucosidase activities. The GDE 4-alpha-glucanotransferases of mammals show stringent donor specificity; only alpha-glucans with an alpha-1,6-linked maltotetraosyl or maltotriosyl branch function as donors of a maltotriosyl or maltosyl residue. In this study, we investigated the acceptor specificity of the 4-alpha-glucanotransferases using methyl alpha-maltooligosides, p-nitrophenyl alpha-maltooligosides, and pyridylaminated maltooligosaccharides of various sizes as the acceptor substrates, and phosphorylase limit dextrin as the donor substrate. High-performance liquid chromatography analysis of the transfer products indicated that maltotriosyl and maltosyl residues were specifically transferred from phosphorylase limit dextrin to acceptors with a maltopentaosyl residue comprising a nonreducing-end. These results suggest that the acceptor binding sites in the active sites of mammalian GDE 4-alpha-glucanotransferases are composed of tandem subsites that are geometrically complementary to five glucose residues.     

Structural features of the Nostoc punctiforme debranching enzyme reveal the basis of its mechanism and substrate specificity

The debranching enzyme Nostoc punctiforme debranching enzyme (NPDE) from the cyanobacterium Nostoc punctiforme (PCC73102) hydrolyzes the α‐1,6 glycosidic linkages of malto‐oligosaccharides. Despite its high homology to cyclodextrin/pullulan (CD/PUL)‐hydrolyzing enzymes from glycosyl hydrolase 13 family (GH‐13), NPDE exhibits a unique catalytic preference for longer malto‐oligosaccharides (>G8), performing hydrolysis without the transgylcosylation or CD‐hydrolyzing activities of other GH‐13 enzymes. To investigate the molecular basis for the property of NPDE, we determined the structure of NPDE at 2.37‐Å resolution. NPDE lacks the typical N‐terminal domain of other CD/PUL‐hydrolyzing enzymes and forms an elongated dimer in a head‐to‐head configuration. The unique orientation of residues 25–55 in NPDE yields an extended substrate binding groove from the catalytic center to the dimeric interface. The substrate binding groove with a lengthy cavity beyond the ?1 subsite exhibits a suitable architecture for binding longer malto‐oligosaccharides (>G8). These structural results may provide a molecular basis for the substrate specificity and catalytic function of this cyanobacterial enzyme, distinguishing it from the classical neopullulanases and CD/PUL‐hydrolyzing enzymes. Proteins 2010. © 2009 Wiley‐Liss, Inc.     

Glycogen debranching enzyme: purification, antibody characterization, and immunoblot analyses of type III glycogen storage disease.

Type III glycogen storage disease is caused by a deficiency of glycogen debranching-enzyme activity. Many patients with this disease have both liver and muscle involvement, whereas others have only liver involvement without clinical or laboratory evidence of myopathy. To improve our understanding of the molecular basis of the disease, debranching enzyme was purified 238-fold from porcine skeletal muscle. In sodium dodecyl sulfate-polyacrylamide gel electrophoresis the purified enzyme gave a single band with a relative molecular weight of 160,000 that migrated to the same position as purified rabbit-muscle debranching enzyme. Antiserum against porcine debranching enzyme was prepared in rabbit. The antiserum reacted against porcine debranching enzyme with a single precipitin line and demonstrated a reaction having complete identity to those of both the enzyme present in crude muscle and the enzyme present in liver extracts. Incubation of antiserum with purified porcine debranching enzyme inhibited almost all enzyme activity, whereas such treatment with preimmune serum had little effect. The antiserum also inhibited debranching-enzyme activity in crude liver extracts from both pigs and humans to the same extent as was observed in muscle. Immunoblot analysis probed with anti-porcine-muscle debranching-enzyme antiserum showed that the antiserum can detect debranching enzyme in both human muscle and human liver. The bands detected in human samples by the antiserum were the same size as the one detected in porcine muscle. Five patients with Type III and six patients with other types of glycogen storage disease were subjected to immunoblot analysis. Although anti-porcine antiserum detected specific bands in all liver and muscle samples from patients with other types of glycogen storage disease (Types I, II, and IX), the antiserum detected no cross-reactive material in any of the liver or muscle samples from patients with Type III glycogen storage disease. These data indicate (1) immunochemical similarity of debranching enzyme in liver and muscle and (2) that deficiency of debranching-enzyme activity in Type III glycogen storage disease is due to absence of debrancher protein in the patients that we studied.