Glycobiology Advance Access originally published online on November 22, 2005
Glycobiology 2006 16(3):258-270; doi:10.1093/glycob/cwj060
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Substitution of the N-glycan function in glycosyltransferases by specific amino acids: ST3Gal-V as a model enzyme
2 Department of Biomembrane and Biofunctional Chemistry, Graduate School of Pharmaceutical Science, Frontier Research Center for Post-Genomic Science and Technology; 3 Core Research for Evaluational Science and Technology Program (CREST), Japan Science and Technology Agency (JST), Graduate School of Pharmaceutical Science, Frontier Research Center for Post-Genomic Science and Technology; and 4 Department of Structural Biology, Graduate School of Pharmaceutical Science, Hokkaido University, Kita 21-jo, Nishi 11-choume, Kita-ku, Sapporo 001-0021, Japan; 5 Department of Bioscience and Biotechnology, Graduate School of Bioresource and Bioenvironmental Sciences, Kyushu University, Higashi-ku, Fukuoka 812-8581, Japan; and 6 Pharmacodynamics, Meiji Pharmaceutical University, 2-522-1 Noshio, Kiyose, Tokyo 204-8588, Japan
1 To whom correspondence should be addressed; Tohoku Pharmaceutical University, 4-4-1, Komatsujima, Aoba-Ku, Sendai, 981-8558, Miyagi, Japan; e-mail: inokuchi{at}kinou02.pharm.hokudai.ac.jp
Received on October 12, 2005; revised on November 8, 2005; accepted on November 9, 2005
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMethods and materialsReferences |
|---|
The sialyltranferase ST3Gal-V transfers a sialic acid to lactosylceramide. We investigated the role of each of the N-glycans modifying mouse ST3Gal-V (mST3Gal-V) by measuring the in vitro enzyme activity of Chinese hamster ovary (CHO) cells transfected with ST3Gal-V cDNA or its mutants. By examining mutants of mST3Gal-V, in which each asparagine was replaced with glutamine (N180Q, N224Q, N334Q), we determined that all three sites are N-glycosylated and that each N-glycan is required for enzyme activity. Despite their importance, N-glycosylation sites in ST3Gal-V are not conserved among species. Therefore, we considered whether the function in the activity that is performed in mST3Gal-V by the N-glycan could be substituted for by specific amino acid residues selected from the ST3Gal-V of other species or from related sialyltransferases (ST3Gal-I, -II, -III, and -IV), placed at or near the glycosylation sites. To this end, we constructed a series of interspecies mutants for mST3Gal-V, specifically, mST3Gal-V-H177D-N180S (medaka or tetraodon type), mST3Gal-V-N224K (human type), and mST3Gal-V-T336Q (zebrafish type). The ST3Gal-V activity of these mutants was quite similar to that of the wild-type enzyme. Thus, we have demonstrated here that the N-glycans on mST3Gal-V are required for activity but can be substituted for specific amino acid residues placed at or near the glycosylation sites. We named this method SUNGA (substitution of N-glycan functions in glycosyltransferases by specific amino acids). Furthermore, we verified that the ST3Gal-V mutant created using the SUNGA method maintains its high activity when expressed in Escherichia coli thereby establishing the usefulness of the SUNGA method in exploring the function of N-glycans in vivo.
Key words: ST3Gal-V / N-glycan / glycosyltransferase / sialyltransferase / glycotechnology
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMethods and materialsReferences |
|---|
Many glycosyltransferases are modified by N-linked carbohydrate chains. Depletion of an N-glycan by site-directed mutagenesis of N to Q in the glycosyltransferase consensus sequence NXS/T, or by treatment with tunicamycin, often results in a loss of enzyme activity (Table I). Because N-glycans are involved in the stabilization of proper conformation, this decrease may be the result of a change in the conformation of the enzyme into an inactive form (Trombetta, 2003
). However, comparisons of the N-glycosylation sites of various glycosyltransferases reveal a surprising amount of discrepancy. For example, in the sialyltransferase family presented in Table II, the number of N-glycosylation sites in each enzyme varies among species, and on average 76.5% of all N-glycosylation sites are not conserved among the listed enzymes. In addition, in 37.1% of these sites, the lack of conservation is because of a single amino acid alteration. Considering these facts, it is perhaps possible that in sialyltransferases a single amino acid in place of the N at an N-glycosylation site, or at an adjacent residue, can function as a substitute for the N-glycan in maintaining enzyme conformation and activity.
|
|
In this study, we examined the effects of site-directed mutagenesis in ST3Gal-V at each of its N-glycosylation sites or adjacent amino acid residues. The selection of the substituting residues was based on comparisons between the amino acid sequences of ST3Gal-V from various species or between members of the sialyltransferase family. This technique succeeded in identifying a specific single amino acid substitution for each site that preserves the in vitro enzyme activity. We named this method SUNGA (substitution of N-glycan functions in glycosyltransferases by specific amino acids). We describe here our successful mutations using the SUNGA method to eliminate each N-glycan from mST3Gal-V without decreasing its in vitro enzyme activity. In addition, we demonstrate that the ST3Gal-V mutant that was created using the SUNGA method maintains its high activity when produced in Escherichia coli. Thus, SUNGA has a distinct advantage over the original N to Q substitution and will be very useful not only to study the fundamental significance of N-glycans but also in glycotechnology and structural biology of bioactive glycoproteins.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMethods and materialsReferences |
|---|
mST3Gal-V has N-glycans of the complex type
We transiently transfected mST3Gal-V cDNA into Chinese hamster ovary (CHO) cells and then analyzed the cell lysates by immunoblotting with an anti-ST3Gal-V antibody that recognizes the C-terminal 55 amino acids of mST3Gal-V. The results shown in Figure 1 revealed that the ST3Gal-V protein had been modified, since the molecular mass of the stained band was higher than the mass predicted from the amino acid sequence (40 kDa). We considered the possibility that the transfected ST3Gal-V was glycosylated. Melkerson-Watson and Sweeley (1991)
reported that N-glycans carried on ST3Gal-V purified from rat liver contain galactose, GlcNAc, and NeuAc, suggesting the presence of complex type glycans. However, Bieberich et al. (2002) demonstrated that overexpressed ST3Gal-V fused to a FLAG-yellow fluorescent protein (YFP) epitope tag had only Endo H-sensitive N-glycans. We performed endoglycosidase digestions of the cell lysates, and, as expected, a shift in the mobility of the ST3Gal-V was observed, clearly indicating the presence of oligosaccharides. Moreover, the results indicated that mST3Gal-V carries both Endo H-sensitive oligosaccharides, which are high–mannose-type glycans modified in the endoplasmic reticulum (ER), and peptide N-glycosidase F (PNGase F)-sensitive oligosaccharides, which are modified in the Golgi (Figure 1). We also confirmed that overexpressed ST3Gal-V fused to a 3x FLAG tag has only Endo H-sensitive N-glycans, as we were unable to detect any Endo H-resistant N-glycans (data not shown). Therefore, the tag protein may interfere the transport from ER to Golgi. These results strongly indicate that N-glycans on wild-type ST3Gal-V are of the complex type.
|
N-Glycans having a high-mannose core are essential for mST3Gal-V activity
To examine whether the N-glycans on mST3Gal-V are necessary for its enzyme activity, we used a series of inhibitors for N-glycan processing. In immunoblots of lysates from cells treated with tunicamycin, which inhibits the formation of dolichol pyrophosphate-N-acetyl glucosamine, mST3Gal-V was detected as a single band with a molecular mass lower than that of the protein from untreated cells (Figure 2A). This band was not affected further by Endo H treatment, indicating that the N-glycosylation had been completely abolished. Lysates from cells treated with the mannosidase inhibitor kifnecine or with castanospermine, a glucosidase I and II inhibitor, also exhibited a single band, which was Endo H sensitive (Figure 2A), indicate that mST3Gal-V treated with kifnecine or castanospermine has only N-glycans of the high-mannose type. Furthermore, the ST3Gal-V enzyme activity was remarkably reduced by the tunicamycin treatment, but was not affected by the kifnecine or castanospermine (Figure 2B). This suggests that N-glycans with a high-mannose core (at least) are essential for ST3Gal-V activity.
|
Each N-glycan is necessary for ST3Gal-V activity
Three potential glycosylation sites (Asn-X-Ser/Thr) exist in mST3Gal-V, the asparagine residues at amino acid 180, 224, and 334 (Figure 3). Site-directed mutagenesis was performed to study the role of the individual sites. N to Q mutants for each site (N180Q, N224Q, N334Q) and three additional mutants (N180, 224, 334Q) were generated from wild-type mST3Gal-V and transiently transfected into CHO cells. Cell lysates were collected 24 h after transfection and analyzed by immunoblotting. As shown in Figure 4A, the mutants migrated differently from the wild-type mST3Gal-V, according to the number of oligosaccharide chains attached. The wild-type protein migrated to 
42–48 kDa, but removal of the N-glycosylation sites resulted in decreases in the molecular mass of 2–3 kDa for the single mutants and 8–14 kDa for the triple mutant. Thus, the mass of the triple mutant was, as expected, comparable with that of the wild-type enzyme that had been treated with tunicamycin. The immunoblots also revealed that the expression levels of N334Q and of the triple mutant were about half that of the wild-type transfectant (Figure 4A, lower panel). Similarly, in pulse chase experiments, the initial amount of the N334Q mutant labeled with [35S]-methionine was about half that of the wild type, and no recovery in the expression levels of the mutant was observed after treatment with the proteosome protease inhibitor MG132 (data not shown). Moreover, the decreased expression of these mutants was not because of the decreased reactivity of this antibody but because FLAG-tagged 334N mutants detected by anti-FLAG antibody were similar to the results of Figure 4A, lower panel. Therefore, the reduced expression was likely caused by instability in the mRNA because of the mutation. Additionally, when we assessed the trypsin sensitivity of all the N
Q mutants (single and triple replacement) and the wild type, there were no differences in their degradation time courses (data not shown). Therefore, serious misfolding because of conformational changes may not occur in the deglycosylated mutants.
|
|
The enzyme activities were also compared between the wild-type mST3Gal-V and the mutants. The activity was almost abolished in the triple mutant, and the single mutants exhibited only 10–30% of the activity of the wild-type protein (Figure 4B). This clearly demonstrated that each N-glycan is necessary for enzyme activity.
Site-specific amino acid residues can replace the function of the N-glycans
The results in Figure 4 suggested that every N-glycan in mST3Gal-V is necessary for its activity. However, as shown in Figure 3, not every N-glycosylation site is conserved among species. For instance, the 177HVGNES in mST3Gal-V is replaced by 211DVGSRT in medaka and by 216DAGSHT in tetraodon. Likewise, the 224NES in mST3Gal-V is instead 224KES in human and dog; 225KES in bovine; 225RTP in zebra fish; 258KKP in medaka; and 263KQP in tetraodon. Finally, the 334NVT in mST3Gal-V is changed to 334NVQ in zebra fish, 362DVS in medaka, and 367DIS in tetraodon. Additionally, we can compare the sialyl motifs L and VS in ST3Gal-V with those in the other alpha 2-3 sialyltransferases (ST3Gal-I, -II, -III, and -IV) and see a similar lack of conservation (Figure 5); the 177HVGNES in mST3Gal-V is replaced by 183/193/201/160DVG(S/T)(K/R)T in ST3Gal-I, -II, -III, and -IV, and the 334NVT in mST3Gal-V by 276NIQ in ST3Gal-III and 242/252DAD in ST3Gal-I and -II. This lack of conservation led us to consider the possibility that in some enzymes certain amino acids in place of (or near) the asparagine at an N-glycosylation site might function similarly to an N-glycan. Thus, we compared the functionality of alternate sequences for each site using our newly established SUNGA method. For this method, we selected amino acids based on the comparisons in Figure 6A and generated a series of mST3Gal-V mutants (N180K, N180S, H177D-N180S, H177D, N224K, N224D, N334K, and T336Q) by site-directed mutagenesis. We first examined the differences in the mutated proteins by immunoblotting. As shown in Figure 6B, the N180S and H177D-N180S mutants were each detected as a band of 42 kDa, similar to the N180Q mutant, but the N180K mutant was detected as two bands of 40 and 42 kDa. The molecular mass of the H177D mutant was the same as that of the wild-type protein. We also examined the enzyme activity of these mutants and found that the activities of the N180Q and N180K mutants were decreased to 20% relative to the activity of the wild-type control and the activity of the N180S mutant decreased to 50% (Figure 6C). However, the H177D-N180S double mutant, which was based on the medaka and tetraodon sequences, exhibited nearly 100% activity. Moreover, the H177D single mutant exhibited 150% activity.
|
|
Mutations in the other N-glycosylation sites conferred a similar range of variation. The expression levels and the molecular masses of the N224K and N224D mutants were the same as those of the N224Q mutant (Figure 6B, middle panel). In the enzyme assays, the activity of the N224K mutant was quite similar to that of the wild-type control. However, the activity of the N224D mutant was greatly reduced, similar to that of the N224Q mutant (Figure 6C). In experiments regarding the final site, the expression levels and the molecular masses of the N334K and T336Q mutants were the same as those of the N334Q mutant (Figure 6B, bottom panel). Likewise, the activity of the N334K mutant was equal to that of the N334Q mutant. However, the activity of the T336Q mutant was higher than that of the wild-type transferase (Figure 6C). Together, these results indicate that alternate amino acid sequences at an N-glycosylation site can perform the same function as an N-glycan, that of maintaining the proper conformation. Thus, using the SUNGA method we have demonstrated that partially de-N-glycosylated mST3Gal-V can maintain its full enzyme activity in vitro and have identified which amino acids can act in place of the glycan.
To examine further the potential usefulness of the SUNGA method, we generated sugar-free mST3Gal-V mutant (H177D-N180S-N224K-T336Q) and examined its expression and activities. As shown in Figure 7A, this non-glycosylated mutant migrated with a mass comparable with that of the wild-type cells that had been treated with tunicamycin (Figure 2A). In enzyme assays, the H177D-N180S-N224K-T336Q mutant exhibited 55% activity (Figure 7B). Furthermore, the activity was 8- to 10-fold greater than the activity of the N180, 224, and 334Q mutants.
|
Production of sugar-free ST3Gal-V mutant in E. coli
To attain mass production of bioactive glycolipids and glycoproteins, it is desirable to produce active glycosyltransferases in E. coli. However, many glycosyltransferases require carbohydrate chains for in vitro activity (Table I). Here, we tried to express the sugar-free mST3Gal-V mutant protein (H177D-N180S-N224K-T336Q) that we created using the SUNGA method, in E. coli. To obtain a soluble enzyme, we expressed the 
TM version of the wild-type or mutant protein, N-terminally labeled with a His-tag, together with in a plasmid containing three chaperones, using a cold-shock expression system. In the presence of chaperones (groES, groEL, and Tig), both wild and mutant ST3Gal-V proteins were completely recovered in the soluble fraction (Figure 8A). The soluble fraction was subjected to Ni-NTA-agarose chromatography and eluted with 150 mM imidazole, and samples were then separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE). As shown in Figure 8B, the equal amounts of ST3Gal-V and its mutant were purified with chaperones.
|
After undergoing dialysis to remove the imidazole, additional samples were examined for GM3-producing ability using LacCer and CMP-NeuAc as substrates. The products were separated by high performance TLC (HPTLC) and transblotted onto a polyvinylidene difluoride (PVDF) membrane. GM3 was stained with an anti-GM3 antibody (M2590). As shown in Figure 8C, the amount of GM3 produced by the mutant enzyme (His-TM-H177D-N180S-N224K-T336Q) was much greater than that produced by the wild-type enzyme (His-TM-mST3Gal-V wild type).
Comparison of the secretion of mST3Gal-V wild type and mutants
The glycosyltransferases are believed to be localized specifically in the Golgi, where they synthesize glycochains. Some glycosyltransferases have a proteolysis-sensitive region in the "stem." This stem region is often cleaved by endogenous proteases so that the glycosyltransferases are secreted out of the cell; for instance, BACE1 was recently identified as a protease involved in the secretion of ST6Gal-I (Kitazume et al., 2001, 2003). It has been well documented that the proteolytic cleavage and secretion of glycosyltransferases into body fluids are affected by various pathological conditions, such as malignant transformation and inflammation (Colley, 1997). Thus, the significance of secretion has been an important subject to investigate for glycosyltransferases.
Little is known regarding the secretion of ST3Gal-V. We examined the secretion in HEK293 cells transiently transfected with a pcDNA3 vector or mST3Gal-V. Cell extracts and culture supernatants were analyzed by immunoblotting with the anti-ST3Gal-V antibody (Figure 9A and B). mST3Gal-V also has N-glycans of Endo H-resistant in HEK293T cell (Figure 9A). In the culture medium, ST3Gal-V was detected as two bands of 45 and 38 kDa (Figure 9B). The broad band of 45 kDa shifted to 38 kDa after treatment with PNGase F, indicating that the ST3Gal-V of 45 kDa was N-glycosylated with an Endo H-resistant glycan (data not shown).
|
Next, we examined the amount of secretion to the medium of each mutant after PNGase F treatment. The amount of secretion of the N224Q and N224K mutants was quite similar to that of the wild-type protein (Figure 9C). The N180Q mutation caused a reduction in the amount of secretion to 20% relative to the wild-type levels and the H177D-N180S mutant to 50%. Surprisingly, the N334Q and T336Q mutations caused an increase in the amount of secretion to 300% as compared with the wild type. These results suggest that the N-glycan on 180N of ST3Gal-V is necessary for secretion to the medium, whereas the N-glycan on 334N is important for remaining in the cells. Moreover, the amount of secretion of the H177D-N180S mutant, which retained its full enzyme activity (Figure 6B) increased and was approximately twice that of the N180Q mutant (Figure 9C, inset), which exhibited only 20% activity. These results provide evidence that not only the N-glycan on 180N but also the surrounding protein conformation is important for ST3Gal-V activity and secretion.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMethods and materialsReferences |
|---|
Recent achievements in cDNA cloning and the functional identification of a wide variety of glycosyltransferases have made possible certain insights into the homology of these enzymes and the characterization of common features among species and within enzyme subfamilies. In some previous reports, summarized in Table I, the importance of N-glycans for glycosyltransferase activity was demonstrated, either by the depletion of sugar chains (by tunicamycin treatment) or by the site-directed mutagenesis of N to Q in the consensus sequence NXS/T. However, for each glycosyltransferase, we recognized considerable differences among species in the position and number of N-glycosylation sites (Table II). For example, the possible N-glycosylation sites in an individual sialyltransferase vary greatly among species. It is conceivable that the functional regulation of glycosyltransferases by N-glycosylation may have been altered during the evolutional process. Thus, we were interested in the effects of interspecies substitutions of amino acid residues at or near the N-glycosylation sites, which result in differences in the number of sugar chains. Consequently, we considered the possibility that specific amino acid residues may be able to substitute for the function of an N-glycan, at least in part. This theory gave rise to our newly established method called SUNGA, in which site-directed mutagenesis replaces the N, or nearby amino acids, of N-glycosylation sites with potentially active amino acids selected from nonconserved sites. We selected ST3Gal-V as a model enzyme to investigate the biological significance of the N-glycan modification, because this enzyme has no conserved N-glycosylation sites, and the number of possible N-glycosylation site(s) also varies (from 0 to 3) among species (Table II; Figure 3).
We have demonstrated that mST3Gal-V has N-glycans of the complex type (Figure 1) and that the presence of at least high–mannose-type sugar chains is essential for the enzyme activity (Figure 2). Although N to Q mutational analysis has proven that each N-glycan in mST3Gal-V is important for its enzyme activity (Figure 4), we succeeded in removing each N-glycan without affecting the in vitro enzyme activity by employing the SUNGA method (Figure 6). Moreover, wild-type mST3Gal-V produced in E. coli lost its enzyme activity, as expected, however, the mutant protein ST3Gal-V-H177D-N180S-N224K-T336Q developed by SUNGA proved to have sufficient GM3-producing ability (Figure 8C).
Thus, the SUNGA method has proven to be useful in preparing an enzyme that mimics the proper conformation supported by N-glycans. In a recent report on ST3Gal-I, in which amino acids used for substitution were selected by comparisons of proteins of various species, the authors claimed that N-glycans were not important for activity (Jeanneau et al., 2004). However, that report had no data on the activity of mutants carrying an N to Q substitution or on the use of N-glycosylation inhibitors, so it is not clear whether their amino acid substitution was able to mimic the functions of the N-glycan in vitro.
The details of our successful mutations using the SUNGA method to eliminate each N-glycan from mST3Gal-V without decreasing its in vitro enzyme activity are listed below:
- The 177HVGNKT region in the sialyl motif L of mST3Gal-V was replaced by 177DVGSRT, which is the corresponding sequence of the ST3Gal-V of medaka and tetraodon (Figures 3 and 6A) and of related sialyltransferases, such as ST3Gal-I, -II, -III, and -IV (Figure 5).
- The 224NES residues in mST3Gal-V were mutated to 224KES, the corresponding sequence in the human, zebrafish, medaka, and tetraodon enzymes.
- The 334NVT in the sialyl motif VS of mST3Gal-V was mutated to 334NVQ, the sequence found in zebrafish and ST3Gal-III.
- The complete replacement of all the N-glycosylation sites of mST3Gal-V, using a combination of the successful mutations described above, resulted in an enzyme which retained an enzyme activity =50% that of the wild-type enzyme.
- A successful E. coli production system capable of providing an active mST3Gal-V mutant protein (H177D-N180S-N224K-T336Q) has been achieved.
Because N224K mutant maintains the activity, it will be expected that N224R mutant can hold the activity. Also 334NVT in mST3Gal-V is changed to 362DVS in medaka and 367DIS in tetraodon, we assume that 224NR or 334ND mutation can exhibit the activity. N-glycans on 180N and 334N are included in the sialyl motif L and VS, respectively, which are highly conserved in sialyltransferase family. The silalyl motif L has been shown to bind to the donor substrate (CMP-sialic acid) (Datta and Paulson, 1995), whereas the sialyl motif S participates in the binding of both donor and acceptor substrates (Datta et al., 1998). The exact role for sialyl motif VS in the catalytic process has not been identified. Because the substituted sequences exist in these motifs among species, we assume that the substrate specificity does not change between wild type and mutants. Thus, SUNGA is a practical strategy for eliminating N-glycans without losing enzyme activity, one that capitalizes on the evolutionary alterations of amino acid sequences.
Furthermore, we have demonstrated further advantages of the SUNGA method in addition to maintaining enzyme activity by supporting proper conformation in place of N-glycans. Using several mutants, we demonstrated that the N-glycan modifying 180N in mST3Gal-V is necessary for secretion to the medium, whereas the N-glycan on 334N is important for remaining in the cell (Figure 9C). Moreover, depletion of the N-glycan at 180N by the N180Q mutation, which caused a decrease in activity to 10%, or the H177D-N180S mutation, which did not affect activity, resulted in a significant increase in secretion (Figure 9C, inset). These results indicate that the loss of the N-glycan in the N180Q mutant caused both abnormal and inactive conformation resulting in the 80% decrease in secretion, whereas in the H177D-N180S mutant the 50% decreased secretion was due solely to the absence of the N-glycan, since the active conformation was apparently preserved. In other words, we were able to explore functional role(s) for N-glycan and its maintenance of proper conformation, that is, interaction with cellular lectins. Thus, SUNGA has a distinct advantage over the original N to Q substitution.
Because there is no N-glycosylation site in the ST3Gal-V of medaka (Table II), this enzyme must function naturally without an N-glycan. Thus, ST3Gal-V in other species might have acquired the N-glycan during the evolutionary process. In principle, the conformation of the N-glycan attachment site and that of the corresponding nonglycosylated site appears to be similar enough to maintain enzyme activity. Based on this concept (Figure 10), SUNGA is a useful method to explore the fundamental significance of N-glycans (such as in conformation, enzyme activity, stability, subcellular localization, secretion, etc.) by comparing amino acid sequences acquired at the N-glycosylation site in later evolved species and corresponding sequences in early species, which have no N-glycans.
|
The SUNGA method will likely be applicable to many fields of glycotechnology. Progress in the structural analysis of bioactive glycoproteins is still slow because of a general hindrance in crystallization caused by carbohydrate chains. To solve this issue, it is highly desirable to generate bioactive proteins that maintain proper conformation in the absence of sugars. Soon we will explore the general rules of the SUNGA method as they apply to various functional glycoproteins. We expect that the SUNGA method will be exceedingly useful in general for the structural and functional analyses of glycoproteins.
|
Methods and materials |
|---|
|
TopAbstractIntroductionResultsDiscussionMethods and materialsReferences |
|---|
Site-directed mutagenesis of mST3Gal-V
Each mutagenesis of mouse ST3Gal-V (mST3Gal-V) was performed using the QuickChange site-directed mutagenesis system (Stratagene, La Jolla, CA). Using wild-type pcDNA3.1 Zeo(+)-mST3Gal-V, each amino acid substitution was carried out following polymerase chain reaction (PCR) amplification. The PCR was performed in 50 µL of buffer (10 mM KCl, 10 mM (NH4)2SO4, 20 mM Tris–HCl [pH 8.8], 2 mM MgSO4, 0.1% TritonX-100, 0.1 mg/mL bovine serum albumin [BSA], and 10 mM dNTP) and 2.5 units of Pfu DNA polymerase. Each reaction was carried out using 125 ng of a primer specific for each mutation (Table III). PCRs were performed for 16 cycles (30 s at 95°C, 1 min at 55°C, and 12.5 min at 68°C). The PCR products were each treated with 10 units of DpnI endonuclease and were transfected into E. coli supercompetent cells. All the mST3Gal-V mutated plasmids were sequenced in both directions to confirm the presence of the incorporated mutation.
|
Plasmid construction
The pSU141 (TM-mST3Gal-V wild type) was constructed using pcDNA3.1 Zeo(+)-mST3Gal-V and the primers 5'-GGATCCGCTGAAGAATGTGACATGAAAAGAAT-3' and 5'-AAGCTTTCAGTGGATGCCGCCGCTGAGGTCC-3'. The resulting fragments were cloned into pGEM-T Easy (Promega, Madison, WI) to generate the pSU136, and the 0.99-kb BamHI–HindIII fragment of pSU136 was then cloned into the BamHI–HindIII site of pCold I DNA (TakaRa, Shiga, Japan), producing pSU141 (TM-mST3Gal-V wild type). The pSU142 (TM-mST3Gal-V-H117D-N180S-N224K-T336Q mutant) was constructed using pcDNA3.1 Zeo(+)-mST3Gal-V-H117D-N180S-N224K-T336Q mutant and the primers 5'-GGATCCGCTGAAGAATGTGACATGAAAAGAAT-3' and 5'-AAGCTTTCAGTGGATGCCG CCGCTGAGGTCC-3'. The resulting fragments were cloned into pGEM-T Easy to generate the pSU137, and the 0.99-kb BamHI–HindIII fragment of pSU137 was then cloned into the BamHI–HindIII site of pCold I DNA, producing pSU142 (TM-mST3Gal-V-H117D-N180S-N224K-T336Q mutant).
Cell culture and transfection
CHO cells were cultured in Ham’s F-12 medium (Sigma, St. Louis, MO) and HEK293 cells in Dulbecco’s modified Eagle’s medium (Sigma), each supplemented with 10% fetal bovine serum (complete medium). Cells were cultured at 37°C in a humidified 5% CO2 atmosphere. Cells were transfected using Lipofectamine 2000 (Invitrogen, Carlsbad, CA), according to the manufacturer’s instructions.
Inhibitor treatment
CHO cells were transiently transfected with mST3Gal-V. Six hours after transfection, the medium was replaced with fresh culture medium supplemented or not with 2.5 µg/mL tunicamycin (Sigma), 5 µg/mL kifnecine (Calbiochem, San Diego, CA), or 1 mM castanospermine (Calbiochem). Eighteen hours later, the cells were harvested.
Preparation of culture supernatants and cell lysates
Twenty-four hours after transfection with the mST3Gal-V cDNA, culture media were collected and subjected to centrifugation at 13,000 x g for 5 min. The supernatants were treated with 100 µL of a 0.2% deoxycholate solution (Sigma) for 10 min at 0°C, then an equal volume of 10% (w/v) TCA was added, and the mixture was incubated for another 20 min at 0°C. Protein precipitates were washed with acetone and suspended in 1x SDS sample buffer (62.5 mM Tris–HCl [pH 6.8], 2% SDS, 10% glycerol, 0.001% bromophenol blue (BPB), and 5% 2-mercaptoethanol). Cells attached to the culture plate were washed twice with ice-cold phosphate-buffered saline (PBS) and lysed with extraction buffer (50 mM Tris–Hcl [pH 7.4], 150 mM NaCl, 2 mM NaF, 1 mM ethylenediamine tetra-acetic acid, 1 mM ethylene glycol bis (2-aminoethyl ether)-tetra acetic acid, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, and 1x protease inhibitor cocktail (Complete; Roche Molecular Biochemicals, Mannheim, Germany). After a 10-min incubation at 0°C, cell lysates were centrifuged for 5 min at 15,000 x g to remove insoluble proteins. The supernatants were diluted with 5x sample buffer (312.5 mM Tris–HCl [pH 6.8], 10% SDS, 50% glycerol, 0.005% BPB, and 25% 2-mercaptoethanol) and boiled for 5 min. Endo H (New England Biolabs, Beverly, MA) or PNGase F (New England Biolabs) digestion was performed on the lysates at 37°C for 1 h, according to the manufacturer’s recommended procedure. All samples were analyzed by immunoblotting.
Immunoblotting
Immunoblots were performed on the cell lysates and culture supernatants prepared as described above. Protein concentrations were determined with a BCA protein assay kit (Pierce Chemical Company, Rockford, IL). Equal amounts of protein were separated by SDS–PAGE and transferred to an Immobilon PVDF membrane (Millipore, Bedford, MA). The membrane was then incubated for 1 h with a 1:1000 dilution of an anti-ST3Gal-V antibody that had been raised against a glutathione-S-transferase fusion protein encompassing the C-terminal 55 amino acids. After a wash, the membrane was incubated for 1 h with a 1:5000 dilution of horseradish peroxidase-conjugated donkey anti-rabbit IgG F(ab')2 fragment (Amersham Bioscience, Piscataway, NJ). Labeling was detected by the ELC detection method (Amersham Bioscience).
Assay of enzymatic activity
CHO cells were transfected with a plasmid vector containing wild-type or mutated mST3Gal-V. After 24 h, the cells were harvested and lysed with a cell suspension solution (15 mM sodium cacodylate [pH 6.2], 5% glycerol, 0.1% Lubrol, and 1x protease inhibitor cocktail) by sonication. After a 10-min incubation at 0°C, cell lysates were centrifuged for 5 min at 15,000 x g to remove insoluble proteins, and the supernatants were examined for enzymatic activity as described previously (Nara et al., 1994). Specifically, an acceptor, bovine milk lactosylceramide (0.02 µmol) (LacCer; Matreya, Pleasant Gap, PA) was dissolved in chloroform/methanol (1/1, v/v), dried under room temperature, and then suspended in 10 µL of 4x reaction mixture (400 mM sodium cacodylate [pH 6.2], 40 mM MgCl2, 1.6% Triton X-100) by sonication for 1 min. An aliquot (10 µL) of this suspension was incubated with the enzyme and 24 nmol CMP-NeuAc containing 375 nCi CMP-[14C]NeuAc (PerkinElmer, Boston, MA), in a final volume of 40 µL at 37°C. After 120 min, the products, including GM3, were partially purified on a Sep-Pak Plus C18 cartridge (Water Associates, Milford, MA) and then analyzed by HPTLC. The relative enzyme activities in Figures 2, 4, 6–
8 were calculated using the radioactivities associated with GM3, and the corresponding protein levels determined by immunoblotting with an anti-ST3Gal-V antibody after PNGaseF treatment of the sample. Data are presented as a percent relative to the activity of the untreated wild-type transfectant.
Alternatively, a nonradioisotope method was developed to measure the GM3-synthesizing ability of the ST3Gal-V proteins generated in E. coli. Partially purified ST3Gal-V proteins were incubated with 0.5 mM LacCer and 0.75 mM CMP-NeuAc at 37°C for 2 h. The products were separated by HPTLC and transblotted onto a PVDF membrane. Semi-quantitative immunological detection of GM3 was performed with an anti-GM3 antibody (M2590) and peroxidase-conjugated anti-IgM using ECL chemiluminescence.
TM-mST3Gal-V expression and purification in E. coli
Either pSU141 (His-TM-mST3Gal-V wild type) or pSU142 (His-TM-mST3Gal-V-H177D-N180S-N224K-T336Q mutant) was transfected into E. coil BL21 cells, which contain the chaperone (groES, groEL, and Tig) expression plasmid pG-Tf2. The transformed cells were grown at 37°C in LB (1% tryptone, 0.5% yeast extract, 1% NaCl [pH 7,2]) medium. At mid-log phase, the culture was shifted to 15°C, and 0.01 mM isopropyl I-b-D-thiogalactopyranoside was added to induce protein expression. After a 48-h incubation, cells were washed and suspended in PBS containing a 1x protease inhibitor cocktail. After sonication, cells were separated into soluble (S) and insoluble (I) fractions by centrifugation at 20,000 x g for 15 min. The soluble fractions were subjected to Ni-NTA-agarose chromatography. The protein bound on the Ni-NTA-agarose were washed with 50 and 100 mM imidazole and then eluted with 150 mM imidazole.
|
Abbreviations |
|---|
BPB, bromophenol blue; CHO, Chinese hamster ovary; HPTLC, high performance TLC; mST3Gal-V, mouse ST3Gal-V; PCR, polymerase chain reaction; PNGase F, peptide N-glycosidase F; PVDF, polyvinylidene difluoride; SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; SUNGA, substitution of N-glycan functions in glycosyltransferases by specific amino acids; TLC, thin-layer chromatography
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMethods and materialsReferences |
|---|
Baboval, T., Koul, O., and Smith, F.I. (2000) N-glycosylation site occupancy of rat alpha-1,3-fucosyltransferase IV and the effect of glycosylation on enzymatic activity. Biochim. Biophys. Acta, 1475, 383–389.[Medline]
Bieberich, E., MacKinnon, S., Silva, J., Li, D.D., Tencomnao, T., Irwin, L., Kapitonov, D., and Yu, R.K. (2002) Regulation of ganglioside biosynthesis by enzyme complex formation of glycosyltransferases. Biochemistry, 41, 11479–11487.[CrossRef][Medline]
Chen, C. and Colley, K.J. (2000) Minimal structural and glycosylation requirements for ST6Gal I activity and trafficking. Glycobiology, 10, 531–583.
Christensen, L.L., Bross, P., and Orntoft, T.F. (2000a) Glycosylation of the N-terminal potential N-glycosylation sites in the human alpha1,3-fucosyltransferase V and -VI (hFucTV and -VI). Glycoconj. J., 17, 859–865.[CrossRef][ISI][Medline]
Christensen, L.L., Jensen, U.B., Bross, P., and Orntoft, T.F. (2000b) The C-terminal N-glycosylation sites of the human alpha1,3/4-fucosyltransferase III, -V, and -VI (hFucTIII-V, adn -VI) are necessary for the expression of full enzyme activity. Glycobiology, 10, 931–939.
Colley, K.J. (1997) Golgi localization of glycosyltransferases: more questions than answers. Glycobiology, 7, 1–13.
Datta, A.K. and Paulson, J.C. (1995) The sialyltransferase "sialylmotif " participates in binding the donor substrate CMP-NeuAc. J. Biol. Chem., 270, 1497–1500.
Datta, A.K., Sinha, A., and Paulson, J.C. (1998) Mutation of the sialyltransferase S-sialylmotif alters the kinetics of the donor and acceptor substrates. J. Biol. Chem., 273, 9608–9614.
Eckhardt, M., Fewou, S.N., Ackermann, I., and Gieselmann, V. (2002) N-glycosylation is required for full enzymic activity of the murine galactosylceramide sulphotransferase. Biochem. J., 368, 317–324.[CrossRef][ISI][Medline]
Fast, D.G., Jamieson, J.C., and McCaffrey, G. (1993) The role of the carbohydrate chains of Gal beta-1,4-GlcNAc alpha 2,6-sialyltransferase for enzyme activity. Biochim. Biophys. Acta, 1202, 325–330.[CrossRef][Medline]
Haraguchi, M., Yamashiro, S., Furukawa, K., Takamiya, K., and Shiku, H. (1995) The effects of the site-directed removal of N-glycosylation sites from beta-1,4-N-acetylgalactosaminyltransferase on its function. Biochem. J., 312, 273–280.[Medline]
Jeanneau, C., Chazalet, V., Auge, C., Soumpasis, D.M., Harduin-Lepers, A., Delannoy, P., Imberty, A., and Breton, C. (2004) Structure-function analysis of the human sialyltransferase ST3Gal I: role of n-glycosylation and a novel conserved sialylmotif. J. Biol. Chem., 279, 13461–13468.
Kitazume, S., Tachida, Y., Oka, R., Shirotani, K., Saido, T.C., and Hashimoto, Y. (2001) Alzheimer’s beta-secretase, beta-site amyloid precursor protein-cleaving enzyme, is responsible for cleavage secretion of a Golgi-resident sialyltransferase. Proc. Natl. Acad. Sci. U. S. A., 98, 13554–13559.
Kitazume, S., Tachida, Y., Oka, R., Kotani, N., Ogawa, K., Suzuki, M., Dohmae, N., Takio, K., Saido, T.C., and Hashimoto, Y. (2003) Characterization of alpha 2,6-sialyltransferase cleavage by Alzheimer’s beta-secretase (BACE1). J. Biol. Chem., 278, 14865–14871.
Malissard, M., Borsig, L., Di Marco, S., Grutter, M.G., Kragl, U., Wandrey, C., and Berger, E.G. (1996) Recombinant soluble beta-1,4-galactosyltransferases expressed in Saccharomyces cerevisiae. Purification, characterization and comparison with human enzyme. Eur. J. Biochem., 239, 340–348.[ISI][Medline]
Martina, J.A., Daniotti, J.L., and Maccioni, H.J. (1998) Influence of N-glycosylation and N-glycan trimming on the activity and intracellular traffic of GD3 synthase. J. Biol. Chem., 273, 3725–3731.
Martina, J.A., Daniotti, J.L., and Maccioni, H.J. (2000) GM1 synthase depends on N-glycosylation for enzyme activity and trafficking to the Golgi complex. Neurochem. Res., 25, 725–731.[CrossRef][ISI][Medline]
Melkerson-Watson, L.J. and Sweeley, C.C. (1991) Purification to apparent homogeneity by immunoaffinity chromatography and partial characterization of the GM3 ganglioside-forming enzyme, CMP-sialic acid:lactosylceramide alpha 2,3-sialyltransferase (SAT-1), from rat liver Golgi. J. Biol. Chem., 266, 4448–4457.
Minowa, M.T., Oguri, S., Yoshida, A., Hara, T., Iwamatsu, A., Ikenaga, H., and Takeuchi, M. (1998) cDNA cloning and expression of bovine UDP-N-acetylglucosamine: alpha1,3-D-mannoside beta1,4-N-acetylglucosaminyltransferase IV. J. Biol. Chem., 273, 11556–11562.
Morais, V.A., Costa, M.T., and Costa, J. (2003) N-glycosylation of recombinant human fucosyltransferase III is required for its in vivo folding in mammalian and insect cells. Biochim. Biophys. Acta, 1619, 133–138.[Medline]
Muhlenhoff, M., Manegold, A., Windfuhr, M., Gotza, B., and Gerardy-Schahn, R. (2001) The impact of N-glycosylation on the functions of polysialyltransferases. J. Biol. Chem., 276, 34066–34073.
Nagai, K., Ihara, Y., Wada, Y., and Taniguchi, N. (1997) N-glycosylation is requisite for the enzyme activity and Golgi retention of N-acetylglucosaminyltransferase III. Glycobiology, 7, 769–776.
Nara, K., Watanabe, Y., Maruyama, K., Kasahara, K., Nagai, Y., and Sanai, Y. (1994) Expression cloning of a CMP-NeuAc: NeuAc alpha 2-3Gal beta 1-4Glc beta 1–1?Cer alpha 2,8-sialyltransferase (GD3 synthase) from human melanoma cells. Proc. Natl. Acad. Sci. U. S. A., 91, 7952–7956.
Thompson, J.D., Higgins, D.G., and Gibson, T.J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res., 22, 4673–4680.
Trombetta, E.S. (2003) The contribution of N-glycans and their processing in the endoplasmic reticulum to glycoprotein biosynthesis. Glycobiology, 13, 77R–91R.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Uemura, F. Feng, M. Kume, K. Yamada, K. Kabayama, S.-I. Nishimura, Y. Igarashi, and J.-I. Inokuchi Cell growth arrest by sialic acid clusters in ganglioside GM3 mimetic polymers Glycobiology, June 1, 2007; 17(6): 568 - 577. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||