Molecular Biology and Evolution 19:1451-1463 (2002)
© 2002 Society for Molecular Biology and Evolution
Common Origin and Evolution of Glycosyltransferases Using Dol-P-monosaccharides as Donor Substrate
INSERM U504, University of Paris Sud XI, Villejuif, France
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResults and DiscussionAcknowledgementsReferences |
|---|
On the basis of the analysis of 64 glycosyltransferases from 14 species we propose that several successive duplications of a common ancestral gene, followed by divergent evolution, have generated the mannosyltransferases and the glucosyltransferases involved in asparagine-linked glycosylation (ALG) and phosphatidyl-inositol glycan anchor (PIG or GPI), which use lipid-related donor and acceptor substrates. Long and short conserved peptide motifs were found in all enzymes. Conserved and identical amino acid positions were found for the
2/6- and the 3/4-mannosyltransferases and for the 2/3-glucosyltransferases, suggesting unique ancestors for these three superfamilies. The three members of the 2-mannosyltransferase family (ALG9, PIG-B, and SMP3) and the two members of the 3-glucosyltransferase family (ALG6 and ALG8) shared 11 and 30 identical amino acid positions, respectively, suggesting that these enzymes have also originated by duplication and divergent evolution. This model predicts a common genetic origin for ALG and PIG enzymes using dolichyl-phospho-monosaccharide (Dol-P-monosaccharide) donors, which might be related to similar spatial orientation of the hydroxyl acceptors. On the basis of the multiple sequence analysis and the prediction of transmembrane topology we propose that the endoplasmic reticulum glycosyltransferases using Dol-P-monosaccharides as donor substrate have a multispan transmembrane topology with a first large luminal conserved loop containing the long motif and a small cytosolic conserved loop containing the short motif, different from the classical type II glycosyltransferases, which are anchored in the Golgi by a single transmembrane domain. |
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionAcknowledgementsReferences |
|---|
Asparagine-linked glycans (N-glycans) are engaged in a large panel of cellular and tissular functions (Varki 1993
). They are involved in (1) quality control of proteins synthesized in the endoplasmic reticulum (ER) (Helenius and Aebi 2001
), (2) intracellular trafficking of some glycoproteins along the secretory pathway (Hauri et al. 2000
), (3) lysosomal delivery of acid hydrolases (Kornfeld 1990
), (4) membrane targeting of some proteins in polarized cells (Scheiffele, Peränen, and Simons 1995
), and (5) they act as recognition signals at the cell surface and participate in morphogenesis and development (Takahashi, Honda, and Nishikawa 2000
).
The process of N-glycosylation starts in the cytosolic face of the ER by the successive assembly of two N-acetylglucosamines and five mannoses on the lipid carrier dolichylpyrophosphate. Then the lipid-linked heptasaccharide is translocated across the membrane to the lumen of the ER by the RFT1 flippase (Helenius et al. 2002
), where consecutive additions of four residues of mannose and three glucose residues complete the oligosaccharide, which is in turn transferred en bloc from the dolichol carrier to asparagine, in the sequence Asn-X-Ser/Thr of the nascent protein chain. The N-glycosylation process is completed by a series of trimming and processing reactions ending in the late Golgi compartment (Kornfeld and Kornfeld 1985
; Burda and Aebi 1999
).
The assembly of the initial heptasaccharide on the cytosolic face of the ER is carried out by glycosyltransferases using UDP-GlcNAc or GDP-Man as donor substrates, whereas the sequential addition in the ER lumen of the following four mannoses and three glucoses is carried out by glycosyltransferases using Dol-P-Man and Dol-P-Glc as donor substrates (fig. 1
). These last two donor substrates are synthesized in the cytosol from GDP-Man and UDP-Glc. The existence of specific flippases able to transport the dolichol-based donor substrates from the cytosol to the luminal side of the membrane has been proposed but they have not yet been identified (Schenk, Fernandez, and Waechter 2001
), and it is possible that more than one protein contributes to these flippase activities (Anand et al. 2001
).
|
The phosphatidyl-inositol glycan anchor (PIG or GPI) is a posttranslational modification frequently observed in proteins expressed in the plasma membrane (Cole and Hart 1997
). GPI is synthesized in the ER membrane, where it substitutes the C-terminus of some newly synthesized proteins. The four mannoses of the PIG anchor are also added in the ER (Maeda et al. 2001
). The first mannose is added by PIG-M which is an 4-mannosyltransferase, the second mannose is added by an 6-mannosyltransferase which has not yet been cloned, and the third and fourth mannoses are added by the PIG-B and SMP3 2-mannosyltransferases, respectively. These four enzymes use the same Dol-P-Man donor substrate and make linkages similar to those of the corresponding asparagine-linked glycosylation (ALG) mannosyltransferases of the N-glycan series (fig. 1
). In spite of these functional similarities, the ALG and PIG mannosyltransferases have always been considered separately, and no attempts had been made, in the past, to find a common evolutionary path for these two series of enzymes. The aims of this work were (1) to find conserved peptide motifs, potential candidates for interaction with acceptor or donor substrates, (2) to find eventual relations between ALG and PIG mannosyltransferases using the same donor substrate, and (3) to see if some common characters of these enzymes could be related to structural features of the linkage to the acceptor substrates and if these could help to understand their evolutionary path.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionAcknowledgementsReferences |
|---|
Nomenclature
We consider that the glycosyltransferases which use the same donor and acceptor substrates and make the same sugar linkage (i.e.,
2-, 3-, 4- or 6-mannosyltransferases and 2- or 3-glucosyltransferases) belong to the same family and that families sharing common peptide motifs, identical amino acid positions, and similar transmembrane domain (TMD) topology can be clustered in superfamilies (i.e., 2/6-mannosyltransferases, 3/4-mannosyltransferases, or 2/3-glucosyltransferases).
Sequence Retrieval
Saccharomyces cerevisiae or human (or both) DNA and protein glycosyltransferase sequences were retrieved from the literature. Orthologous protein sequences from other species were first searched with gapped-BLAST. The cloning of human ALG12 (AJ303120) has been reported by Chantret et al. (2002)
. The human EST databank was searched with rat alg10 and with S. cerevisiae rft1 (U15087). Contigs were built for both series of human ESTs by alignment with HUCAP (Huang 1992
). Primers were designed in the EST contigs (sense 5'-CAGGAGTAGGTTCTTGGGCAGTGGC-3' and antisense 5'-AATGTAATTTGAAGACCACCACTGCACC-3' for ALG10, and sense 5'-GGCATTTCCTGGTGTCTGAGCCTG-3' and antisense 5'-CACAGAACTACCCATAGCTGGTCC-3' for RFT1). The corresponding human 1,582-bp ALG10 (AJ312278) and 1,761-bp RFT1 (AJ318099) cDNA sequences were amplified from an expression library (Cailleau-Thomas et al. 2000
), and they were cloned in a PCR3.1 expression vector (Eukaryotic TA cloning kit from Invitrogen), and sequenced in both strands by the dideoxy chain termination method with the T7 DNA polymerase (kit Amersham-Pharmacia-Biotech) and submitted to EMBL. The cDNA sequences of the coding sequences of Mus musculus alg12, Drosophila melanogaster alg10 and Caenorhabditis elegans pig-B were deduced from HUCAP alignments made with more than two concordant EST overlaps per position.
Conserved Peptide Motifs
After retrieval of all the sequences, the best-conserved peptide patterns for each family were determined with CLUSTALW 1.8 (Thomson, Higgins, and Gibson 1994
) and used in a final search with the corresponding human sequence in an iterative PHI-BLAST search with default parameters, until convergence (Altschul et al. 1997
). The seed expression patterns for the PHI-BLAST were: H[KQ]EXRF[ILMV][IYFLSV][YPLV] for the 2/6-mannosyltransferase superfamily; [VL]X[YF]T[DEK][IV]D[YW]X[VIAT] for the 3/4-mannosyltransferase superfamily; W[GT]LDYPP[LF][TF]A[FYW] for the 3-glucosyltransferase family; and LADNRH[YF][TL]FY for the 2-glucosyltransferase family. A combined PHI-PSI-BLAST was also used for the 2/3-glucosyltransferase superfamily with human ALG10 and the initial PHI-BLAST seed pattern: [WP]X[LI][DT][YT][PFL]PX[TFIL][AY]. Only complete sequences comprising both the long and the short conserved peptide motifs were used for this study, but the presence of numerous EST and partial sequences in other species suggests that these enzymes are widely distributed among other animals and plants and even some bacteria.
Transmembrane Domains
Helical TMDs were predicted by PHD-htm for each enzyme (table 1 
) (Rost, Fariselli, and Casadio 1996
). This analysis gave a similar, but not identical multispan TMD structure for all individual sequences (ranging from 8–14 TMDs, table 1
). Then, because the best computer-assisted TMD prediction methods have about 86% accuracy (Rost, Fariselli, and Casadio 1996
), we searched for consensus TMD topologies. Individual TMDs predicted by PHD-htm were added to each species sequence on the CLUSTALW alignments for each enzyme, and TMDs detected at marginal significance were added or deleted, when needed, to get the best overall consensus TMD fit with the other species. Helical segments of 15–24 amino acids were considered as single helical TMDs, and longer hydrophobic helical domains of more than 25 amino acids were considered as double TMDs entering and leaving the membrane from the same side. This artificial consensus TMD prediction of all species for each enzyme was drawn, and the loop invariant amino acids were added to the figures. The high stringent criteria of total identity for each individual amino acid position was selected on purpose for visualization, but the less stringent analysis based on conserved amino acid positions at 50% leads to similar conclusions. The analysis of conserved amino acid positions and the phylogeny were derived independently from the transmembrane topology.
|
|
Phylogeny
Peptide sequences of mannosyltransferases and glucosyltransferases were aligned with CLUSTALW. Two strategies for selection of informative positions were used to make the trees: (1) selection of the two conserved peptide motifs of each sequence (fig. 2 ), followed by CLUSTALW alignment and global gap removal, which gave blocks of 43 of the original 53 sites (81%) for the 45 species of mannosyltransferases and 48 of the 57 original sites (84%) for the 17 species of glucosyltransferases; and (2) computer selection of conserved blocks with GBLOCKS (Castresana 2000
) with parameters adjusted for maximum block retrieval in the CLUSTALW alignments of each of the three superfamilies, which gave 106 sites (9% of the original 1,132) for the 28 species of the 2/6-mannosyltransferase superfamily, 150 sites (25% of the original 616) for the 17 species of the 3/4-mannosyltransferase superfamily, and 137 sites (20% of the original 668) for the 17 species of the 2/3-glucosyltransferase superfamily. Both selection methods gave trees with similar topology, showing that the additional sites recruited by GBLOCKS do not bias the phylogeny. Distance matrices were generated with the observed difference of the neighbor-joining method from the PHYLO_WIN package (Galtier, Gouy, and Gautier 1996
). The sequence of alg8 of C. elegans, which has a frameshift and lacks the short motif, and the sequence of alg10 of A. thaliana, which lacks half of the short motif, were excluded from the phylogeny study. Five hundred sets of data were used for bootstrap calculations.
|
|
Results and Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionAcknowledgementsReferences |
|---|
Conserved Peptide Motifs
We found two highly conserved peptide motifs in each of the mannosyltransferases and glucosyltransferases, by protein sequence alignment with CLUSTALW. The long motif (41–44 amino acids) is located close to the amino terminus and the short motif (9–10 amino acids) is located in the carboxy terminal side of the enzymes. The distance between these two motifs is relatively well conserved in the three superfamilies (median 252, range 215–381) (fig. 2A–C ).
The 2- and 6-Mannosyltransferase Superfamily Using Dol-P-Man as Donor Substrate
The 6-mannosyltransferases (ALG12) (Burda et al. 1999
) are clustered in the same glycosyltransferase-22 family of the Carbohydrate Active enZyme database (CAZy) (http://afmb.cnrs-mrs.fr/
pedro/CAZY/db.html) together with three 2-mannosyltransferases, one belonging to the same asparagine-linked glycosylation series (ALG9) and two enzymes involved in the synthesis of the PIG anchor (PIG-B and SMP3) (Ferguson 1992
). These four enzymes were known to have a short common peptide pattern: H[KQ]EXRF (Canivenc-Gansel et al. 1998
), which is part of the short conserved peptide motif shown in figure 2A
. After five iterative runs with the human ALG12 protein sequence, the PHI-BLAST converges and retrieves the 28 enzymes of the 2/6-mannosyltransferase superfamily, including all the ALG12, ALG9, PIG-B, and SMP3 enzymes (table 1
).
The alignment of the sequences of the homologous ALG12 enzymes from four different species (S. cerevisiae, Schizosaccharomyces pombe, D. melanogaster, and C. elegans) showed another highly conserved peptide pattern (TKVEESF) in the long conserved peptide motif of ALG12 (fig. 2A
), which helped us to clone the mouse and the human ALG12 genes. Unlike the short motif, which is highly conserved in the four enzymes of this superfamily, the long motif has few positions conserved in all four families, but it has a large proportion of amino acid positions conserved among the 6-mannosyltransferases on one side and among the three 2-mannosyltransferase families on the other side, suggesting the existence of two conserved long peptide motifs, one specific for the 6-mannosyltransferases (ALG12) and another specific for the 2-mannosyltransferases (ALG9, PIG-B and SMP3) (fig. 2A
). The phylogenetic tree of all mannosyltransferases supports the idea that the three 2-mannosyltransferases have a common genetic origin and are more closely related among themselves than to the 6-mannosyltransferase family (fig. 3
).
|
The enzymes of this
2/6-mannosyltransferase superfamily have a consensus multispan TMD structure with an average of 12 spans. The long motif is flanked by TMD-1 and TMD-2, whereas the short motif is flanked by TMD-10 and TMD-11 (fig. 4
). PIG-B (Takahashi et al. 1996
) and ALG9 (Burda et al. 1996
) glycosyltransferases add an 2-mannose to a sterically similar mannose acceptor (fig. 1 ), and they share eleven identical amino acid positions (solid and gray circles in fig. 4
). Three of the eight ALG9 and three of the ten PIG-B enzymes have an ER retention signal (table 1 ) (Teasdale and Jackson 1996
), suggesting that their COOH terminus is in the cytosol. The SMP3 enzymes make a further elongation of a fourth mannose (Grimme et al. 2001
) also linked in 1
2 in PIG (fig. 1
). They share the same eleven identical amino acids with the two other 2-mannosyltransferases (fig. 4 ), but none of the four known SMP3 enzymes has ER retention signals (table 1
).
|
It is interesting to note that a large proportion of the conserved amino acids in the
2-mannosyltransferases are located in the long loop between the first two TMDs, where we found the ALG12-specific highly conserved pattern (TKVEESF). In addition, the four identical amino acids, which are specific of the 2-mannosyltransferase activity (gray circles in fig. 4
), are also in this first loop, suggesting that it might be involved in 2- and 6-mannosyltransferase active sites.
Seven of the eleven amino acids, identical among 2-mannosyltransferases, are also shared by the 6-mannosyltransferases (solid circles, fig. 4
). This might be related to the fact that the hydroxyl groups on C2 and C6 of the -D-mannose acceptor are on the same side of the molecule, and they may offer some similarities of acceptor surface to the corresponding 2- and 6-mannosyltransferases (fig. 5
), although they are not very close in space. We have previously found a similar phenomenon between 2-fucosyltransferases and 6-fucosyltransferases which share three conserved peptide motifs (Breton, Oriol, and Imberty 1998
; Chazalet et al. 2001
). In addition, conserved peptide motifs were also found between 3-fucosyltransferases and 4-fucosyltransferases (Oriol et al. 1999
). This last observation prompted us to ascertain whether the 3- and the 4-mannosyltransferases using Dol-P-Man as donor substrate have also conserved common peptide motifs.
|
The
3- and 4-Mannosyltransferase Superfamily Using Dol-P-Man as Donor SubstrateAfter three iterative runs with the corresponding pattern and either of the human ALG3 (CAZy glycosyltransferase-58 family) or PIG-M (CAZy glycosyltransferase-50 family), the PHI-BLAST reaches convergence and retrieves the 17 enzymes of the
3/4-mannosyltransferase superfamily (table 1
). Several conserved amino acid positions were found for both the long and the short motifs of the 3/4-mannosyltransferase superfamily (fig. 2B
). The phylogenetic tree of all mannosyltransferases show that the ALG3 family and the PIG-M family constitute separate clusters (fig. 3
) and are different enough from the cluster of 2/6-mannosyltransferases to be considered as an outgroup defining the position of a possible root for the tree between the two superfamilies of 2/6- and 3/4-mannosyl-transferases.
The TMD prediction also suggests a multispan structure with an average of 12 spans, with the long motif in the first loop and the short motif between TMD-10 and TMD-11 (fig. 6
). Seven identical amino acid positions are shared by the 3- and the 4-mannosyltransferases (solid circles in fig. 6
). Five of these seven identical positions are in the first large luminal loop, which is equivalent to the first loop of the 2/6-mannosyltransferase superfamily, suggesting again that this first luminal loop is particularly well conserved and may play a role in the active site of the enzymes and that there are some similarities between the consensus TMD topology of the two superfamilies of 2/6-mannosyltransferases and 3/4-mannosyltransferases. In addition, the presence of a small conserved loop, in the cytosolic side of the membrane, between TMD-10 and TMD-11 of 3- and 4-mannosyltransferases, supports the idea of conserved TMD topology between 2/6-mannosyltransferases (fig. 4
) and 3/4-mannosyltransferases (fig. 6
). Six of the eight 3-mannosyltransferases (ALG3) and five of the nine 4-mannosyltransferases (PIG-M) have an ER retention signal (table 1
), in good agreement with the consensus TMD topology prediction (fig. 6
), which suggests that both NH2 and COOH terminus are cytosolic. PIG-M belongs to the PIG-anchor and ALG3 to the N-glycan series, suggesting that these two series of enzymes have also a common genetic origin.
|
The hydroxyls on C3 of the
-D-mannose and C4 of the -D-glucosamine acceptors have the same equatorial orientation and offer a similar equatorial approach to the corresponding 3-mannosyltransferases (ALG3) and 4-mannosyltransferases (PIG-M), especially if the glucosamine is rotated 180° around its horizontal axis (fig. 5 ). Consequently, we expected more common features between 3- and 4-mannosyltransferases, which share the same equatorial approach to the acceptor and the same donor and less common features between the two superfamilies of 2/6- and 3/4-mannosyltransferases because they only share the donor substrate (Dol-P-Man).
The 2 and 3-Glucosyltransferase Superfamily Using Dol-P-Glc as Donor Substrate
After two iterative runs, the PHI-BLAST with either of human ALG6 or ALG8 converges and retrieves the 12 sequences of the 3-glucosyltransferase family, comprising the ALG6 and the ALG8 enzymes. These two enzymes are closely related as shown in the phylogenetic tree of figure 7
. Their close similarity is also illustrated by 30 identical amino acid positions among ALG6 and ALG8 (fig. 8
). Four out of the six ALG6 and two out of the six ALG8 enzymes (Stagljar, Hessen, and Aebi 1994
; Stanchi et al. 2001
) have a putative ER retention signal (table 1
), suggesting that their COOH-end is in the cytosol, as predicted by the consensus TMD topology (fig. 8
).
|
|
Either of PHI-BLAST or PSI-BLAST alone with ALG10 retrieved only the ALG10 enzymes of the different species, but the combined PHI-PSI-BLAST with human ALG10 retrieves ALG10, ALG8, and ALG6 enzymes of the
2/3-glucosyltransferase superfamily. Few similarities exist between ALG10 and the 3-glucosyltransferases, but the short and long motifs are well identified, among all the different species for ALG10 (gray and black areas of the ALG10 family in fig. 2C
). The phylogenetic tree confirms this impression, showing a cluster of ALG10 enzymes at high distance from the two 3-glucosyltransferases, ALG6 and ALG8 (fig. 7
). The alg10 gene was first described in yeast (Burda and Aebi 1998
), then in rat (Hoshi et al. 1998
), and we have cloned the Drosophila (AJ431376) and the human (AJ312278) ALG10 orthologous genes. None of the ALG10 enzymes has ER retention signals, but all of them have three identical amino acid positions with ALG6 and ALG8, plus a glutamic acid (E) identical with ALG6 in the long motif, and a histidine (H) identical with ALG8 in the short motif (solid circles fig. 8
). The transmembrane topology of the last three glucosyltransferases suggests a consensus TMD organization similar to the aforementioned TMD structure of mannosyltransferases with about 12 TMD. A large conserved luminal loop between TMD-1 and TMD-2 and a small conserved loop, in the cytosolic side of the membrane, between TMD-8 and TMD-9 are present (fig. 8 ).
The scheme of figure 9
illustrates identical equatorial orientations of the hydroxyls on C3 of the last -D-mannose and the first -D-glucose, which are acceptors for ALG6 and ALG8, respectively. Unlike this, a different spatial orientation can be seen for the hydroxyl on C2 of the second -D-glucose, which is the acceptor substrate for ALG10. Therefore, we expect more common features between ALG6 and ALG8, which share both acceptor and donor substrates, than between the 2-glucosyltransferase (ALG10) and the two 3-glucosyltransferases (ALG6 and ALG8) because they have different acceptor substrates and they only share the donor substrate (Dol-P-Glc). Comparison of the OH on C3 (acceptors for ALG6 and ALG8) and the OH on C2 (acceptor for ALG10, fig. 9
) with the OH on C3, C4, C2, and C6 (acceptors for mannosyltransferases, fig. 5
), suggests that the acceptor OH of the two families of glucosyltransferases are as different from each other as the corresponding acceptors for the two mannosyltransferase superfamilies.
|
Potential Functions of TMD and Conserved Loops
The transmembrane topology of the glycosyltransferases, using Dol-P-monosaccharides as donor substrate, suggests a common consensus multispan transmembrane topology. The enzymes have an average of 12 putative helical TMD, and both the NH2 and COOH-ends are in the cytosolic side of the membrane (figs. 4 , 6 , and 8 ). But they still are a statistical prediction, and we cannot be certain that all the putative TMDs span the membrane in vivo, for example only 7 of the 11 TMDs predicted for the protein O-mannosyltransferase (Pmt1p) were shown to span the membrane (Straahl-Bolsinger and Scheinost 1999
). However, if some of the putative -helical TMDs do not span the membrane, the second most probable topology with both extremes in the cytosol would be a 10 TMD, and if there is an odd number of TMD, the NH2 and the COOH terminus would be on different sides of the membrane. None of these possibilities can be formally ruled out for individual sequences because the consensus topology is only an average artificial prediction. Nevertheless, we can say that all these enzymes have a multispan TMD topology, similar to the general structure of sugar transport proteins such as glucose-6-P-transporter (G6PT) (Pan, Lin, and Chou 1999
) or glucose transporter 1 (Glut1) (Sato and Mueckler 1999
). In fact, the great majority of sugar transporters have a multispan structure with 10 or 12 TMDs (Maiden et al. 1987
; Marger and Saier 1993
). It is possible that each of the enzymes under study contributes also to the translocation across the membrane of the Dol-P-monosaccharide, which is synthesized in the cytosol and has to reach the acceptor substrate oligosaccharides in the lumen of the ER (Rush and Waechter 1998
; Rush et al. 1998
). A similar distribution of the putative TMD along the peptide chain is an additional structural feature in favor of a common genetic origin for a series of related enzymes. Our consensus TMD topology suggests that the first, large and highly conserved loop, is on the luminal side of the membrane and contains identical amino acid positions specific for each linkage. These properties are compatible with the contribution of this first loop to the glycosyltransferase active site.
For the sake of clarity, only identical amino acid positions located in the extramembrane loops are shown in figures 4
, 6
, and 8
, but between 30% and 42% of the total amino acids of these enzymes is located inside the putative helical TMD. The comparative analysis of identical amino acid positions in the loops and in the helical TMD shows that they are not randomly distributed. In general, there are fewer identical amino acid positions inside the TMD (8%–35%) than in the loops (65%–92%), and the proportion of identical amino acids in TMDs are characteristic and different for the 2/6-mannosyltransferase (8%–19%), the 3/4-mannosyltransferase (21%–27%) and the 2/3-glucosyltransferase (35%), in favor of a common origin for each of these three superfamilies of glycosyltransferases.
The mutations of these ER glycosyltransferases, known to be responsible of congenital disorders of glycosylation (CDG), are also not randomly distributed. Only one out of eight mutations is in a short loop and seven are in TMDs (solid stars in figs. 4 , 6
, and 8
). A similar difference in favor of a TMD location has been reported for the mutations inactivating the Golgi transporter of CMP-sialic acid (Eckhardt, Gotza, and Gerardy-Schahn 1999
), and this protein is also a candidate for establishing a new CDG of type II (Mollicone et al., personal communication).
Divergent Evolutionary Model
All together the common structural features of the enzymes using Dol-P-monosaccharides as donor substrate suggest that they might have followed a common evolutionary path. The duplication of a hypothetical ancestral gene, encoding for a protein with the long and short conserved peptide motifs and the consensus multispan TMD topology, followed by divergent evolution, might have generated the two main mannosyltransferase and glucosyltransferase genes (fig. 10
). A second duplication within the mannosyltransferases might have originated the 2/6- and the 3/4-mannosyltransferase superfamilies. A duplication of the 2/6-mannosyltransferase gene may be at the origin of the 6-mannosyltransferase genes (ALG12) and the ancestor of the 2-mannosyltransferase genes. Two further duplications of the 2-mannosyltransferase branch may be at the origin of SMP3, PIG-B, and ALG9 genes. A duplication of the gene of the 3/4-mannosyltransferase superfamily may have originated ALG3 and PIG-M genes. Similar duplication events may have originated the 2-glucosyltransferase genes (ALG10) and the ancestor of 3-glucosyltransferase genes in the glucosyltransferase branch. Finally, a last and more recent duplication must be at the origin of the present two enzymes with 3-glucosyltransferase activity (ALG6 and ALG8) because they display the highest score of sequence identity (30 amino acids, fig. 8
).
|
The compilation of the data reported previously and the identical and conserved amino acid positions in each of the superfamilies, families, and enzymes, constitute indirect evidence for a common genetic origin followed by divergent evolution of these enzymes. The existence of the more ancestral precursor genes (the two empty circles in fig. 10 ) is supported by the relative positions and sizes of the long and the short conserved peptide motifs, but no identical or conserved amino acid positions were found between mannosyltransferases and glucosyltransferases, nor between the
2/6 and the 3/4-superfamilies of mannosyltransferases. The relative positions of the duplications reflect the increasing proportion of identical and conserved amino acids from left to right. All the enzymes on the right side of the tree have been cloned and experimentally tested, whereas the existence of a single common ancestor on the left is only a working hypothesis. However, if the single common ancestor hypothesis is not confirmed, a very similar evolutionary model with two non-related ancestors, one for the -mannosyltransferases and another for the -glucosyltransferases, or three ancestors, one for each superfamily, can be imagined. Irrespective of considering one, two, or three ancestors, the tree depicted in figure 10
must be quite ancient because these nine glycosyltranferases have been found in plants, yeasts, worms, insects, and vertebrates.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionAcknowledgementsReferences |
|---|
The research was partially supported by the Association for Research on Cancer (ARC) grant 5348, the French network for Recombinant Glycosyltransferases (GT-rec), and the French INSERM/AFM network 4MR29F for CDG. We are grateful to Stuart E. H. Moore for critical reading of the manuscript and helpful discussions.
|
Footnotes |
|---|
Pierre Capy, Reviewing Editor
Abbreviations: ALG, Asparagine-Linked Glycosylation; CAZy, Carbohydrate Active enZyme database; CDG, Congenital Disorder of Glycosylation; Dol, dolichol; ER, endoplasmic reticulum; Man, mannose; Glc, glucose; GlcNAc, N-acetylglucosamine; PIG, Phosphatidyl-Inositol Glycan anchor (GPI-anchor); TMD, transmembrane domain. 
Keywords: glucosyltransferase
GPI-anchor
mannosyltransferase
N-glycan
transmembrane topology
phylogeny
Address for correspondence and reprints: Rafael Oriol, INSERM U504, University of Paris Sud XI, 16 Avenue Paul Vaillant-Couturier, 94807 Villejuif Cedex, France. E-mail: oriol{at}infobiogen.fr
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResults and DiscussionAcknowledgementsReferences |
|---|
Altschul S., F. Stephen, T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, D. J. Lipman, 1997 Gapped BLAST and PSI-BLAST: a new generation of protein database search programs Nucleic Acid Res 25:3389-3402
Anand M., J. S. Rush, S. Ray, M. A. Doucey, J. Weik, F. E. Ware, J. Hofsteenge, C. J. Waechter, M. A. Lehrman, 2001 Requirement of the Lec35 gene for all known classes of monosaccharide-P-Dolichol-dependent glycosyltransferase reactions in mammals Mol. Biol. Cell 12:487-501
Breton C., R. Oriol, A. Imberty, 1998 Conserved structural features in eukaryotic and prokaryotic fucosyltransferases Glycobiology 8:87-94
Burda P., M. Aebi, 1998 The ALG10 locus of Saccharomyces cerevisiae encodes the -1,2 glucosyltransferase of the endoplasmic reticulum: the terminal glucose of the lipid-linked oligosaccharide is required for efficient N-linked glycosylation Glycobiology 8:455-462
———. 1999 The dolichol pathway of N-linked glycosylation Biochim. Biophys. Acta 1426:239-257[Medline]
Burda P., S. T. Heesen, A. Brachat, A. Wach, A. Dusterhoft, M. Aebi, 1996 Stepwise assembly of the lipid-linked oligosaccharide in the endoplasmic reticulum of Saccharomyces cerevisiae: identification of the ALG9 gene encoding a putative mannosyl transferase Proc. Natl. Acad. Sci. USA 93:7160-7165
Burda P., C. A. Jakob, J. Beinhauer, J. H. Hegemann, M. Aebi, 1999 Orderly assembly of the asymmetrically branched lipid-linked oligosaccharide in the endoplasmic reticulum is ensured by the substrate specificity of the individual glycosyltransferases Glycobiology 9:617-625
Cailleau-Thomas A., P. Coullin, J. J. Candelier, L. Balanzino, B. Mennesson, R. Oriol, R. Mollicone, 2000 FUT4 and FUT9 genes are expressed early in human embryogenesis Glycobiology 10:789-802
Canivenc-Gansel E., I. Imhof, F. Reggiori, P. Burda, A. Conzelmann, A. Benachour, 1998 GPI anchor biosynthesis in yeast: phosphoethanolamine is attached to the 1,4-linked mannose of the complete precursor glycosphingolipid Glycobiology 8:761-770
Castresana J., 2000 Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis Mol. Biol. Evol 17:540-552
Chantret I., T. Dupre, C. Delenda, et al. (15 co-authors) 2002 Congenital disorder of glycosylation type Ig is defined by a deficiency in dolichyl-P-mannose: Man7GlcNAc2-PP-dolichyl mannosyltransferase J. Biol. Chem 277:25815-25822.
Chazalet V., K. Uehara, R. A. Geremia, C. Breton, 2001 Identification of essential amino acids in the Azorhizobium caulinodans fucosyltransferase NodZ J. Bacteriol 183:7067-7075
Cole R. N., G. W. Hart, 1997 Glycosyl-phosphatidylinositol anchors: structure, biosynthesis and function Pp. 69–88 in J. Montreuil et al., ed. Glycoproteins II. Elsevier, Amsterdam
de Lonlay P., N. Seta, S. Barrot, et al. (22 co-authors) 2001 A broad spectrum of clinical presentations in congenital disorders of glycosylation I: a series of 26 cases J. Med. Genet 38:14-19
Eckhardt M., B. Gotza, R. Gerardy-Schahn, 1999 Membrane topology of the mammalian CMP-sialic acid transporter J. Biol. Chem 274:8779-8787
Ferguson M. A. J., 1992 Glycosyl-phosphatidylinositol membrane anchors: the tale of a tail Biochem. Soc. Trans 20:243-256[ISI][Medline]
Galtier N., M. Gouy, C. Gautier, 1996 Two graphic tools for sequence alignment and molecular phylogeny Comput. Appl. Biosci 12:543-548
Grimme S. J., B. A. Westfall, J. M. Wiedman, C. H. Taron, P. Orlean, 2001 The essential smp3 protein is required for addition of the side branching fourth mannose during assembly of yeast glycosylphosphatidylinositols J. Biol. Chem 276:27731-27739
Grünewald S., T. Imbach, K. Huijbe, et al. (14 co-authors) 2000 Clinical and biochemical characteristics of congenital disorder of glycosylation type Ic, the first recognized endoplasmic reticulum defect in N-glycan synthesis Ann. Neurol 47:776-781[ISI][Medline]
Hanefeld F., C. Körner, U. Holzbach-Erberle, K. von Figura, 2000 Congenital disorder of glycosylation-Ic: case report and genetic defect Neuropediatrics 31:57-59[ISI][Medline]
Hauri H. P., C. Appenzeller, F. Kuhn, O. Nufer, 2000 Lectins and traffic in the secretory pathway FEBS Lett 476:32-37[ISI][Medline]
Helenius A., M. Aebi, 2001 Intracellular functions of N-linked glycans Science 291:2364-2369
Helenius J., D. T. W. Ng, C. L. Marolda, P. Walter, M. A. Valvano, M. Aebi, 2002 Translocation of lipid-linked oligosaccharides across the ER membrane requires Rft1 protein Nature 415:447-450[Medline]
Hoshi N., H. Takahashi, M. Shahidullah, S. Yokoyama, H. Higashida, 1998 KCR1, a membrane protein that facilitates functional expression of non-inactivating K+ currents associates with rat EAG voltage-dependent K+ channels J. Biol. Chem 273:23080-23085
Huang X., 1992 A contig assembly program based on sensitive detection of fragment overlaps Genomics 14:18-25[ISI][Medline]
Imbach T., P. Burda, P. Kuhner, R. A. Wevers, M. Aebi, E. G. Berger, 1999 A mutation in the human ortholog of the Saccharomyces cerevisiae ALG6 gene causes carbohydrate-deficient glycoprotein syndrome type-Ic Proc. Natl. Acad. Sci. USA 96:6982-6987
Imbach T., S. Grünewald, B. Schenk, et al. (12 co-authors) 2000 Multi-allelic origin of congenital disorder of glycosylation (CDG)-Ic Hum. Genet 106:538-545[ISI][Medline]
Körner C., R. Knauer, U. Stephani, T. Marquard, L. Lehle, K. von Figura, 1999 Carbohydrate deficient glycoprotein syndrome type IV: deficiency of dolichol-P-Man:Man5GlcNAc2-PP-dolichyl mannosyltransferase EMBO J 18:6816-6822[ISI][Medline]
Kornfeld S., 1990 Lysosomal enzyme targeting Biochem. Soc. Trans 18:367-374[ISI][Medline]
Kornfeld R., S. Kornfeld, 1985 Assembly of asparagine-linked oligosaccharides Annu. Rev. Biochem 54:631-664[ISI][Medline]
Maeda Y., R. Watanabe, C. L. Harris, Y. Hong, K. Ohishi, K. Kinoshita, T. Kinoshita, 2001 PIG-M transfers the first mannose to glycosylphosphatidylinositol on the lumenal side of the ER EMBO J 20:250-261[ISI][Medline]
Maiden M. C., E. O. Davis, S. A. Baldwin, D. C. M. N. Moore, P. J. Henderson, 1987 Mammalian and bacterial sugar transport proteins are homologous Nature 325:641-643[Medline]
Marger M. D., M. H. Saier, 1993 A major superfamily of transmembrane facilitators that catalyse uniport, symport and antiport Trends Biochem. Sci 18:248-249[ISI][Medline]
Oriol R., R. Mollicone, A. Cailleau, L. Balanzino, C. Breton, 1999 Divergent evolution of fucosyltransferase genes from vertebrates, invertebrates, and bacteria Glycobiology 9:323-334
Pan C. J., B. Lin, J. Y. Chou, 1999 Transmembrane topology of human glucose-6-phosphate transporter J. Biol. Chem 274:13865-13869
Rost B., P. Fariselli, R. Casadio, 1996 Topology prediction for helical transmembrane proteins at 86% accuracy Protein Sci 7:1704-1718
Rush J. S., K. van Leyen, O. Ouerfelli, B. Wolucka, C. J. Waechter, 1998 Transbilayer movement of Glc-P-dolichol and its function as a glucosyl donor: protein-mediated transport of a water-soluble analog into sealed ER vesicles from pig brain Glycobiology 8:1195-1205
Rush J. S., C. J. Waechter, 1998 Topological studies on the enzymes catalyzing the biosynthesis of Glc-P-dolichol and the triglucosyl cap of Glc3Man9GlcNAc2-P-P-dolichol in microsomal vesicles from pig brain: use of the processing glucosidase I/II as latency markers Glycobiology 8:1207-1213
Sato M., M. Mueckler, 1999 A conserved amino acid motif (R-X-G-R-R) in the Glut1 glucose transporter is an important determinant of membrane topology J. Biol. Chem 274:24721-24725
Scheiffele P., J. Peränen, K. Simons, 1995 N-glycans as apical sorting signals in epithelial cells Nature 378:96-98[Medline]
Schenk B., F. Fernandez, C. Waechter, 2001 The ins(ide) and outs(ide) of dolichyl phosphate biosynthesis and recycling in the endoplasmic reticulum Glycobiology 11:61R-70R
Sharma C. B., R. Knauer, L. Lehle, 2001 Biosynthesis of lipid-linked oligosaccharides in yeast: the ALG3 gene encodes the Dol-P-Man:Man5GlcNAc2-PP-Dol mannosyltransferase Biol. Chem 382:321-328[ISI][Medline]
Stagljar I., S. T. Hessen, M. Aebi, 1994 New phenotype of mutations deficient in glycosylation of the lipid-linked oligosaccharide: cloning of the ALG8 locus Proc. Natl. Acad. Sci. USA 91:5977-5981
Stanchi F., E. Bertocco, S. Toppo, R. Dioguardi, B. Simionati, N. Cannata, R. Zimbello, G. Lanfranchi, G. Valle, 2001 Characterization of 16 novel human genes showing high similarity to yeast sequences Yeast 18:69-80[ISI][Medline]
Straahl-Bolsinger S., A. Scheinost, 1999 Transmembrane topology of Pmt1p, a member of an evolutionarily conserved family of protein O-mannosyltransferases J. Biol. Chem 274:9068-9075
Takahashi M., N. Inoue, K. Ohishi, Y. Maeda, N. Nakamura, Y. Endo, T. Fujita, J. Takeda, T. Kinoshita, 1996 PIG-B, a membrane protein of the endoplasmic reticulum with a large lumenal domain is involved in transferring the third mannose of the GPI anchor EMBO J 15:4254-4261[ISI][Medline]
Takahashi T., R. Honda, Y. Nishikawa, 2000 Cloning of the human cDNA which can complement the defect of the yeast mannosyltransferase I-deficient mutant alg1 Glycobiology 10:321-327
Teasdale R. D., M. R. Jackson, 1996 Signal-mediated sorting of membrane proteins between the endoplasmic reticulum and the Golgi apparatus Annu. Rev. Cell Develop. Biol 12:27-54[ISI][Medline]
Thomson J. D., D. G. Higgins, T. J. Gibson, 1994 CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice Nucleic Acid Res 22:4673-4680
Varki A., 1993 Biological roles of oligosaccharides, all of the theories are correct Glycobiology 3:97-130
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. A. Grabinska, S. K. Ghosh, Z. Guan, J. Cui, C. R. H. Raetz, P. W. Robbins, and J. Samuelson Dolichyl-Phosphate-Glucose Is Used To Make O-Glycans on Glycoproteins of Trichomonas vaginalis Eukaryot. Cell, August 1, 2008; 7(8): 1344 - 1351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Manthri, M L. S Guther, L. Izquierdo, A. Acosta-Serrano, and M. A J Ferguson Deletion of the TbALG3 gene demonstrates site-specific N-glycosylation and N-glycan processing in Trypanosoma brucei Glycobiology, May 1, 2008; 18(5): 367 - 383. [Abstract] [Full Text] [PDF]
|
|
|
|
 |