Glycobiology Advance Access originally published online on June 8, 2006
Glycobiology 2006 16(10):947-958; doi:10.1093/glycob/cwl008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Identification of core 1 O-glycan T-synthase from Caenorhabditis elegans
Department of Biochemistry and Molecular Biology, Oklahoma Center for Medical Glycobiology, University of Oklahoma Health Sciences Center, 975 N.E. 10th Street, BRC Rm. 417, Oklahoma City, OK 73104
1 To whom correspondence should be addressed; e-mail: richard-cummings{at}ouhsc.edu
Received on February 27, 2006; revised on May 25, 2006; accepted on May 26, 2006
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
The common O-glycan core structure in animal glycoproteins is the core 1 disaccharide Galß1-3GalNAc
1-Ser/Thr, which is generated by the addition of Gal to GalNAc1-Ser/Thr by core 1 UDP--galactose (UDP-Gal):GalNAc1-Ser/Thr ß1,3-galactosyltransferase (core 1 ß3-Gal-T or T-synthase, EC2.4.1.122). Although O-glycans play important roles in vertebrates, much remains to be learned from model organisms such as the free-living nematode Caenorhabditis elegans, which offer many advantages in exploring O-glycan structure/function. Here, we report the cloning and enzymatic characterization of T-synthase from C. elegans (Ce-T-synthase). A putative C. elegans gene for T-synthase, C38H2.2, was identified in GenBank by a BlastP search using the human T-synthase protein sequence. The full-length cDNA for Ce-T-synthase, which was generated by polymerase chain reaction using a C. elegans cDNA library as the template, contains 1170 bp including the stop TAA. The cDNA encodes a protein of 389 amino acids with typical type II membrane topology and a remarkable 42.7% identity to the human T-synthase. Ce-T-synthase has seven Cys residues in the lumenal domain including six conserved Cys residues in all orthologs. The Ce-T-synthase has four potential N-glycosylation sequons, whereas the mammalian orthologs lack N-glycosylation sequons. Only one gene for Ce-T-synthase was identified in the genome-wide search, and it contains eight exons. Promoter analysis of the Ce-T-synthase using green fluorescent protein (GFP) constructs shows that the gene is expressed at all developmental stages and appears to be in all cells. Unexpectedly, only minimal activity was recovered in the recombinant, soluble Ce-T-synthase secreted from a wide variety of mammalian cell lines, whereas robust enzyme activity was recovered in the soluble Ce-T-synthase expressed in Hi-5 insect cells. Vertebrate T-synthase requires the molecular chaperone Cosmc, but our results show that Ce-T-synthase does not require Cosmc and might require invertebrate-specific factors for the formation of the optimally active enzyme. These results show that the Ce-T-synthase is a functional ortholog to the human T-synthase in generating core 1 O-glycans and open new avenues to explore O-glycan function in this model organism. Key words: Caenorhabditis elegans / cloning / core 1 O-glycan / galactosyltransferase / T-synthase
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
O-Glycans are common posttranslational modifications of glycoproteins (Brockhausen, Schutzbach, et al., 1998
; Schachter, 2000; Lamblin et al., 2001; Ten Hagen et al., 2003). The core 1 structure Galß1-3GalNAc1-Ser/Thr is the most common of such O-glycans and is a precursor to more complex O-glycans such as extended core 1 and core 2 structures. The core 1 structure is synthesized by the addition of galactose from UDP--galactose (UDP-Gal) to GalNAc1-Ser/Thr, a reaction catalyzed by the enzyme core 1 UDP-Gal:GalNAc1-Ser/Thr ß1,3-galactosyltransferase (core 1 ß3-Gal-T or T-synthase, EC2.4.1.122) (Ju, Brewer, et al., 2002; Ju, Cummings, et al., 2002). Such core 1-derived O-glycans have many important biological roles, including key roles in leukocyte inflammation (McEver and Cummings, 1997), lymphocyte homing (Tsuboi and Fukuda, 2001; Lowe and Marth, 2003; Rosen, 2004), and animal development (Ten Hagen and Tran, 2002; Xia et al., 2004). In addition, O-glycans can serve as ligands for adhesion by many microorganisms (Carlyon et al., 2003; Yago et al., 2003). The reduced expression of T-synthase results in the expression of the abnormal truncated GalNAc1-Ser/Thr structure known as the Tn antigen, which is associated with several human autoimmune diseases including Tn syndrome (Ju and Cummings, 2005), IgA nephropathy, Henoch-Schönlein purpura (Julian and Novak, 2004), and progressive properties of some tumors such as colon and breast carcinomas (Springer, 1997).
The T-synthase was initially purified from rat liver, leading to the cloning of the cDNA and gene for T-synthase from humans, rats, mice, and Caenorhabditis elegans (Ju, Brewer, et al., 2002; Ju, Cummings, et al., 2002). The cDNA for T-synthase in mammals encodes a 363-amino acid transmembrane protein with type II topology. Unlike most other glycosyltransferases, mammalian T-synthase lacks potential N-glycosylation sequons. Unexpectedly, we found that the expression of active vertebrate T-synthase in animal cells requires a unique molecular chaperone, Cosmc, that promotes correct folding and the acquisition of the activity of the T-synthase (Ju and Cummings, 2002, 2005). However, no ortholog of Cosmc is found in the known genomes of several invertebrates such as C. elegans and Drosophila.
The nonparasitic free-living nematode C. elegans is a useful model system to study glycan functions (Altmann et al., 2001; Fan et al., 2005), because many of the glycosyltransferase families in humans are represented in the nematode (DeBose-Boyd et al., 1998; Oriol et al., 1999; Kawar et al., 2002; Zheng et al., 2002; Sarkar et al., 2006). Several studies have now identified common N- and O-glycans of C. elegans and shown that they have many features in common with vertebrate glycans, especially in terms of core glycan biosynthesis (Guerardel et al., 2001; Cipollo et al., 2002; Haslam et al., 2002; Natsuka et al., 2002; Haslam and Dell, 2003; Zhu et al., 2004; Cipollo et al., 2005). However, the terminal glycan structures in C. elegans appear to diverge in many respects from vertebrate glycans. For example, C. elegans glycans lack sialic acid and contain unusual additions of fucose and xylose, along with methylated fucose residues (Haslam and Dell, 2003; Cipollo et al., 2005).
However, some O-glycans from C. elegans contain a typical core 1 structure (Guerardel et al., 2001), in addition to other highly unusual O-glycan core structures and modifications. These results imply that C. elegans contains the common enzymatic machinery to synthesize GalNAc1-Ser/Thr from the family of polypeptide GalNAc-transferases and Galß1-3GalNAc1-Ser/Thr from a putative T-synthase activity. Here, we describe our studies on the C. elegans T-synthase and demonstrate that the organism contains a single gene encoding the T-synthase. Although the Ce-T-synthase is similar to the mammalian and vertebrate T-synthases in generating the core 1 O-glycan, the Ce-T-synthase does not require the molecular chaperone Cosmc that was identified for the mammalian and vertebrate T-synthases and may utilize a protein-folding pathway unique to the T-synthase from invertebrate cells.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Isolation of the cDNA and gene for Ce-T-synthase
A putative C. elegans gene, C38H2.2, for T-synthase was identified in GenBank by a BlastP search using the human T-synthase protein sequence. Using the information from the gene sequence and an expressed sequence tag (EST) clone in the C. elegans database, the full-length cDNA for Ce-T-synthase was generated by polymerase chain reaction (PCR) using a C. elegans cDNA library as the template. The full-length cDNA contains 1170 bp including the stop TAA. Sequencing of the PCR products revealed, however, that although the 3' end perfectly matched the cDNA predicted by C38H2.2, the 5' end only matched partially. Using this 5'-end-matched sequence of this PCR product, we identified a C. elegans EST clone (C41813 [GenBank] ) by a BlastN dbEST search. We compared the sequences of the PCR product from the 5' end of cDNA that was predicted to be encoded by C38H2.2 and the EST clone C41813 [GenBank] , which confirmed that the 5' end of the cDNA predicted in the BlastP search from C38H2.2 is incorrect. We then redesigned primers based on the EST clone C41813 [GenBank] sequence, and using these primers, we performed PCR with the cDNA library of C. elegans. This generated a 620-bp product. The sequence analyses of the EST clone C41813 [GenBank] and this PCR product, along with the sequence of the 3'-half PCR product, gave the sequence of the full-length cDNA for the putative Ce-T-synthase (Figure 1A).
|
The cDNA for this putative Ce-T-synthase predicts a protein of 389 amino acids. The hydropathy plot (Figure 1B) predicts that the protein is a type II membrane protein with a short N-terminal cytoplasmic domain of six amino acids (or five residues minus Met), a transmembrane domain (TM) of 24 amino acids, and the C-terminal stem and catalytic domains comprising 359 amino acids. The protein has four predicted N-glycosylation sites at Asn residues 84, 85, 152, and 360. This is in contrast to the mammalian T-synthases that have no predicted sites of N-glycosylation. The Ce-T-synthase gene is contained in cosmid C38H2.2, which encodes the only gene we found in the C. elegans database, with a significant homology to mammalian T-synthase. The gene is composed of eight exons, as shown in Figure 1C.
The putative Ce-T-synthase has significant homology to both vertebrate and invertebrate T-synthases (Figure 2). There is a remarkable 42.7% identity to the human T-synthase. All T-synthases from these different species share very high similarity or are very conserved. The most conserved region is located between 80 and 330 amino acids (amino acid number of human T-synthase) and between 110 and 350 amino acids (Ce-T-synthase number). All T-synthases share a DDD motif at the middle of the protein and a CCSD sequence close to the C-terminus. As indicated above, although none of the mammalian T-synthases contain N-glycosylation sequons, a highly active T-synthase from Drosophila (CG9520) was recently reported (Muller et al., 2005), and it has a single potential N-glycosylation site. Zebrafish T-synthase contains a single N-glycosylation site very close to its C-terminal, whereas Xenopus has three potential N-glycosylation sites, two of them in the central region of the primary sequence and one at the C-terminus. In comparing the potential N-glycosylation sites in T-synthases from invertebrates to lower vertebrates, however, we did not observe any site conservation. Thus, a common theme among the mammalian T-synthases appears to be their lack of N-glycosylation sites, whereas T-synthases from invertebrates contain potential sites in nonhomologous positions.
|
Expression of wild-type and soluble, N-terminal HPC4-epitope-tagged Ce-T-synthase
To confirm that the Ce-T-synthase is an authentic core 1 UDP-Gal:GalNAc1-Ser/Thr ß1,3-galactosyltransferase, we prepared a recombinant full-length form of the protein. A baculovirus vector was used to express the recombinant protein in infected Hi-5 insect cells. Cell extracts expressing the full-length protein had high T-synthase activity toward the acceptor GalNAc1-O-phenyl compared with uninfected control cells (Figure 3A). We also prepared a soluble form of the Ce-T-synthase lacking the cytoplasmic domain and TM at the N-terminus and instead containing the HPC4-epitope tag. This was constructed in a vector containing a signal sequence of the human transferrin to promote the secretion of the recombinant enzyme. Culture media Hi-5 cells from cells expressing the soluble, HPC4-tagged Ce-T-synthase had high levels of activity compared with control media from uninfected cells, and the activity was captured by the immobilized monoclonal antibody HPC4 to the peptide epitope (Figure 3B). These studies indicate that the Ce-T-synthase can transfer Gal from UDP-Gal to the acceptor GalNAc1-O-phenyl.
|
Product characterization by O-glycanase digestion
To confirm that the activity of the Ce-T-synthase is able to generate the core 1 O-glycan, we tested the recombinant soluble Ce-T-synthase, along with human recombinant T-synthase, against the acceptor glycopeptide shown in Figure 4. This glycopeptide is based on the N-terminal sequence from human PSGL-1 and contains a single residue of GalNAc linked to Thr. The incubation of this glycopeptide with both human and Ce-T-synthase and UDP-[3H]-Gal resulted in significant product formation, and the products were entirely cleaved by O-glycanase (endo -N-acetylgalactosaminidase), an enzyme specific to the core 1 disaccharide Galß1-3GalNAc1-Ser/Thr (Endo and Kobata, 1976). These results demonstrate that the Ce-T-synthase is an authentic ß1,3-galactosyltransferase similar to the human T-synthase and is capable of synthesizing the core 1 O-glycan.
|
Characterization of Ce-T-synthase expressed in different systems
In earlier studies, we found that active recombinant human T-synthase could not be recovered following expression in Hi-5 insect cells, and active enzyme could only be recovered by expression in mammalian cells. We subsequently discovered that mammalian cells have a unique molecular chaperone, which we named Cosmc, which is specifically required for folding and the acquisition of activity of the mammalian T-synthase (Ju and Cummings, 2002). Mutations in the Cosmc gene result in the loss of T-synthase activity, as seen for patients with Tn syndrome (Ju and Cummings, 2005). We therefore tested the expression of the Ce-T-synthase in mammalian cell lines compared with Hi-5 insect cells. For these studies, we used human 293T cells, clonal tumor cell lines LSC and LSB that were derived from LS174T human colonic cancer cells, and wild-type Chinese hamster ovary (CHO K1) cells, along with selected mutant CHO cells having altered glycosylation—CHO Lec1, Lec2, and Lec8 (Figure 5). Human 293T cells are known to be capable of generating recombinant human T-synthase, whereas the LSC cell line lacks T-synthase activity (Brockhausen, Yang, et al., 1998) because of a mutation in the Cosmc gene (mistakenly termed C1Gal-T2) (Kudo et al., 2002) and is unable to support the production of active T-synthase. By contrast, the LSB cell line has normal T-synthase activity (Brockhausen, Yang, et al., 1998) and normal Cosmc and supports the production of active T-synthase. We utilized the Lec mutants of CHO cells, because it is known that insect cells generate glycoproteins with truncated N-glycans, usually short oligomannose-type chains (Ailor and Betenbaugh, 1999). Truncated N-glycans are produced by these Lec mutant cell lines. Lec1 cells lack N-acetylglucosaminyl transferase I and generate only oligomannose-type N-glycans (Stanley et al., 1975). Lec2 cells lack the transporter for cytidine monophosphate (CMP)-sialic acid and consequently synthesize complex- and hybrid-type N- and O-glycans lacking sialic acid (Stanley et al., 1980; Deutscher et al., 1984). Lec8 cells lack the transporter for UDP-Gal and consequently synthesize complex- and hybrid-type N- and O-glycans lacking galactose and sialic acid (Deutscher and Hirschberg, 1986).
|
In all cell types, recombinant Ce-T-synthase was expressed and secreted into media, as determined by western blotting (Figure 5), and the recombinant protein was in the expected range of 45–50 kDa. To test whether the Ce-T-synthase was N-glycosylated and to assess the types of N-glycans generated, we treated the recombinant Ce-T-synthase with either N-glycanase or endo N-acetylglucosaminidase H (endo H). Endo H cleaves only oligomannose- and hybrid-type N-glycans (Tarentino et al., 1978), whereas N-glycanase can cleave all N-glycans in mammalian cells (Elder and Alexander, 1982).
Unexpectedly, very high levels of enzyme activity were recovered in recombinant Ce-T-synthase from Hi-5 insect cells, but only very low levels of activity were obtained from the other cell lines, even after normalizing by the amounts of proteins observed in western blots. An exception was the enzyme from 293T cells, which showed a similar if not a higher level of expression of T-synthase compared with Hi-5 cells and exhibited only 
20% of the activity compared with the enzyme from Hi-5 cells (Figure 5). The enzyme from Hi-5 cells had only small-sized oligomannose-type chains, because there was only a change of <2 kDa upon deglycosylation by either endo H or N-glycanase. By contrast, the enzyme from 293T cells and CHO cells was largely resistant to endo H and clearly sensitive to N-glycanase (Figure 5), suggesting that the enzyme from those cells acquired complex-type N-glycans.
We reasoned that the low level of enzyme activity in Ce-T-synthase expressed in mammalian cells might be due to the abnormal N-glycosylation and presence of complex-type N-glycans that might not be normally found on the C. elegans glycoproteins. For instance, it has been shown that most N-glycans from C. elegans glycoproteins contain various glycans, ranging from large-size to truncated oligomannose-type structures (Cipollo et al., 2002, 2005; Natsuka et al., 2002), the latter of which is related to those identified in insect cells. However, the recombinant Ce-T-synthase expressed in Lec mutants lacking sialic acid and/or galactose and generating oligomannose-type N-glycans did not have appreciable activity and was comparable with the low activity obtained in the enzyme from wild-type CHO cells. The sensitivity of the enzyme from Lec mutants to endoglycosidases was consistent with the expected glycosylation status of the enzyme, as described above.
The results suggest that the Ce-T-synthase can be efficiently expressed in insect cells as well as in mammalian cell lines, but those mammalian cell lines do not support the acquisition of activity of the enzyme. The opposite is true for the human T-synthase, which cannot be expressed in insect cells but can be expressed actively in mammalian cells lines that have the functional specific chaperone Cosmc. Thus, it is possible that insect cells have an unknown chaperone that may facilitate folding and the acquisition of activity of the Ce-T-synthase, but mammalian cells lack this invertebrate chaperone. It is also possible that the Ce-T-synthase is posttranslationally modified in some unknown manner in Hi-5 insect cells, which promotes activity, whereas mammalian cells might not add this modification. Further studies are underway to address this phenomenon.
Ce-T-synthase promoter analysis and developmental expression
To define the expression of the Ce-T-synthase, we generated reporter constructs driving the expression of the green fluorescent protein (GFP) in the vector pPD95.67/Ce-T-syn-prom. This construct contained the entire 3.2 kb region immediately upstream to 5' region of the gene. The expression of this construct in transgenic worms was straightforward, and all adult worms and larvae were brightly fluorescent, and all cells appeared to fluoresce, indicating that the promoter is active in all cells at all stages (Figure 6A–F). Particularly, high apparent expression was observed in intestinal muscle and rectal gland cells. To confirm the expression of the Ce-T-synthase during development, we performed quantitative RT–PCR to analyze the mRNA expression level at the different stages (Figure 6G). The results indicated that there is an equivalent expression of the Ce-T-synthase gene at all developmental stages tested. In this regard, the results are similar to mammalian T-synthase, which was found to be expressed in all cells and tissues during development to adulthood (Ju, Brewer, et al., 2002; Xia et al., 2004).
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
The results presented here provide several new insights into the biosynthesis of animal cell glycoproteins. We have identified the core 1 ß3-galactosyltransferase (T-synthase) from C. elegans that is capable of utilizing UDP-Gal as the donor for the transfer of Gal residues to GalNAc
-1-O-phenyl and GalNAc1-Ser/Thr for synthesizing core 1 structure Galß1-3GalNAc1-R. The enzyme has 43% identity to human T-synthase, with most of the homology concentrated in the putative C-terminal catalytic domain. Thus, all available data indicate that the Ce-T-synthase reported in here is an authentic core 1 ß3-galactosyltransferase (T-synthase) with analogous activity to the human T-synthase. Like mammalian T-synthases, the Ce-T-synthase also contains a DDD motif. More interestingly, all T-synthases from different species contain a CCSD sequence, which may be a hallmark for T-synthase or T-synthase-related proteins, including Cosmc (Ju and Cummings, 2002). All vertebrate and invertebrate T-synthases contain six conserved Cys in the catalytic domain, whereas Ce-T-synthase, along with other invertebrate T-synthases (Figure 2), contains an additional conserved Cys residue in the catalytic domain. Interestingly, the Ce-T-synthase lacks two Cys residues in its TM compared with the mammalian enzymes, which all appear to contain these two conserved Cys residues. The Ce-T-synthase behaves as a homodimer in nonreducing SDS–PAGE and as a monomer upon reduction (data not shown), similar to the mammalian T-synthase, indicating that disulfides in the catalytic domain may contribute to covalent dimerization. This is consistent with the observation that the soluble form of the Ce-T-synthase also behaved as a homodimer in nonreducing SDS–PAGE (data not shown).
Interestingly, unlike most other glycosyltransferases, which occur in families, such as the polypeptide GalNAc-transferase family with over a dozen family members (Ten Hagen et al., 2003) and the 1,2FT family with over 20 family members (Oriol et al., 1999; Javaud et al., 2003), only a single gene-coding T-synthase was found in C. elegans. This is similar to the situation in mammals, where there is a single T-synthase gene (Ju, Brewer, et al., 2002; Xia et al., 2004). Although there are eight sequences with 1520% homology to Ce-T-synthase found in WormBase, all of them lack the CCSD motif, which is a unique sequence conserved in all T-synthases from different species. Thus, until recombinant forms of these potential homologs are produced and analyzed for enzyme activity, it is not possible to conclude whether they are functionally related to the Ce-T-synthase. It is also possible that some of these homologs might function as molecular chaperones for the Ce-T-synthase, much in the same way that human Cosmc, which has about 15% homology to human T-synthase, functions as a chaperone for the T-synthase (Ju and Cummings, 2002, 2005). For example, although four apparent T-synthase genes have been reported in Drosophila related to the human T-synthase (Muller et al., 2005), it appears that only one gene product has high levels of T-synthase activity. Interestingly, we recently discovered there are three pseudogenes for human T-synthase (unpublished observation). Our results also indicate that the Ce-T-synthase is expressed at all developmental stages of C. elegans and in all cells, with particularly high apparent expression in intestinal muscle and rectal gland cells. We reported earlier that mammalian T-synthase is expressed in all cells and tissues (Ju, Brewer, et al., 2002; Ju, Cummings, et al., 2002) and that null mice lacking the T-synthase lack extended O-glycans and die at embryonic day 14 (Xia et al., 2004).
Mammalian T-synthase expressed in insect cells is not active, because it requires the specific molecular chaperone Cosmc for folding of the protein and the acquisition of activity (Ju and Cummings, 2002, 2005). Interestingly, we were unable to identify orthologs of Cosmc in invertebrates such as C. elegans and Drosophila. Thus, it is especially interesting that Ce-T-synthase expressed in insect cells is highly active, whereas it is poorly active when expressed in mammalian cells. There are many possible reasons for this divergence in expression and activity. The Ce-T-synthase is N-glycosylated and may undergo protein folding by interactions with molecular chaperones, such as calnexin/calreticulin, ERp57 (Dejgaard et al., 2004; Helenius and Aebi, 2004; Bedard et al., 2005), that recognize N-glycans and perhaps specific chaperones that are only expressed in invertebrate cells. Also, the least homologous regions between the Ce-T-synthase and mammalian T-synthases are the cytoplasmic domain and the so-called stem region between the TM and the catalytic domains. The stem region may influence the protein-folding process or may require different folding processes, as has been found for a ß4-galactosyltransferase (Boeggeman et al., 2003). It will be interesting to study this in the future by generating chimeras between the mammalian and C. elegans T-synthases. An interesting aspect of our study is that the Ce-T-synthase can be expressed at high levels directly in baculovirus-infected Hi-5 cells, thus making this enzyme more practical for large-scale enzymatic studies and biosynthesis of O-glycans requiring recombinant T-synthase. By contrast, the large-scale use of recombinant human T-synthase is problematic, because its production in insect cells requires co-expression with Cosmc, and even in mammalian cells, Cosmc may be rate limiting for efficient T-synthase production (Ju and Cummings, 2002).
Our results indicate that the Ce-T-synthase is expressed ubiquitously in the nematode, which suggests that core 1 O-glycans may be present in most tissues in C. elegans. Indeed, it has been shown that C. elegans glycoproteins contain core 1 type O-glycan structures, but in addition, they contain other many unusual O-glycan modifications, including the potentially novel O-glycan GlcAß1-3GalNAc-ol (Guerardel et al., 2001). Thus, although core 1 O-glycans may contribute to the overall O-glycan structures in C. elegans, there are clearly many complex O-glycans produced by this organism. In relation to this, in preliminary studies, we attempted to explore the function of the Ce-T-synthase by RNA interference (RNAi) approaches. However, we did not observe a phenotype generated by RNAi to the Ce-T-synthase (data not shown), whereas in control worms treated with RNAi to Talin (Cram et al. 2003), we observed visible defects of locomotion, and worms showed the characteristic "zig-zag" shape compared with the normal locomotion of worms in a smooth "S"-shape path (data not shown). These observations are consistent with the lack of a phenotype for Ce-T-synthase by RNAi experiments as reported in WormBase http://wormbase.org/. Among the many possibilities, it is conceivable that (1) RNAi is not efficient in suppressing the expression of the Ce-T-synthase, and a small amount of residual activity is sufficient; (2) another active core 1 ß3-Gal-T may exist in C. elegans; or (3) the deficiency of core 1 O-glycans may be compensated for by other O-glycosylation pathways in C. elegans. It should be noted again that the gene-encoding Ce-T-syn is unique, and no homologs were found in the C. elegans genome, although hypothetically it is possible that an unrelated gene might encode an enzyme with an activity related to the Ce-T-synthase and capable of generating core 1 O-glycans. Future studies will be aimed at defining the O-glycan structures in worms treated with RNAi to the Ce-T-synthase or worms having deletions in the Ce-T-syn gene to define the effects of deficiency of T-synthase activity on O-glycan structures.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Materials
All chemicals and reagents used in this study, unless otherwise indicated, were from Sigma-Aldrich (St. Louis, MO). The C. elegans cDNA library was a gift from Dr. Robert Barstead (Oklahoma Medical Research Foundation, Oklahoma City, OK). The QIA Quick gel extraction kit and plasmid miniprep kit were from Qiagen (Valencia, CA). Restriction enzymes were from New England Biolabs (Beverly, MA). The pCR2.1 vector was from Invitrogen (Carlsbad, CA). The pcDNA3.1(+)-TH was a gift from Dr. Alireza R. Rezaie (Department of Biochemistry and Molecular Biology, St. Louis University School of Medicine, St. Louis, MO). FuGENE 6 and Complete Protease Inhibitor Cocktail were from Roche (Indianapolis, IN). BaculoGold transfection kit was purchased from BD Biosciences (San Diego, CA). UDP-Gal was purchased from Calbiochem (San Diego, CA). UDP-Gal[6-3H] 20 Ci/mmol was purchased from American Radiolabeled Chemicals (St. Louis, MO). N-Glycanase and endo H were obtained from Glyko (Novato, CA). Purified O-glycanase from Diplococcus pneumoniae was obtained from Roche Molecular Biochemicals. HighSignal West Pico Chemiluminescent Substrate was from Pierce (Rockford, IL). Radiolabeled nucleotide sugars were obtained from New England Nuclear (Boston, MA) and were diluted with unlabeled nucleotide sugars (Sigma-Aldrich) to give the desired specific radioactivity. Total RNA isolation kit was purchased from Ambion (Austin, TX). Taq DNA polymerase and other PCR components were obtained form Boehringer Mannheim (Indianapolis, IN).
Cloning and sequencing of Ce-T-synthase
The cDNA of the C. elegans core 1 ß3-Gal-T (Ce-T-synthase) was generated by PCR of 5' and 3' portions of Ce-T-synthase using a C. elegans cDNA library as the template. BlastP search using the human T-synthase protein sequence demonstrated that the C. elegans gene in cosmid C38H2.2 was highly homologous. The cDNA for the putative Ce-T-synthase was found that had a size of 1.26 kb as predicted. For the isolation of its cDNA, two PCR reactions covering both 5' and 3' ends with 170-bp overlap were carried out using primers based on the cDNA from C38H2.2 using C. elegans cDNA library. The 5'-end PCR primer set utilized the forward primer 5'-CCACCATGAAGAGACGGTTTTGTCG-3' and the reverse primer 5'-CGGTGAATGAGCCAATAG-3'. The 3'-end PCR primer set utilized the forward primer 5'-AGGATGCAGAACTCCCAG-3' and the reverse primer 5'-GCTTATACAGCAACTTCCGGCC-3'. The PCR products were ligated with pCR2.1 vector and cloned and sequenced. The PCR product of the 3' end perfectly matched the cDNA predicted by C38H2.2. By contrast, the PCR product of the 5' end only matched partially. Using 5'-end-matched sequence of this PCR product allowed the identification of a C. elegans EST clone (C41813
[GenBank]
) by a BlastN dbEST search. The analysis of the PCR product from the 5' end of cDNA predicted to be encoded by C38H2.2, and the comparison of that sequence to the EST sequence of the EST clone C41813
[GenBank]
confirmed that the 5' end of the cDNA predicted in the BlastP search from C38H2.2 is incorrect. Therefore, we redesigned primers based on the EST clone C41813
[GenBank]
sequence. The primer set utilized the forward primer 5'-CCACCATGGCAAACTGGCCACGTGTTTC-3' and the reverse primer 5'-CGGTGAATGAGCCAATAG-3'. With these primers for the 5' end of the EST clone, we performed a PCR with the cDNA library of C. elegans and generated a 620-bp product. The analysis of the sequences of the EST clone C41813
[GenBank]
and this PCR product, along with the sequence of the 3'-half PCR product, gave the sequence of the full-length cDNA for the Ce-T-synthase.
Protein assay
Protein was determined by the bicinchoninic acid (BCA) assay (Pierce) according to the manufacturer’s protocol using bovine serum albumin as a standard.
Assay for activity of Ce-T-synthase
Ce-T-synthase was assayed using GalNAc1-O-phenyl as the acceptor as previously described (Ju, Cummings, et al., 2002). Briefly, the assay was carried by a 50-µL reaction containing 50 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffer (pH 6.5), 1 mM GalNAc-1-phenyl, 400 µM UDP-Gal, 60,000 cpm of UDP-[3H]-Gal, 20 mM MnCl2, 2 mM ATP, and 2030 µL of enzyme. After incubation at 37°C for 60 min, the reaction mixture was loaded onto a C18 cartridge (50 mg). After washing with 12 mL of water, the bound material was eluted with 1 mL of n-butanol and mixed with 10 mL of scintillation liquid. The radioactivity was counted, and the activity of T-synthase was calculated. Each sample was assayed in triplicates, and each experiment was repeated twice.
Construct a plasmid expressing full-length Ce-T-synthase
The full-length Ce-T-synthase expression vector was constructed by subcloning the complete open reading frame (ORF) of Ce-T-synthase from pCR2.1 vector prepared into pVL1393, an insect cell expression vector.
Construction of a plasmid-expressing soluble, N-terminal HPC4-epitope-tagged Ce-T-synthase
For mammalian cell expression of Ce-T-synthase, the plasmid-expressing soluble, N-terminal HPC4-epitope-tagged Ce-T-synthase was constructed by cloning of a DNA fragment of BamHI/XhoI from a plasmid containing full-length Ce-T-synthase cDNA into pcDNA3.1(+)-TH fragment digested with BamHI and XhoI in frame. Then, the whole ORF was subcloned into pVL1393 insect cell expression vector. The vector construct was confirmed by sequencing.
Baculovirus preparation and expression of Ce-T-synthase in Hi-5 insect cells
Baculovirus encoding full-length and soluble, HPC4-tagged Ce-T-synthase was prepared by co-transfection of Sf-9 insect cells with baculovirus-linearized DNA and the plasmid with pVL1393 encoding either full-length or soluble Ce-T-synthase as previously described (Ju and Cummings, 2002). The Ce-T-synthase was expressed in Hi-5 insect cells by infection with the baculovirus.
Transfection of mammalian cells with a plasmid-expressing soluble, HPC4-tagged Ce-T-synthase
Mammalian cells including 293T, LSC, LSB, CHO K1, Lec1, Lec2, and Lec8 were cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS) at 37°C and 5% CO2 and transfected with a plasmid in a backbone of pcDNA3.1(+) encoding soluble, HPC4-tagged Ce-T-synthase using FuGENE 6, according to the manufacturer’s instructions. At 72 h after transfection, the media were collected for T-synthase activity assay and western blot analysis.
Capture of soluble T-synthase on HPC4-UltraLink
Ce-T-synthase in 10 mL of media collected from transfected mammalian cells or infected Hi-5 was captured on 100 µL of HPC4-UltraLink. After washing, 40 µL of beads was then directly assayed for Ce-T-synthase in triplicates plus a blank using GalNAc1-O-phenyl as the acceptor. The remaining beads were eluted with 150 µL of 25 mM Tris–HCl, pH 8.0, containing 10 mM ethylenediaminetetraacetic acid (EDTA). The eluted T-synthase was subjected to endoglycosidase (N-glycanase and endo H) digestion according to the manufacturer’s instruction.
Western blot of soluble, HPC4-epitope-tagged Ce-T-synthase
Ce-T-synthase purified from media containing soluble, HPC4-epitope-tagged Ce-T-synthase secreted by mammalian cells and Hi-5 cells was treated with endoglycosidases, electrophoresed on SDS–PAGE (4–20%), and transferred to a nitrocellulose membrane. Ce-T-synthase was detected by immunoblotting with mouse monoclonal antibody HPC4 as previously described (Ju, Brewer, et al., 2002).
Identification of the product generated by the HPC4-epitope-tagged T-synthase
The HPC4-epitope-tagged Ce-T-synthase in 20 mL of media from Hi-5 cells infected with baculovirus expressing the soluble, HPC4-tagged Ce-T-synthase was captured by incubation with 200 µL of HPC4-UltraLink beads as mentioned previously. One hundred micrograms of PSGL-1 glycopeptide (4-GP-1) in 100 µL containing 100 mM MES (pH 6.5), 500 µM UDP-[3H]-Gal (200,000 cpm), and 20 mM MnCl2 was incubated with 50 µL of the beads. As a positive control, a reaction was set up as the same as above except that human soluble, HPC4-tagged T-synthase (2 nmol/h) was used as the enzyme source (Ju, Brewer, et al., 2002). After an incubation of 16 h at 37°C, the reaction mixtures were loaded onto C18 cartridges (50 mg) following the standard T-synthase assay protocols (Ju, Brewer, et al., 2002). The bound material was eluted with 1.0 mL of methanol, and a 0.05-mL portion was mixed with 4 mL of Scintiverse-BD and radioactivity determined in a liquid scintillation counter. The residual 0.95 mL from each sample was dried in a speed-vac concentrator. The dried material was redissolved in 200 µL of 50 mM of Tris–HCl (pH 7.0) containing 150 mM NaCl. The redissolved material (100 µL) was treated with 5 mU of O-glycanase at 37°C for 24 h. One-half of this reaction mixture was then loaded onto C18 cartridges (50 mg). The cartridges were washed with 6 mL of water, and bound material was eluted with 1 mL of methanol. The eluates were mixed with 10 mL of Scintiverse-BD, and radioactivity was determined by liquid scintillation counting.
Construction of GFP-promoter analysis construct—PPD95.67/Ce-T-syn
Ce-T-synthase promoter (3.2 kb) from immediately upstream of the putative gene was obtained by PCR using genomic DNA as the template and PCR primers: 5'-GGATACGTTCTCCGACTGC-3' (forward) and GCTC TAGATCACCAGCACATCT-GTAGTTAATG-3' (reverse) containing an XbaI site for cloning into the vector. The fragment of 3.2 kb obtained by PstI/XbaI digestion was subcloned into the GFP-promoter analysis vector PPD95.67.
Microinjection and selection of the transgenic worm
The mixture of PPD95.67/Ce-T-syn (10 ng/µL) and pBX plasmid (10 ng/µL) was microinjected into the valve of L1L2-stage C. elegans with pha-1 mutation as previously described, and transgenic worms were selected at growing at room temperature (23°C), and GFP fluorescence was observed under a fluorescent microscope as previously described (Zheng et al., 2002).
Quantitative real-time PCR of T-synthase from different stages of C. elegans
Caenorhabditis elegans were cultured on plates, and L1, L2–L4, and young adult worms were harvested, and total RNAs were isolated using total RNA isolation kit. The first-strand cDNA was used as a template in PCR reactions to amplify Ce-T-synthase in a 50-µL reaction containing 3 µL of first-strand cDNA. The primers for C. elegans T-synthase are 5'-TTGGCGTTCTTCTCGGTTTG-3' (forward) and 5'-CCCTCGGTTGGCTCGTAGTAT-3' (reverse). The product for Ce-T-synthase is 100 bp. As a control, C. elegans -actin was amplified using -actin gene-specific primer pair 5'-ATCGTCCTCGACTCTGGAGAT-3' and 5'-TCACGTCCAGCCAAGTCAAG-3' C. elegans -actin gene in the identical conditions described. To confirm the absence of nonspecific amplification, we analyzed the PCR products using 1.5% agarose gel.
RNAi of Ce-T-synthase
RNAi experiment was carried out by feeding C. elegans with Escherichia coli expressing double-stranded RNA of a part of Ce-T-synthase gene. The construct for RNAi was made by the ligation of a DNA fragment of 483 bp from Ce-T-synthase coding region generated by EagI and EcoRV digestion to a dual T7-promoter-containing vector L4440 DoubleT-7script fragment generated with NotI and SmaI digestion. The plasmid generated for RNAi (L4440/Ce-T-syn) was confirmed by DNA sequencing. Escherichia coli strain HT115(DE3) was transformed with L4440/Ce-T-syn and selected on Luria-Bertani/Amp plate. A single colony was picked and cultured in 5 mL of LB media containing 50 µg/mL of Amp for 14 h. Then, 100 µg/mL of isopropyl-1-thio-ß-D-galactopyranoside (IPTG) (final concentration) was added for induction for 4 h. The plasmid from 500 µL of the bacteria was isolated and confirmed by restriction enzyme digestion. The remaining bacteria were planted on LB plate in a strip manner and cultured for 48 h at room temperature. Four worms at the L1–L2 stages were transferred to the plate loaded with L4440/Ce-T-syn containing E. coli and cultured at 16°C for 4 days. The next generation of the worm at L1–L2-stage worms was transferred to a new LB plate (four worms/plate) loaded with the E. coli transformed with L4440/Ce-T-syn and continued culturing for 4–6 days. The eggs and worms were observed and recorded under microscope carefully to assess their number, size, movement, mortality, and growth. The RNAi control experiments were performed under the same procedures, except for the plasmid used was the empty vector L4440 DoubleT-7script. The positive control was worms fed E. coli transformed with L4440/Tal.1 expressing double-stranded RNA of 300 bp at the 5' end of the C. elegans Talin gene.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
None declared.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
We thank Dr. Anne Leppänen for the donation of the acceptor glycopeptide from PSGL-1. This work and the Open Access publication charges were supported by a National Institutes of Health Grant RO1 CH/HD54832-01 to R.D.C.
|
Abbreviations |
|---|
Ce-T-synthase, Caenorhabditis elegans T-synthase; CHO, Chinese hamster ovary; core 1 O-glycan, Galß1-3GalNAc
1-Ser/Thr; core 1 ß3-Gal-T or T-synthase, core 1 UDP-Gal: GalNAc1-Ser/Thr ß1,3-galactosyltransferase; endo H, endo N-acetylglucosaminidase H; EST, expressed sequence tag; GFP, green fluorescent protein; PCR, polymerase chain reaction; RNAi, RNA interference; TM, transmembrane domain |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgmentsReferences |
|---|
Ailor, E. and Betenbaugh, M.J. (1999) Modifying secretion and post-translational processing in insect cells. Curr. Opin. Biotechnol., 10, 142–145.[CrossRef][ISI][Medline]
Altmann, F., Fabini, G., Ahorn, H., and Wilson, I.B. (2001) Genetic model organisms in the study of N-glycans. Biochimie, 83, 703–712.[Medline]
Bedard, K., Szabo, E., Michalak, M., and Opas, M. (2005) Cellular functions of endoplasmic reticulum chaperones calreticulin, calnexin, and ERp57. Int. Rev. Cytol., 245, 91–121.[ISI][Medline]
Boeggeman, E.E., Ramakrishnan, B., and Qasba, P.K. (2003) The N-terminal stem region of bovine and human beta1,4-galactosyltransferase I increases the in vitro folding efficiency of their catalytic domain from inclusion bodies. Protein Expr. Purif., 30, 219–229.[CrossRef][ISI][Medline]
Brockhausen, I., Schutzbach, J., and Kuhns, W. (1998) Glycoproteins and their relationship to human disease. Acta Anat. (Basel), 161, 36–78.[CrossRef][ISI][Medline]
Brockhausen, I., Yang, J., Dickinson, N., Ogata, S., and Itzkowitz, S.H. (1998) Enzymatic basis for sialyl-Tn expression in human colon cancer cells. Glycoconj. J., 15, 595–603.[CrossRef][ISI][Medline]
Carlyon, J.A., Akkoyunlu, M., Xia, L., Yago, T., Wang, T., Cummings, R.D., McEver, R.P., and Fikrig, E. (2003) Murine neutrophils require alpha1,3-fucosylation but not PSGL-1 for productive infection with Anaplasma phagocytophilum. Blood, 102, 3387–3395.
Cipollo, J.F., Awad, A.M., Costello, C.E., and Hirschberg, C.B. (2005) N-Glycans of Caenorhabditis elegans are specific to developmental stages. J. Biol. Chem., 280, 26063–26072.
Cipollo, J.F., Costello, C.E., and Hirschberg, C.B. (2002) The fine structure of Caenorhabditis elegans N-glycans. J. Biol. Chem., 277, 49143–49157.
Cram, E.J., Clark, S.G., and Schwarzbauer, J.E. (2003) Talin loss-of-function uncovers roles in cell contractility and migration in C. elegans. J. Cell Sci., 116, 3871–3878.
DeBose-Boyd, R.A., Nyame, A.K., and Cummings, R.D. (1998) Molecular cloning and characterization of an alpha1,3 fucosyltransferase, CEFT-1, from Caenorhabditis elegans. Glycobiology, 8, 905–917.
Dejgaard, S., Nicolay, J., Taheri, M., Thomas, D.Y., and Bergeron, J.J. (2004) The ER glycoprotein quality control system. Curr. Issues Mol. Biol., 6, 29–42.[ISI][Medline]
Deutscher, S.L. and Hirschberg, C.B. (1986) Mechanism of galactosylation in the Golgi apparatus. A Chinese hamster ovary cell mutant deficient in translocation of UDP-galactose across Golgi vesicle membranes. J. Biol. Chem., 261, 96–100.
Deutscher, S.L., Nuwayhid, N., Stanley, P., Briles, E.I., and Hirschberg, C.B. (1984) Translocation across Golgi vesicle membranes: a CHO glycosylation mutant deficient in CMP-sialic acid transport. Cell, 39, 295–299.[CrossRef][ISI][Medline]
Elder, J.H. and Alexander, S. (1982) Endo-beta-N-acetylglucosaminidase F: endoglycosidase from Flavobacterium meningosepticum that cleaves both high-mannose and complex glycoproteins. Proc. Natl. Acad. Sci. U. S. A., 79, 4540–4544.
Endo, Y. and Kobata, A. (1976) Partial purification and characterization of an endo-alpha-N-acetylgalactosaminidase from the culture of medium of Diplococcus pneumoniae. J. Biochem. (Tokyo), 80, 1–8.
Fan, X., She, Y.M., Bagshaw, R.D., Callahan, J.W., Schachter, H., and Mahuran, D.J. (2005) Identification of the hydrophobic glycoproteins of Caenorhabditis elegans. Glycobiology, 15, 952–964.
Guerardel, Y., Balanzino, L., Maes, E., Leroy, Y., Coddeville, B., Oriol, R., and Strecker, G. (2001) The nematode Caenorhabditis elegans synthesizes unusual O-linked glycans: identification of glucose-substituted mucin-type O-glycans and short chondroitin-like oligosaccharides. Biochem. J., 357, 167–182.[CrossRef][ISI][Medline]
Haslam, S.M. and Dell, A. (2003) Hallmarks of Caenorhabditis elegans N-glycosylation: complexity and controversy. Biochimie, 85, 25–32.[Medline]
Haslam, S.M., Gems, D., Morris, H.R., and Dell, A. (2002) The glycomes of Caenorhabditis elegans and other model organisms. Biochem. Soc. Symp., 117–134.
Helenius, A. and Aebi, M. (2004) Roles of N-linked glycans in the endoplasmic reticulum. Annu. Rev. Biochem., 73, 1019–1049.[CrossRef][ISI][Medline]
Javaud, C., Dupuy, F., Maftah, A., Julien, R., and Petit, J.M. (2003) The fucosyltransferase gene family: an amazing summary of the underlying mechanisms of gene evolution. Genetica, 118, 157–170.[CrossRef][ISI][Medline]
Ju, T., Brewer, K., D’Souza, A., Cummings, R.D., and Canfield, W.M. (2002) Cloning and expression of human core 1 beta1,3-galactosyltransferase. J. Biol. Chem., 277, 178–186.
Ju, T. and Cummings, R.D. (2002) A unique molecular chaperone Cosmc required for activity of the mammalian core 1 beta 3-galactosyltransferase. Proc. Natl. Acad. Sci. U. S. A., 99, 16613–16618.
Ju, T. and Cummings, R.D. (2005) Protein glycosylation: chaperone mutation in Tn syndrome. Nature, 437, 1252.[CrossRef][Medline]
Ju, T., Cummings, R.D., and Canfield, W.M. (2002) Purification, characterization, and subunit structure of rat core 1 beta1,3-galactosyltransferase. J. Biol. Chem., 277, 169–177.
Julian, B.A. and Novak, J. (2004) IgA nephropathy: an update. Curr. Opin. Nephrol. Hypertens., 13, 171–179.[ISI][Medline]
Kawar, Z.S., Van Die, I., and Cummings, R.D. (2002) Molecular cloning and enzymatic characterization of a UDP-GalNAc:GlcNAc(beta)-R beta1,4-N-acetylgalactosaminyltransferase from Caenorhabditis elegans. J. Biol. Chem., 277, 34924–34932.
Kudo, T., Iwai, T., Kubota, T., Iwasaki, H., Takayma, Y., Hiruma, T., Inaba, N., Zhang, Y., Gotoh, M., Togayachi, A., and Narimatsu, H. (2002) Molecular cloning and characterization of a novel UDP-Gal: GalNAc(alpha) peptide beta 1,3-galactosyltransferase (C1Gal-T2), an enzyme synthesizing a core 1 structure of O-glycan. J. Biol. Chem., 277, 47724–47731.
Lamblin, G., Degroote, S., Perini, J.M., Delmotte, P., Scharfman, A., Davril, M., Lo-Guidice, J.M., Houdret, N., Dumur, V., Klein, A., and Rousse, P. (2001) Human airway mucin glycosylation: a combinatory of carbohydrate determinants which vary in cystic fibrosis. Glycoconj. J., 18, 661–684.[CrossRef][ISI][Medline]
Lowe, J.B. and Marth, J.D. (2003) A genetic approach to mammalian glycan function. Annu. Rev. Biochem., 72, 643–691.[CrossRef][ISI][Medline]
McEver, R.P. and Cummings, R.D. (1997) Perspectives series: cell adhesion in vascular biology. Role of PSGL-1 binding to selectins in leukocyte recruitment. J. Clin. Invest., 100, 485–491.[ISI][Medline]
Muller, R., Hulsmeier, A.J., Altmann, F., Ten Hagen, K., Tiemeyer, M., and Hennet, T. (2005) Characterization of mucin-type core-1 beta1-3 galactosyltransferase homologous enzymes in Drosophila melanogaster. FEBS J., 272, 4295–4305.[CrossRef][Medline]
Natsuka, S., Adachi, J., Kawaguchi, M., Nakakita, S., Hase, S., Ichikawa, A., and Ikura, K. (2002) Structural analysis of N-linked glycans in Caenorhabditis elegans. J. Biochem. (Tokyo), 131, 807–813.
Oriol, R., Mollicone, R., Cailleau, A., Balanzino, L., and Breton, C. (1999) Divergent evolution of fucosyltransferase genes from vertebrates, invertebrates, and bacteria. Glycobiology, 9, 323–334.