In Vitro Polymerization of Heparan Sulfate Backbone by the EXT Proteins
2003; Elsevier BV; Volume: 278; Issue: 42 Linguagem: Inglês
10.1074/jbc.m308314200
ISSN1083-351X
AutoresMarta Busse, Marion Kusche‐Gullberg,
Tópico(s)Connective tissue disorders research
ResumoMultiple exosotoses is a dominantly inherited bone disorder caused by defects in EXT1 and EXT2, genes encoding glycosyltransferases involved in heparan sulfate chain elongation. Heparan sulfate polymerization occurs by the alternating addition of glucuronic acid and N-acetylglucosamine units to the nonreducing end of the polysaccharide. EXT1 and EXT2 are suggested to be dual glucuronyl/N-acetylglucosaminyltransferases, and a heterooligomeric complex of EXT1 and EXT2 (EXT1/2) is considered to be the biological functional polymerization unit. Here, we have investigated the in vitro polymerization capacities of recombinant soluble EXT1, EXT2, and EXT1/2 complex on exogenous oligosaccharide acceptors derived from Escherichia coli K5 capsular polysaccharide. Incubations of recombinant EXT1 or EXT1/2 complex with 3H-labeled oligosaccharide acceptors and the appropriate nucleotide sugars resulted in conversion of the acceptors to higher molecular weight compounds but with different efficacies for EXT1 and EXT1/2. In contrast, incubations with recombinant EXT2 resulted in the addition of a single glucuronic acid but no further polymerization. These results indicate that EXT1 alone and the EXT1/2 heterocomplex can act as heparan sulfate polymerases in vitro without the addition of additional auxiliary proteins. Multiple exosotoses is a dominantly inherited bone disorder caused by defects in EXT1 and EXT2, genes encoding glycosyltransferases involved in heparan sulfate chain elongation. Heparan sulfate polymerization occurs by the alternating addition of glucuronic acid and N-acetylglucosamine units to the nonreducing end of the polysaccharide. EXT1 and EXT2 are suggested to be dual glucuronyl/N-acetylglucosaminyltransferases, and a heterooligomeric complex of EXT1 and EXT2 (EXT1/2) is considered to be the biological functional polymerization unit. Here, we have investigated the in vitro polymerization capacities of recombinant soluble EXT1, EXT2, and EXT1/2 complex on exogenous oligosaccharide acceptors derived from Escherichia coli K5 capsular polysaccharide. Incubations of recombinant EXT1 or EXT1/2 complex with 3H-labeled oligosaccharide acceptors and the appropriate nucleotide sugars resulted in conversion of the acceptors to higher molecular weight compounds but with different efficacies for EXT1 and EXT1/2. In contrast, incubations with recombinant EXT2 resulted in the addition of a single glucuronic acid but no further polymerization. These results indicate that EXT1 alone and the EXT1/2 heterocomplex can act as heparan sulfate polymerases in vitro without the addition of additional auxiliary proteins. Heparan sulfate (HS) 1The abbreviations used are: HS, heparan sulfate; GlcA, d-glucuronic acid; aManR, 2,5-anhydro-d-mannitol (formed by reduction of terminal 2,5-anhydromannose residues with NaBH4). proteoglycans, ubiquitous on cell surfaces and in the extracellular matrix, are composed of extended polysaccharide (glycosaminoglycan) chains covalently attached to various core proteins. An ever-growing number of biological processes are regulated by the interaction of proteins with HS. These interactions play important roles in normal physiological processes, such as organogenesis, angiogenesis, blood coagulation, growth factor signaling, lipid metabolism, etc. but also in pathological processes like tumor metastasis, viral/bacterial adherence, and invasion (1Sasisekharan R. Shriver Z. Venkataraman G. Narayanasami U. Nat. Rev. Cancer. 2002; 2: 521-528Crossref PubMed Scopus (564) Google Scholar, 2Iozzo R.V. J. Clin. Invest. 2001; 108: 165-167Crossref PubMed Scopus (234) Google Scholar, 3Gallagher J.T. J. Clin. Invest. 2001; 108: 357-361Crossref PubMed Scopus (286) Google Scholar, 4Perrimon N. Bernfield M. Nature. 2000; 404: 725-728Crossref PubMed Scopus (667) Google Scholar). The biosynthesis of a HS proteoglycan is a complex event that includes core protein synthesis, polysaccharide-protein linkage formation, polymer formation, and polysaccharide modification reactions (reviewed in Refs. 5Esko J.D. Selleck S.B. Annu. Rev. Biochem. 2002; 71: 435-471Crossref PubMed Scopus (1270) Google Scholar, 6Esko J.D. Lindahl U. J. Clin. Invest. 2001; 108: 169-173Crossref PubMed Scopus (801) Google Scholar, 7Sugahara K. Kitagawa H. Curr. Opin. Struct. Biol. 2000; 10: 518-527Crossref PubMed Scopus (359) Google Scholar, 8Lindahl U. Kusche-Gullberg M. Kjellen L. J. Biol. Chem. 1998; 273: 24979-24982Abstract Full Text Full Text PDF PubMed Scopus (580) Google Scholar). The HS-protein linkage region consists of the tetrasaccharide glucuronic acid-galactose-galactose-xylose (GlcAβ1,3Galβ1,3Galβ1,4Xylβ-), where Xyl is linked to a serine residue in the protein core. Onto this linkage region an α-linked GlcNAc is added, and thereafter polysaccharide assembly continues with the formation of a (GlcAβ1,4GlcNAcα1,4) n polymer that is modified by selective C5 epimerization of GlcA residues to l-iduronic acid units and introduction of N- and O-sulfate groups. The additions of the first five sugar units closest to the protein require the action of at least five separate enzymes (7Sugahara K. Kitagawa H. Curr. Opin. Struct. Biol. 2000; 10: 518-527Crossref PubMed Scopus (359) Google Scholar). The following polymerization step is catalyzed by the HS copolymerase, which is believed to be a complex of EXT1 and EXT2 proteins (reviewed in Ref. 9Zak B.M. Crawford B.E. Esko J.D. Biochim. Biophys. Acta. 2002; 1573: 346-355Crossref PubMed Scopus (149) Google Scholar). Mutational defects in either EXT1 or EXT2 cause hereditary multiple exostoses, an autosomal dominant disorder characterized by a cartilage-capped bony outgrowth at the ends of the long bones (10Ahn J. Ludecke H.J. Lindow S. Horton W.A. Lee B. Wagner M.J. Horsthemke B. Wells D.E. Nat. Genet. 1995; 11: 137-143Crossref PubMed Scopus (382) Google Scholar, 11Stickens D. Clines G. Burbee D. Ramos P. Thomas S. Hogue D. Hecht J.T. Lovett M. Evans G.A. Nat. Genet. 1996; 14: 25-32Crossref PubMed Scopus (297) Google Scholar, 12Wuyts W. Van Hul W. Wauters J. Nemtsova M. Reyniers E. Van Hul E.V. De Boulle K. de Vries B.B. Hendrickx J. Herrygers I. Bossuyt P. Balemans W. Fransen E. Vits L. Coucke P. Nowak N.J. Shows T.B. Mallet L. van den Ouweland A.M. McGaughran J. Halley D.J. Willems P.J. Hum. Mol. Genet. 1996; 5: 1547-1557Crossref PubMed Scopus (167) Google Scholar). The EXT family contains three additional members, designated EXTL1–3 (EXT-like 1–3). The EXTL genes have not been linked to hereditary multiple extoses but belong to the EXT family of proteins based on amino acid sequence similarity to EXT1 and EXT2. All three EXTL proteins express glycosyltransferase activities related to HS biosynthesis; however, their role in HS formation in vivo remains undefined (13Kitagawa H. Shimakawa H. Sugahara K. J. Biol. Chem. 1999; 274: 13933-13937Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 14Kim B.T. Kitagawa H. Tamura J. Saito T. Kusche-Gullberg M. Lindahl U. Sugahara K. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 7176-7181Crossref PubMed Scopus (140) Google Scholar). The mechanism of chain elongation in heparin/heparan sulfate biosynthesis involves the alternating transfer of GlcA and GlcNAc from their respective UDP-derivatives to the nonreducing end of the growing polymer. Although solubilized microsomal enzymes as well as purified EXT proteins can catalyze the transfer of single GlcA or GlcNAc residues to exogenous substrates in vitro, actual polymerization on exogenous substrates has not been demonstrated (15Lidholt K. Riesenfeld J. Jacobsson K.G. Feingold D.S. Lindahl U. Biochem. J. 1988; 254: 571-578Crossref PubMed Scopus (23) Google Scholar, 16Lind T. Tufaro F. McCormick C. Lindahl U. Lidholt K. J. Biol. Chem. 1998; 273: 26265-26268Abstract Full Text Full Text PDF PubMed Scopus (355) Google Scholar, 17Wei G. Bai X. Gabb M.M. Bame K.J. Koshy T.I. Spear P.G. Esko J.D. J. Biol. Chem. 2000; 275: 27733-27740Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 18Senay C. Lind T. Muguruma K. Tone Y. Kitagawa H. Sugahara K. Lidholt K. Lindahl U. Kusche-Gullberg M. EMBO Rep. 2000; 1: 282-286Crossref PubMed Scopus (143) Google Scholar, 19McCormick C. Duncan G. Goutsos K.T. Tufaro F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 668-673Crossref PubMed Scopus (373) Google Scholar). The previous results suggest that the biologically functional unit in HS polymerization is a complex containing EXT1 and EXT2 (18Senay C. Lind T. Muguruma K. Tone Y. Kitagawa H. Sugahara K. Lidholt K. Lindahl U. Kusche-Gullberg M. EMBO Rep. 2000; 1: 282-286Crossref PubMed Scopus (143) Google Scholar, 19McCormick C. Duncan G. Goutsos K.T. Tufaro F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 668-673Crossref PubMed Scopus (373) Google Scholar). In this study, we examined the ability of EXT1 and EXT2 and of the EXT1/2 complex to elongate the HS backbone in vitro. We found that both EXT1 alone and the EXT1/2 complex were able to sequentially add GlcA and GlcNAc to 3H-labeled oligosaccharide substrates, albeit with apparently different kinetics. Construction of Expression Plasmids—Truncated forms of human EXT2 lacking the first N-terminal 46 amino acids were amplified by PCR using a human placenta cDNA as a template. Primers used for p3FLAGCMV9 (Sigma) were 5′-ATATGCGGCCGCTCCCCATTCTATCGAGT-3′ (forward) and 5′-ACTGGGATCCTAAGCTGCCAATGTTGGGGAA-3′ (reverse), and those for pSecTagC (Invitrogen) were 5′-GGCCAAGCTGGCCCCCCATTCTATCGA-3′ (forward) and 5′-GCGCTCTAGATAAGCTGCCAATGTTGGGGAAG-3′ (reverse). A truncated form of human EXT1 (amino acids 29–746) was amplified by PCR from a previously constructed plasmid (18Senay C. Lind T. Muguruma K. Tone Y. Kitagawa H. Sugahara K. Lidholt K. Lindahl U. Kusche-Gullberg M. EMBO Rep. 2000; 1: 282-286Crossref PubMed Scopus (143) Google Scholar), with primers 5′-ATATGCGGCCGCTAGGGCATCGAGGA-3′ (forward) and 5′-GCCTGGATCCAAGTCGCTCAATGTCTCG-3 (reverse). All of the PCRs were carried out using the Expand High Fidelity PCR System (Roche Applied Science). PCR products were cloned in-frame either with the preprotrypsin leader sequence and the N-terminal 3-FLAG tag (p3FLAGCMV9) or with the Ig κ-chain leader sequence and a C-terminal Myc-His tag (pSecTagC). Expression of Truncated EXT Proteins—HEK 293 cells were transfected using Lipofectin (Invitrogen). Stable clones expressing p3FLAGCMV9 constructs were selected in 800 μg/ml G-418 sulfate (Invitrogen). For co-expression of EXT1 and EXT2 (EXT1/2), one of the stable EXT1p3FLAGCMV9 clones was transfected with pSeqTagC EXT2, and double EXT1/2 transfectants were selected by addition of 200 μg/ml Zeocin (Invitrogen). The stable clones were maintained in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% (v/v) fetal calf serum (Invitrogen), 100 μg streptomycin sulfate/ml, 100 units penicillin G/ml, and 400 μg geneticin/ml or 400 μg geneticin/ml together with 200 μg/ml Zeocin. To isolate expressed proteins, the medium was changed to serum-free Optimem-1 with Glutamax-I (Invitrogen). Conditioned medium was collected after 48 h, and protein expression was monitored by Western blotting as described below. FLAG-tagged bacterial alkaline phosphatase (Met 3× FLAG-BAP; Sigma) was used to estimate the relative amount of purified enzymes. Recombinant proteins were captured on anti-FLAG agarose (Sigma) during overnight incubation at 4 °C, washed three times in phosphate-buffered saline, and used as the enzyme source in transferase and polymerase reactions. Analysis of Expressed Truncated EXT Proteins—The proteins present in the conditioned medium from HEK 293-transfected cells were precipitated with 11% trichloroacetic acid, separated on a 10% polyacrylamide gel, and transferred to a polyvinylidene difluoride membrane (Millipore). The antibodies used were anti-FLAG M2 monoclonal antibody (Sigma) detected using horseradish peroxidase-conjugated rabbit anti-mouse IgG (Dako), rabbit anti-Myc (kindly provided by E. Fries, Uppsala University) detected using horseradish peroxidase-conjugated swine anti-rabbit antibodies (Dako), and a horseradish peroxidase-conjugated anti-His monoclonal antibody (Sigma). The membranes were analyzed for chemiluminescence according to the manufacturer's instructions (ECL Plus detection kit; Amersham Biosciences). For isoelectrofocusing and gel filtration analysis, conditioned medium was concentrated using Microcon YM-10 devices (Amicon) and then applied to a precast isoelectrofocusing gel, pI range 3–10 (Bio-Rad), or a Superose 12 column (Amersham Biosciences). Polymerization Reaction—Size-defined oligosaccharide acceptors were prepared from a purified capsular polysaccharide of Escherichia coli K5, with the structure (GlcAβ1,4GlcNAcα1,4) n . The polysaccharide was partially N-deacetylated by incubation in 2 m NaOH (68 °C, 40 min) followed by deaminative cleavage (pH 3.9) (20Shaklee P.N. Conrad H.E. Biochem. J. 1984; 217: 187-197Crossref PubMed Scopus (83) Google Scholar). The generated fragments were reduced with NaB3H4 to yield reducing terminal 3H-labeled 2,5-anhydromannitol (aManR) residues. Labeled oligosaccharides (GlcA-[GlcNAc-GlcA]n-[3H]aManR) were size-separated on a Bio-Gel P-10 (Bio-Rad) column (1 × 200 cm) in 0.5 m NH4HCO3. Fractions corresponding to peaks of oligosaccharides ranging from tetra- to octadecasaccharides were pooled separately and lyophilized. Labeled [Glc-NAc-GlcA]5-[3H]aManR acceptor was prepared by digestion of GlcA-[GlcNAc-GlcA]5-[3H]aManR with β-glucuronidase (Sigma) as described below. Agarose-bound proteins were incubated in 20 mm MnCl2, 10 mm MgCl2, 5 mm CaCl2, 60 mm NaCl, 20 mm Hepes, pH 7.0, containing 16 mm unlabeled UDP-GlcNAc, 16 mm unlabeled UDP-GlcA, and 17 μm3H-end-labeled K5 oligosaccharide substrate. Reaction mixtures (100–200 μl) were incubated at 37 °C for 24 h, and then the reactions were stopped by precipitation with 5% trichloroacetic acid. After centrifugation the supernatants were neutralized with NaOH and again centrifuged, and the reaction products present in the supernatants were separated on Bio Gel P10 columns. Glycosyltransferase Assays—GlcNAc and GlcA transferase activities were measured essentially as described (18Senay C. Lind T. Muguruma K. Tone Y. Kitagawa H. Sugahara K. Lidholt K. Lindahl U. Kusche-Gullberg M. EMBO Rep. 2000; 1: 282-286Crossref PubMed Scopus (143) Google Scholar, 21Lind T. Lindahl U. Lidholt K. J. Biol. Chem. 1993; 268: 20705-20708Abstract Full Text PDF PubMed Google Scholar) by incubating anti-FLAG agarose-bound enzyme protein preparations with 0.125 μCi of 14C-labeled UDP sugars (62.5 μCi/μmol; prepared by mixing radiolabeled and unlabeled UDP sugars) and 45 μg of oligosaccharide acceptor at 37 °C for 2 h. Oligosaccharide acceptors were generated from K5 polysaccharide, as described (22Lidholt K. Lindahl U. Biochem. J. 1992; 287: 21-29Crossref PubMed Scopus (60) Google Scholar). Labeled products were isolated by gel chromatography and quantified by scintillation counting. Enzymatic Digestions—Digestion with heparitinase I (4.2.2.8; Seikagaku Corporation) was carried out according to the instructions of the manufacturer. Purified polymerization products were incubated with α-N-acetylglucosaminidase (Oxford GlycoSciences) and/or β-d-glucuronidase in 50 mm sodium acetate, pH 5.0, containing 100 μg/ml bovine serum albumin. The samples were incubated for ∼16 h according to the manufacturer's protocol. The EXT proteins are Golgi transmembrane type II proteins with an N-terminal cytoplasmic tail, a transmembrane domain, a stem region, and a large luminal catalytic domain. Thus the enzymes are believed to execute their function in a membrane bound state in vivo. However, because it has previously been shown that truncated soluble recombinant EXT proteins also are enzymatically active (18Senay C. Lind T. Muguruma K. Tone Y. Kitagawa H. Sugahara K. Lidholt K. Lindahl U. Kusche-Gullberg M. EMBO Rep. 2000; 1: 282-286Crossref PubMed Scopus (143) Google Scholar), we expressed soluble forms of the EXT proteins to facilitate their purification. We stably transfected HEK 293 cells with N-terminally FLAG-tagged human EXT1 and EXT2 constructs lacking the transmembrane domains, yielding soluble fusion proteins that were released into the culture medium (EXT1 and EXT2; Fig. 1). For co-expression of EXT1 and EXT2, cells expressing FLAG-tagged EXT1 were stably transfected with a C-terminally Myc-His-tagged EXT2 construct (EXT1/2 in Fig. 1; see "Experimental Procedures"). To confirm the association of EXT1 and EXT2 into a co-secreted FLAG-EXT1/Myc-His-EXT2 complex, following co-expression, the complex was affinity adsorbed from medium onto anti-FLAG agarose and visualized by Western blotting with anti-His antibodies. Myc-His-tagged EXT2 was readily detected in the medium of the FLAG-EXT1/Myc-His-EXT2 co-expressing cells, indicating that EXT1 and EXT2 formed a secreted heterocomplex (EXT1/2; Fig. 1) that could be captured on anti-FLAG agarose (data not shown, but see below). Soluble affinity-tagged EXT proteins, expressed separately or co-secreted, were affinity captured on anti-FLAG agarose beads, and the enzyme-containing beads were used as enzyme source. In Vitro Polymerization of Heparan Sulfate Backbone—For polymerization we used small amounts of 3H-end-labeled size-defined K5 oligosaccharides as substrates and an excess of unlabeled nucleotide sugar precursors, in a molar ratio of 1:1000 (K5/UDP sugars). The substrates used were a 3H-labeled 10-mer containing a nonreducing terminal GlcA as acceptor and a 3H-labeled 11-mer containing a nonreducing terminal GlcNAc residue (GlcA-[GlcNAc-GlcA]4-[3H]aManR and [GlcNAc-GlcA]5-[3H]aManR, respectively; see "Experimental Procedures"). In vitro chain elongation of the 10-mer proceeds by the initial addition of a GlcNAc residue, followed by alternating addition of GlcA and GlcNAc residues to the nonreducing end of the acceptor. Accordingly, chain elongation of an 11-mer starts by the addition of a GlcA unit followed by the sequential addition of GlcNAc and GlcA units. Consequently, polymerization products with nonreducing end GlcNAc residues are odd-numbered, and those ending with GlcA units are even-numbered. The polymerization products formed in incubations of recombinant EXT proteins with the 10-mer are shown in Fig. 2A. Incubations with EXT1 resulted in extensive polymerization generating products containing mostly 20 or more sugar units. The reaction products generated with the EXT1/2 complex as a catalyst were shorter and contained distinct even-numbered products, implying a more efficient incorporation of GlcA than of GlcNAc units. When instead an 11-mer, containing a nonreducing terminal GlcNAc, was used as a substrate, both EXT1 and the EXT1/2 complex likewise elongated the oligosaccharide, although the proportion of smaller reaction products was higher than seen with the 10-mer (Fig. 2B). Interestingly, EXT1/2 again mostly generated even-numbered oligosaccharides, suggesting that one GlcA unit was efficiently added to the 11-mer, and thereafter chain elongation continued in a similar manner as in Fig. 2A. EXT1, on the other hand, produced both even- and odd-numbered products (Fig. 3A). In contrast to EXT1 and EXT1/2, no significant chain elongation was observed in incubations with EXT2. Nevertheless, EXT2 was able to catalyze the addition of a single GlcA to the 11-mer, as indicated by the small shoulder in Fig. 2B. No chain elongation was observed after incubation with mock beads (i.e. anti-FLAG agarose beads reacted with the medium from vector-transfected cells).Fig. 3Restricted polymerization by recombinant EXT1 and EXT1/2 complex. Affinity-purified EXT1 (A) and EXT1/2 (B) were incubated for 24 h with a 3H-reducing end-labeled 10-mer acceptor substrate and a mixture of unlabeled UDP-GlcNAc and UDP-GlcA as described under "Experimental Procedures." The products were fractionated by gel chromatography on Bio Gel P-10 eluted with 0.5 m NH4HCO3 as described in the legend to Fig. 2. The fractions corresponding to V 0 (<30-mers) and 15- and 14-mers were pooled as indicated by the horizontal bars, lyophilized, and analyzed further. The numbers above the various peaks indicate even-numbered oligosaccharide units.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The different modes of polymerization kinetics are more evident in Fig. 3 where the polymerization reactions were less extensive than in Fig. 2. EXT1 clearly generated even- and odd-numbered products in fairly similar proportions, indicating that GlcNAc and GlcA residues were incorporated with similar efficiency. The elution profile of the EXT1/2 complexcatalyzed polymerization again contained polymers of largely even-numbered oligosaccharides. We were intrigued by the fact that EXT2 appeared to modify the catalytic properties of EXT1. To determine whether the observed activity required complex formation of EXT1 and EXT2, singly expressed EXT1 and EXT2 were combined in equal amounts, adsorbed to anti-FLAG agarose, and incubated with the 10-mer and the 11-mer. The resultant gel chromatography patterns resembled that of EXT1 alone, indicating that the catalytic activity of EXT1/2 was due to complex formation of the two proteins (not shown). Analysis of Reaction Products—The nature of the polymerization products was determined by enzymatic digestions. The 14-mer, 15-mer, and V 0 material obtained after incubation of the 10-mer with EXT1 and the V 0 material from incubations of the 10-mer with EXT1/2 were pooled as indicated in Fig. 3. Each of the reaction products were digested with β-glucuronidase alone or in combination with α-N-acetylglucosaminidase and analyzed by gel chromatography on Superdex 30 (Fig. 4). All of the reaction products were digested to monosaccharide ([3H]aManR) or small oligosaccharides by the mixed exoglycosidases, indicating that they were composed of alternating GlcA-GlcNAc units with the predicted linkage configurations (Fig. 4, A and C). Digestion with β-glucuronidase alone converted the 14-mer into a 13-mer but did not change the elution position of the 15-mer, further demonstrating that the incorporated sugars were transferred to the nonreducing end of the expected acceptor structure (Fig. 4, A and B). Finally, all of the products were completely degraded to disaccharides by heparitinase I (data not shown). Glycosyltransferase Activities of Expressed EXT Proteins— We wanted to examine whether the elution patterns shown in Figs. 2 and 3 related to the in vitro transferase activities (single sugar transfer reactions performed in the presence of only one of the two donor nucleotide substrates). FLAG affinity-purified EXT proteins were analyzed for GlcA and GlcNAc transferase activities, with radiolabeled UDP-GlcA and a GlcNAc[GlcA-GlcNAc] n oligosaccharide acceptor (measuring GlcA transferase activity) or with radiolabeled UDP-GlcNAc and a [GlcA-GlcNAc] n acceptor (measuring GlcNAc transferase activity) (see "Experimental Procedures"). The results clearly demonstrated both transferase activities for EXT1 and EXT1/2, whereas only low activities were detected with EXT2 (Fig. 5). The GlcA transferase activity of EXT2 was low but readily detected, whereas the GlcNAc transferase activity was even lower and in some preparations virtually absent. Similar to the polymerization reaction, complex formation of EXT1 and EXT2 influenced the transferase activities. In particular, the GlcNAc transferase activity of EXT1/2 was lower in relation to the GlcA transferase activity, compared with EXT1 alone. Homo- and Heterooligomeric Complex Formation of Expressed EXT Proteins—The available information suggests that EXT1 and EXT2 form both homooligomeric and heterooligomeric complexes (18Senay C. Lind T. Muguruma K. Tone Y. Kitagawa H. Sugahara K. Lidholt K. Lindahl U. Kusche-Gullberg M. EMBO Rep. 2000; 1: 282-286Crossref PubMed Scopus (143) Google Scholar, 19McCormick C. Duncan G. Goutsos K.T. Tufaro F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 668-673Crossref PubMed Scopus (373) Google Scholar, 23Kobayashi S. Morimoto K. Shimizu T. Takahashi M. Kurosawa H. Shirasawa T. Biochem. Biophys. Res. Commun. 2000; 268: 860-867Crossref PubMed Scopus (60) Google Scholar). To determine whether, in our system, the secreted truncated proteins oligomerized, conditioned media from EXT1, EXT2, and EXT1/2 overexpressing cells were analyzed by gel filtration on a Superose-12 column. Concentrated media from FLAG-EXT1, FLAG-EXT2, or FLAG-EXT1/Myc-His-EXT2 (Fig. 1) expressing cells were applied to the column, and individual fractions were analyzed by SDS-PAGE and Western blotting with anti-FLAG and anti-Myc antibodies. EXT2 eluted in the position of a monomeric protein (80 kDa), whereas both EXT1 and EXT1/2 eluted as dimers (160 kDa) (Fig. 6). Identical fractions from the EXT1/2 chromatogram reacted with both anti-FLAG and anti-Myc antibodies, showing that co-expression of EXT1 (FLAG) and EXT2 (Myc-His) yielded a heterocomplex. Interactions between the expressed and endogenous EXT proteins could conceivably also affect analyses of affinity-purified recombinant proteins. We lack specific antibodies against the EXT proteins; therefore the presence of endogenous EXT protein in complex with expressed proteins was examined by isoelectrofocusing followed by immunoblotting. Bands corresponding to the theoretically predicted isoelectric point (pI) for EXT1, EXT2, and EXT1/2, pI 9, 6, and 7.5, respectively, were detected for the respective expressed proteins (data not shown). The absence of any detectable bands corresponding to pI 7.5 in the conditioned medium of EXT1 or EXT2 transfected cells indicated that the majority of the expressed recombinant EXT1 and EXT2 proteins do not form heterooligomeric complexes with endogenous EXT proteins. EXT1 and EXT2 are both GlcA/GlcNAc transferases, and evidence suggests that EXT1 and EXT2 interact to form the biologically active HS polymerase (19McCormick C. Duncan G. Goutsos K.T. Tufaro F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 668-673Crossref PubMed Scopus (373) Google Scholar). Lack of either EXT1 or EXT2 disrupts HS synthesis, indicating that both proteins are required for proper polymerization (9Zak B.M. Crawford B.E. Esko J.D. Biochim. Biophys. Acta. 2002; 1573: 346-355Crossref PubMed Scopus (149) Google Scholar, 24McCormick C. Leduc Y. Martindale D. Mattison K. Esford L.E. Dyer A.P. Tufaro F. Nat. Genet. 1998; 19: 158-161Crossref PubMed Scopus (330) Google Scholar, 25Lin X. Wei G. Shi Z. Dryer L. Esko J.D. Wells D.E. Matzuk M.M. Dev. Biol. 2000; 224: 299-311Crossref PubMed Scopus (351) Google Scholar, 26Toyoda H. Kinoshita-Toyoda A. Selleck S.B. J. Biol. Chem. 2000; 275: 2269-2275Abstract Full Text Full Text PDF PubMed Scopus (251) Google Scholar). From the results presented in this study, it is apparent that EXT1 alone, as well as EXT1/2, can catalyze the in vitro polymerization of the HS backbone structure on an oligosaccharide primer. No additional factors such as membrane structures, lipids, or other proteins, seem to be required for the in vitro polymerization reaction. Both EXT1 and EXT1/2 catalyzed the formation of intermediate sized and long chains, but the kinetics of the EXT1-catalyzed polymerization reaction appears to be different from that of EXT1/2. EXT1 alone seemed to catalyze the incorporation of GlcNAc and GlcA residues with similar efficiency. When, on the other hand, the EXT1/2 complex catalyzed polymerization, GlcA residues appeared to be more efficiently incorporated than GlcNAc units. When a single sugar nucleotide (UDP-GlcA or UDP-GlcNAc) was used as a sugar donor in the transferase assays, complex formation of EXT1 and EXT2 resulted in less efficient transfer of GlcNAc as compared with EXT1 alone (Fig. 5). The most reasonable explanation for the difference in mode of polymerization, as indicated by the elution profiles in Figs. 2 and 3, is that the in vitro polymerization process reflects the single sugar transferase activities and that complex formation changed the catalysis. The effect of EXT2 on EXT1 activity is intriguing. EXT2 appeared to interfere with the addition of GlcNAc such that the EXT1/2 complex polymerized less efficiently than EXT1 alone. Our data indicate that EXT2 harbors low GlcA transferase activity but no readily detected GlcNac transferase activity. Both EXT1 and EXT2 have been reported to form homodi/oligomers (19McCormick C. Duncan G. Goutsos K.T. Tufaro F. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 668-673Crossref PubMed Scopus (373) Google Scholar, 23Kobayashi S. Morimoto K. Shimizu T. Takahashi M. Kurosawa H. Shirasawa T. Biochem. Biophys. Res. Commun. 2000; 268: 860-867Crossref PubMed Scopus (60) Google Scholar). It is possible that the EXT proteins need to di/oligomerize to catalyze the polymerization reaction. In our study, EXT2 did not seem to dimerize, and we cannot exclude the possibility that this is the reason for the observed low transferase activity. On the other hand, it is tempting to speculate that EXT2 is not involved in the actual elongation of the HS backbone but serves as a chaperon and delivers EXT1 to the Golgi apparatus. So far no additional GlcA transferases involved in actual HS chain elongation are known. Both EXTL1 and EXTL3 have been shown to be able to transfer GlcNAc (but not GlcA) to K5 oligosaccharide acceptors and are likely to be involved in HS biosynthesis (14Kim B.T. Kitagawa H. Tamura J. Saito T. Kusche-Gullberg M. Lindahl U. Sugahara K. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 7176-7181Crossref PubMed Scopus (140) Google Scholar). Of the two, only EXTL3 is ubiquitously expressed in human tissues (27Wise C.A. Clines G.A. Massa H. Trask B.J. Lovett M. Genome Res. 1997; 7: 10-16Crossref PubMed Scopus (121) Google Scholar, 28Saito T. Seki N. Yamauchi M. Tsuji S. Hayashi A. Kozuma S. Hori T. Biochem. Biophys. Res. Commun. 1998; 243: 61-66Crossref PubMed Scopus (51) Google Scholar) and therefore more likely to be part of a universal HS chain-elongating complex. The more restricted mRNA expression of EXTL1 may indicate a more specialized role for this enzyme. Our present findings raise a number of questions. By what mechanism does the interaction between EXT1 and EXT2 promote enzyme translocation and function? It is noted that EXT1 transfected into EXT1-deficient mutant l-cells remained localized to the endoplasmic reticulum (24McCormick C. Leduc Y. Martindale D. Mattison K. Esford L.E. Dyer A.P. Tufaro F. Nat. Genet. 1998; 19: 158-161Crossref PubMed Scopus (330) Google Scholar), whereas HS chain generation is believed to take place in the Golgi (5Esko J.D. Selleck S.B. Annu. Rev. Biochem. 2002; 71: 435-471Crossref PubMed Scopus (1270) Google Scholar, 8Lindahl U. Kusche-Gullberg M. Kjellen L. J. Biol. Chem. 1998; 273: 24979-24982Abstract Full Text Full Text PDF PubMed Scopus (580) Google Scholar). How does the EXT proteins relate to the proteoglycan core protein and the other enzymes required for HS assembly? The exact coupling between polysaccharide chain elongation and modification has not yet been defined, but previous studies have shown that heparin/HS chain elongation is stimulated by concomitant N-sulfation of the product (29Lidholt K. Kjellén L. Lindahl U. Biochem. J. 1989; 261: 999-1007Crossref PubMed Scopus (53) Google Scholar, 30Pikas D.S. Eriksson I. Kjellén L. Biochemistry. 2000; 39: 4552-4558Crossref PubMed Scopus (56) Google Scholar). Finally, how does tumor formation caused by defective EXT proteins relate to HS biosynthesis? These and similar problems can hopefully be approached after a more detailed analysis, at the molecular and cellular levels, of the EXT proteins and their various interactions. In particular, it would be of interest to assess the effect of N-sulfation and the role of EXTL proteins on the in vitro polymerization process. We thank Dr. Ulf Lindahl and Krzysztof Wicher for helpful suggestions and critical reading of the manuscript and Helena Grundberg for excellent technical assistance.
Referência(s)