Abstract
A cDNA clone containing a 2,150-bp insert was isolated from a bovine lung λgt10 cDNA library by cross-species hybridization using a DNA probe generated by polymerase chain reaction (PCR) employing a human cDNA that encodes mucin core 2 β6-N-acetylglucosaminyltransferase (hC2TF) as the template. The bovine cDNA (bcDNA) insert was devoid of 220 bp of the 5′ portion of the C2TF open reading frame (ORF), as predicted from the human counterpart. Southern blotting analysis suggested that the coding region of this C2TF gene is in one exon. To construct a full-length bovine C2TF (bC2TF) cDNA, a genomic DNA fragment containing the 5′ portion of the ORF of the bC2TF gene was cloned from a λEMBL bovine genomic DNA library and ligated to the 5′ end of the cloned cDNA insert. DNA sequence analysis showed that the complete ORF of bC2TF gene was 1,281 bp in length, which corresponds to a polypeptide of 427 amino acids. Catalytically active bC2TF was expressed in sf21 insect cells infected with recombinant baculovirus containing the ORF of the bC2TF gene. The recombinant bC2TF catalyzed the synthesis of core 2, but not core 4 and blood group I structures. Western blotting analysis showed that the recombinant bC2TF migrated with the same mobility (∼ 55 kD) as the native bovine tracheal C2TF. Immunohistochemical analysis showed that in bovine trachea, the bC2TF was present at the surface epithelium and in the submucosal glands, with the latter being the major site of distribution.
Airway epithelial mucins are secreted as high molecular weight glycoproteins consisting of 80–90% carbohydrate and 10–20% protein (1). Airway epithelial mucins play an important role in lung defense by trapping airborne microorganisms through covalently linked carbohydrates (2, 3). Mucin carbohydrates are heterogeneous in chain length, charge, and structures (4). Carbohydrate structure can be regulated by the relative expression of chain-elongation and chain-termination glycosyltransferases that act on the same branch points (5, 6). Most mucins contain N-acetylgalactosamine (GalNAc), galactose (Gal), N-acetylglucosamine (GlcNAc), fucose, and neuraminic acid. The maximal size of oligosaccharides without GlcNAc is limited to pentasaccharides, such as the sialylated oligosaccharide, found in porcine submaxillary mucin, that exhibits blood group A activity (7). Thus, GlcNAc transferases, which synthesize N-acetylglucosaminides, play key roles in the elongation of mucin-type oligosaccharides. β6-GlcNAc transferases are responsible for the synthesis of three branched mucin-type oligosaccharides, i.e., core 2, core 4, and blood group I (8), from core 1, core 3, and blood group i structures, respectively. These core structures are as follows:
Galβ1–3GalNAc Core 1
Galβ1–3(GlcNAcβ1–6)GalNAc Core 2
GlcNAcβ1–3GalNAc Core 3
GlcNAcβ1–3(GlcNAcβ1–6)GalNAc Core 4
GlcNAcβ1–3Gal Blood group i
GlcNAcβ1–3(GlcNAcβ1–6)Gal Blood group I
Mucin core 2 β6-N-acetylglucosaminyltransferase is involved in the regulation of an early branching, mucin-type carbohydrate (6). Enhancement of the activity of this enzyme is associated with longer carbohydrate chain length, activation of T lymphocytes, and malignant transformation (9, 10). Whereas core 1, core 2, core 3, and core 4 structures are found only in serine/threonine-linked mucin-type carbohydrate, blood groups i and I can be found in both mucin-type and asparagine-linked serum-type glycans (1, 11-13).
To study further the role of β6-GlcNAc transferases in their regulation of mucin carbohydrate synthesis, it is necessary to purify and characterize these enzymes. However, purification of glycosyltransferases is not only time consuming but also technically difficult. Rapid advancement in molecular biotechnology has provided a new way to generate recombinant enzymes for biochemical studies. For example, cDNAs encoding core 2 N-acetylglucosaminyltransferase (C2TF) (EC 2.4.1.102) (11, 12) and blood group I (13) β6-GlcNAc transferases have been cloned. These studies have helped us understand the protein domains of these enzymes, which would have been difficult to attain by classic means.
In this article, we report the cloning of a full-length C2TF cDNA from bovine lung by cross-species hybridization. A fully active bC2TF was expressed in sf21 insect cells. In addition, we employed monoclonal antibody (mAb) generated with recombinant bC2TF to show the distribution of this enzyme in bovine trachea and other tissues.
Restriction enzymes, alkaline phosphatase, T4 DNA ligase, an Erase-a-Base DNA deletion kit, and a Wizard DNA prep kit were purchased from Promega (Madison, WI). A polymerase chain reaction (PCR) kit and random priming DNA-labeling kit were from Boehringer Mannheim (Indianapolis, IN). Bluescript II SK+ plasmid was from Stratagene (La Jolla, CA) and pET 23-c plasmid was from Novagen (Madison, WI). The Escherichia coli X-press system, pRSET vector, and the metal ion affinity column were purchased from Invitrogen (San Diego, CA) and the DNA Elutip-d, from Schleicher & Schuell GmbH (Keene, NH). The enhanced chemiluminescence (ECL) kit was obtained from Eastman Kodak (New Haven, CT). The bovine genomic DNA, the λEMBL3-SP6/T7 bovine genomic DNA library, the λgt10 bovine lung cDNA library, the site-directed mutagenesis kit, the baculovirus expression system and the sf21 insect cells were obtained from Clontech (Palo Alto, CA). [35S]dATP, [32P]dCTP and an ECL–Western blotting detection kit were from Amersham (Arlington Heights, IL). Fluorescein-conjugated rabbit anti-mouse immunoglobulin (Ig) and horseradish peroxidase-conjugated goat anti-mouse Ig were purchased from Cappel (Durham, NC). A Sequenase 2.0 kit was from United States Biochemical (Cleveland, OH). Other chemicals were of molecular biological grade unless otherwise specified.
Three oligonucleotide primers were synthesized (Genosys Biotech, Woodlands, TX) on the basis of the sequence of a human promyelocytic core 2 β6-GlcNAc TF (hC2TF) cDNA (11): (1) 5′-GGACACCTGACGACTATATAAACATG-3′ (nucleotides 263–288); (2) 5′-GGTGGAAGAAGCGGTATGAGG-3′ (nucleotides 761–781); (3) 5′-CCTCATACCGCTTCTTCCACC-3′ (nucleotides 781–761); and (4) 5′-GCGCAGCATCCAGTTCAAGTCACC-3′ (nucleotides 1179–1156). The PCR reaction was performed initially at 94°C for 5 min, which was followed by 30 cycles of the following program: 55°C (1 min), 72°C (2 min), and 94°C (1 min). The reaction was terminated after incubation at 72°C for 5 min. DNA fragments of 917 and 519 bp were synthesized with paired primers 1 and 3 and 1 and 2, respectively. The PCR products were purified with the Wizard DNA prep kit and labeled by [α-32P]dCTP with the DNA random priming labeling kit. The labeled DNA probes were purified by Elutip-d column and used for screening and Southern blotting analysis.
BamHI, EcoRI, and HindIII were used individually to digest 10 μg of bovine genomic DNA. All digestions were carried out at room temperature for 18 h. The digested DNA was purified by phenol–chloroform extraction and ethanol precipitation, run in an 0.8% agarose gel, and blotted onto nitrocellulose membrane by standard protocol (14). The blotted membrane was incubated with 20 ml of prehybridization solution, which contained 1 M NaCl, 5% dextran sulfate (Sigma, St. Louis, MO), 1% sodium dodecyl sulfate (SDS), and 1 μg of salmon sperm DNA, at 65°C for 4 h. Freshly made 917-bp 32P-labeled PCR fragment (1.5 × 106 cpm) was then added to the prehybridization solution and incubated at 65°C for 18 h. The membrane was successively washed with 3×, 1×, and 0.3× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate) (30 min each wash) and exposed to X-ray film at −70°C for 4 wk.
The 917-bp 32P-labeled PCR fragment was employed as the probe to screen a bovine lung λgt10 cDNA library according to the protocol provided by Clontech. The positive clones were purified by two more rounds of subscreening and amplification on agarose plates. The recombinant phage DNA was extracted from the clones, using the Clontech protocol with some modifications: The amplified phages were recovered from the plates and treated with 1 μg/ml RNase and DNase at 37°C for 30 min. The treated sample was mixed with an equal volume of 20% polyethylene glycol (PEG)–2 M NaCl, incubated in ice for 1 h, and centrifuged at 9,000 × g for 10 min. The phage pellet was resuspended in 0.5 ml of deionized H2O and recentrifuged at 9,000 × g for 10 min. The supernatant was treated with 15 μl of proteinase K solution (1.7 mg/ml) containing 0.33% SDS and 0.17 M EDTA at 37°C for 30 min. The DNA was then purified by phenol–chloroform extraction and ethanol precipitation. The inserts of these recombinant phage DNA were characterized by PCR analysis and restriction mapping. These inserts were confirmed by DNA sequencing after being cloned into the Bluescript II SK+ plasmid.
For screening the bovine genomic DNA library, the 519-bp DNA fragment was used as the probe. The positive clone was analyzed by restriction mapping.
The dideoxy chain termination method of Sanger and coworkers (15) was used for DNA sequencing, using a Sequenase 2.0 kit. Both primer extension and overlapping deletion methods were applied to sequence the cloned cDNA and genomic DNA inserts. The oligonucleotides for primer extension sequencing were synthesized in the Department of Pathology of the University of North Carolina (Chapel Hill, NC). Overlapping deletion subclones were constructed in the Bluescript II SK+ plasmid using the Erase-a-Base system.
For construction of the recombinant expression plasmid containing the complete open reading frame (ORF) of the bC2TF gene, a site-directed mutation was performed with a transformer site-directed mutagenesis kit (Promega). A 24-base oligonucleotide mutation primer corresponding to the upstream sequence (−50 to ∼ −73 bp) of the bC2TF gene with a single base substitution of A to T to create an external EcoRI site and a selection primer (26 bp, corresponding to the polylinker region of Bluescript II SK+ plasmid, with a base change to remove the XbaI site) were used. The mutagenesis was performed according to the instruction provided with the kit.
Two recombinant baculovirus transfer vectors were constructed (Figure 3). One of the vectors contains the complete ORF of bC2TF DNA cloned into the EcoRI site of the plasmid whereas the other has partial bC2TF ORF DNA cloned into the XbaI and BglII sites. The complete ORF DNA was obtained by partial EcoRI digestion of the mutated bC2TF DNA, which has an internal EcoRI site located at +98 bp of the C2TF DNA, and cloned into a baculovirus transfer vector via the EcoRI site. The recombinant vector containing partial C2TF ORF DNA was constructed as follows: first, the bC2TF DNA was extensively digested with EcoRI and the resultant bC2TF DNA, devoid of the first 98 bp of the bC2TF ORF, was subcloned into the pET23-c plasmid downstream of the T7 tag (11 amino acids [aa]) to form a 5′-minus-bC2TF-T7 fusion protein vector. The DNA fragment containing the partial C2TF DNA and T7 tag sequence was then excised from this vector by XbaI and BglII and cloned into the XbaI and BglII sites of the baculovirus transfer vector to form a recombinant transfer vector containing the DNA sequence of T7-partial-bC2TF fusion protein.
Fig. 3. The strategy for cloning of an expression plasmid containing a complete ORF of bovine lung C2TF cDNA. First, the bC2TF cDNA devoid of 220 bp at the 5′ end was ligated with the cloned bC2TF genomic DNA at the BamHI site and an EcoRI site was created upsteam from the start site of the bC2TF DNA insert. The DNA was then partially digested by EcoRI and the fragment containing the complete coding sequence of the bC2TF gene was cloned into the EcoRI site of a baculovirus transfer vector. For construction of an expression plasmid containing bC2TF cDNA devoid of 98 bp at the 5′ end, the cDNA was completely digested with EcoRI, and the large fragment without the first 98-bp portion of bC2TF cDNA was cloned into the EcoRI site of pET 23-c downstream to the T7 tag sequence. The insert was then excised by digestion with XbaI and BglII and cloned into the corresponding sites of the transfer vector.
[More] [Minimize]For preparation of the recombinant baculoviruses, the sf21 insect cells were cultured in a 60-mm dish at 27°C with 3.5 ml of insect cell culture medium supplemented with 10% fetal bovine serum. When the cells became 80–90% confluent, each dish of the cultured cells was cotransfected with 10 μg of purified recombinant transfer vector DNA and 20 μg of Bsu36I-digested baculovirus DNA (according to the Clontech baculovirus expression system manual) and cultured at 27°C. On Day 5 posttransfection, the conditioned medium was harvested and inoculated into freshly cultured sf21 cells. The resultant recombinant baculovirus containing bC2TF and 5′-minus-bC2TF DNA inserts were purified by limiting dilution in this insect cell line.
Enzyme activities were assayed as described (16) with some modification. After the sf21 cells had been infected with recombinant or wild-type baculoviruses, the culture suspension was subjected to a low-speed spin (500 × g, 5 min). The cell pellet was suspended at a final concentration of 5 × 105 cells/ml in 0.25 M sucrose–0.1 M N-morpholinopropane sulfonic acid (MOPS) buffer, pH 7.5. The cells were then lysed by treatment with 0.2% Triton X-100 at 4°C for 1 h. In all assays, 25 μl of the samples (culture supernatant or cell lysate) was mixed with 25 μl of reaction mixture, which contained 100 mM MOPS (pH 7.5), 10 mM MnCl2, 2% Tween 20, 3 mM ATP, 1 mg of bovine serum albumin (BSA), 4 mM UDP-[14C]GlcNAc (1,200 dpm/nmol), and 4 mM Galβ1–3GalNAcαbenzyl (Bzl) or 4 mM GlcNAcβ1– 3GalNAcαparanitrophenol (PNP) or 4 mM GlcNAcβ1– 3GalβMe. After the reaction mixtures had been incubated at 37°C for 60 min, the reaction was terminated by the addition of 0.6 ml of 10 mM ZnCl2. The products were isolated either by the C18 cartridge extraction method (16) when Galβ1–3GalNAcαBzl and GlcNAcβ1–3GalNAcαPNP were used as the acceptors, or by anion-exchange chromatography (17) when GlcNAcβ1–3GalβMe was the acceptor. The radioactivity was determined by liquid scintillation spectrometry.
For structural analysis, the C2TF product was prepared using a total of 2 × 106 sf21 cells infected with recombinant baculovirus containing bC2TF insert. The cell pellet was suspended and lysed in 1 ml of buffer, which was then mixed with 1 ml of reaction mixture and incubated at 37°C for 4 h. The product was isolated on a C18 cartridge column and further purified two times by Bio-Gel P-4 (1.5 × 90 cm) gel-filtration chromatography. The structure of the purified sugar product was analyzed by 1H NMR spectrometry at the Complex Carbohydrate Research Center at the University of Georgia (Athens, GA).
The murine monoclonal antibodies against bC2TF were raised with recombinant bC2TF-polyhistidine fusion protein prepared in the E. coli X-press system. The bC2TF cDNA EcoRI digest, which is devoid of the 5′ 98-bp cDNA fragment corresponding to the N-terminal 33 amino acids, was cloned into the pRSET vector and expressed in E. coli. The recombinant bC2TF-polyhistidine fusion protein of 51 kD was purified on a metal ion affinity column and used as immunogen. Hybridomas that produced antibodies against bC2TF were identified by ELISA, using the recombinant bC2TF-polyhistidine fusion protein, and confirmed by immunofluorescence in sf21 cells infected with recombinant baculovirus containing the full-length bC2TF cDNA but no polyhistidine tag.
Western blotting analysis was performed as described previously (18) with some modification. The virus-infected sf21 cells were harvested from the culture suspension by brief centrifugation. The cell pellet and supernatant were directly mixed with the SDS– gel sample buffer. After boiling for 5 min, both supernatant and lysed cell samples were subjected to 10% polyacrylamide-SDS gel electrophoresis and electroblotted onto a 0.45-μm Immobilon-P membrane (Millipore, Bedford, MA). For the detection of the enzyme present in the tissue, bovine tracheal epithelium and submucosal gland samples were recovered by scraping the tracheal epithelium, which had been sliced 0.3–1 mm deep with a surgical scalpel. The samples were then treated with Laemlli sample buffer, electrophoresed, and blotted as described previously. The blotted membranes were blocked with 3% skim milk in phosphate-buffered saline (PBS) for 2 h at room temperature. The membrane was incubated first with mAb supernatant for 3 h and then with diluted peroxidase-conjugated goat anti-mouse IgG (diluted 1:10,000 with 3% skim milk in PBS) for 2 h. After each incubation, the membranes were washed three times with 0.2% Tween 20 in PBS. Finally, the membranes were visualized by the ECL method according to the protocol provided by the manufacturer.
For detection of recombinant bC2TF, the sf21 cells were infected with recombinant virus. Forty-eight hours after viral infection, aliquots of the infected cell suspensions were spread onto glass slides and air dried. The dried cells were fixed with cold acetone for 10 min and rinsed three times with PBS. The cells were then treated with bC2TF MAb followed by treatment with fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse Ig. Samples were examined under a Nikon fluorescence microscope. For detection of native bC2TF, 6-mm-thick frozen sections of bovine tracheas were used.
Screening a λgt10 cDNA library (approximately 2 × 105 recombinant phages) yielded two positive cDNA clones. PCR and restriction mapping showed that both clones contained identical cDNA inserts of about 2.2 kb with an internal EcoRI site. This insert was excised by partial EcoRI digestion and subcloned into the EcoRI site of the Bluescript II SK+ plasmid. Sequencing of this insert by the primer extension method as well as by the Exo III deletion method showed that the insert was 2,151 bp in length (Figure 1) and lacked about 220 bp of the 5′ end of the bC2TF open reading frame as predicted from human myeloid C2TF cDNA (11).
Fig. 1. Nucleotide sequence of bC2TF cDNA and the deduced amino acid sequence: The transmembrane domain of bC2TF is underlined. The asterisks at Asn-58 and Asn-95 denote the potential N-glycosylation sites. The sequence information from −65 to +220 bp was obtained from genomic DNA of bC2TF.
[More] [Minimize]Southern blotting analysis showed that only one band was present in all three restriction enzyme digests of bovine genomic DNA (Figure 2). This result suggested that the coding region of the bovine lung C2TF gene, like the ORF of the human myeloid C2TF gene (19), also is present in one exon. On the basis of this result, a bovine λEMBL3-SP6/T7 genomic DNA library was screened by using the 519-bp PCR fragment as the probe. One positive clone was obtained after screening 1.8 × 106 recombinant phages. BamHI digestion of this genomic clone generated four DNA fragments of 2.5, 3.3, 4.2, and 5.6 kb. These fragments were subcloned into the Bluescript II SK+ plasmid. The resulted recombinant plasmid clones were characterized by double digestion with HindIII and BamHI. The result suggests that the 3.3-kb BamHI fragment contains the 5′ portion of the ORF of the bovine lung C2TF gene.
Fig. 2. Southern blotting analysis of bovine genomic DNA using a 519-bp PCR fragment as the probe. Lane 1, λ phage HindIII-digested DNA marker; lanes 2–4, bovine genomic DNA digested by EcoRI, BamHI, and HindIII, respectively; lane 5, untreated bovine genomic DNA control. Only one band was identified in each digested DNA lane: lane 2, 1.3 kb; lane 3, 6.9 kb; lane 4, 3.4 kb.
[More] [Minimize]A hybrid DNA with the complete ORF of the bovine lung C2TF gene was constructed by ligating the genomic DNA fragment from the genomic DNA clone with the cloned cDNA fragment via the BamHI site (Figure 3). The nucleotide sequence of this hybrid DNA and its corresponding amino acid sequence are shown in Figure 1. The deduced amino acid sequence of this bC2TF protein consists of 427 aa and contains two potential N-glycosylation sites, Asn-X-Ser/Thr (X ≠ proline or asparagine) (11, 20), at Asn-58 and Asn-95. The calculated molecular mass of the apoprotein of this bC2TF is about 49.5 kD.
Figure 3 illustrates the strategy for the cloning and expression of the complete ORF of the bC2TF gene. To clone the DNA fragment that contained the complete coding sequence of bC2TF, an EcoRI site was created at −64 bp upstream of the translation start codon by site-directed mutagenesis. A DNA fragment that contained the complete ORF of the bC2TF gene was obtained by partial EcoRI digestion of the mutated DNA and cloned into the baculoviral transfer vector. After cotransfection of sf21 cells with this recombinant transfer vector and Bsu36I-digested baculovirus DNA, the resultant recombinant baculovirus was further amplified by repeated passage in the insect cells. The C2TF activity was detected in the infected cells after the third passage. The bC2TF activity in the infected cells reached 8 × 10−5 units/mg protein (1 unit = mmol/min at 37°C) at 48 h postinfection. No enzyme activity was detected in the culture medium. Analysis of the acceptor specificity showed that the recombinant bC2TF can transfer GlcNAc from UDP-GlcNAc only to core 1 disaccharide acceptor forming trisaccharide product, but not to core 3 and blood group i disaccharide acceptors. Analysis of the trisaccharide product by 1H NMR showed that the NMR spectrum was identical to that of Galβ3(GlcNAcβ6)GalNAcαBzl, core 2 structure (16, 21).
Expression of the partial bC2TF protein using recombinant 5′-minus-C2TF baculovirus (Figure 4) was demonstrated by immunoassay, but no C2TF activity could be detected in the conditioned medium or the infected cells.
Fig. 4. Analysis of the recombinant bovine C2TF by 10% polyacrylamide SDS gel electrophoresis as stained with Coomassie blue (A) and Western blot probed with bC2TF mAbs (B). Lane 1, protein molecular weight markers; lanes 2–4, lysates of sf21 cells infected with recombinant baculovirus containing full-length bC2TF sequence, recombinant baculovirus containing bC2TF cDNA devoid of 98 bp at the 5′ end, and wild-type baculovirus, respectively; lanes 5–7, conditioned media corresponding to lanes 2–4 of infected cells. No recombinant protein band was detected in a Coomassie blue-stained polyacrylamide gel (a heavy band in all three conditioned media was identified as bovine serum albumin in culture medium). The recombinant bC2TF (lane 2) and partial bC2TF (lane 3) were demonstrated by Western blotting assay.
[More] [Minimize]A mixture of mAbs 5, 12, 79, and 93 was used to identify the recombinant protein. Immunoreactive bands were demonstrated in recombinant virus-infected cells, but not in wild-type virus-infected cells. Culture supernatant recovered from both recombinant virus- and wild-type virus- infected cells also did not show any immunoreactivity (Figure 4). The recombinant bC2TF expressed by recombinant virus that contained the full-length bC2TF DNA insert had a molecular mass of 55 kD and the recombinant enzyme expressed by 5′-minus-bC2TF virus had a mass of 51 kD. These mAbs also detected a 55-kD protein band in the sample containing bovine tracheal epithelium and submucosal glands (Figure 5).
Fig. 5. Comparison of recombinant bC2TF with native bovine tracheal C2TF by Western blotting assay. Lane 1, full-length recombinant bC2TF produced in sf21 cells infected by baculovirus; lane 2, homogenate of bovine tracheal epithelium containing surface epithelium and submucosal glands. Both lanes have a common 55-kD band. A smaller, degraded bC2TF was seen in infected cells (lane 1).
[More] [Minimize]Immunofluorescence staining of sf21 cells showed intense fluorescence in cells infected by recombinant baculoviruses containing either the complete bC2TF ORF or the partial bC2TF ORF (missing 98 bp at the 5′ end). The recombinant enzyme was detected in the infected cells as early as 18 h after infection. The fluorescence was concentrated in some areas of the cytoplasm and on the cell surface (Figure 6A, part a). No fluorescence was observed in wild-type baculovirus-infected cells (Figure 6A, part b).
Fig. 6. Immunofluorescence analysis of bC2TF in (A) sf21 cells infected with baculovirus and (B) frozen sections of bovine trachea using a mixture of bC2TF mAbs 5 and 79. In (A), the fluorescence was observed in (a) sf21 cells infected with baculovirus containing recombinant full-length bC2TF DNA but not in (b) wild-type baculovirus-infected cells. In (B), the fluorescence was mainly present in (a) the submucosal glands (arrows) and in (b) a few isolated epithelial cells (arrowhead) of the surface epithelium. (c) and (d) are bright-field areas corresponding to (a) and (b), respectively. (c) and (d) show the typical tracheal submucosal glands and surface epithelium. Bar: 5 μm.
[More] [Minimize]Immunofluorescence staining of frozen sections of bovine trachea with mAbs 5 and 79 showed strong fluorescence in the submucosal glands (Figure 6B, part a) and in the surface epithelial layer. However, fluorescence was observed only in a few isolated cells in certain areas (Figure 6B, part b). In addition to trachea, this enzyme was also detected in the lung, bronchus, tongue, submaxillary gland, small intestine, and colon, but not in kidney, spleen, heart, liver, and muscle (data not shown).
We have cloned and constructed a bC2TF DNA that contains the complete coding sequence of this enzyme. By using a baculoviral expression system, an enzymatically active recombinant bC2TF was produced. The recombinant enzyme can convert core 1, Galβ3GalNAc, to core 2, Galβ3 (GlcNAcβ6)GalNAc, but not core 3 to core 4 or blood group i to I structures. In addition, the size and tissue localization of the native enzyme in bovine airways were demonstrated by using C2TF mAbs raised against recombinant bC2TF.
From the cloned bC2TF DNA, a polypeptide of 427 amino acids was deduced (Figure 1). Like most mammalian glycosyltransferases (13, 22, 23), bC2TF is a type II transmembrane protein containing a short cytoplasmic NH2-terminal segment (9 aa) followed by transmembrane, stem, and catalytic (carboxy-terminal) domains. There are two consensus N-glycosylation sites, Asn-X-Ser/Thr, at Asn-58 and Asn-95 in the stem region. The asparagine in the Asn(52)-Pro-Ser sequence is not likely to be glycosylated because glycosylation at asparagine has never been found when amino acid X is proline or asparagine (11, 20). The nonglycosylated form of this enzyme should have a molecular mass of about 49.5 kD, but Western blot analysis demonstrated that the actual size of both the recombinant (the slow-moving species) and native bC2TF is 55 kD (Figure 5). The faster moving species of the recombinant enzyme has an estimated molecular mass of 49–50 kD, consistent with the size of the unglycosylated enzyme. Such a difference in molecular mass may be attributed to the acquisition by these two bC2TFs of two complex-type N-linked carbohydrates during posttranslational modification. This modification may be important for preserving the enzymatic activity of bC2TF. For example, cloning of the complete coding sequence of this enzyme into an E. coli expression system produced a protein of expected size as identified by polyacrylamide gel electrophoresis (PAGE), but no core 2 enzyme activity could be detected in either the transformed bacteria or culture supernatant. These results suggest that posttranslational modification of bovine C2TF is crucial for preserving enzyme activity and that this modification may be achieved only in eukaryotic expression systems. The baculovirus expression system has been proven to be a useful eukaryotic expression system. In addition to its capability for expressing protein at a high level, preserving biologic activity, and ease of scaling up the production of recombinant protein, the insect cell expression system can also perform several posttranslational modifications, such as signal cleavage (24), proteolytic cleavage (25), N-glycosylation (26), and O-glycosylation (27). Many biologically active proteins have been expressed with this system (28), including α1,3-Gal-transferase (EC 2.4.1.124), β1,2Glc-NAc-transferase (EC 2.4.1.101), and peptidyl GalNAc-transferase (EC 2.4.1.41) (29-31).
Although there are many reports of successful heterologous expression of chimeric fusion glycosyltransferases in mammalian cell systems (11, 32, 33), our attempts to express catalytically active 5′-minus chimera fusion enzyme have failed. In this study, expressed protein of an expected size (51 kD) was present in the cells infected with recombinant virus containing 5′-minus bC2TF sequence, but no enzyme activity was found in these cells or their culture supernatant. In the 5′-minus bC2TF construct, only the N-terminal 33 amino acids, which corresponded to the cytoplasmic and transmembrane domains of the enzyme, were eliminated. Usually, these domains of glycosyltransferases are spaced far from the catalytic domain by a stem domain (60–120 aa in length) and are involved in targeting the enzyme to their destination, the Golgi complex (34). With 5′-minus recombinant virus, an inactive recombinant bC2TF protein of the expected size was detected inside the cells but not in the conditioned medium. These results suggest that (1) in addition to their role in targeting the enzyme to the Golgi complex, the cytoplasmic and transmembrane domains of bC2TF may also be involved in preserving the conformation of the catalytic domain of the enzyme. Deletion of these domains may cause steric change of stem and catalytic domains and lead to the loss of enzyme activity; and (2) elimination of both cytoplasmic and transmembrane domains of bC2TF does not cause the release of the soluble product from the cell because the cytoplasmic domain may be needed to guide the protein through various cellular membrane compartments. It was noted, however, that when the portion of the human C2TF cDNA that encodes the cytoplasmic and transmembrane domains plus the five adjacent amino acids of the stem region is replaced with a cDNA encoding a signal peptide and the IgG-binding domain of protein A, the fusion protein secreted into the medium retained C2TF activity (11). This result suggests that the signal peptide and the IgG-binding domain of protein A stabilize the C2TF activity, which raises the question of whether a specific amino sequence in this peptide fragment is responsible for this effect. This question remains to be answered.
Immunofluorescence localization of the recombinant bC2TF in sf21 cells infected with recombinant baculovirus containing full-length C2TF cDNA indicates that this enzyme is found not only intracellularly but also at the cell surface (Figure 6). Low-resolution light microscopy precludes identification of this enzyme in a specific intracellular organelle. But the uneven distribution of intracellular fluorescence intensity suggests that this recombinant enzyme is localized in certain intracellular organelles, presumably the Golgi apparatus. The intense fluorescence staining of the surface of the infected sf21 cells suggests that this enzyme may also localize at the cell surface. It has been well established that GlcNAc:β1–4 galactosyltransferase (EC 2.4.1.90) is found at the surface of sperm and plays an important role in the fertilization of eggs (35). The potential physiologic role of this cell surface bC2TF is unknown.
Searches of the Genetics Computer Group (GCG; University of Wisconsin, Madison, WI) database did not reveal much sequence similarity between bC2TF and known DNA or protein sequences, including those of previously cloned glycosyltransferases, except for the sequences of human (11, 19) and mouse muscle (12) C2TF (Figure 7). bC2TF is one amino acid shorter than hC2TF and mC2TF (428 aa). The sequence similarity between human myeloid C2TF and bovine lung C2TF is 86% in amino acid sequence, 84% in cDNA coding region (nucleotides 1–1284), and 74% in the 3′ untranslated region (nucleotides 1285–1888). This enzyme was found in many mucus secretory tissues, including tracheobronchial epithelium, gastrointestinal epithelium, colon, and submaxillary gland, but not in nonmucus secretory tissues, such as kidney, spleen, heart, and liver (data not shown). In bovine trachea, C2TF was present mainly in the submucosal glands although a few surface epithelial cells in certain areas were also detected, which suggests that bovine airway mucins containing core 2 structures are mainly synthesized in the glands. The distribution of this enzyme in the airways is consistent with the periodic acid–Schiff (PAS) staining pattern of the mucus cells in bovine trachea, in which only 1–2% of the surface epithelial cells were goblet cells, whereas submucosal glands contain abundant numbers of mucus cells.
Fig. 7. Comparison of amino acid sequences of bovine lung C2TF, human myeloid C2TF, and mouse muscle C2TF. Dots denote amino acids in the hC2TF and mC2TF sequences that are identical to those in the bC2TF sequence. The mismatched amino acids are indicated in hC2TF and mC2TF sequences. The underbar between aa 372 and 373 of bC2TF denotes no amino acid in that position, and therefore, bC2TF is one amino acid shorter than human and mouse C2TF.
[More] [Minimize]Previously, we reported the purification of a bovine tracheal epithelial β6-GlcNAc transferase (EC 2.4.1.148) to apparent homogeneity by UDP-GlcNAc affinity column and showed that the purified enzyme catalyzes has a molecular mass of 68 kD and the synthesis of three core structures: core 2, core 4, and blood group I (16). In this study, we tried to use bC2TF cDNA as a probe to clone the cDNA encoding this β6-GlcNAc transferase under various hybridization conditions. No clone of interest was detected. Western blotting analysis also showed only one C2TF protein band with a molecular mass of 55 kD in tracheal samples. These results suggest that C2TF and β6-GlcNAc transferase differ markedly at both the DNA and protein levels. Hence, alternative cloning strategies, such as screening the cDNA library constructed from bovine tracheal epithelium instead of whole lung or using amino acid sequence from the purified enzyme, may be employed in order to clone the cDNA that encodes the β6-GlcNAc transferase with a broad acceptor specificity (16).
The authors wish to thank Drs. M. F. A. Bierhuizen and M. Fukuda (University of California at San Diego) for providing the human C2 GlcNAc TF cDNA, Drs. Billie Moats-Staats and Brent Weston (University of North Carolina at Chapel Hill) for advice and assistance in cloning, and the Cystic Fibrosis Foundation (P-579) and the NIH (RO1 HL 48282) for grant support. The authors acknowledge that the 1H NMR analysis of the trisaccharide product was performed at the Complex Carbohydrate Research Center (at the University of Georgia), which was funded by NIH Grant No. 2-P41-RR05351-06. The nucleotide sequence reported in this article has been submitted to the GenBank/EMBL Data Bank with accession number U41320.
| 1. | Boat, T. F., P.-W. Cheng, and M. Leigh. 1993. In Biochemistry of Mucus in Airway Secretions. T. Takashima and S. Shimura, editors. Marcel Dekker, New York. 217–282. |
| 2. | Rademacher T. W., Parekh R. B., Dwek R. A.Glycobiology. Annu. Rev. Biochem.571988785838 |
| 3. | Olden K., Bernard B. A., Humphries M. J., Yeo T. K., Yeo F. T., White S. L., Parent J. B.Function of glycoprotein glycans. Trends Biochem. Sci.1019857882 |
| 4. | Strous G. J., Dekker J.Mucin-type glycoprotein. Crit. Rev. Biochem. Mol. Biol.2719925759 |
| 5. | Wingert W., Cheng P.-W.Mucin biosynthesis: characterization of rabbit small intestinal UDP-GlcNAc: galactoseβ3N-acetylgalactosamine (GlcNAc-GalNAc) β6 N-acetylglucosaminyltransferase. Biochemistry231984690697 |
| 6. | Piller F., Piller V., Fox R. I., Fukuda M.T lymphocytic leukemia expresses complex, branched O-linked oligosaccharides on a major sialoglycoprotein leukosialin. J. Biol. Chem.26319881514615150 |
| 7. | Carlson D. M.Structures and immunochemical properties of oligosaccharides isolated from pig submaxillary mucins. J. Biol. Chem.2431968616626 |
| 8. | Fukuda M., Carlson S. R., Klock J. C., Dell A.Structure of O-linked oligosaccharides isolated from normal granulocytes, chronic myelogenous leukemia cells and acute myelogenous leukemia cells. J. Biol. Chem.26119891279612806 |
| 9. | Piller F., Le Deist F., Weinberg K. I., Parkman R., Fukuda M.Altered O-glycan synthesis in lymphocytes from patient with Wiskott– Aldrich syndrome. J. Exp. Med.173199115011510 |
| 10. | Yousefi S., Higgins E., Daoling Z., Pollex-Kruger A., Hindsgaul O., Dennis J. W.Increased UDP-GlcNAc:Galβ1–3GalNAc-R (GlcNAc to GalNAc)β1–6-N-acetylglucosaminyltransferase activity in metastatic murine tumor cell lines. J. Biol. Chem.266199117721782 |
| 11. | Bierhuizen M. F. A., Fukuda M.Expression cloning of a cDNA encoding UDP-GlcNAc:Galβ1–3-GalNAc-R (GlcNAc to GalNAc) β1– 6GlcNAc transferase by gene transfer into CHO cells expressing polyoma large tumor antigen. Proc. Natl. Acad. Sci. USA89199293269330 |
| 12. | Warren, C. E., D. S. Smookler, and J. W. Dennis. 1995. UDP-GlcNAc:Galβ1– 3GalNAc-R (GlcNAc to GalNAc) β1–6GlcNAc transferase (core 2-GnT) mouse cDNA sequence. GenBank Accession No. U19265 |
| 13. | Bierhuizen M. F. A., Mattei M. G., Fukuda M.Expression of the developmental I antigen by a cloned human cDNA encoding a member of a β-1,6-N-acetylglucosaminyltransferase gene family. Genes Dev.71993468474 |
| 14. | Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. |
| 15. | Sanger F., Nockln S., Coulson A. R.DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA74197754635467 |
| 16. | Ropp P. A., Little M. R., Cheng P.-W.Mucin biosynthesis: purification and characterization of a mucin β6 N-acetylglucosaminyltransferase. J. Biol. Chem.26619912386323871 |
| 17. | Brockhausen I., Matta K. L., Orr J., Schachter H., Koenderman A., Van Den Eijden D.Mucin synthesis: conversion of R1-β1–3Gal-R2 to β1–3(GlcNAcβ1–6)Gal-R2 and of R1β1–3GalNAc-R2 to R1β1–3 (GlcNAc β1–6) GalNAc-R2 by a β6-N-acetylglucosaminyltransferase in pig gastric mucosa. Eur. J. Biochem.1571986463471 |
| 18. | Li C. M., Cheng P.-W., Adler K. B.Production and characterization of monoclonal antibodies against guinea pig tracheal mucins. Hybridoma131993281287 |
| 19. | Bierhuizen M. F. A., Maemura K., Kudo S., Fukuda M.Genomic organization of core 2 and I branching β-1,6-N-acetylglucosaminyltransferases: implication for evolution of the β-1,6-N-acetylglucosaminyltransferase gene family. Glycobiology51995417425 |
| 20. | Voet, D., and J. G. Voet. 1990. Biochemistry. John Wiley & Sons, New York. 264. |
| 21. | Cheng P.-W., Wingert W., Little M., Wei R.Mucin biosynthesis: properties of a bovine tracheal mucin β6-GlcNAc transferase. Biochem. J.2271985405412 |
| 22. | Kleene R., Berger E. G.The molecular and cell biology of glycosyltransferases. Biochim. Biophys. Acta11541993283325 |
| 23. | Paulson J. C., Colley K. J.Glycosyltransferases: structure, localization, and control of cell type-specific glycosylation. J. Biol. Chem.26419891761517618 |
| 24. | Bustos M. M., Luckow V. A., Griffing L. R., Summers M. D., Hall T. C.Expression, glycosylation, and secretion of phaseolin in a baculovirus system. Plant Mol. Biol.101988475488 |
| 25. | Lebacq A. M., Kasprzyk P. G., Raum M. G., Van Wyke K., Lebacq J. A., Battey J. F.Posttranslational processing of endogenous and baculovirus-expressed human gastric-releasing peptide precursor. Mol. Cell Biol.8198831293135 |
| 26. | Kornfeld R., Kornfeld S.Assembly of asparagine-linked oligosaccharides. Annu. Rev. Biochem.541985631664 |
| 27. | Thomsen D. R., Post L. E., Elhammer A. P.Structure of O-glycosidically linked oligosaccharides synthesized by insect cell line Sf 9. J. Cell. Biochem.4319906779 |
| 28. | O'Reilly, D. R., L. K. Miller, and V. A. Luckow. 1994. Baculovirus Expression Vectors—A Laboratory Manual. Oxford University Press, New York. 295. |
| 29. | Joziasse D. H., Shaper N. L., Salyer L. S., Van Den Eijnden D. H., Van der Spoel A. C., Shaper J. H.α1–3 Galactosyltransferase: the use of recombinant enzyme for synthesis of α-galactosylated glycoconjugates. Eur. J. Biochem.19119907583 |
| 30. | Sarkar M., Schachter H.Expression of UDP-GlcNAc:α3-d-mannoside β1,2 GlcNAc TF I in insect cells. Glycobiology21992483491 |
| 31. | Homa F. L., Hollander T., Lehman D. J., Thomsen D. R., Elhammer A. P.Isolation and expression of a cDNA clone encoding a bovine UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase. J. Biol. Chem.26819931260912616 |
| 32. | Potvin B., Kumar R., Howard D. R., Stanley P.Transfection of a human α-(1,3) fucosyltransferase gene into chinese hamster ovary cells. J. Biol. Chem.265199016151622 |
| 33. | Weston B. W., Nair R. P., Larsen R. D., Lowe J. B.Isolation of a novel human α(1,3) fucosyltransferase gene and molecular comparison to the human Lewis blood group α(1,3/1,4) fucosyltransferase gene. J. Biol. Chem.267199241524160 |
| 34. | Shaper J. H., Shaper N. L.Enzymes associated with glycosylation. Curr. Opin. Struct. Biol.21992701709 |
| 35. | Shur B. D.Cell surface β1,4 galactosyltransferase: twenty years later. Glycobiology11991563575 |
