| Glycobiology | Pages |
Cloning and functional expression of the human GlcNAc-1-P transferase, the enzyme for the committed step of the dolichol cycle, by heterologous complementation in Saccharomyces cerevisiae
Introduction
Results
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
Cloning and functional expression of the human GlcNAc-1-P transferase, the enzyme for the committed step of the dolichol cycle, by heterologous complementation in Saccharomyces cerevisiae
The gene for the human dolichol cycle GlcNAc-1-P transferase (ALG7/GPT) was cloned by screening a human lung fibroblast cDNA library. The library was constructed in a Saccharomyces cerevisiae expression vector, and the positive clone was identified by complementation of the conditional lethal S.cerevisiae strain YPH-A7-GAL. This strain was constructed by replacing the endogenous promoter of the GPT-gene by the stringently regulated GAL1-promoter. This construct allows to specifically suppress the endogenous enzyme activity. The insert of the positive clone displayed an open reading frame of 1200 nucleotides, coding for a putative protein of 400 amino acids with a calculated molecular weight of 44.7 kDa. The deduced protein sequence shows a homology of over 90% when compared with other mammalian GPT sequences, thus resembling the close phylogenetic relationship between mammalian species. This homology however decreases to 40-50% when compared to more distantly related organisms such as S.cerevisiae, Schizosaccharomyces pombe, or Leishmania amazonensis. Biochemical characterization of the recombinant protein showed that it is functionally expressed in the S.cerevisiae strain YPH-A7-GAL. GlcNAc- and GlcNAc2-PP-Dolichol biosynthesis could be shown with isolated S.cerevisiae membranes from cells harboring the recombinant plasmid and grown on glucose thus suppressing transcription of the endogenous gene. Synthesis could be stimulated by dolicholphosphate and was inhibited by tunicamycin. These results show that we have cloned the human GlcNAc-1-P transferase by heterologous complementation in S.cerevisiae, a strategy that may be useful for the cloning and characterization of glycosyltransferases from a variety of organisms.
Key words: glycosyltransferase/dolichol cycle/conditional lethality/complementation cloning/heterologous expression
Introduction
Asparagine-linked glycosylation (N-glycosylation) is initiated in all eucaryotic cells with the synthesis of lipid-linked oligosaccharides in a cyclic pathway, the dolichol-cycle. The lipid carrier for the growing oligosaccharide chain is the isoprenoid dolicholphosphate (Dol-P), anchoring the dolichol-cycle intermediates in the ER-membrane. The evolutionary conserved precursor oligosaccharide Dol-PP-GlcNAc2Man9Glc3 is assembled in a stepwise fashion and subsequently transferred onto protein by the oligosaccharyltransferase complex (for review, see Kornfeld and Kornfeld, 1985; Schachter, 1995). These protein bound oligosaccharides then undergo a complex process of remodeling called trimming on their passage through the secretory pathway (Kornfeld and Kornfeld, 1985; Schachter, 1995), yielding proteins that can exhibit a variety of species and cell type specific oligosaccharides. Furthermore, N-glycosylation plays an important role in protein folding and quality control in the ER (Sousa et al., 1992; Hammond et al., 1994; Hammond and Helenius, 1995) and is crucial for trafficking of proteins through the secretory pathway (for review, see Varki, 1993; Fiedler and Simons, 1995). Transgenic mouse models could also show that correct processing of N-glycans is a prerequisite for normal embryogenesis (Ioffe and Stanley, 1994; Metzler et al., 1994), making this system a powerful tool for studying the functions of N-glycans in vivo as reviewed in Marth (1994) and Stanley and Ioffe (1995).
The synthesis of GlcNAc-PP-Dol from dolichol phosphate and UDP-GlcNAc is the first and committed step in the dolichol cycle during which lipid linked synthesis of oligosaccharide precursors for subsequent N-glycosylation of proteins takes place as reviewed by Lehrman (1991). This reaction is catalyzed by the enzyme N-acetylglucosamine-1-phosphate transferase (GPT, ALG 7; for review, see Lehrman, 1991) and can be inhibited by the antibiotic tunicamycin (for review, see Schwarz and Datema, 1980, 1982; Elbein, 1987; Kaushal and Elbein, 1994). The GPT-gene has been cloned from S.cerevisiae (Rine et al., 1983), Schizosaccharomyces pombe (Zou et al., 1995), Leishmania amazonensis (Liu and Chang, 1992), mouse (Rajput et al. 1992 ), and CHO cells (Lehrman et al., 1988; Scocca and Krag, 1990; Zhu and Lehrman, 1990), and the organization of the mammalian genes has been elucidated (Rajput et al., 1994; Scocca et al., 1995).
Despite complex structural requirements on the protein level and the high functional conservation throughout evolution, sequence comparison between distantly related taxa such as mammals and S.cerevisiae, S.pombe, or L.amazonensis revealed an only moderate overall conservation on the amino acid sequence level (Zou et al., 1995). This finding has also been demonstrated for other glycosyltransferases (Schönbächler et al., 1995; Mazhari-Tabrizi et al., 1996) and will make the cloning of such enzymes by hybridization or PCR from sources such as protozoan parasites difficult. We therefore decided to use the approach of heterologous complementation in S.cerevisiae for the cloning of glycosyltransferases.
Complementation in S.cerevisiae has been used to clone, e.g., human cell cyclus genes (Elledge and Spottswood, 1991) and even plant genes (Minet et al., 1992). A prerequisite for such an approach are S.cerevisiae mutants with a selectable (conditional lethal) phenotype, mostly temperature-sensitive (ts) strains. Such strains have been used to clone several glycosyltransferases from S.cerevisiae by homologous complementation. These strains, however, are often difficult to handle in complementation screens and tend to revert with a certain, sometimes high frequency. We therefore constructed a conditional lethal S.cerevisiae strain (YPH-A7-GAL) by chromosomal promoter replacement, bringing the S.cerevisiae gene under the control of the tightly regulated GAL1 promoter. In this construct transcription of the endogenous gene is specifically suppressed by simply shifting these cells on medium containing glucose as sole carbohydrate source. Such a strain not only allows the cloning of essential genes by complementation but also allows to monitor the enzyme activity of the recombinant protein without the need of sometimes tedious protein purification and allows the use of complementation screens in cases where there are no ts-strains available. In this article we describe the cloning and sequencing of a human GPT-cDNA clone which is able to complement a conditional lethal S.cerevisiae strain for the GlcNAc-1-P transferase and show that the human enzyme can be expressed in an active form in S.cerevisiae.
The nucleotide sequence reported in this article has been submitted to the EMBL Data Bank with accession number Z82022
Results
Construction of the conditional lethal S.cerevisiae strain YPH-A7-GAL (YPH 499 [alg7::HIS3/GAL1-Alg7])
Our goal was to construct a conditional lethal S.cerevisiae mutant, which can be used easily for heterologous complementation screens. We decided to generate such a strain by bringing the expression of the GPT gene under the control of the stringently regulated GAL1-promoter. This promoter is induced in the presence of galactose and the absence of glucose but is tightly repressed in the presence of glucose (Johnston and Davies, 1984). This should allow turning off the expression the GPT gene brought under control of this promoter. Promoter replacement is achieved by a one step procedure, in which the target promoter is exchanged by a selection marker/promoter HIS3/GAL1-cassette (Lorenz et al., 1995). This His GALl cassette was constructed by cloning of the GAL1-promoter-fragment from p416GAL1 into pRS17 (Sikorski and Hieter, 1989; Mumberg et al., 1994) to yield the vector pGAL1/HIS3. Target sequences for the homologous recombination were chosen 200 bp upstream of the corresponding ATG-start codon in the promoter region and at the sequence including the ATG-start codon of the coding region respectively (Hartog and Bishop, 1987). Because homologous recombination in S.cerevisiae is very efficient, the flanking regions for recombination can be reduced to 50 bp, which allows to generate the HIS3/GAL1-replacement cassette by the polymerase chain reaction (Baudin et al., 1993; Wach et al., 1994; Manivasakam et al., 1995). For the amplification of the HIS3/GAL1 cassette and replacement of the endogenous promoter the primers were chosen as follows: One part of the primer (19-mer) is needed for the amplification of the His3/Gal1 cassette in the pGAL1/HIS3 vector. The second part of the primer (a 50-mer overhang) must be complementary to sequences adjacent to the chromosomal region targeted for replacement. Thus, the endogenous promoter of the gene of interest should be replaced by the Gal1-promoter via homologous recombination after transformation with the amplified His-Gal fragment. After transformation of S.cerevisiae with the PCR-generated His-Gal cassette transformants can be selected on minimal medium lacking histidine and containing galactose. Strains carrying the correct integration can be verified by Southern blot or whole cell PCR and should show the expected conditional lethal phenotype when shifted to glucose-containing medium.
The primer sequences for amplification of the His-Gal cassette were chosen as follows. The numbers indicate the nucleotide positions in the S.cerevisiae DNA sequence with the adenosine of the initiation codon being defined as position +1: (1) sequence for integration upstream the coding region (nucleotides -253 to -200), 5[prime]GTTACGAATACAAACACAGAGGTTGATGCATGAATTTTTTCTAGCTACTACCAGGGCGAATTGGAGCTCCAC; (2) sequence for integration at the initiation codon (nucleotides +1 to +253), 5[prime]TGGAATAGTAGATTAAGCATGTGATAAGTGCCAGTGAAAAAAGTCGCAACATGGGGATCCACTAGTTCTAG. The 19 bp sequences at the 3[prime] ends of these oligonucleotides, which are homologous to sequences of the vector pGAL1/HIS3 adjacent to the GAL1/HIS3-cassette serving as template for amplification, are printed in italics. The following 50 nucleotides in the 5[prime] direction are homologous to segments in the promoter region to be replaced by homologous recombination. Transformation of 1 µg of the respective PCR-fragment into the haploid strain YPH 499 yielded 18 transformants on galactose containing plates lacking histidine. Fifteen of the transformants showed the expected conditional lethal phenotype when plated under nonpermissive conditions on glucose medium. The correct insertion of the His-Gal fragment into genomic DNA was confirmed by whole cell PCR using primers 5[prime] and 3[prime] adjacent to the integration sites. The sizes of the PCR fragments were verified by agarose gel electrophoresis with corresponding fragments from YPH 499 as control. To finally confirm that the lethality was due to repression of the gene in question, the strain YPH-A7-GAL was transformed with the plasmid pJR 39 harboring the wild type S.cerevisiae GPT gene (Rine et al., 1983) which resulted in the expected rescue of the conditional lethal phenotype.
Isolation of a human cDNA fragment complementing the conditional lethal defect of the S.cerevisiae strain YPH-A7-GAL
The strain YPH-A7-Gal was transformed with the human lung fibroblast cDNA library as described in Materials and methods; 250,000 recombinant clones were plated and grown under nonpermissive conditions (SD medium). Forty-eight initial clones were obtained growing on glucose as sole carbohydrate source. In order to eliminate false positives these clones were plated on SGR-medium containing 5-fluoroorotic acid to screen for loss of plasmid (Boeke et al., 1987). Colonies were subsequently streaked out again on SD medium. The eight clones that lost their ability to grow on glucose were analyzed further.
Recombinant plasmids were isolated, amplified in E.coli, and used to retransform the strain YPH-A7-GAL. One of these clones (clone HA7-8) was able to repeatedly rescue the conditionally lethal phenotype of YPH-A7-GAL and thus chosen for sequencing and biochemical analysis.
Sequence analysis of the human GPT-gene
Restriction analysis of clone HA7-8 revealed a cDNA insert of 1.3 kb. The DNA sequence of both strands was determined by double strand sequencing after generation of subclones in SK-vectors. For regions that were not readily accessible by standard subcloning, specific DNA primers were synthesized to span those gaps.
An open reading frame of 1200 base pairs was identified revealing a protein sequence of 400 amino acids with a calculated molecular weight of 44.7 kDa and an isoelectric point of 8.06. Figure 1 shows the cDNA and derived amino acid sequence of the putative GlcNAc-1-P-transferase. The predicted cleavage site for the signal peptidase was determined according to Von Heijne (1983). Sequence comparison revealed a homology of 93% between the human and the hamster sequence on the amino acid level whereas the homology between the human and the S.cerevisiae protein is 42%. The two potential dolichol binding sequences are identical between the human and the hamster gene in respect to sequence and position within the protein. Four recognition sites for N-glycosylation are present in the deduced protein sequence (amino acids 39-41, 138-140, 352-354, 370-372). Three of these sites are located at comparable positions between both sequences (138-140, 352-354, 370-372). At the first recognition sequence (39-31) the threonine is changed to leucine in the hamster sequence, whereas at position 317-319 the asparagine is replaced by a serine in the human protein.
Figure
Northern blot analysis of clone HA7-8
To show that the cDNA insert of clone HA7-8 is of human origin, and to obtain a preliminary assessment of a potential transcript heterogeneity, Northern blots were performed as described in Materials and methods. When poly(A)+ RNA as well as total RNA from human lung fibroblasts was probed with the digoxygenin labeled cDNA insert, a band in the 2 kb range could be detected with poly(A)+ RNA under stringent (Figure 2, lane 2) and nonstringent conditions. Upon overexposure no additional bands appeared, and a faint band could also be detected with total RNA (data not shown). When poly(A)+ RNAs from the protozoan parasites T.brucei, P.falciparum,and T.gondii were probed with the human cDNA fragment no signal could be detected (data not shown). These data indicate that human lung fibroblasts, in contrast to S.cerevisiae, L.amazonensis and CHO cells do not show transcript heterogeneity of the GPT-gene and that the RNA for this enzyme does not represent one of the major transcripts in this cell line.
Figure
The cloned human GlcNAc-1-P transferase is active in S.cerevisiae
The capability of the human cDNA clone to repeatedly complement the conditional lethal phenotype of the strain YPH-A7-GAL strongly suggests that the coding sequence indeed represents the enzyme GlcNAc-1-P transferase. To prove this we performed enzyme assays as described in Materials and methods. Cells were grown in SD or SGR medium overnight to completely suppress the endogenous GPT activity under nonpermissive conditions. Membrane preparations were then assayed for GPT activity.
To demonstrate the identity of the radiolabeled reaction products, C/M extractable glycolipids were analyzed by TLC as shown in Figure 3. Figure 3A shows such an analysis for the conditional lethal strain YPH-A7-GAL grown in SGR medium (permissive conditions). The two peaks of radioactive glycolipids represent GlcNAc-PP-Dol (peak 1) and GlcNAc2-PP-Dol (peak 2) as demonstrated below. Figure 3B shows the results for YPH-A7-GAL grown under nonpermissive conditions (SD medium) demonstrating that the endogenous enzyme activity can be suppressed to a nondetectable level. Incubation of YPH-A7-GAL harboring plasmid HA7-8 (YPH-A7-GAL [HA7-8]) under permissive conditions (Figure 3C) yields an identical glycolipid pattern as compared to Figure 3A. Incubation of this construct under nonpermissive conditions fully restored GPT-activity as shown in Figure 3D although the relative amounts of GlcNAc-PP-Dol versus GlcNAc2-PP-Dol vary between the different growth conditions. The same effect was seen when membranes from wild type S.cerevisiae cells (YPH 499) grown in SGR or SD medium were analyzed (data not shown).
Figure
To finally confirm that the radioactivity in the C/M extractable compounds are GlcNAc or GlcNAc2 the corresponding areas were scraped off the TLC plates. Glycolipids from peaks 1 and 2 were eluted from the TLC material and subjected to acid hydrolysis. This treatment liberates the carbohydrate moiety from dolicholpyrophosphate which then can be analyzed by gel filtration on Bio Gel P2. Figure 4 shows the result of such an analysis for peaks 1 and 2. The radiolabeled material from both peaks comigrates with a GlcNAc and GlcNAc2 standard respectively, thus demonstrating clearly that the enzyme activity of clone HA7-8 is the human GPT.
Figure
The recombinant GPT gene confers a higher tolerance towards tunicamycin
Membranes from wild type cells (YPH 499) and cells harboring the human GPT gene (YPH 499 [HA7-8]) were incubated with concentrations of tunicamycin ranging from 0.1 pg/ml to 1 µg/ml prior to the addition of [3H] UDP-GlcNAc. Assays were performed in triplicate and the result of such an experiment is shown in Figure 5. The diagram clearly demonstrates that expression of the recombinant protein from a multicopyexpression vector such as pRS 426 MET confers a higher tolerance toward this inhibitor due to elevated enzyme synthesis, thus showing a significant gene dose effect. With membrane preparations from wild type cells, a 50% inhibition is achieved at a tunicamycin concentration of 0.05 ng/ml. In the case of cells expressing the recombinant enzyme, concentrations of 1-1.5 ng/ml are necessary to obtain the same degree of inhibition.
Figure
Discussion
We have cloned the gene for the human GlcNAc-1-P transferase (GPT/Alg7) from an expression cDNA library by heterologous complementation in S.cerevisiae. The approach of heterologous complementation in this microorganism has been used for a wide variety of functionally conserved genes from several organisms. In the case of glycosyltransferases, the cloning of homologous genes from evolutionary distantly related species such as S.cerevisiae, protozoa, and mammals showed that the degree of conservation on the sequence level not necessarily reflects their functional conservation among eucaryotes (Zhu and Lehrman, 1990; Hamburger et al., 1995; Zou et al., 1995; Mazhari et al., 1996). This may hamper the cloning of such enzymes from, e.g., protozoan sources by hybridization or PCR. Since a number of glycosyltransferase genes for N-glycosylation and GPI-anchor biosynthesis have been cloned by homologous complementation in S.cerevisiae or mammalian cell lines (for review, see Schachter, 1995; Varki, 1995; Takeda and Kinoshita, 1995; Eckert et al., 1997), we decided to use the approach of heterologous complementation in S.cerevisiae for the cloning of such enzymes. Using this strategy we previously cloned the gene for the Dol-P-Man synthase from the protozoan parasite Trypanosoma brucei (Mazhari et al., 1996). The results presented here demonstrate that this approach can be applied to a variety of sources since conditional lethal S.cerevisiae strains for essential genes can be generated by chromosomal promoter replacement as shown by the construction of the strain YPH-A7-GAL. The human cDNA clone HA7-8 was identified as harboring the coding sequence for the human GPT gene by its ability to complement the conditional lethal defect of this strain YPH-A7-GAL and by its ability to exhibit GPT activity in vitro. Sequence analysis showed an open reading frame coding for 400 amino acids being in agreement with other published GPT sequences which range from 408 to 466 amino acids (Zou et al., 1995). Northern analysis revealed a single transcript in the 2 kb range. Single transcripts have also been found for mouse (Rajput et al., 1992) and S.pombe (Zou et al., 1995) RNAs, whereas S.cervisiae (Kukuruzinska and Robbins, 1987), L. amazonensis (Liu and Chang, 1992), and CHO cells (Scocca and Krag, 1990) showed multiple GPT transcripts.
Analysis of the chromosomal organization showed a complex structure for the two rodent genes with the coding sequence being interrupted by eight introns. In S.cerevisiae, S.pombe, and L.amazonensis the GPT-gene was found to be intronless. The high homology of over 90% between the mammalian sequences clearly reflects the close phylogenetic relationship between these species. The three mammalian sequences harbor two potential dolichol recognition sequences which in the case of the hamster gene were both shown to be essential for enzyme function (Datta and Lehrman, 1993). The protein from S.cerevisiae contains only one such sequence (Zhu and Lehrman, 1993), and the L. amazonensis enzyme lacks a sequence resembling this consensus (Liu and Chang, 1992). In the case of another glycosyltransferase, the Dol-P-Man synthase, it could be shown that in the S.cerevisiae protein, which harbors one potential dolichol recognition sequence, this sequence is not necessary for enzyme activity per se (Schutzbach et al., 1993) and that the enzyme from Trypanosoma brucei completely lacks such a sequence (Mazhari-Tabrizi et al., 1996). It therefore remains to be elucidated whether these sequences indeed are necessary for dolichol recognition and binding or whether they serve another purpose.
Four potential sites for N-glycosylation are present in the mammalian proteins. Three of these sites (amino acids 138-140, 352-354, and 370-372 ) are conserved with respect to the hamster protein whereas the recognition site at position 39-41 is not present in the hamster protein due to a change from threonine to leucine. The site at position 325-327 in the hamster sequence is absent in the human protein due to a change from asparagine to serine at position 317 in the human sequence. The positions of these sites in the mouse protein are comparable to those of the hamster sequence. The L.amazonensis sequence shows two, the S.cerevisiae and S.pombe proteins only one site for N-glycosylation indicating that even the conservation of such recognition sites is rather poor between enzymes from distantly related taxa, raising the question of their functional relevance.
As expected, the homology on the amino acid sequence level is high between the three mammalian sequences (>90%) but drops significantly when the other species are included in these comparisons, ranging from 32% (L.amazonensis-S.cerevisiae/CHO cells) to 50% (S.pombe-S.cerevisiae). This divergence in the protein sequence between distantly related taxa may make the detection of GPT sequences by hybridization with heterologous probes impossible as was the case when L.amazonensis DNA was probed with the Alg7 gene from S.cerevisiae (Liu and Chang, 1992). A similar result was obtained when a S.pombe genomic library was screened with a S.cerevisiae Dol-P-Man synthase probe (Zou et al., 1995), demonstrating the apparently low degree of sequence conservation for at least some glycosyltransferases. These data emphasize that, especially in the light of a sometimes very biased codon usage in certain protozoa, heterologous complementation may present a valuable alternative to the screening by hybridization. Sequence alignment of the previously published GPT sequences (Zou et al., 1995) show that there are invariant amino acids present in all five species but only two regions consisting of five contiguous amino acids (positions 240-244 and 291-295 in the human protein) shared by all sequences. Additional sequence information from other species will determine whether these two regions indeed are invariant and can be used for PCR-cloning.
To finally prove that we cloned the human GPT sequence we performed in vitro assays using isolated membranes from YPH-A7-GAL cells that were grown under nonpermissive conditions. This allowed suppression of the endogenous enzyme activity and thus characterization of the cloned gene. GlcNAc- and GlcNAc2-PP-Dol synthesis could be demonstrated by TLC analysis of the radioactive labeled reaction products as well as by demonstrating that, after acid hydrolysis, the radioactivity found in the glycolipid fraction was comigrating with GlcNAc and GlcNAc2 standards in Bio Gel P2 column chromatography. This in vitro synthesis of the first dolichol cycle intermediates could also be inhibited by tunicamycin, a specific inhibitor of GPT activity. Furthermore, membranes prepared from S.cerevisiae cells harboring the plasmid containing the human GPT gene showed a higher tolerance toward tunicamycin due to the higher amount of protein being made, showing a clear gene-dose effect.
These data show that heterologous complementation in S.cerevisiae can be used for the cloning of essential genes from various sources where the conservation on the sequence level between distantly related taxa is not very high. In addition, our data demonstrate that conditional lethal S.cerevisiae strains in which the endogenous enzyme activity is suppressed by simply shifting such cells to a medium containing glucose as sole carbohydrate source can be used to specifically monitor the activity of a recombinant protein. Therefore, such constructs can be highly useful for the determination of enzyme properties such as susceptibilities toward inhibitors especially if purification of such proteins for in vitro assays in an active form cannot be achieved easily and might also provide a powerful system for drug screening.
Materials and methods
Chemicals
UDP-[3H]-GlcNAc, (36.5 Ci/mmol) was purchased from NEN.
Dolicholmonophosphate (C80-105) and Lyticase (300 U/mg) were from Sigma, tunicamycin was from Calbiochem. TLC plates were purchased from Merck, Bio Gel P2 was from Bio-Rad, and chemicals were analytical grade unless stated otherwise.
Restriction endonucleases and other DNA modifying enzymes used in recombinant DNA experiments were from Boehringer-Mannheim, New England Biolabs, or Stratagene and were used in accordance with the manufacturer's instructions.
Strains and media
The following Saccharomyces cerevisiae and E. coli strains were used in this work. YPH 499 (Mat a; ura3-52; lys2-801amber; ade2-101ochre; trp1-[Delta]63; his3-[Delta]200; leu2-[Delta]1; Sikorski and Hieter 1989; Stratagene) was used as a 'wild type control" in respect to GPT-activity. E.coli strain XL1-blue (Stratagene) was used for subcloning and other standard recombinant DNA procedures.
S.cerevisiae strains were grown in YPAD medium (1% Bacto yeast extract, 2% Bactopeptone, 2% dextrose, 4 mg/l adenine) or SD medium (0.17% Bacto yeast nitrogen base and 2% dextrose) containing the nutritional supplements necessary to complement strain auxotrophies or allow selection of transformants. YPH-A7-GAL was maintained on SGR-His medium (4% galactose, 2% raffinose, 0.17% Bacto yeast nitrogen base, 0.5% ammonium sulfate) where dextrose is replaced by galactose/raffinose as carbohydrate source.E.coli strains were grown in LB medium, supplemented with ampicillin (100 µg/ml) when necessary.Recombinant DNA methods
All standard DNA techniques were carried out as described (Sambrook et al., 1989) unless mentioned otherwise. Plasmid isolation from E. coli was routinely performed using Quiaprep columns (Qiagen). Plasmid isolation from S.cerevisiae followed the protocol of Hoffman and Winston (1987). Transformation of S.cerevisiae was performed according to Ito et al. (1983). Restriction fragments were separated by agarose gel electrophoresis in Tris-acetate buffer. Individual restriction fragments were isolated with the QIAquick Gel extraction Kit (Qiagen).
Sequencing was performed by PCR-cycle sequencing using the ABI PRISM Dye Terminator Cycle Sequencing Kit from PERKIN ELMER. Sequencing PCR reactions were performed using the PERKIN ELMER Gene Amp 2400 PCR System. Sequence analysis was performed with the ABI Prism 377 DNA Sequencer. Computer analysis of sequence data was performed with the Heidelberg Unix Sequence Analysis Resources (HUSAR), release 3.0. using the Gap program according to the algorithm of Needleman and Wunsch (1970)
PCR conditions
PCR reactions were performed using the Hotwax Optistart Kit for PCR Optimization from Invitrogen in accordance with the manufacturer's instructions. 10 ng Template (pGAL1/HIS3); 50 ng of each primer (70 bp); 10 µl 5× buffer, pH 8.5; 5 µl 10 mM dNTPs; 1U Taq polymerase; H2O ad 50 µl + hot wax Mg2+ Beads (final MgCl2 concentration 2.5 mM). PCR conditions were as follows. Denaturation, annealing, and extension: 1 min 94°C, 1 min 60°C, 2 min 72°C, 35 cycles (cycle extension: 2 sec/cycle). Whole cell PCR was performed using YPH499 as wild type control and the corresponding promoter mutants for verification of the correct insertion of the His Gal cassette. The His GALl cassette was constructed by cloning of the GAL1-promoter-fragment from p416GAL1 into pRS17 (Sikorski and Hieter, 1989; Mumberg et al., 1994).
Construction of a cDNA library from human lung fibroblasts
W138 cells (Hayflick, 1965) were starved by serum deprivation for 72 h followed by stimulation with 10% FCS. RNA was isolated after 12, 14, and 16 h according to Chomczynski and Sacchi (1987). Poly(A)+ RNA was isolated with the Poly(A)+ Quick mRNA Purification Kit from Stratagene according the manufacturer's instructions.
The cDNA synthesis was performed by using the Zap-cDNA synthesis kit from Stratagene for unidirectional cloning of cDNA inserts. The cDNA was ligated into the S.cerevisiae expression vector pRS 416-Met (Mumberg et al., 1995) which had been digested with the restriction enzymes EcoRI and XhoI. This library was then transformed into E.coli DH5[alpha] by electroporation (Electromax Gibco/BRL) and amplified.
Nucleic acid hybridization
Northern transfers were essentially performed as described previously (Sambrook et al., 1988). Briefly: RNA was separated on a 1% formaldehyde agarose gel followed by capillary transfer onto nylon membranes (Boehringer Mannheim) and UV crosslinking for 1.25 min (Stratalinker, Stratagene). Hybridization was performed in 5× SSC/50% formamide at 45°C and 52°C, respectively, using the digoxigenin labeled human GPT-fragment as probe (Digoxigenin Labeling and Detection Kit, Boehringer Mannheim) followed by two washes at room temperature in 2× SSC/0.2% SDS and 2 additional washes at 45°C and 52°C, respectively (0.1× SSC/0,1% SDS; 15 min). Visualization of hybridization signals was then performed by fluorographic detection using the Boehringer CSPD kit according to the manufacturers instructions. Digoxigenin labeled RNA fragments (Boehringer Mannheim) were used as size markers
Assay of GPT-activity
S.cerevisiae cells were grown overnight either in SD or SGR liquid cultures and harvested by centrifugation. Membranes were prepared from 3 × 108 cells as described previously (Rine et al., 1982). Cell viability was checked by plating serial dilutions on SGR plates. Enzyme assays were performed as described previously (Lehrman et al., 1988). Briefly, 10 µl of purified S.cerevisiae membranes (equivalent to 3 × 107 cells) were preincubated for 15 min at 30°C in 10 mM Tris-HCl, pH 8.0, 10 mM MgCl, 0.04% NP 40 and 160 µg/ml dolicholphosphate before adding 0.5 µCI UDP-[3H]-GlcNAc (5 µl). The total reaction volume was 100 µl. The incubation time was 10 min at 30° C.
Incubations were terminated by the addition of 1 ml ice cold chloroform/methanol (CM, 3/1, v/v), vortexed ,and centrifuged. The liquid phase was transferred to a fresh tube, and the pellet was discarded. Phase separation was induced by the addition of 200 µl 4 mM MgCl2 followed by centrifugation. The aqueous phase was aspirated and the organic phase was subjected to repeated Folch washing (Sharma et al., 1974; Gerold et al., 1994). An aliquot of the Folch washed material was taken for determination of the radioactivity in the organic phase by scintillation counting. Samples of the radiolabeled, chloroform/methanol-extractable reaction products were analyzed by thin layer chromatography on Silica 60 plates (Merck) using chloroform/methanol/acetic acid/water (25:15:4:2) as solvent. Plates were screened for radioactivity using a Berthold LB 2842 Automatic TLC Linear Analyzer.
To finally demonstrate the nature of the lipid-linked radioactivity, the radiolabeled material was purified by DEAE-cellulose chromatography and subjected to acid hydrolysis (2 M HCl/1-propanol, 1:1, v/v, 15 min, 50°C) as described by McDowell and Schwarz (1988). The reaction products were analyzed by gel filtration on Bio Gel P2 columns.
Acknowledgments
This work was supported by grants from the Deutsche Forschungsgemeinschaft to R.T.S. (Sonderforschungsbereich 286), PROCOPE/DAAD, Fonds der Chemischen Industrie, Hessisches Ministerium für Wissenschaft und Kunst and P. E. Kempkes Foundation Marburg, Germany. R.M. thanks the Friedrich Ebert Stiftung for a doctoral fellowship (St. Nr. 171859). We are grateful to Dr. Jasper Rine for providing the yeast alg7 clone and to Dr. Mark A. Lehrman for providing the hamster GPT clone, which were used as positive controls. We also thank Dr. P. Gerold for critically reading the manuscript.
Abbreviations
Dol-P, dolicholphosphate; ER, endoplasmic reticulum; GlcNAc, N-acetylglucosamine; GPT/ALG 7, N-acetylglucosamine-1- phosphate transferase; ts, temperature sensitive.
References
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 18 Jan 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Noffz, S. Keppler-Ross, and N. Dean Hetero-oligomeric interactions between early glycosyltransferases of the dolichol cycle Glycobiology, May 1, 2009; 19(5): 472 - 478. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Shams-Eldin, B. Chaban, S. Niehus, R. T. Schwarz, and K. F. Jarrell Identification of the Archaeal alg7 Gene Homolog (Encoding N-Acetylglucosamine-1-Phosphate Transferase) of the N-Linked Glycosylation System by Cross-Domain Complementation in Saccharomyces cerevisiae J. Bacteriol., March 15, 2008; 190(6): 2217 - 2220. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Takahashi, R. Honda, and Y. Nishikawa Cloning of the human cDNA which can complement the defect of the yeast mannosyltransferase I-deficient mutant alg 1 Glycobiology, March 1, 2000; 10(3): 321 - 327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Strahl-Bolsinger and A. Scheinost Transmembrane Topology of Pmt1p, a Member of an Evolutionarily Conserved Family of Protein O-Mannosyltransferases J. Biol. Chem., March 26, 1999; 274(13): 9068 - 9075. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.A. Kukuruzinska and K. Lennon Protein N-Glycosylation: Molecular Genetics and Functional Significance Critical Reviews in Oral Biology & Medicine, January 1, 1998; 9(4): 415 - 448. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||