HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 40, 31407-31413, October 6, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/40/31407    most recent M004524200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Garinot-Schneider, C. Articles by Geremia, R. A. Search for Related Content PubMed PubMed Citation Articles by Garinot-Schneider, C. Articles by Geremia, R. A. Social Bookmarking                      What's this?

Identification of Essential Amino Acid Residues in the Sinorhizobium meliloti Glucosyltransferase ExoM^*

Carole Garinot-Schneider, Annemarie C. Lellouch, and Roberto A. Geremia^§

From the Centre de Recherches sur les Macromolécules Végétales, CNRS, Affiliated with the Joseph Fourier University, BP 53X, Grenoble 38041, Cedex 9, France

Received for publication, May 25, 2000, and in revised form, July 24, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ExoM is a (1-4)-glucosyltransferase involved in the assembly of the repeat unit of the exopolysaccharide succinoglycan from Sinorhizobium meliloti. By comparing the sequence of ExoM to those of other members of the Pfam Glyco Domain 2 family, most notably SpsA (Bacillus subtilis) for whom the three-dimensional structure has been resolved, three potentially important aspartic acid residues of ExoM were identified. Single substitutions of each of the Asp amino acids at positions 44, 96, and 187 with Ala resulted in the loss of mutant recombinant protein activity in vitro as well as the loss of succinoglycan production in an in vivo rescue assay. Mutants harboring Glu instead of Asp-44 or Asp-96 possessed no in vitro activity but could restore succinoglycan production in vivo. However, replacement of Asp-187 with Glu completely inactivated ExoM as judged by both the in vitro and in vivo assays. These results indicate that Asp-44, Asp-96, and Asp-187 are essential for the activity of ExoM. Furthermore, these data are consistent with the functions proposed for each of the analogous aspartic acids of SpsA based on the SpsA-UDP structure, namely, that Asp-44 and Asp-96 are involved in UDP substrate binding and that Asp-187 is the catalytic base in the glycosyltransferase reaction.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Bacterial polysaccharides composed of repeat unit oligosaccharides are a major component of bacteria cell surfaces and are involved in many biological processes such as adhesion, immune evasion, and plant-bacteria symbiosis. This group of polysaccharides includes the O-antigen portion of lipopolysaccharides, capsular polysaccharides, and extracellular polysaccharides (EPS).¹ Each of these structures is biosynthesized from an oligosaccharide repeat unit that has been assembled on a polyisoprenyl-pyrophosphate lipid anchor, most likely at the cytoplasmic face of the inner membrane, by the sequential action of specific glycosyltransferases. The oligosaccharide is subsequently translocated, polymerized, and either secreted (EPS) or ligated to another membrane anchor as in the case of lipopolysaccharides and capsular polysaccharides (for a review see Ref. 1). Bacterial cell surface-associated polysaccharides are extremely diverse in their composition. This diversity is reflected in the large number of genes thought to encode glycosyltransferases that have been identified to date. However, both the details of the catalytic mechanism and the structural features dictating the substrate specificity for this broad class of enzymes remain poorly defined. A greater understanding of bacterial glycosyltransferase activity is necessary for biotechnological applications such as chemoenzymatic oligosaccharide production or polysaccharide engineering.

One example of a bacterial exopolysaccharide, whose biological function has been relatively well defined, is succinoglycan or EPS I produced by the nitrogen-fixing bacterium Sinorhizobium meliloti. Succinoglycan, a polysaccharide composed of an octasaccharide repeat unit (2), has been shown to be necessary for the invasion of the host root nodules during the establishment of the symbiotic relationship with leguminous plants (3, 4). Of the 26 genes known to be required for succinoglycan biosynthesis (5-10), at least 6 are thought to be glycosyltransferases involved in the assembly of the polyprenyl-pyrophosphate octasaccharide (6). We have recently demonstrated that a recombinant form of one of these six putative glycosyltransferases, ExoM, is capable of transferring glucose from UDP-glucose to the native polyprenyl-pyrophosphate trisaccharide substrate (Glc1-4Glc1-3GalP-P-lipid) in vitro (11), thus confirming its function as a nonprocessive (1-4)- glucosyltransferase.

Several recent glycosyltransferase classification schemes, notably the Pfam Protein Family Database of Batemann et al. (12), have shown that the N-terminal 100-amino acid region of ExoM contains significant sequence similarity with a large family of both prokaryotic and eukaryotic glycosyltransferases that utilize a nucleotide diphospho-hexose sugar donor (12-15). With a few exceptions, the majority of these enzymes function with inversion of the anomeric configuration. In accordance with the Pfam Database, the N-terminal region will hereafter be referred to as the glycosyltransferase-2 domain (GT2Domain) and ExoM will be referred to as belonging to the GT2Domain family.

The recent publication of the crystal structure of SpsA of Bacillus subtilis bound to UDP provides the first structural information for GT2Domain family glycosyltransferases (16). SpsA is a UDP binding protein of unknown activity involved in spore coat formation but which has sequence similarity to members of the GT2Domain family. In the SpsA-UDP structure, two of the 6 residues (Asp-39 and Asp-99) involved in the UDP binding exist in clusters of amino acids that are highly conserved among members of the GT2Domain family (16). In addition, 2 Asp residues, Asp-158 and Asp-191, were found bound to a free glycerol molecule, which was in turn bound to the distal pyrophosphate moiety of UDP via Mn²⁺, in a fashion that would mimic the binding of the sugar acceptor. Because of the appropriate geometric relationship between Asp-191, the Mn²⁺ ion, and the UDP-phosphate and the fact that only Asp-191 is conserved among family members, Charnock and Davies (16) speculated that this Asp residue might function as the catalytic base in the transferase reaction. However, because no functional assay exists for the SpsA activity, this very interesting hypothesis was not tested.

To further our understanding of the ExoM reaction mechanism, we have re-analyzed the sequence comparisons between SpsA and ExoM to verify that the ExoM Asp-44 and Asp-96 residues correspond to the nucleotide binding aspartic acid residues SpsA Asp-39 and Asp-99, respectively. In addition we have identified ExoM Asp-187 as being in the homologous location as the potential catalytic base, SpsA Asp-191. We found that replacement of each of the three positions by alanine (D44A, D96A, and the newly identified D187A) abolished glucosyltransferase activity in vitro and resulted in the loss of the ability to restore succinoglycan production in an in vivo rescue assay. However, for the conservative mutations aspartic acid to glutamic acid, complete loss of activity both in the in vitro and the in vivo assays was observed only for D187E, a finding consistent with this residue playing a role in catalysis. Interestingly, we found that the substitution D98A of the DXD motif (17-19) resulted only in a partial loss of activity, suggesting that in ExoM this residue does not play an important functional role.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Plasmids, Bacterial Strains, and Media-- See Table I for a description of the plasmids and genotypes of all strains employed. Cloning work was done using Escherichia coli strains XL1-Blue and JM109 cells. Recombinant proteins were expressed in E. coli BL21(DE3)/pLysS. E. coli was grown in Luria-Bertani (LB) medium at 37 °C. S. meliloti was grown in LB medium supplemented with 2.5 mM MgSO₄ and 2.5 mM CaCl₂ at 30 °C. For exopolysaccharide production, S. meliloti was grown in modified M9 medium supplemented with 1% mannitol, 0.1% glutamate, 0.75 mM MgSO₄, and 0.75 mM CaCl₂. Antibiotics were used at the following concentrations: kanamycin, 30 µg/ml; chloramphenicol, 34 µg/ml; neomycin, 200 µg/ml; gentamicin, 20 µg/ml; spectinomycin, 50 µg/ml; trimethoprim, 500 µg/ml; tetracycline, 12.5 µg/ml; streptomycin, 500 µg/ml. Oligonucleotide primers for PCR were purchased from Cybergène (Saint-Malo, France). Restriction enzymes and T4 ligase were from New England BioLabs (Beverly, MA). Pfu polymerase was from Stratagene (La Jolla, CA). UDP-[U-¹⁴C]glucose was from Amersham Pharmacia Biotech (Les Ulis, France). The pET29b expression vector and S-tag Western blotting kit were from Novagen (Madison, WI).

Sequence Analysis-- Sequences were retrieved from the ExPASy server on the Web, the accession numbers at the Swiss-Prot or TrEMBL data bases were: P33695 (ExoM), P39621 (SpsA), P33691 (ExoA), P33700 (ExoU), P33697 (ExoO), P33702 (ExoW), Q9XBL5 (AceP). The Pfam GT2Domain family was accessed on the Web. BLAST (20) was performed at the NCBI server using advanced BLAST default settings, except for the expected item, which was set to 1000. HCA plots (21) were obtained at the DrawHCA server. The secondary structure predictions were created by accessing the Jpred2 server (22) and using ExoM sequence as query.

Site-directed Mutagenesis-- Site-directed mutagenesis of the exoM gene was carried out by PCR using the "megaprimer" method (23), with pALC1 as the template. Appropriated mutations were generated by using synthetic oligonucleotides specifying the desired change. A mutagenic primer was used in a PCR reaction with the universal primer or the reverse primer, whichever was appropriate. After isolation of the first product with the Qiaex kit (Qiagen, GmbH, Germany), a second PCR reaction was performed using the first PCR product as a megaprimer together with either the reverse or universal primer, whichever was appropriate. The second PCR product containing the desired mutation was then purified and restricted with SacI and HindIII. The 535-bp SacI-HindIII fragment containing the desired mutation was then ligated with the SacI-HindIII 5794-bp fragment of pLGC2. This construction encodes a Stag-ExoM fusion protein. Mutated DNA was fully sequenced (Genome Express SA, Grenoble, France) to confirm the PCR fidelity.

Expression of the Mutated ExoM Proteins-- Culture and expression of ExoM in E. coli BL21(DE3)/pLysS harboring plasmid pLGC2 with different mutations was performed as described previously (11). Briefly, isopropyl--D-thiogalactopyranoside, 0.4 mM, was added to 50-ml cultures at A₆₀₀ of 0.5 and cultured for 2 h at 37 °C. The cells were recovered by centrifugation, washed, and resuspended in 70 mM Tris-HCl/10 mM EDTA (pH 8.2) at 40 OD equivalents/ml (approximately 2 ml). Cells were disrupted by two passages through a French press (18,000 p.s.i.). Unbroken cells and inclusion bodies were removed by a 15-min centrifugation (5,000 × g) at 4 °C. The 5,000 × g supernatant was centrifuged for 45 min at 50,000 × g to obtain the membrane fraction. This fraction was resuspended in the same volume of buffer and used immediately to measure the activity and then stored at 20 °C.

ExoM Activity-- ExoM activity was determined as described previously (11). Briefly, membrane fractions containing equivalent amounts of ExoM, as judged by Coomassie Blue-staining SDS-PAGE, were incubated with freshly prepared glycolipid acceptor in the presence of UDP-[U-¹⁴C]glucose. The glycolipid acceptor was extracted, and incorporated radioactivity was measured by liquid scintillation. When comparing wild type and mutant forms of ExoM, radioactivity recovered from BL21(DE3)/pLysS E. coli strain, harboring pET29b, was taken as background and subtracted from reported values. The results reported are the average of at least four assays. The oligosaccharide was released from the pyrophosphate-lipid by mild acid hydrolysis and analyzed by thin-layer chromatography (TLC). 10,000 cpm of each sample supplemented with 50 µg of malto-oligosaccharides (dimer to pentamer) was spotted in a TLC plate (Silica Gel 60, 0.25 mm, Sigma). Malto-oligosaccharides were used as standards. The TLC plate was developed twice in 1-propanol:nitromethane:water (50:20:20, v/v). Standards were revealed by charring after treatment with 5% H₂SO₄ in ethanol. Photographic films with a Transcreen LE (Hyperfilm max, Amersham Pharmacia Biotech) were exposed 24-72 h before development.

In Vivo Activity of ExoM Mutants-- The 1120-bp XbaI fragment containing the promoterless S-tag ExoM fusion open reading frame with the ribosomal binding site was cloned into the XbaI site of pBBR1MCS-3 under the control of P_Lac promoter-generating pCS3. The different S. meliloti strains containing mutated exoM forms were obtained by mating S. meliloti 1021 exoM::Tn5 and E. coli S-17 cells harboring the appropriate pCS3.

Quantification of Succinoglycan Production-- EPS-I production by S. meliloti mutants was detected by the Calcofluor white staining method as described previously (24). For succinoglycan quantification, S. meliloti strains were grown for 6 days in 25 ml of M9-glutamate-mannitol, CaCl₂, MgSO₄ medium (described above) at 30 °C.

Exopolysaccharide was precipitated from the culture supernatant by addition of 1 M NaCl in the initial volume and 2 volumes of ethanol. Sugars were quantified by the phenol method (25). EPS-I production of S. meliloti harboring pBBR-MCS-3 was taken as background and subtracted from reported values. Results for succinoglycan production represent the average of at least two experiments.

Western Blotting on Proteins Extracted from S. meliloti-- Total protein extracts (equivalent OD) from 2-day cultures of S. meliloti, containing mutant ExoM, were loaded onto a 12% SDS-PAGE gel. The presence of the N-terminal S-tag epitope fusion was confirmed by transfer of the gel to a polyvinylidene difluoride membrane and subsequently treated with the Novagen S-tag Western blotting kit with alkaline phosphatase S-protein conjugate following the manufacturer's instructions.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Identification of Conserved Asp Residue in ExoM-- ExoM is a 309-amino acid glucosyltransferase with two predicted C-terminal transmembrane helices (Fig. 1A). In the work of Saxena et al. (14), where a limited number of inverting glycosyltransferases, including the S. meliloti Exo glucosyltransferases, were analyzed by HCA, two strictly conserved aspartic acid residues were identified in an N-terminal region, which was then named Domain A. In this scheme, ExoM residues Asp-44 and Asp-96 were identified as strictly conserved residues, which may be of functional importance. The recent Pfam Database (12) defines a larger glycosyltransferase family, the GT2Domain family, which comprises 348 proteins, including ExoM. The defining feature of this family is an N-terminal region of sequence similarity, the GT2Domain, which is equivalent to the Domain A defined by Saxena et al. (14). Like ExoM, SpsA is a member of this larger family. In the crystal structure of SpsA, 3 aspartic acid residues are involved in UDP binding (16), 2 of which, Asp-39 and Asp-99, are found in conserved regions within the GT2Domain. SpsA Asp-39 binds the uracil ring, whereas SpsA Asp-99 binds the pyrophosphate via one molecule of Mn²⁺. We will refer to these conserved regions within the GT2Domain as regions I and II. To determine whether the UDP binding aspartic acid residues of SpsA (Asp-39 and Asp-99) correspond to the previously described conserved aspartic acid residues in ExoM (Asp-44 and Asp-96), we have re-examined the sequence similarities between these two proteins using several different methods, such as BLAST, HCA, and secondary structure prediction programs. In addition, we wished to identify an aspartic acid in ExoM, which might correspond to the SpsA Asp-191, found outside the GT2Domain in what we will term region III and which has been proposed to be a catalytic base in the transferase reaction.

View larger version (71K): [in this window] [in a new window]   Fig. 1.   Conserved regions of ExoM. A, location of conserved aspartic residues in three regions in ExoM: GT2Domain is in light gray, regions I, II, and III in black, and potential transmembrane domains (TM) in dark gray. The mutated amino acids are indicated. B, HCA alignment of ExoM, SpsA, and AceP. In the HCA plots, the protein sequences are written on a duplicated -helical net, and the contour of clusters of hydrophobic residues is automatically drawn. The standard one-letter code for amino acids is used except for proline, glycine, serine, and threonine, which are represented by , , , and , respectively. The dark circles show conserved Asp. In AceP, bold circles show the Asp residue equivalent to ExoM Asp-96, whereas the dark gray circle indicates the residue equivalent to Asp-99 of SpsA. The predicted secondary structure of ExoM, SpsA, and AceP using the software Jetnetpred from the Web site Jpred2, is shown, the arrows indicate -sheets, and twisted lines represent -helixes. For SpsA, a predicted secondary structure element absent in the structure was omitted from the figure.

Despite their belonging to the same GT2Domain family, ExoM and SpsA share low sequence similarity. Therefore, we decided to compare each protein to a third protein, which has a greater degree of sequence similarity to ExoM and SpsA than ExoM and SpsA have to each other. Using BLAST searches, we identified several proteins that matched both the ExoM or SpsA queries. Among the common matches, we have selected AceP (Acetobacter xylinus) for the comparison, because this protein has been confirmed biochemically to be a 1-6 glucosyltransferase (26). We then used HCA and secondary structure prediction to identify aspartic acids conserved among these three proteins. The equivalent amino acids in the other S. meliloti Exo glucosyltransferases were located in a similar way (Fig. 2). In region I of the GT2 domain, both HCA and secondary structure prediction programs predict that ExoM Asp-44 and AceP Asp-42 are located after a -sheet (Fig. 1B). This result is compatible with the position of Asp-39 in the SpsA crystal structure. In region II, the interpretation of the HCA analysis is more ambiguous, because the number of residues separating the relevant aspartic acids (SpsA Asp-99 and ExoM Asp-96) from the neighboring hydrophobic cluster is different. However, both the HCA and the secondary structure prediction of ExoM predict that Asp-96 is the first amino acid in a loop following a -sheet (Fig. 1B). Interestingly, SpsA Asp-99 is found in exactly such a position in the SpsA crystal structure. Although the same considerations are also valid for the other Exo glucosyltransferases (Fig. 2), it is not possible to unambiguously identify this position in AceP, because HCA analysis provides two possible interpretations. AceP Asp-95 and Asp-96 are located at the same distance from the preceding hydrophobic cluster as SpsA Asp-98 and Asp-99; whereas AceP Asp-93 is in the same position with respect to the nearest hydrophobic cluster as is ExoM Asp-96 (Fig. 1B). Finally, by the same methods we have identified ExoM Asp-187 and AceP Asp-193 as being equivalent to SpsA Asp-191. HCA and secondary structure analysis predict that these ExoM and AceP residues are located just before an -helix (Fig. 1B), as is SpsA Asp-191 in the SpsA crystal structure. We therefore postulate that ExoM Asp-187 is also a potential catalytic residue in the transfer reaction. To test these hypothesis, we undertook site-directed mutagenesis of the conserved aspartic acids in regions I, II, and III.

View larger version (48K): [in this window] [in a new window]   Fig. 2.   Conserved regions in Exo glucosyltransferases. Manual alignments based on HCA analysis of regions I, II, and III in Exo glycosyltransferases showing conserved Asp residues (in boldface).

Characterization of Mutant Forms of ExoM Expressed in E. coli-- Each conserved aspartic acid residue within regions I, II, and III, as well as certain neighboring amino acids (Fig. 1A), were mutated to alanine, and these ExoM mutants were subsequently expressed as recombinant proteins in E. coli with an N-terminal S-tag fusion to facilitate their detection. 14 of the 16 mutants were expressed in E. coli strain BL21(DE3)/pLysS at levels equivalent to the wild type ExoM. However, the two mutants N45A and D96E were not detected on Coomassie Blue-stained SDS-PAGE gels (Fig. 3). Subsequent Western blotting against the S-tag revealed that these mutant proteins are degraded when expressed in E. coli (data not shown). The glucosyltransferase activity of each mutant was determined using total membrane fractions containing equivalent quantities of ExoM protein. The amount of ¹⁴C-labeled Glc transferred from UDP-[¹⁴C]Glc to the native rhizobial glycolipid acceptor (see "Materials and Methods") was quantified. The activity of the mutants was expressed as a percentage of the activity of the wild-type ExoM (Table II). In region I, the substitution of the highly conserved residue Asp-44 with alanine resulted in a complete loss of the activity. Similar data were obtained for the conservative change D44E. Mutations of the neighboring amino acids, for example V42I and D46A, resulted in little if any loss of activity. Ala-43 was replaced with Asp, because in the other Exo glucosyltransferases Asp occupies this position. However the A43D mutation resulted in a complete loss of activity.

View larger version (37K): [in this window] [in a new window]   Fig. 3.   Expression of the mutated versions of ExoM. Proteins were produced by E. coli BL21(DE3)/pLysS harboring pEt29b () or mutant versions of pLGC2 as described under "Materials and Methods." SDS-12% PAGE gel of the 50,000 × g pellet fraction of the different mutants is shown. Proteins were stained with Coomassie Brilliant Blue. The arrows indicate the position of the ExoM-S-tag fusion protein (345 amino acid residues, molecular mass 37.5 kDa). The lower molecular weight bands are endogenous E. coli proteins, and are present in the absence of ExoM (lane pET).

In region II, surrounding the highly conserved Asp-96, each of the residues from Leu-95 to Glu-99 were substituted by alanine. Each of the mutants L95A, D97A, and D98A displayed reduced activity in vitro (around 30%), whereas the mutant E99A retained 100% activity. Only the mutation of the conserved Asp, D96A, resulted in complete loss of activity. Unfortunately, the mutant protein containing the conservative D96E mutation was not stably expressed in E. coli (Fig. 3) but, as will be presented below, was stably expressed in S. meliloti. Asp-96 and Asp-98 together comprise the DXD motif, which has been shown to be essential for activity in other -glycosyltransferases (17). Our data suggest that, in vitro, only Asp-96 is critical for activity. The activity of the mutant proteins D97A and D98A was found to decrease after storage at 20 °C, as compared with the wild type ExoM protein, suggesting that these residues are involved in structural stability.

In region III, the replacement of Asp-187 by alanine or by glutamate resulted in complete loss of in vitro activity. Finally, to verify that the radioactivity incorporated into the glycolipid extract during the assay of each of the partially active mutants (V42I, D46A, F94V, L95A, D97A, D98A, and E99A) corresponds to the expected tetrasaccharide product, we have analyzed the reaction products by TLC. Each of the reaction products was found to comigrate with the authentic ExoM product (Fig. 4).

View larger version (45K): [in this window] [in a new window]   Fig. 4.   TLC of oligosaccharides formed in vitro by the action of the active mutant forms of ExoM. The oligosaccharides formed by the ExoM-S-tag fusion proteins were obtained as described under "Materials and Methods," separated by TLC, and located by fluorography. MBR and OBR are the oligosaccharides accumulated by S. meliloti exoMexoBexoR (Rm8278) and exoOexoBexoR (Rm10003) strains, respectively. G2, G3, and G4 indicates maltose, maltotriose, and maltotetraose, respectively.

Effect of the Amino Acids Substitution of ExoM on Succinoglycan Production in Vivo-- To further assess the role of each of the mutations, the wild-type and mutant forms of S-tag-ExoM fusion were introduced into a strain of S. meliloti in which exoM has been interrupted and therefore produces no succinoglycan. Expression of ExoM-S-tag fusion protein in the exoM S. meliloti strain restores succinoglycan production (Table II). This in vivo complementation assay is more sensitive than the in vitro assay, because lower levels of enzyme activity, which are yet physiologically significant, may be detected.

Each of the mutant ExoM-S-tag fusions was cloned into the broad host vector pBBRMCS-3 and introduced into the S. meliloti strain by bi-parental mating. Blotting against the S-tag showed that all the mutants except D46A and N45A were stably expressed in S. meliloti (Fig. 5). The amount of succinoglycan produced by each transconjugate strain was determined. The relative amount of succinoglycan produced by each of the mutant forms of ExoM as compared with the wild type was taken as a measure of the in vivo activity (Table II). Each of the mutant forms showing in vitro activity, except D46A, which was not expressed (Fig. 5), was capable of restoring succinoglycan production to between 60 and 180% of wild type production. The substitutions of the conserved aspartic acids, D44A, D96A, and D187A, which were expressed in E. coli but inactive in vitro, were expressed in S. meliloti and were also not able to restore succinoglycan production in vivo. However, of the conservative mutations D44E, D96E, and D187E, only D187E was completely incapable of restoring succinoglycan production. S. meliloti, harboring ExoM D44E and ExoM D96E, was able to produce 77 and 41% of the wild type succinoglycan level, respectively, demonstrating that these conservative mutations do not completely inactivate the protein.

View larger version (40K): [in this window] [in a new window]   Fig. 5.   Western blots using S-tag detection of all the ExoM mutants extracted from S. meliloti. S. meliloti strains harboring the different ExoM forms were obtained as described under "Materials and Methods," separated by SDS-PAGE, and transferred to a polyvinylidene difluoride membrane. The ExoM-S-tag fusion proteins were located using S-protein alkaline phosphatase conjugate.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this study we have employed two separate functional assays to identify three aspartic acid residues, Asp-44, Asp-96, and Asp-187, that are involved in the activity of the (1-4)-glucosyltransferase ExoM. These results represent a first step toward the understanding of ExoM's mechanism. A general mechanism for glycosyl transfer has been inferred from the extensive biochemical and crystallographic study of glycosyl hydrolases. Inverting glycosyl transfer is believed to proceed via a direct nucleophilic attack of the UDP-sugar anomeric carbon by the acceptor hydroxyl group (14). Such a mechanism implies the participation of a single acidic amino acid residue, such as Asp or Glu, that would catalyze the reaction by activating the attacking hydroxyl group. No sequence comparison study or functional analysis of site-directed mutants of any member of the GT2Domain family has yet led to the identification of this catalytic residue. The recent publication of the three-dimensional structure of SpsA provided the first real indication of where this residue might be located (16).

To date, the crystal structures of only three glycosyltransferases have been solved; the phage T4 -glucosyltransferase (27), the bovine 1,4-galactosyltransferase T1 (28), and the nucleotide diphospho-sugar transferase SpsA (16). The structure of each of these enzymes in complex with UDP has also been solved, thus providing a clear definition of the UDP binding interactions. In the case of the bovine 1,4-galactosyltransferase T1, which does not belong to the GT2Domain family, a highly conserved cluster of acidic residues, distinct from the DXD motif region, is found in the substrate binding pocket (28). Site-directed mutagenesis studies have demonstrated that alteration of certain residues within the cluster, notably Asp-318 and Asp-320, result in a dramatic reduction of the transferase activity without significant alteration of the substrate binding (29). These results together suggest that these acidic residues are directly involved in catalysis, however, because neither the galactose portion of the UDP-Gal donor nor the acceptor substrate are visible in the crystal structure, no further conclusions about the function of these residues can be drawn (28, 29). The fortuitous definition of a glycerol molecule in the SpsA structure allowed previous authors to speculate about possible acceptor substrate binding interactions, notably the presence of SpsA Asp-191, which they postulate was correctly positioned to serve as a catalytic base in the transferase reaction (16). Although SpsA clearly binds UDP and has sequence similarity to other members of the GT2Domain family such as ExoM, the exact transferase reaction catalyzed by SpsA is unknown, and therefore, this hypothesis has not been substantiated experimentally. A careful comparison of ExoM and SpsA sequences using BLAST, HCA, and secondary structure analysis allowed us to identify ExoM Asp-187, which we now believe to be analogous to the SpsA Asp-191. Replacement of Asp-187 by either Ala or Glu abolished the activity of ExoM, as judged by the in vivo (glucosyl-transfer) or the in vitro (succinoglycan recovery) assays. These results clearly demonstrate that ExoM Asp-187 is critical for the glycosyltransferase activity. In the case of inverting glycosyl hydrolases, it has been shown that the catalytic Asp residue could not be changed to Glu, because the distance is critical for a good position of the amino acid to act in the catalysis (30). Therefore, the fact that altering either the charge or the side chain length completely abolishes activity, strongly supports the proposal of Charnock and Davies that SpsA Asp-191, and thus ExoM Asp-187, is the catalytic base in this glycosyltranferase reaction.

We were also able to confirm that 2 residues of ExoM, Asp-44 and Asp-96, identified in previous sequence comparison studies (14), correspond to the SpsA residues Asp-39 and Asp-99. Each of these two highly conserved Asp residues in ExoM, as well as certain neighboring residues, were substituted with Ala. Only the mutations concerning the sites of interest, D44A and D96A, were found to be inactive in both the in vitro (glucosyl-transfer) and the in vivo (succinoglycan recovery) assays. In testing the conservative mutants D44E and D96E, no in vitro activity was detected, however, very significant succinoglycan production was recovered when these mutants were expressed in the exoM S. meliloti strain. Clearly, both the D44E and D96E mutants retain physiologically significant levels of activity that was not detected in the less sensitive in vitro assay. These results, which indicate that the presence of a charged amino acid, but not the side chain length, is critical for activity, are consistent with the idea that ExoM Asp-44 and Asp- 96 are involved in UDP binding in a manner similar to their SpsA counterparts Asp-39 and Asp-99. These findings also emphasize the fact that, in the absence of a very sensitive in vitro assay, it is important to have a complementary measure of the enzyme activity.

A consensus sequence, called the DXD motif has been identified in a wide variety of glycosyltransferases belonging to different families (17-19), including the GT2Domain family. The ExoM sequence ⁹⁶DDD⁹⁸ corresponds to this motif. In SpsA, the sequence ⁹⁷TDD⁹⁹ has been suggested to be a DXD motif, albeit in a modified form (31). Site-directed mutagenesis studies of several different glycosyltransferases, notably the yeast 1,3-mannosyltransferase MNN1 (17), the bacterial bifunctional glycosyltransferase KfiC (32), and the UDP-GalNAc polypeptide N-acetyl-galactosaminyltransferase (33), have shown that the residues at the first and third position of this motif are essential for glycosyltransferase activity. Furthermore, site-directed mutagenesis and biochemical labeling experiments of the clostridial toxin glucosaminyltransferases (18) as well as the crystal structures of the bovine GalTI and SpsA in complex with UDP (16, 28), reveal that the DXD motif is directly involved in nucleotide-sugar binding. In SpsA, the nucleotide sugar-binding sequence is in fact ⁹⁷TDD⁹⁹. SpsA Asp-99, the last residue of the motif, interacts through Mn²⁺ with phosphate, whereas Asp-98 interacts with the ribose group. It is not known whether the Mn²⁺ is required for activity, nor is the relative functional importance of the binding interactions of Asp-99 versus Asp-98 known. Based on atomic distances, the interaction Asp-99-Mn²⁺ is predicted to be stronger than Asp-98-ribose. In the equivalent region of ExoM, where DXD is ⁹⁶DDD⁹⁸, substitution of Asp-96 abolishes enzymatic activity, whereas substitution of either Asp-97 or Asp-98 by Ala results only in reduced activity. Because ExoM also requires a divalent cation to function, Mg²⁺ in preference to Mn²⁺,² it is tempting to speculate that Asp-96 could also bind to the phosphate via a divalent cation and possibly stabilize the nucleotide diphosphate-leaving group during the nucleophilic attack. Asp-97 and/or Asp-98 may interact with the ribose moiety. If so, the function of these amino acids with regard to their position in the consensus sequence is inverted when compared with that of SpsA, but in the same order when compared with relative position and function of the residues of the DXD motif within the bovine (1-4)-galactosyltransferase structure. Pfam alignments of the DXD region of the GT2Domain of many GT2 family members reveals that this region is often very rich in Asp residues, making it very difficult to assign functions based on sequence alignments. This is particularly true in the case of AceP and KfiC. Several studies have predicted that the DXD motif is located in a loop (14, 33), which is indeed the case in the bovine -GalT and the SpsA crystal structures. HCA and secondary structure analyses also predict that ExoM Asp-96 should be in a loop. It is clear that the DXD motif exists in most glycosyltransferases, and all examples shown to date implicate this sequence in nucleotide sugar binding. However, the cases studied so far suggest that the interactions between this consensus sequence and the nucleotide sugar substrate could vary from enzyme to enzyme. Given that the motif is found in a loop that can assume many different conformations and that the motif is often surrounded by other acidic residues that could also interact with the substrate, it is not surprising that this motif may be able to assume different modes of interaction. Further crystallographic and enzymological study of glycosyltransferases will be necessary to resolve this point. The present analysis shows biochemical evidence that 3 Asp residues are involved in the activity of ExoM and has allowed the potential identification of the catalytic residue Asp-187. The results obtained are consistent with the role of the essential Asp identified in the structure of SpsA.

	ACKNOWLEDGEMENTS

We thank Valerie Chazalet for her expert technical assistance.

	FOOTNOTES

^* This work was supported in part by the program "Physique et chimie du vivant" (CNRS).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of a postdoctoral fellowship from the "Société de secours des Amis des Sciences."

^§ To whom correspondence should be addressed: CERMAV/CNRS BP 53. Grenoble cedex 9, France. Tel.: 33-04-76-03-76-47; Fax: 33-04-76-54-72-03; E-mail: roberto.geremia@cermav.cnrs.fr.

Published, JBC Papers in Press, July 24, 2000, DOI 10.1074/jbc.M004524200

² C. Garinot-Schneider, unpublished result.

	ABBREVIATIONS

The abbreviations used are: EPS, extracellular polysaccharide; GT2Domain, glycosyltransferase-2 domain; HCA, hydrophobic cluster analysis; PCR, polymerase chain reaction; bp, base pair(s); PAGE, polyacrylamide gel electrophoresis; TLC, thin layer chromatography.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				K. M. Blair, L. Turner, J. T. Winkelman, H. C. Berg, and D. B. Kearns A Molecular Clutch Disables Flagella in the Bacillus subtilis Biofilm Science, June 20, 2008; 320(5883): 1636 - 1638. [Abstract] [Full Text] [PDF]

				L. L. Oglesby, S. Jain, and D. E. Ohman Membrane topology and roles of Pseudomonas aeruginosa Alg8 and Alg44 in alginate polymerization Microbiology, June 1, 2008; 154(6): 1605 - 1615. [Abstract] [Full Text] [PDF]

				J. E. Pak, P. Arnoux, S. Zhou, P. Sivarajah, M. Satkunarajah, X. Xing, and J. M. Rini X-ray Crystal Structure of Leukocyte Type Core 2 beta1,6-N-Acetylglucosaminyltransferase: EVIDENCE FOR A CONVERGENCE OF METAL ION-INDEPENDENT GLYCOSYLTRANSFERASE MECHANISM J. Biol. Chem., September 8, 2006; 281(36): 26693 - 26701. [Abstract] [Full Text] [PDF]

				A. E. Ciocchini, M. S. Roset, G. Briones, N. I. de Iannino, and R. A. Ugalde Identification of active site residues of the inverting glycosyltransferase Cgs required for the synthesis of cyclic {beta}-1,2-glucan, a Brucella abortus virulence factor Glycobiology, July 1, 2006; 16(7): 679 - 691. [Abstract] [Full Text] [PDF]

				W. Jing and P. L. DeAngelis Analysis of the two active sites of the hyaluronan synthase and the chondroitin synthase of Pasteurella multocida Glycobiology, October 1, 2003; 13(10): 661 - 671. [Abstract] [Full Text] [PDF]

				T. Karnezis, V. C. Epa, B. A. Stone, and V. A. Stanisich Topological characterization of an inner membrane (1->3)-{beta}-D-glucan (curdlan) synthase from Agrobacterium sp. strain ATCC31749 Glycobiology, October 1, 2003; 13(10): 693 - 706. [Abstract] [Full Text] [PDF]

				O. L. Franco and D. J. Rigden Fold recognition analysis of glycosyltransferase families: further members of structural superfamilies Glycobiology, October 1, 2003; 13(10): 707 - 712. [Abstract] [Full Text] [PDF]

				H. van der Wel, S. Z. Fisher, and C. M. West A Bifunctional Diglycosyltransferase Forms the Fucalpha 1,2Galbeta 1,3-Disaccharide on Skp1 in the Cytoplasm of Dictyostelium J. Biol. Chem., November 22, 2002; 277(48): 46527 - 46534. [Abstract] [Full Text] [PDF]

				C. M. West, H. van der Wel, and E. A. Gaucher Complex glycosylation of Skp1 in Dictyostelium: implications for the modification of other eukaryotic cytoplasmic and nuclear proteins Glycobiology, February 1, 2002; 12(2): 17R - 27R. [Abstract] [Full Text] [PDF]

				S. J. Charnock, B. Henrissat, and G. J. Davies Three-Dimensional Structures of UDP-Sugar Glycosyltransferases Illuminate the Biosynthesis of Plant Polysaccharides Plant Physiology, February 1, 2001; 125(2): 527 - 531. [Full Text]

W. J. Keenleyside, A. J. Clarke, and C. Whitfield Identification of Residues Involved in Catalytic Activity of the Inverting Glycosyl Transferase WbbE from Salmonella enterica Serovar Borreze J. Bacteri