The Eukaryotic UDP-N-Acetylglucosamine Pyrophosphorylases
1998; Elsevier BV; Volume: 273; Issue: 23 Linguagem: Inglês
10.1074/jbc.273.23.14392
ISSN1083-351X
AutoresToshiyuki Mio, Tomio Yabe, Mikio Arisawa, Hisafumi Yamada‐Okabe,
Tópico(s)Biochemical and Molecular Research
ResumoA search of the yeast data base for a protein homologous to Escherichia coliUDP-N-acetylglucosamine pyrophosphorylase yieldedUAP1 (UDP-N-acetylglucosaminepyrophosphorylase), the Saccharomyces cerevisiae gene for UDP-N-acetylglucosamine pyrophosphorylase. The Candida albicans and human homologs were also cloned by screening a C. albicans genomic library and a human testis cDNA library, respectively. Sequence analysis revealed that the human UAP1 cDNA was identical to previously reported AGX1. A null mutation of the S. cerevisiae UAP1 (ScUAP1) gene was lethal, and when expressed under the control of ScUAP1 promoter, bothC. albicans and Homo sapiens UAP1(CaUAP1 and HsUAP1) rescued theScUAP1-deficient S. cerevisiae cells. All the recombinant ScUap1p, CaUap1p, and HsUap1p possessed UDP-N-acetylglucosamine pyrophosphorylase activitiesin vitro. The yeast Uap1p utilizedN-acetylglucosamine-1-phosphate as the substrate, and together with Agm1p, it produced UDP-N-acetylglucosamine from N-acetylglucosamine-6-phosphate. These results demonstrate that the UAP1 genes indeed specify eukaryotic UDP-GlcNAc pyrophosphorylase and that phosphomutase reaction precedes uridyltransfer. Sequence comparison with other UDP-sugar pyrophosphorylases revealed that amino acid residues, Gly112, Gly114, Thr115, Arg116, Pro122, and Lys123 of ScUap1p are highly conserved in UDP-sugar pyrophosphorylases reported to date. Among these amino acids, alanine substitution for Gly112, Arg116, or Lys123 severely diminished the activity, suggesting that Gly112, Arg116, or Lys123 are possible catalytic residues of the enzyme. A search of the yeast data base for a protein homologous to Escherichia coliUDP-N-acetylglucosamine pyrophosphorylase yieldedUAP1 (UDP-N-acetylglucosaminepyrophosphorylase), the Saccharomyces cerevisiae gene for UDP-N-acetylglucosamine pyrophosphorylase. The Candida albicans and human homologs were also cloned by screening a C. albicans genomic library and a human testis cDNA library, respectively. Sequence analysis revealed that the human UAP1 cDNA was identical to previously reported AGX1. A null mutation of the S. cerevisiae UAP1 (ScUAP1) gene was lethal, and when expressed under the control of ScUAP1 promoter, bothC. albicans and Homo sapiens UAP1(CaUAP1 and HsUAP1) rescued theScUAP1-deficient S. cerevisiae cells. All the recombinant ScUap1p, CaUap1p, and HsUap1p possessed UDP-N-acetylglucosamine pyrophosphorylase activitiesin vitro. The yeast Uap1p utilizedN-acetylglucosamine-1-phosphate as the substrate, and together with Agm1p, it produced UDP-N-acetylglucosamine from N-acetylglucosamine-6-phosphate. These results demonstrate that the UAP1 genes indeed specify eukaryotic UDP-GlcNAc pyrophosphorylase and that phosphomutase reaction precedes uridyltransfer. Sequence comparison with other UDP-sugar pyrophosphorylases revealed that amino acid residues, Gly112, Gly114, Thr115, Arg116, Pro122, and Lys123 of ScUap1p are highly conserved in UDP-sugar pyrophosphorylases reported to date. Among these amino acids, alanine substitution for Gly112, Arg116, or Lys123 severely diminished the activity, suggesting that Gly112, Arg116, or Lys123 are possible catalytic residues of the enzyme. UDP-N-acetylglucosamine (UDP-GlcNAc 1The abbreviations used are: UDP-GlcNAc, UDP-N-acetylglucosamine; ORF, open reading frame; GST, glutathione S-transferase; GlcN, glucosamine; Fru-6-P, fructose-6-phosphate; GlcN-6-P, glucosamine-6-phosphate; GlcN-1-P, glucosamine-1-phosphate; Man-6-P, mannose-6-phosphate; Man-1-P, mannose-1-phosphate; Gal-1-P, galactose-1-phosphate; Glc-1-P, glucose-1-phosphate; TLC, thin layer chromatography. ) is a ubiquitous and essential metabolite and plays important roles in several metabolic processes. In bacteria, it is known as a major cytoplasmic precursor of cell wall peptide glycan and the disaccharide moiety of lipid A (1Holtje J.V. Schwartz U. Nanninga N. Molecular Cytology of Escherichia coli. Academic Press, Inc., New York1985: 77-119Google Scholar, 2Park J.T. Neidhardt F.C. Ingraham J.L. Low K.B. Magasanik B. Schaechter M. Umbarger H.E. Escherichia coli and Salmonella typhimurinum: Cellular and Molecular Biology. 1. American Society for Microbiology, Washington, D. C.1987: 663-671Google Scholar, 3Raetz C.R.H. Neidhardt F.C. Ingraham J.L. Low K.B. Magasanik B. Schaechter M. Umbarger H.E. Escherichia coli and Salmonella typhimurinum: Cellular and Molecular Biology. 1. American Society for Microbiology, Washington, D. C.1987: 498-503Google Scholar). In eukaryotes, it serves as the substrate of chitin synthase, whose product is shown to be essential for fungal cell wall (4Cabib E. Roberts R. Bowers B. Annu. Rev. Biochem. 1982; 51: 763-793Crossref PubMed Scopus (270) Google Scholar). It is also used in the GlcNAc moiety of N-linked glycosylation and the GPI-anchor of cellular proteins (5Herscovics A. Orlean P. FASEB J. 1993; 7: 540-550Crossref PubMed Scopus (440) Google Scholar). Biosynthesis of UDP-GlcNAc has been extensively studied in bacteria, and it requires the following enzymatic reactions: i) conversion of fructose-6-phosphate (Fru-6-P) into glucosamine-6-phosphate (GlcN-6-P) by glutamine:Fru-6-P amidotransferase; ii) conversion of GlcN-6-P into glucosamine-1-phosphate (GlcN-1-P) by glucosamine (GlcN) phosphate mutase; iii) acetylation of GlcN-1-P by GlcN-1-P acetyltransferase to produce N-acetylglucosamine-1-phosphate (GlcNAc-1-P); and iv) synthesis of UDP-GlcNAc from GlcNAc-1-P and UTP by GlcNAc-1-P uridyltransferase (also called UDP-GlcNAc pyrophosphorylase) (6Dobrogosz W.J. J. Bacteriol. 1968; 95: 578-584Crossref PubMed Google Scholar, 7White R.J. Biochem. J. 1968; 106: 847-858Crossref PubMed Scopus (116) Google Scholar, 8Freese E.B. Cole R.M. Klofat W. Freese E. J. Bacteriol. 1970; 101: 1046-1062Crossref PubMed Google Scholar). The Escherichia coli GlmS gene encodes glutamine:Fru-6-P amidotransferase (9Wu H.C. Wu T.C. J. Bacteriol. 1971; 105: 455-466Crossref PubMed Google Scholar, 10Walker J.E. Gay N.J. Saraste M. Eberle A.N. Biochem. J. 1984; 224: 799-815Crossref PubMed Scopus (107) Google Scholar, 11Dutka-Malen S. Mazodier P. Badet B. Biochimie (Paris). 1988; 70: 287-290Crossref PubMed Scopus (51) Google Scholar). E. coli GlmU specifies a bifunctional protein with GlcN-1-P acetyltransferase and UDP-GlcNAc pyrophosphorylase activities (12Mengin-Lecreulx D. Heijenoort J. J. Bacteriol. 1993; 175: 6150-6157Crossref PubMed Google Scholar, 13Mengin-Lecreulx D. Heijenoort J. J. Bacteriol. 1994; 176: 5788-5795Crossref PubMed Scopus (166) Google Scholar). In yeast Saccharomyces cerevisiae, Fru-6-P is converted either into GlcN-6-P by glutamine:Fru-6-P amidotransferase or into mannose-6-phosphate by phosphomannose isomerase. GFA1 andPMI have been shown to be the genes for glutamine:Fru-6-P amidotransferase and phosphomannose isomerase, respectively (14Watzele G. Tanner W. J. Biol. Chem. 1989; 264: 8753-8758Abstract Full Text PDF PubMed Google Scholar, 15Smith D.J. Proudfoot A. Friedli L. Klig L.S. Paravicini G. Payton M.A. Mol. Cell. Biol. 1992; 12: 2924-2930Crossref PubMed Scopus (58) Google Scholar). Then, GlcN-6-P is N-acetylated by an acetylase to become GlcNAc-6-P, which is further converted into GlcNAc-1-P by GlcNAc phosphate mutase (16Cabib E. Tanner W. Loewus F.A. Encyclopedia of Plant Physiology, Carbohydorate II: Extracellular carbohydorates. 13B. Springer-Verlag, Heidelberg1981: 395-416Google Scholar). S. cerevisiae harbors four different hexosephosphate mutase genes, PGM1 (17Oh D. Hopper J.E. Mol. Cell. Biol. 1990; 10: 1415-1422Crossref PubMed Scopus (48) Google Scholar), PGM2(18Boles E. Liebetrau W. Hofmann M. Zimmermann F.K. Eur. J. Biochem. 1994; 220: 83-96Crossref PubMed Scopus (82) Google Scholar), SEC53 (19Kepes F. Schekman R. J. Biol. Chem. 1988; 263: 9155-9161Abstract Full Text PDF PubMed Google Scholar), and AGM1 (20Hofmann M. Boles E. Zimmermann F.K. Eur. J. Biochem. 1994; 221: 741-747Crossref PubMed Scopus (62) Google Scholar). Among them,AGM1 is responsible for the interconversion of GlcNAc-6-P and GlcNAc-1-P (20Hofmann M. Boles E. Zimmermann F.K. Eur. J. Biochem. 1994; 221: 741-747Crossref PubMed Scopus (62) Google Scholar). Interestingly, Agm1p has dual substrate specificity; it also converts glucose-6-phosphate to glucose-1-phosphate (Glc-1-P) (20Hofmann M. Boles E. Zimmermann F.K. Eur. J. Biochem. 1994; 221: 741-747Crossref PubMed Scopus (62) Google Scholar). Finally, UDP-GlcNAc is produced from GlcNAc-1-P by UDP-GlcNAc pyrophosphorylase. However, the eukaryotic genes for GlcN-6-P acetylase and UDP-GlcNAc pyrophosphorylase remain unidentified. On the other hand, there are three UDP-sugar pyrophosphorylase genes inS. cerevisiae reported to date. GAL7 (21Tajima M. Nogi Y. Fukasawa T. Yeast. 1985; 1: 67-77Crossref PubMed Scopus (147) Google Scholar) andUGP1 (22Daran J.M. Dallies N. Thines-Sempoux D. Paquet V. Francois J. Eur. J. Biochem. 1995; 233: 520-530Crossref PubMed Scopus (107) Google Scholar) encode UDP-galactose (UDP-Gal) pyrophosphorylase and UDP-glucose (UDP-Glc) pyrophosphorylase, respectively. Recently, VIG9 was identified as the GDP-mannose (GDP-Man) pyrophosphorylase gene by functional complementation using the glycosylation defective vig9-1mutant (23Hashimoto H. Sakakibara A. Yamasaki M. Yoda K. J. Biol. Chem. 1997; 272: 16308-16314Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar), and the possible amino acid sequence motif for the active site of UDP-sugar pyrophosphorylase is proposed. Because all of these enzymes preserve substrate specificity to a certain type of sugar, there should be an enzyme specific to GlcNAc-1-P. In an attempt to identify the gene for UDP-GlcNAc pyrophosphorylase, we searched the S. cerevisiae genome data base and found that the protein specified by YDL103C. The Candida albicans and human homologs were also isolated and characterized. From sequence comparison and mutation analysis, the probable catalytic residues of UDP-sugar pyrophosphorylases are proposed. An amino acid sequence motif of LXXGXGTXMXXXXPK whereX represents any amino acid was obtained by comparing the amino acid sequences of E. coli GlmUp (EcGlm1p), S. cerevisiae Ugp1p (ScUgp1p), and Homo sapiens Ugp1p (HsUgp1p), and was used to search the S. cerevisiae genome data bases. The entire open reading frame of ScUAP1(originally designated YDL103C) was amplified by polymerase chain reaction using the S. cerevisiae genomic DNA extracted from strain A451 (MATα can1, aro7, can1, leu2, trp1, ura3) as a template, and cloned at the XbaI site of pUC18 or pYEUra3 (Toyobo) generating pUC-ScUAP1 and pYEU-ScUAP1, respectively. Primers used for polymerase chain reaction were 5′-AGATCTAGAATGACTGACACAAAACAGCT-3′ and 5′-AGATCTAGATTATTTTTCTAATACTATAC-3′. The C. albicans and human homologs of ScUAP1 were cloned by screening a C. albicans genomic DNA library and a human testis cDNA library using the 1.4-kilobaseEcoRI-EcoRI fragment of ScUAP1 as a probe. Hybridization and washing of the filters were carried out under stringent conditions (20 mm sodium phosphate (pH 7.2), 5× SSC (1× SSC contains 150 mm NaCl and 15 mmsodium citrate), 5× Denhardt's solution, 0.1% SDS, 25% formamide at 42 °C for hybridization; 0.1× SSC and 0.1% SDS at 50 °C for washing). Bacterial cells and phages that were strongly hybridized with the probe DNA were collected. After the third screening, DNA was extracted from bacterial cells and phages, and the insert DNA was cloned at the SmaI site of pUC19 for further plasmid construction. Radiolabeling of the probe DNA was performed by the random priming method using [α-32P]dCTP (24Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual.2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar), and DNA sequencing was carried out as described elsewhere (24Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual.2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989Google Scholar). Construction of the C. albicans genomic DNA library was already reported (25Yamada-Okabe T. Shimmi O. Doi R. Mizumoto K. Arisawa M. Yamada-Okabe H. Microbiology. 1996; 142: 2515-2523Crossref PubMed Scopus (45) Google Scholar). A human testis cDNA library was purchased fromCLONTECH (USA). The coding regions of ScUAP1, CaUAP1,HsUAP1, and ScAGM1 were cloned at theEcoRI (for ScUAP1 and ScAGM1) orSmaI (for CaUAP1 and HsUAP1) site of pGEX2T (26Smith D.B. Johnson K.S. Gene (Amst.). 1988; 67: 31-40Crossref PubMed Scopus (5047) Google Scholar), and the resulting plasmids were transfected into E. coli JM109 to let them express recombinant yeast and human proteins as a fusion product with glutathione S-transferase (GST). Induction and expression of the recombinant Uap1 proteins was carried out with isopropyl β-d-thio-galactopyranoside as described (25Yamada-Okabe T. Shimmi O. Doi R. Mizumoto K. Arisawa M. Yamada-Okabe H. Microbiology. 1996; 142: 2515-2523Crossref PubMed Scopus (45) Google Scholar, 26Smith D.B. Johnson K.S. Gene (Amst.). 1988; 67: 31-40Crossref PubMed Scopus (5047) Google Scholar). At 4 h after the addition of isopropyl-β-d-thio-galactopyranoside, the bacterial cells were harvested, suspended in a buffer containing 20 mmTris-HCl (pH 7.5), 0.5 mm EDTA, 50 mm NaCl, 10 mm β-mercaptoethanol, 10%(v/v) glycerol, 1 mm phenylmethylsulfonyl fluoride, and lysed by sonication. After cell debris were removed by centrifugation at 15,000 ×g at 4 °C for 30 min, GST-Uap1 and GST-Agm1 fusion proteins were purified by glutathione Sepharose CL-4B column chromatography, as described (26Smith D.B. Johnson K.S. Gene (Amst.). 1988; 67: 31-40Crossref PubMed Scopus (5047) Google Scholar) and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The primers used for amplifying the ScAGM1 open reading frame (ORF) were 5′-CGGGAATTCATAAGGTTGATTACGAGCAAT-3′ and 5′-ATTGAATTCTCAAGCAGATGCCTTAACGTG-3′. An assay for UDP- GlcNAc pyrophosphorylase was performed in a 20 μl standard reaction mixture containing 50 mm Tris-HCl (pH 8.3), 5 mm MgCl2, 20 μm GlcNAc-1-P, 10% (v/v) glycerol, and 0.1 μm [α-32P]UTP (specific activity 1 × 103 cpm/pmol) and approximately 0.1 μg of the indicated recombinant proteins at 30 °C for 10 min. 2 μl of each reaction mixture were spotted onto polyethyleneimine cellulose plates, and nucleotide sugars were separated by thin layer chromatography (TLC) in a solution that was prepared by mixing 6 g of Na2B4O7/10H2O, 3 g of H3BO3, and 25 ml of ethylene glycol in 70 ml of H2O (27Randerath K. Randerath E. Anal. Biochem. 1965; 13: 575-579Crossref PubMed Scopus (41) Google Scholar). The radioactive spots were visualized by autoradiography. An alternative high flux assay was carried out in 90 μl of reaction mixture containing 50 mm Tris-HCl (pH 8.3), 5 mm MgCl2, 25 μm UTP, 20 μm GlcNAc-1-P, 10% (v/v) glycerol, 1 mmdithiothreitol, 0.4 units/ml pyrophosphatase (Sigma), and approximately 0.1 μg of the recombinant enzyme. After incubation at 30 °C for 10 min, 100 μl of the color reagent containing 0.03% (w/v) malachite green, 0.2% (w/v) ammonium molybdate, and 0.05% (v/v) Triton X-100 in 0.7 n HCl was added to the reaction mixture, which was followed by incubation at room temperature for 5 min. Inorganic phosphate derived from the pyrophosphate and thereby representing the enzyme activity was quantified by measuring optimal density at 655 nm. The entire ORF ofScUAP1 was cloned at the XbaI site of pUC18 and pYEUra3 (downstream of the GAL1 promoter), generating pUC-ScUAP1 and pYEU-ScUAP1, respectively. Then the 539-base pairStuI-BalI region of the ScUAP1 ORF in pUC-ScUAP1 was excised and replaced by the LEU2 gene, generating pUC-ScUAP1L. The haploid strain YPH499 (MATa ura3, lys2, ade2, trp1,his3, leu2) was transformed with pYEU-ScUAP1, and ura+ transformants were further transfected with pUC-ScUAP1L that had been digested with XbaI. The resulting ura+ leu+ transformants, which grew in galactose medium but not in glucose medium, were collected and used asuap1Δ strain (MATa ura3, lys2,ade2, trp1, his3, leu2, uap1Δ::LEU2 UAP1-URA3). The S. cerevisiae agm1Δ null mutant strain was obtained by a means similar to that for the ScUAP1 depletion. The entire ORF of ScAGM1 was cloned at the XbaI site of pUC18 and pYEUra3, generating pUC-ScAGM1 and pYEU-ScAGM1, respectively. The 1.4-kilobase BalI-BglII region of theScAGM1 ORF in pUC-ScAGM1 was replaced by LEU2, generating pUC-AGM1L. YPH499 cells were transformed with pYEU-ScAGM1 and then with pUC-ScAGM1L that had been previously digested with XbaI. The resulting ura+ leu+transformants, which grew in galactose medium but died in glucose medium, were collected and used as agm1Δ strain (MATa ura3, lys2, ade2,trp1, his3, leu2, agm1Δ::LEU2 AGM1-URA3). To test the ability of CaUAP1, HsUAP1, and the mutantScUAP1 to complement ScUAP1, the entire ORFs ofCaUAP1, HsUAP1, and the mutant ScUAP1 were cloned in pRS414–1 where a 2.0-kilobase BglII-XbaI fragment encompassing the ScUAP1 promoter was inserted at the BamHI site of pRS414 (Stratagene). Thus, the transcription of CaUAP1 and HsUAP1 from this plasmid was under the control of the ScUAP1 promoter. The resulting plasmids were transfected into uap1Δ cells. After selection of trp+ cells in the presence of galactose, they were transferred to plates containing glucose and further cultured for 3 days. A series of the ScUAP1mutants harboring an alanine substitution for Gly111, Gly112, Gly114, Thr115, Arg116, Leu117, Pro122, or Lys123 were generated by the oligonucleotide-directed dual amber method as described (28Hashimoto-Gotoh T. Mizuno T. Ogasahara Y. Nakagawa M. Gene (Amst.). 1995; 152: 271-275Crossref PubMed Scopus (79) Google Scholar) with Mutan-Express KmTM(Takara). The entire ORF of the ScUAP1 gene was cloned at the EcoRI site of pKF18k (Takara) using EcoRI linker and hybridized with oligonucleotides containing the indicated mutations. The resulting mutant ScUAP1 genes were excised from the vector and ligated at the EcoRI site of pGEX-2T and the BamHI site of pRS414–1. All the mutations were confirmed by sequencing the DNA. Three distinct UDP-sugar pyrophosphorylase activities are present in yeast. In S. cerevisiae, the GAL7 (21Tajima M. Nogi Y. Fukasawa T. Yeast. 1985; 1: 67-77Crossref PubMed Scopus (147) Google Scholar), UGP1(22Daran J.M. Dallies N. Thines-Sempoux D. Paquet V. Francois J. Eur. J. Biochem. 1995; 233: 520-530Crossref PubMed Scopus (107) Google Scholar), and VIG9 (23Hashimoto H. Sakakibara A. Yamasaki M. Yoda K. J. Biol. Chem. 1997; 272: 16308-16314Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar) genes have been shown to encode UDP-Gal pyrophosphorylase, UDP-Glc pyrophosphorylase, and GDP-Man pyrophosphorylase, respectively, but the gene for UDP-GlcNAc remains to be established. Comparison of the amino acid sequences between E. coli UDP-GlcNAc pyrophosphorylase (GlmUp) and S. cerevisiae UDP-Glc pyrophosphorylase (Ugp1p) identified an amino acid sequence motif, L(X)2GXGTXM(X)4PK, where X represents any amino acid. In an attempt to identify the S. cerevisiae UDP-GlcNAc pyrophosphorylase gene, we searched the yeast data base and found that PSA1 andYDL103C could encode proteins with a sequence similar to the above amino acid motif (Fig. 1).PSA1 is identical to VIG9, which has been shown to be the GDP-Man pyrophosphorylase gene. Accordingly, we asked whetherYDL103C specifies UDP-GlcNAc pyrophosphorylase. The Ydl103c protein was expressed in E. coli as a fusion protein with GST and purified by affinity column chromatography using glutathione-Sepharose CL-4B. The purified GST-Ydl103c fusion protein produced [32P]UDP-GlcNAc when incubated with GlcNAc-1-P and [α-32P]UTP, whereas GST alone did not (Fig.2). The above result demonstrates thatYDL103C is a gene for UDP-GlcNAc pyrophosphorylase, and, therefore, the gene was designated ScUAP1 (the S. cerevisiae UDP-GlcNAc pyrophosphorylase gene1).Figure 2Expression and enzyme activities of the yeast and human UDP-GlcNAc pyrophosphorylases. The yeast and human UDP-GlcNAc pyrophosphorylases were expressed in E. coli as a fusion with GST and purified with glutathione-Sepharose beads.A, approximately 1 μg of the purified recombinant proteins were separated on a 10% SDS-polyacrylamide gel and stained with Coomassie Brilliant Blue. Lane 1, GST; lane 2,GST-ScUap1p; lane 3, GST-CaUap1p; lane 4,GST-HsUap1p; lane 5, GST-ScAgm1p. The positions of the protein size markers are indicated in kDa. B, approximately 0.1 μg of the purified recombinant proteins were incubated with [α-32P]UTP and GlcNAc-1-P. The reaction products were separated by polyethyleneimine cellulose TLC and visualized by autoradiography. The positions of the UDP-GlcNAc, UDP-Glc, and UTP that were visualized under UV light are indicated. Lane 1, no protein; lane 2, GST; lane 3, GST-ScUap1p;lane 4, GST-CaUap1p; lane 5, GST-HsUap1p.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Because uridyltransfer to GlcNAc-1-P releases pyrophosphate from UTP, we developed a conventional high-flux assay by adding pyrophosphatase to the reaction mixture, which allows us to estimate the enzyme activity from the amounts of inorganic phosphates produced after the hydrolysis of pyrophosphates in the reaction mixture. By this assay, it was demonstrated that the GST-ScUap1 fusion protein converted GlcNAc-1-P to UDP-GlcNAc in a dose-dependent manner (see below). Moreover, UTP was essential for the production of UDP-GlcNAc by ScUap1p; none of ATP, GTP, and CTP were used as the substrate (not shown). Because UDP-GlcNAc is an essential metabolite serving as a precursor of cell wall chitin, protein N-glycosylation, and GPI anchor in yeast (4Cabib E. Roberts R. Bowers B. Annu. Rev. Biochem. 1982; 51: 763-793Crossref PubMed Scopus (270) Google Scholar, 5Herscovics A. Orlean P. FASEB J. 1993; 7: 540-550Crossref PubMed Scopus (440) Google Scholar), ScUAP1 may be an essential gene for viability if it is the only UDP-GlcNAc pyrophosphorylase gene in S. cerevisiae. The S. cerevisiae uap1Δ null mutant strain in which the endogenous UAP1 gene was disrupted, but where episomal copies of UAP1 whose transcription was under the control of GAL1 promoter were maintained, grew on galactose plates but died on glucose plates. The cells of S. cerevisiae uap1Δ null mutant displayed an aberrant morphology; most of the yeast cells fully swelled and some were lysed, which is a phenotype quite similar to that caused by a null mutation ofAGM1, the gene for GlcNAc phosphate mutase (Fig.3). This is suggestive that theScUAP1 is a sole UDP-GlcNAc pyrophosphorylase gene inS. cerevisiae and that the most apparent defect resulting from depletion of the functional UAP1 occurred in the cell wall. To gain more insight into the characteristics of UDP-GlcNAc pyrophosphorylase, we intended to isolate theScUAP1 homologs from the pathogenic fungus C. albicans as well as from human. By screening a C. albicans genomic DNA library and a human testis cDNA library with ScUAP1 DNA as a probe, CaUAP1, andHsUAP1, C. albicans (Ca) and the human (Hs) homologs of ScUAP1, were cloned and sequenced. The predicted products of ScUAP1, CaUAP1, and HsUAP1are highly related to each other (Fig. 1). Interestingly, the clonedHsUAP1 cDNA was identical to the previously reportedAGX1 cDNA whose product is implicated as being an antigen causing male infertility (29Diekman A.B. Goldberg E. Biol. Reprod. 1994; 50: 1087-1093Crossref PubMed Scopus (39) Google Scholar). Both of the recombinant CaUap1p and HsUap1p, which were expressed in E. coli as a fusion with GST, possessed UDP-GlcNAc pyrophosphorylase activities (Fig. 2), confirming that CaUAP1 and HsUAP1 indeed specify UDP-GlcNAc pyrophosphorylase. Furthermore, expression ofCaUAP1 or HsUAP1 under the control of theScUAP1 promoter supported the growth of theUAP1-deficient S. cerevisiae cells even in the presence of glucose. Thus, it appears that both the C. albicans and human UAP1 functionally complementScUAP1 (Fig. 4). We next examined the substrate specificity of UDP-GlcNAc pyrophosphorylase using ScUap1p. ScUap1p reproducibly converted GlcNAc-1-P into UDP-GlcNAc in the presence of UTP but did not utilize GlcNAc-6-P, galactose-1-phosphate (Gal-1-P) or mannose-1-phosphate (Man-1-P) as a substrate (Fig. 5 A). Unexpectedly, the enzyme generated a spot whose mobility corresponded to that of UDP-Glc from Glc-1-P, indicating the dual substrate utility of Uap1p. However, Glc-1-P was much less efficient as shown in Fig.5 B. Consequently, ScUAP1 did not complementScUGP1 (data not shown). It is believed that the interconversion of GlcNAc-6-P and GlcNAc-1-P precedes the uridyltransfer in vivo. This prompted us to add the yeast GlcNAc phosphate mutase (ScAgm1p) to the reaction mixture. As shown in Fig. 6 A, together with ScAgm1p, ScUap1p produced UDP-GlcNAc from GlcNAc-6-P, whereas ScAgm1p alone did not. It is also demonstrated that hexosephosphate mutases require Glc-1,6-P2 either as an activator or a cofactor for the catalytic reaction (30Guha S.K. Rose Z.B. Arch. Biochem. Biophys. 1985; 243: 168-173Crossref PubMed Scopus (22) Google Scholar, 31Oesterhelt C. Schnarrenberger C. Gross W. FEBS Lett. 1997; 401: 35-37Crossref PubMed Scopus (15) Google Scholar). Therefore, we examined the effect of Glc-1,6-P2 on the synthesis of UDP-GlcNAc from GlcNAc-6-P. The TLC analysis of the products indicated that Glc-1,6-P2 was not essential for the interconversion of GlcNAc-6-P and GlcNAc-1-P, because UDP-GlcNAc was produced from GlcNAc-6-P by Agm1p and Uap1p even in the absence of Glc-1,6-P2 (Fig. 6 A). However, further assessment of the importance of Glc-1,6-P2 for the reaction by ScAgm1p revealed that the enhancement by Glc-1,6-P2 of the mutase reaction was significant when the ScAgm1p concentration was rather low (Fig. 6 B). Comparison of the amino acid sequences among UDP-sugar pyrophosphorylases revealed that the region between amino acid positions 111 and 123 of ScUap1p shares significant sequence identity with other UDP-sugar pyrophosphorylases (Fig.7). To verify the importance of this region for the catalytic activity, the highly conserved amino acids in this region, Gly111, Gly112, Gly114, Thr115, Arg116, Leu117, Pro122, and Lys123 were replaced by alanine. As was done for the wild type ScUap1p, all the mutant enzymes were expressed as a fusion with GST and purified by affinity column chromatography (Fig.8 A). Although Gly114, Thr115, and Pro122 are also highly conserved in known UDP-sugar pyrophosphorylases, replacement of these amino acids by alanine only weakly impaired the enzyme activity. In contrast, substitution of alanine for Gly112, Arg116, or Lys123 severely diminished the activity (Fig. 8 B). Furthermore, G112A but not other mutants displayed a higher K m value to GlcNAc-1-P (TableI), and all of G112A, R116A, and K123A failed to rescue the S. cerevisiae uap1Δ null mutant (data not shown). None of the mutations significantly affected theK m values in response to UTP (Table I). Taken together, it was proposed that Gly112 serves as a binding site for GlcNAc-1-P, and that Gly112, Arg116, and Lys123 are possible catalytic residues.Figure 8Effects on ScUap1p activity of alanine substitution for the conserved amino acids. The ScUap1 mutant proteins harboring an alanine substitution for each of the amino acids that are highly conserved in UDP-sugar pyrophosphorylases were expressed as a fusion with GST and purified with glutathione-Sepharose beads. A, approximately 1 μg of the wild type and the indicated mutant proteins were separated on a 10% SDS-polyacrylamide gel (PAGE) and stained with Coomassie Brilliant Blue. The position of GST-ScUap1p is indicated by the arrowhead.B, approximately 0.1 μg of the purified GST and the indicated mutant proteins were incubated with GlcNAc-1-P and UTP, and the amounts of the released inorganic phosphate that represent the enzyme activities were determined with malachite green and ammonium molybdate.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Table ICharacteristics of the mutant ScUap1pMutantK mk catk cat/K mK mto UTPμmmin−1μm−1min−1μmWild type13.52925.1868.4120.58G111A33.5780.052.3844.67G112A124.201.060.0171.36G114A19.37318.6216.4524.57T115A15.06526.5434.9523.28R116A10.920.900.0816.72L117A29.33136.314.6554.03P122A11.78264.1822.4430.73K123A18.495.090.2817.62The K m and k cat values of the wild type and mutant enzymes to GlcNAc-1-P were determined from the amounts of pyrophosphate released from UTP. The K mvalues to UTP were also indicated in the right column. Open table in a new tab The K m and k cat values of the wild type and mutant enzymes to GlcNAc-1-P were determined from the amounts of pyrophosphate released from UTP. The K mvalues to UTP were also indicated in the right column. In this paper, we have identified the eukaryotic UDP-GlcNAc pyrophosphorylase genes. The expected amino acid sequences of the yeast and human enzymes are well conserved, and both C. albicansand human enzymes functionally complement S. cerevisiae UAP1. Although the yeast enzyme catalyzed uridyltransfer to Glc-1-P, ScUap1p displayed a reasonable substrate specificity to GlcNAc-1-P, because the enzyme utilized Glc-1-P much less efficiently than GlcNAc-1-P. In fact, overexpression of ScUAP1 did not overcome the lethal phenotype caused by a depletion of UGP1in S. cerevisiae. Moreover, the enzyme did not recognize GlcNAc-6-P, but together with ScAgm1p, it produced UDP-GlcNAc from GlcNAc-6-P, demonstrating that the GlcNAc phosphate mutase reaction precedes uridyltransfer in UDP-GlcNAc biosynthesis. However, we cannot rule out the possibility that the results of the TLC assays and the high flux assays may not be exactly the same, because in the high flux assay the reverse reaction was eliminated by pyrophosphatase. Both UDP-GlcNAc pyrophosphorylase and GlcN-1-P acetyltransferase activities are authentic in E. coli GlmUp (9Wu H.C. Wu T.C. J. Bacteriol. 1971; 105: 455-466Crossref PubMed Google Scholar). It has also been demonstrated that the N-terminal region is responsible for the uridyltransfer, and acetylase activity resides in the C-terminal half of GlmUp (13Mengin-Lecreulx D. Heijenoort J. J. Bacteriol. 1994; 176: 5788-5795Crossref PubMed Scopus (166) Google Scholar). Unlike bacterial UDP-GlcNAc pyrophosphorylase, the eukaryotic enzymes seem not to be bifunctional, because ScUap1p did not utilize GlcN-1-P as the substrate and the C-terminal portion of GlmUp showed no significant sequence homology to any UDP-GlcNAc pyrophosphorylase. Thus, it is likely that in yeast, GlcN-6-P is first acetylated by an as yet unidentified enzyme and then the mutase reaction generates GlcNAc-1-P. Phosphomannomutase and phosphoglucomutase require a sugar biphosphate as a cofactor, which serves as a phosphate donor necessary to activate the enzyme by phosphorylation (30Guha S.K. Rose Z.B. Arch. Biochem. Biophys. 1985; 243: 168-173Crossref PubMed Scopus (22) Google Scholar, 31Oesterhelt C. Schnarrenberger C. Gross W. FEBS Lett. 1997; 401: 35-37Crossref PubMed Scopus (15) Google Scholar). In this study, ScAgm1p was able to produce GlcNAc-1-P even in the absence of cofactor, Glc1,6-P2, if a sufficient amount of ScAgm1p was present. One possible explanation for this discrepancy is that a small portion of the recombinant ScAgm1p was already phosphorylated and thereby activated. However, this hypothesis is inconsistent with the recent report by Oesterhelt et al. (31Oesterhelt C. Schnarrenberger C. Gross W. FEBS Lett. 1997; 401: 35-37Crossref PubMed Scopus (15) Google Scholar) that the plant and yeast enzymes utilize a sugar diphosphate as a co-substrate. Sequence comparisons of the UDP-sugar transferases revealed that there is a region where the amino acid sequence is highly conserved among most of the known UDP-sugar pyrophosphorylases. Alanine substitution for Gly112, Arg116, or Lys123severely diminished the enzyme activity and ability to complement the wild type ScUAP1 gene, strongly suggesting that these amino acids are catalytic residues. Among these three amino acids, Gly112 was shown to be a possible binding site to GlcNAc-1-P, because G112A displayed an increased K mvalue. In human UDP-Glc pyrophosphorylase, it was demonstrated that a single mutation of Gly115 to Asp drastically impaired the enzyme activity and caused cellular UDP-Glc deficiency (32Flores-Diaz M. Alape-Giron A. Persson B. Pollesello P. Moos M. von Eichel-Streiber C. Thelestam M. Florin I. J. Biol. Chem. 1997; 272: 23784-23791Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar). Sequence comparison of the Uap and Ugp proteins reveals that Gly115of the human Ugp1p corresponds to Gly112 of the yeast Ugp1p. Thus, it is likely that Gly115 of the human Ugp1p also serves as a Glc-1-P binding site. Curiously, Gal7p, which is known as UDP-Gal pyrophosphorylase, shares no significant sequence homology to known UDP-sugar pyrophosphorylases, and the conserved amino acids essential for the catalytic activity of ScUap1p are not found in Gal7p (21Tajima M. Nogi Y. Fukasawa T. Yeast. 1985; 1: 67-77Crossref PubMed Scopus (147) Google Scholar). This may imply that the catalytic mechanism of Gal7p differs from those of other UDP-sugar pyrophosphorylases. The human UDP-GlcNAc pyrophosphorylase cDNA turned to be identical to the AGX1 cDNA. Although the physiological function ofAGX1 remains to be established, it encodes an unknown antigen expressed in infertile males and is implicated in antibody-mediated human infertility (29Diekman A.B. Goldberg E. Biol. Reprod. 1994; 50: 1087-1093Crossref PubMed Scopus (39) Google Scholar). AGX1 is abundantly expressed in testes, and only low levels of AGX1 mRNA were detected in placenta, muscle, and liver (29Diekman A.B. Goldberg E. Biol. Reprod. 1994; 50: 1087-1093Crossref PubMed Scopus (39) Google Scholar). The reason why testis expresses a higher level of AGX1 mRNA and how UDP-GlcNAc pyrophosphorylase causes human male infertility await further study. In addition, there is an additional AGXcDNA, termed AGX2, which differs from AGX1 by a 48-base pair insertion in the ORF. The level of AGX2mRNA was not remarkably increased in testis; low but similar levels of AGX2 mRNA were detected in testis, placenta, muscle, and liver (29Diekman A.B. Goldberg E. Biol. Reprod. 1994; 50: 1087-1093Crossref PubMed Scopus (39) Google Scholar). Therefore, it may also be of interest to study how the internal 48-base pair insertion affects the UDP-GlcNAc pyrophosphorylase activity. We thank K. Saitoh for data base search and sequence alignment, Y. Miyazaki and Y. Kitayama for assisting with the experiments, and S. Miwa for reading the manuscript.
Referência(s)