Molecular Biology and Evolution 17:1661-1672 (2000)
© 2000 Society for Molecular Biology and Evolution
ARTICLE |
Ancestral Exonic Organization of FUT8, the Gene Encoding the 
6-Fucosyltransferase, Reveals Successive Peptide Domains Which Suggest a Particular Three-Dimensional Core Structure for the 6-Fucosyltransferase Family
*Unité de Génétique Moléculaire Animale-UR 1061 (INRA/Université de Limoges), Institut des Sciences de la Vie et de la Santé, Faculté des Sciences, Limoges, France;
UPRES1074, Glycobiologie et Biotechnologie, Faculté des Sciences, Limoges, France;
Centre National de la Recherche Scientifique, Laboratoire de Chimie Biologique, Villeneuve d'Ascq, France;
and
§Glycobiologie INSERM U 504/Université Paris Sud XI, Villejuif, France
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionAcknowledgementsliterature cited |
|---|
Based on PCR strategies and expression studies, we define the genomic organization of the FUT8b gene. This gene encodes the only known mammalian enzyme transferring fucose in an
1
6 linkage on the asparagine-branched GlcNAc residue of the chitobiose unit of complex N-glycans. The intron/exon organization of the bovine coding sequence determines five successive functional domains. The first exon encodes a domain homologous to cytoskeleton proteins, the second presents a proline-rich region including a motif XPXPPYXP similar to the peptide ligand of the SH3-domain proteins, the third encodes a gyrase-like domain (an enzyme which can bind nucleotides), and the fourth encodes a peptide sequence homologous to the catalytic domain of proteins transferring sugars. Finally, the last exon encodes a domain homologous to the SH3 conserved motif of the SH2-SH3 protein family. This organization suggests that intramolecular interactions might give a tulip-shaped scaffolding, including the catalytic pocket of the enzyme in the Golgi lumen. Deduced from the published sequence of chromosome 14 (AL109847), the human gene organization of FUT8 seems to be similar to that of bovine FUT8b, although the exon partition is more pronounced (bovine exons 1 and 2 correspond to human exons 1–6). The mosaicism and phylogenetic positions of the 6-fucosyltransferase genes are compared with those of other fucosyltransferase genes. |
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsliterature cited |
|---|
Fucosyltransferases catalyze the transfer of GDP-fucose to oligosaccharide chains linked to proteins, lipids, and sugars (Costache et al. 1997b
). Fucosyl residues are found in four different linkages: 12, 13, 14, and 16. This last linkage appeared during the first steps of N-glycosylation in mammals owing to the GDP-L-Fuc: N-acetyl-ß-d-glucosaminide 6-fucosyltransferase, which transfers a fucosyl residue from GDP-fucose to the asparagine-linked GlcNAc residue of the chitobiose unit glycan-core.
Core-fucosylated N-glycans are widely observed in many mammalian glycoproteins and are especially abundant in brain tissue (Shimizu et al. 1993
). They are also present in insect cells, in which fucose can be either 16 or 13 linked, although difucosylated glycans have also been found in these cells (Staudacher and Marz 1998
). In plants, core fucosylation is always performed by an 3-fucosyltransferase (Fichette-Lainé et al. 1997).
As with the tumor marker -f
toprotein, the level of human 6-fucosylated oligosaccharides is very low in serum glycoproteins, but it increases dramatically in malignant diseases (Miyoshi et al. 1997
; Noda et al. 1998
). Recently, it was reported that core fucosylation affected the antenna flexibility of N-linked biantennary oligosaccharides and also prevented the action of glycoasparaginase (Noronski and Mononen 1997
). Such changes could modify further processing of oligosaccharides by specific glycosyltransferases (Stubbs et al. 1996
; Noronski and Mononen 1997
). Moreover, core fucose would be implicated in the early steps of development, particularly in normal migration of stem cells, as it is required by the 28 sialyltransferase for the polysialylation of the neural cell adhesion molecule (Kojima et al. 1996
).
The first 6-fucosyltransferase was characterized using a partially purified enzyme extract (Longmore and Schachter 1982
). An active secreted form of this enzyme was identified in platelets after blood clotting (Kaminska, Glick, and Koscielak 1998
). Recently, the enzyme was purified from porcine brain (Uozumi et al. 1996
) and a human gastric cancer cell line (Yanagidani et al. 1997
). The partial sequencing of these proteins allowed two highly conserved cDNAs (D89289 and D86723, respectively).
All known fucosyltransferases are type II transmembrane proteins with four distinct domains: cytoplasmic, transmembrane, hypervariable, and catalytic. The catalytic domain resides in the luminal compartment of trans-Golgi vesicles (Paulson and Colley 1989
) and appears to be globular. This hypothesis has been recently reinforced by the determination of the three-dimensional crystal structure of the bovine ß4-galactosyltransferase, whose catalytic domain is a single conical structure with a large pocket at the base (Gastinel, Cambillau, and Bourne 1999
).
Although fucosyltransferases have not yet been crystallized, experiments have been carried out to understand the molecular basis of fucosylation. This has allowed the identification of amino acids involved in different steps of the catalytic process and in acceptor and donor specificities. Truncation experiments on human 3/4-fucosyltransferases have identified the location of the catalytic domain and several amino acids in the hypervariable stem domain that specify the acceptor substrate (Xu, Loc, and Macher 1996
). More precisely, it appeared that amino acids at the two ends of the catalytic domain affected substrate specificity and affinity (Vo et al. 1998
; Nguyen et al. 1998
). Other sequence analyses have illustrated that different conserved peptide motifs specify each fucosyltransferase family and revealed that 2- and 6-fucosyltransferases have common peptide conserved motifs (Breton, Oriol, and Imberty 1998
; Oriol et al. 1999
). Recently, we demonstrated that a single amino acid in the hypervariable stem domain of vertebrate 3/4-fucosyltransferases determines the type-1 or type-2 transfer specificity (Dupuy et al. 1999
). Additionally, it was shown that a few amino acids of the hypervariable domain flanking the transmembrane domain contain signals for sublocalization and stabilization into the Golgi apparatus (Grabenhorst and Conradt 1999
).
The predicted genomic structure of the FUT8-coding sequences differs from that of all other known mammalian fucosyltransferases, which have monoexonic coding sequences (Costache et al. 1997a
; Yamaguchi et al. 1999
). Until now, none of these genomic organizations has been determined. In the present paper, we report the intron/exon organization of the coding sequence of the bovine 6-fucosyltransferase gene and compare it to human and other vertebrate, invertebrate, bacterial, and plant enzymes.
FUT8 reveals a fascinating segmentation in several exons, corresponding to successive functional peptide domains suggesting a possible three-dimensional core-structure. In addition to gene organization, several conserved motifs in different species (mammals, invertebrates, and bacteria) reveal how an ancestral fucosyltransferase gene (Costache et al. 1997a
; Oriol et al. 1999
) has evolved and led to the emergence of the 2- and the 6-fucosyltransferase gene families, which are structurally and functionally divergent.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsliterature cited |
|---|
The oligonucleotides used in this study are listed in table 1 . Before sequencing, PCR products were cloned into the pGEM-T Easy vector (Promega, Madison, Wis.). Transfections of COS-7 cells were carried out using the pcDNAI/Amp expression vector (Invitrogen, Carlsbad, Calif.). Expression of the FUT8b gene into Escherichia coli was performed using the pET-25b (+) expression vector (Novagen, Madison, Wis.). GDP- [14C]fucose (285 mCi/mmol) and [
-32P]dCTP (3,000 Ci/mmol) were purchased from Amersham, United Kingdom. Acceptor substrate for fucosyltransferase assays (Man3GlcNAc4) was obtained from plasma proteins treated successively by pronase, glycoasparaginase, sialidase, and ß-galactosidase. Bovine brain, heart, kidney, lung, and spleen mRNAs were obtained from CLONTECH (Palo Alto, Calif.).
|
Southern Blot Analysis of Genomic DNAs
Bovine, human, and porcine genomic DNAs were subjected to digestion with EcoRI and HindIII and fractionated by electrophoresis through 0.8% (w/v) agarose gel. After depurination (15 min) with 0.25 N HCl and denaturation (30 min) with 0.4 N NaOH, DNA was transferred on Hybond-N+ membranes (Amersham, Arlington Heights, Ill.). Hybridization was carried out with a 223-bp monoexonic probe (probe I) generated by PCR amplification of genomic DNA using S5 and S9 primers (table 1 ). Twenty-five nanograms of probe I were labeled with [
-32P]dCTP by random priming (NEBlot Kit, New England Biolabs). Hybridization was carried out for 12 h at 65°C in a buffer containing 10% (w/v) dextran sulfate, 0.5% (w/v) SDS, 0.5 M NaCl, and 0.2 mg/ml of sheared salmon sperm DNA. Blots were washed three times for 15 min each at 65°C with 2 x SSC/0.1% SDS, 1 x SSC/0.1% SDS, and 0.2 x SSC/0.1% SDS, then analyzed after exposure to X-ray films (Kodak Kyoto, Japan) for 3 days at -80°C.
Rapid Amplification of 5' and 3' cDNA Ends (RACE)
The Marathon cDNA amplification kit (CLONTECH) was used to produce a library of adapter-ligated double-stranded cDNA from bovine brain, heart, kidney, lung, and spleen tissues. The first strand synthesis was carried out at 42°C for 1 h with the 52mer CDS primer, 100 U of the MMLV reverse transcriptase in a total volume of 10 µl, and 1 µg of poly(A)+ mRNA from each tissue as a template. The second-strand synthesis was performed at 16°C for 90 min in a total volume of 80 µl, containing the enzyme mixture (RNase H, E. coli DNA polymerase I, and DNA ligase), the second-strand buffer, the dNTP mixture, and the first-strand reaction product. cDNA ends were then made blunt by adding to the reaction 10 U of T4 DNA polymerase and incubated at 16°C for 45 min. Double-stranded cDNA was phenol/chloroform extracted, ethanol precipitated, and then resuspended in 10 µl of water. Half of this volume was used for ligation of CLONTECH adapter to cDNA ends. The ligation reaction was carried out for 16 h at 16°C in a total volume of 10 µl, using 1 U of T4 DNA ligase. The resulting cDNA library was diluted to a final concentration of 0.2 µg/ml.
The 5' end of FUT8b cDNA was PCR-amplified using 5 µl of the library as template with primers AP1 and P3 (table 1 ). The 50-µl reaction mixture contained 5 nmol of each dNTP, 20 pmol of each primer, 50 nmol MgCl2, and 2.5 U Advantage Taq DNA polymerase (CLONTECH). After a pre-PCR cycle (1 min at 94°C, 15 s at 65°C, and 2 min at 68°C), 35 cycles were performed (10 s at 94°C, 15 s at 65°C, and 50 s at 68°C), followed by a last extension step at 68°C for 5 min. The resulting PCR product was diluted (1/100) and reamplified under the same conditions using nested oligonucleotides AP2 and P4 (table 1 ). The same procedure was followed to identify the 3' end, with the two specific primers P6 and P5 combined with AP1 and AP2, respectively. Final PCR products were cloned and sequenced.
DNA Sequencing
Sequencing was performed using T7 promoter and pUCM13rev primers for DNA cloned into pGEM-T Easy vector, or directly with primers for PCR amplification of long PCR products. A dye labeling chemistry (kit PRISM Ready Reaction Ampli Taq FS) and the ABI Prism 310 Genetic Analyzer (Perkin Elmer, Norwalk, Colo.) were used.
Northern Blot Analysis
Two micrograms of brain, heart, kidney, lung, and spleen mRNAs, obtained from CLONTECH, were denatured and fractionated with 0.8% formaldehyde agarose gel electrophoresis and then blotted onto nylon Hybond-N+ membranes. After prehybridization for 4 h at 42°C in a buffer containing 50% (v/v) deionized formamide, 1% (w/v) SDS, 1 x Denhardt's, 5.5 x SSC, 21.5 mM Na2HPO4, 11% (w/v) dextran sulfate, and 7 µg/ml salmon sperm DNA, the membrane was hybridized overnight at 42°C, with probe I. A bovine ß-actin probe was used as positive control. Membranes were washed three times at 42°C with 2 x SSC/0.1% SDS, 1 x SSC/0.1% SDS, and 0.2 x SSC/0.1% SDS, respectively, for 15 min each and analyzed with a PhosphorImager 445 SI (Molecular Dynamics) and autoradiographed.
Transient Expression of Bovine cDNA
The full-length FUT8b cDNA (2,921 bp) was amplified using forward (COS/1: 5'-GGGGATCCGCGGAGACCGCCTCTG-3') and reverse (COS/2: 5'-TGTCTAGAAAACATTGAGTACTAAAGAATC-3') primers. The PCR product was double-digested by BamHI and XbaI and inserted into the pcDNAI/Amp (pcDNAI/Amp/FUT8b) in order to transiently transfect COS-7 cells. SuperFect transfection reagent (QIAGEN) was used according to the protocol described by the manufacturer. After 48 h, COS-7 cells were harvested and washed with PBS, and proteins were subsequently extracted in lysis buffer (1% [v/v] Triton X100, 10 mM sodium cacodylate [pH 6], 20% [v/v] glycerol, 1 mM DTT) for 2 h at 4°C. The suspension was then centrifuged at 12,000 x g for 10 min at 4°C, and the supernatant was used for assays.
For expression into E. coli (strain AD494 (DE3) pLysS), two primers, Ec1 and Ec2 (table 1 ), including BamHI and NheI restriction sites, respectively, were used to amplify cDNA. The PCR product was then double-digested by the two corresponding restriction enzymes and subsequently introduced into pET-25b(+), a prokaryotic periplasmic expression vector. Transformed bacteria were grown in LB medium supplemented with ampicillin (100 µg/ml) at 37°C for 4 h. Induction of truncated FUT8b expression was performed by adding IPTG (1 mM). Bacteria were pelleted and then resuspended in extraction buffer containing Tris (30 mM, pH 7.9), sucrose 20% (w/v), and EDTA (0.1 mM). After incubation for 10 min at room temperature under slow agitation, the suspension was centrifuged for 10 min at 12,000 x g. Then, MgSO4 (5 mM) was added to the pellet, and the periplasmic fraction used for assays was extracted, after a 5 min incubation at 4°C under shaking, by a new centrifugation of 10 min at 12,000 x g.
6-Fucosyltransferase Enzyme Assays
Assays were made as previously described by Yazawa et al. (1998)
, at 37°C for 4 h in an incubation mixture containing 322 pmol (222,580 cpm) GDP-[14C]fucose, 4 µmol HEPES-NaOH buffer (pH 7.0), 10 nmol acceptor substrate Man3GlcNAc4, and 50 µg of protein in a total volume of 50 µl. Final products were then applied on a PEI-cellulose thin layer chromatography plate (Merck, Darmstadt, Germany) and developed with phosphate buffer (1 mM, pH 8.0). Radioactivity of 6-fucosylated acceptor was estimated by phosphorimaging.
Sequence Analysis
LALIGN (Huang and Miller 1991
) was used for mammalian FUT8 cDNA comparisons. Peptide sequence homologies were analyzed by BLASTP (Altschul et al. 1990
) in the SwissProt data bank. An exhaustive FASTA search of DNA and protein databases (Pearson and Lipman 1988
) was performed with human FUT1 (M35531), human FUT2 (U17894), human Sec1 (U17895), Caenorhabditis elegans 2-fucosyltransferase (Z92830_5), Helicobacter pylori 2-fucosyltransferase (AF076779), Arabidopsis thaliana 2-fucosyltransferase (AF154111_1), human FUT8 (D89289), and Azorhizobium caulinodans nodZ (L18897_8). These searches gave 87 2-fucosyltransferases and 9 6-fucosyltransferases. Putative fucosyltransferase open reading frames (ORFs) were identified in large genomic sequences (human AL109847, containing 186,110 bp, and Drosophila melanogaster AC018043, containing 34,322 bp) with the FGENE program from the Sanger Center genomic analysis Web tools.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsliterature cited |
|---|
Partial cDNA Cloning of Bovine
6-Fucosyltransferase and RACE ExperimentsSense and antisense PCR primers P1 and P2 (table 1 ) were designed in the most conserved regions of human and porcine cDNA sequences. Amplification was performed using lung bovine cDNA, since the existence of bovine core-fucosylated structures in this tissue has already been documented (Norgard-Sumnicht et al. 1995
). Agarose gel electrophoresis of the PCR products revealed only one DNA fragment, which, after subcloning into the T-vector pGEMT-Easy and sequencing, was 1,674 bp long. An identical PCR product was identified in other bovine tissues, such as brain, heart, kidney, spleen, and testis (data not shown). To complete the bovine cDNA sequence, we designed four primers near the 5' (P3, P4) and 3' (P5, P6) ends of the partial cDNA sequence (table 1
). In all of the tested tissues, we identified a unique 5' RACE product of 562 bp. Similar investigations were performed to obtain the 3' untranslated region (UTR) (primer combinations P6/AP1 and P5/AP2; table 1
), which gave only one RACE product of 735 bp. The entire 1,728-bp ORF of the bovine 6-fucosyltransferase gene was then reconstituted. It encodes a 575-aa polypeptide without potential N-glycosylation sites, organized as a type II transmembrane protein, like other known fucosyltransferases (fig. 1
). It contains a short cytoplasmic tail (7 aa), followed by a 18-aa transmembrane domain. The luminal moiety has a catalytic C-terminal domain preceded by a 66-aa proline-rich region (from Pro 292 to Pro 358), already observed in the human and pig FUT8 proteins (Uozumi et al. 1996
; Yanagidani et al. 1997
). Identities with human, mouse, and porcine enzymes were 97%, 96%, and 95%, respectively (fig. 1
). The sequence similarity with the putative C. elegans 6-fucosyltransferase (Wilson et al. 1994
) was about 38% (data not shown); however, the three 6-specific motifs described by Oriol et al. (1999)
were found in all enzymes of vertebrates, invertebrates, plants, and bacteria.
|
Transient Expression of
6-FucosyltransferaseTo ensure that bovine cDNA encoded an
6-fucosyltransferase, the entire cDNA was subcloned into the mammalian expression vector pcDNAI/Amp and transfected into COS-7 cells. The fucosyltransferase activity was measured in the cellular homogenates (table 2
) using ASAG biantennary glycan as acceptor substrate. COS-7 cells transfected with pcDNAI/Amp/FUT8b showed an activity about ninefold higher than the endogenous activity already reported by Clarke and Watkins (1999)
.
|
We also expressed an N-terminal truncated form of
6-fucosyltransferase (devoid of the first 26 aa corresponding to the cytoplasmic and transmembrane domains) into E. coli. Heterologous expression of bovine enzyme in bacterial and mammalian cells confirmed that bovine cDNA encodes a polypeptide able to transfer fucose in an 16 linkage on chitobiose-like substrates (table 2
).
Southern and Northern Blot Analysis
In order to determine how many homologous sequences exist, a monoexonic probe (probe I) was used to sample bovine, human, and porcine genomes. For each digestion, only one band was detected, suggesting that there is only one 6-fucosyltransferase gene in bovine, human, and porcine genomes (fig. 2A
).
|
Northern blot analysis of poly(A)+ mRNA from various bovine tissues showed a unique band at about 3 kb (fig. 2B ). Using ß-actin as a reference, the densitometric analysis revealed that brain presented the highest transcript content.
Exonic Structure of the Coding Sequence
To investigate the genomic structure of the FUT8b coding sequence, we generated a probe containing the entire ORF (1,728 bp) or a part of the fifth exon (probe I; see Materials and Methods). Southern blots probed with the entire ORF revealed more than one band (data not shown). Under the same stringency conditions, only one band was depicted with probe I (fig. 2A
) irrespective of the restriction enzymes used. Subsequent long-distance PCR analysis made with primers designed against ORF revealed variation in amplified fragment size depending on the template used.
Using genomic DNA, PCR products always appeared considerably longer than those obtained using cDNA. Altogether, these results suggested that the FUT8b ORF is interrupted by introns. A long PCR approach in the 5' and 3' directions was retained based on starting primers S1 and S2 (table 1 ). A genomic PCR product of about 5 kb (fig. 3 ), compared with the cDNA counterpart (329 bp), proves that the coding sequence was interrupted by intronic segments. In order to determine the exon/intron organization of the coding sequence, we used successive primers, allowing us to walk along the ORF (S3, S4, S5, S6, S7, S1inv, S2inv, S3inv, S6inv; table 1 ). Sequence analysis of the amplified products allowed us to identify five exons and four introns spanning 40 kb (fig. 4A and B ).
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsliterature cited |
|---|
We report the identification of a bovine
6-fucosyltransferase gene (FUT8b) with an ORF of 1,728 bp. This ORF encodes a 575-aa protein that is highly homologous to human, porcine, and mouse enzymes (fig. 1
). The bovine cDNA encoded an active enzyme (table 2
) which transfers fucose in 16 on the GlcNAc-containing acceptor. This FUT8 is the first mammalian fucosyltransferase expressed as an active enzyme in E. coli.
The genomic organization of the FUT8b coding sequence consists of five exons and four introns (fig. 4A and B
), spanning 40 kb. Exon 1 (375 bp from ATG) corresponds to the cytoplasmic and transmembrane domains and presents in its 3' end a luminal part including a region homologous to cytoskeleton proteins, with up to 61% similarity on 57 aa (table 3
). Exon 2 (707 bp) encodes the stem domain whose 3' end shows a proline-rich region including a XPXPPYXP sequence (residues 298–305) similar to the SH3-binding motif XPXXPPPFXP (in bovines, Y replaces F), usually composed of 10 aa (Pawson 1995
). The following three exons (exon 3, 177 bp; exon 4, 181 bp; exon 5, 288 bp to TAA) contain the catalytic domain. The peptide sequence encoded by exon 5 exhibits a significant homology (up to 52% on 60 aa) with the SH3 domain (table 3
). As recently shown for the postsynaptic density-95 protein (McGee and Bredt 1999
), this SH3 domain could have intramolecular interactions with the proline-rich region. The resulting loop might constitute a tulip-shaped scaffolding, including the catalytic domain corresponding to exons 3 and 4. A similar folded structure has previously been proposed for the crystal structure of the bovine ß4-galactosyltransferase T1 (Gastinel, Cambillau, and Bourne 1999
). This is comparable with bovine exon 3, which encodes a peptide sequence homologous to DNA gyrases (60% on 33 aa). This enzyme binds nucleotides via a specific binding domain crucial for its activity (Cullis, Maxwell, and Weiner 1997
). The peptide sequence encoded by exon 4 presents 57% homology on 27 aa with the glucosamine fructose-6-phosphate aminotransferase (table 3
). The luminal part of FUT8, encoded by exon 1, might be involved in protein-protein interactions with certain Golgi proteins analogous to cytoskeleton proteins (table 3
). These interactions could stabilize the correct orientation of the catalytic loop. The three-dimensional protein structure might also depend on disulfide bridges between the well-conserved cysteines (fig. 1
). Recently, we and others have shown that interactions between amino acids at both ends of the catalytic domain play a crucial role in the core structure of the fucosyltransferases (Holmes et al. 2000).
|
Comparison of the genomic DNA sequence of human chromosome 14 (AL109847), where FUT8 is assigned (Yamaguchi et al. 1999
; Cailleau-Thomas et al. 2000
) with the corresponding cDNA (Yanagidani et al. 1997
), revealed that the human FUT8 coding sequence (cds) has nine exons. The differences with the bovine gene (fig. 5 ) were in the partition of the first two exons, exons 1 and 2, which correspond to the first six human exons. Bovine exons 3 and 4 encode the catalytic domain and have their homologous counterparts in human exons J and K. The last human exon, exon L, encodes an SH3 domain, as does bovine exon 5, that could interact with the upstream proline-rich motif encoded by human exon I.
|
According to the exon/intron organization of genes (fig. 5 ), FUT8 polypeptides could be divided into two parts. The N-terminal part, which includes cytoplasmic, transmembrane, and stem domains, differs between human and bovine species in its exon organization. The C-terminal part overlaps the catalytic domain and includes the three
6-fucosyltransferase conserved peptide motifs. It presents similar organizations at the genomic level for the two species. The bovine gene seems to have lost certain intronic sequences, particularly in the region encoding the putative hypervariable stem domain, which is more divided in the human (six exons) than in the bovine (two exons) or the fruit fly D. melanogaster (three exons) (fig. 6
). The evolution process, by either gain or loss of introns, could have led to the emergence of new biological functions in humans. The possible existence of alternative transcripts that are different in the hypervariable stem region may confer various acceptor substrate specificities. Until now, no transcript of this type has ever been described. However, an alternative splicing mechanism has been reported in the human retina, in which a truncated EST lacking catalytic exons J and K was identified (Costache et al. 1997a
). Such a mature transcript may encode a protein that is without fucose transfer activity but is probably acting as a lectin, since the domain involved in acceptor substrate recognition is conserved. These data suggest that in spite of a major function in core fucosylation, which acts as a quality control point in the sequential glycosylation process, each organism has its own FUT8 organization.
|
Unlike other bovine glycosyltransferase genes, for which expression is developmentally regulated and different transcripts exist for the same gene (Mercier et al. 1999
; Wierinckx et al. 1999
), a unique FUT8b transcript of about 3 kb was identified in all the tissues tested, suggesting a ubiquitous expression.
Sequence comparisons of the 87 2-fucosyltransferases and the 9 6-fucosyltransferases available in GenBank/EBI, including the new bovine sequence, confirmed the existence of the three highly conserved peptide motifs in both enzyme families. The first two motifs have already been reported to have a common conserved amino acid structure in both the 2- and the 6-fucosyltransferase families, whereas the third motif (motif III) appeared to be more specific to each group of enzymes (Oriol et al. 1999
). However, the present analysis on a larger series of sequences with less stringent conditions suggests the existence of two conserved hydrophobic positions in motif III, which, in addition to the highly conserved distance between motifs II and III, also suggest a common origin for motif III (table 4
).
|
All of the eukaryote
2-fucosyltransferases have a monoexonic cds, with the exception of the 22 C. elegans putative 2-fucosyltransferase genes, which have a multiexonic cds. Alternatively, all of the 6-fucosyltransferase genes identified are predicted to be multiexonic. The three C. elegans 6-fucosyltransferase conserved motifs are in the same exon, whereas in mammals (humans and bovines) and in D. melanogaster, distributions were similar and in three different exons (fig. 6
).
In the organisms depicted in figure 6
, the 6-fucosyltransferase genes have evolved independently, as revealed by their structural differences. Nevertheless, in mammals, the enzymes act on the same donor and acceptor substrates, suggesting that the well-conserved peptide motifs I, II, and III (table 4
) might be involved in catalysis. Interestingly, these motifs are homologous to those found in 2-fucosyltransferases, with the spacing between them being remarkably maintained within the two enzyme families. This organization suggests a common origin and an emergence by duplication of an ancestor gene. However, between 2- and 6-fucosyltransferase genes, some amino acids differ in each motif, suggesting separate evolution and divergence after the duplication event.
The exon/intron structure of 6-fucosyltransferase genes is similar to the genomic organization of the mung bean 3-fucosyltransferase gene, in which the ORF comprises four exons (Leiter et al. 1999
). In both cases, enzymes transfer fucose on the asparagine-linked GlcNAc, in 16 linkage for mammals and in 13 linkage for plants. Both gene families have been considered ancestral (Oriol et al. 1999
). If we accept that 2- and 6-fucosyltransferases originated from a single duplication event of a common ancestor, they must have different evolutionary rates, since 20%–30% differences are observed among the different vertebrate 2-fucosyltransferases, whereas only 2%–5% differences are observed among the vertebrate 6-fucosyltransferases.
Our findings are in agreement with the exon theory of gene formation (Gilbert 1987
), which argues that the first proteins were mainly aggregates of short polypeptides and that subsequent genes were initially assembled from small exons. The bovine FUT8b gene is a new and good example of a direct correlation between exons and protein unit domains, as was previously shown for hemoglobin (Go 1981
) and other proteins. It could result from a shuffling of different coding exons, with the introns helping by recombination mechanisms to assemble the first genes (Gilbert, De Souza, and Long 1997
). An intron loss by retrotransposition (Gilbert, De Souza, and Long 1997
; Wierinckx et al. 1999
; Capy 2000
) could have occurred and led to the emergence of the other fucosyltransferases characterized by their monoexonic coding sequences (Oulmouden et al. 1997
; Wierinckx et al. 1999
; Barreaud et al. 2000
). Such a mechanism has already been observed for the human phosphoglycerate kinase gene, for which an autosomal intronless copy exists (McCarrey et al. 1987
), but also for the Drosophila alcohol dehydrogenase gene, a messenger which was retrotransposed and became a part of a newly originated functional gene (Long and Langley 1993
; Long, Wang, and Zhang 1999
). Interestingly, it was reported that in the yeast Saccharomyces cerevisiae, the paucity of introns could be attributed to ancestral retrotransposition mechanisms, based on homolog recombination (Fink 1987
).
In invertebrates (C. elegans and D. melanogaster), the three conserved peptide motifs (motifs I, II, and III) are in three separate exons in the 2-fucosyltransferase genes (data not shown), whereas in 6-fucosyltransferase genes, they reside in the same exon (fig. 6
). In the mammalian (bovine and human) 6-fucosyltransferase genes, the three motifs are on distinct exons. This could signify that some introns have appeared more recently during evolution, or that the gene organization of the invertebrate motifs has been conserved from the common 2/6-fucosyltransferase ancestor.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsliterature cited |
|---|
We thank François Gallet and Anne Wierinckx for their helpful discussions, Marie-Pierre Laforet for her technical assistance in transfection experiments, and Catherine Alonso (Centre National de la Recherche Scientifique technician) for the preparation of oligosaccharide substrate. This work was partially supported by the French network of recombinant Glycosyltransferases (GT-rec) and a Xenotransplantation grant from the European Union (EU).
|
Footnotes |
|---|
Pierre Capy, Reviewing Editor
1 Abbreviations: FUT8, human gene encoding the 6-fucosyltransferase; FUT8b, bovine gene encoding the 6-fucosyltransferase; ORF, open reading frame. 
2 Keywords: fucosyltransferase
exon/intron organization
conserved motifs
3 Address for correspondence and reprints: Raymond Julien, Institut des Sciences de la Vie et de la Santé, Faculté des Sciences, Université de Limoges, 123, avenue Albert Thomas, 87060 Limoges cedex, France. E-mail: rjulien{at}unilim.fr
|
literature cited |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsliterature cited |
|---|
Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403–410.[Web of Science][Medline]
Barreaud, J. P., K. Saunier, J. Souchaire, D. Delourme, A. Oulmouden, R. Oriol, H. Leveziel, R. Julien, and J. M. Petit. 2000. Three bovine -2-fucosyltransferase genes encode enzymes that preferentially transfer fucose on Galß1–3GalNAc acceptor substrates. Glycobiology 10:611–621.
Breton, C., R. Oriol, and A. Imberty. 1998. Conserved structural features in eukaryotic and prokaryotic fucosyltransferases. Glycobiology 8:87–94.
Cailleau-Thomas, A., P. Coullin, J. J. Candelier, L. Balanzino, B. Mennesson, R. Oriol, and R. Mollicone. 2000. FUT4 and FUT9 genes are expressed early in human embryogenesis. Glycobiology 10:789–802.
Capy, P. 2000. Is bigger better than cricket. Science 287:985–986.
Clarke, J. L., and W. M. Watkins. 1999. Expression of human -L-fucosyltransferase gene homologs in monkey kidney COS cells and modification of potential fucosyltransferase acceptor substrates by an endogenous glycosidase. Glycobiology 9:191–202.
Costache, M., P. A. Apoil, A. Cailleau, A. Elmgren, G. Larson, S. Henry, A. Blancher, D. Iordachescu, R. Oriol, and R. Mollicone. 1997a. Evolution of fucosyltransferase genes in vertebrates. J. Biol. Chem. 272:29721–29728.
Costache, M., A. Cailleau, P. Fernandez-Mateos, R. Oriol, and R. Mollicone. 1997b. Advances in molecular genetics of -2- and -3/4-fucosyltransferases. Transfus. Clin. Biol. 4:367–382.
Cullis, P. M., A. Maxwell, and D. P. Weiner. 1997. Exploiting nucleotide triphosphates to probe mechanistic aspects of Escherichia coli DNA gyrase. Biochemistry 36:6059–6068.
Dupuy, F., J. M. Petit, R. Mollicone, R. Oriol, R. Julien, and A. Maftah. 1999. A single amino acid in the hypervariable stem domain of vertebrate 1,3/1,4 fucosyltransferases determines the type 1/type 2 transfer. Characterization of acceptor substrate specificity of the Lewis enzyme by site-directed mutagenesis. J. Biol. Chem. 274:12257–12262.
Fink, G. R. 1987. Pseudogenes in yeast? Cell 49:4–5.
Fitchette-Lainé, A. C., V. Gomord, M. Cabanes, J. C. Michalski, M. Saint-Macary, B. Foucher, B. Cavelier, C. Hawes, P. Lerouge, and L. Faye. 1997. N-glycans harboring the Lewis a epitope are expressed at the surface of plant cells. Plant J. 12:1411–1417.[Web of Science][Medline]
Gastinel, L. N., C. Cambillau, and Y. Bourne. 1999. Crystal structures of the bovine ß4galactosyltransferase catalytic domain and its complex with uridine diphosphogalactose. EMBO J. 18:3546–3557.[Web of Science][Medline]
Gilbert, W. 1987. The exon theory of genes. Cold Spring Harb. Symp. Quant. Biol. 52:901–905.
Gilbert, W, S. J. De Souza, and M. Long. 1997. Origin of genes. Proc. Natl. Acad. Sci. USA 94:7698–7703.
Go, M. 1981. Correlation of DNA exonic regions with protein structural units in haemoglobin. Nature 291:90–93.
Grabenhorst, E., and H. S. Conradt. 1999. The cytoplasmic, transmembrane, and stem regions of glycosyltransferases specify their in vivo functional sublocalization and stability in the Golgi. J. Biol. Chem. 274:36107–36116.
Holmes, E. H., T. Y. Yen, S. Thomas, R. Joshi, A. Nguyen, T. Long, F. Gallet, A. Maftah, R. Julien, and B. A. Macher. 2000. Human 1,3/4 fucosyltransferases: characterization of highly conserved cysteines residues and N-linked glycosylation sites. J. Biol. Chem. 275:24237–24245.
Huang, X., and W. Miller. 1991. A time-efficient, linear-space local similarity algorithm. Adv. Appl. Math. 12:337–357.
Kaminska, J., M. C. Glick, and J. Koscielak. 1998. Purification and characterization of GDP-L-Fuc: N-acetyl ß-D-glucosaminide 1,6 fucosyltransferase from human blood platelets. Glycoconj. J. 15:783–788.[Web of Science][Medline]
Kojima, N., Y. Tachida, Y. Yoshida, and S. Tsuji. 1996. Characterization of mouse ST8Sia II (STX) as a neural cell adhesion molecule-specific polysialic acid synthase. Requirement of core 1,6-linked fucose and a polypeptide chain for polysialylation. J. Biol. Chem. 271:19457–19463.
Leiter, H., J. Mucha, E. Staudacher, R. Grimm, J. Glossl, and F. Altmann. 1999. Purification, cDNA cloning, and expression of GDP-L-Fuc: Asn-linked GlcNAc 1,3-fucosyltransferase from mung beans. J. Biol. Chem. 274:21830–21839.
Lemesle-Varloot, L., B. Henrissat, C. Gaboriaud, V. Bissery, A. Morgat, and J. P. Mornon. 1990. Hydrophobic cluster analysis: procedures to derive structural and functional information from 2-D representation of protein sequence. Biochimie 72:555–574.
Long, M., and C. H. Langley. 1993. Natural selection and the origin of jingwei, a chimeric processed functional gene in Drosophila. Science 260:91–95.
Long, M., W. Wang, and J. Zhang. 1999. Origin of new genes and source for N-terminal domain of the chimerical gene, jingwei, in Drosophila. Gene 238:135–141.
Longmore, G., and H. Schachter. 1982. Product-identification and substrate-specificity studies of the GDP-L-fucose: 2-acetamido-2-deoxy-ß-D-glucoside (FUC goes to Asn-linked GlcNAc) 6--L-fucosyltransferase in a Golgi-rich fraction from porcine liver. Carbohydr. Res. 100:365–392.[Web of Science][Medline]
McCarrey, J. R., and K. Thomas. 1987. Human testis specific PGK gene lacks introns and possesses characteristics of a processed gene. Nature 326:501–505.
McGee, A. W., and D. S. Bredt. 1999. Identification of an intramolecular interaction between the SH3 and guanylate kinase domain of PSD-95. J. Biol. Chem. 274:17431–17436.
Mercier, D., A. Wierinckx, A. Oulmouden, P. F. Gallet, M. M. Palcic, A. Harduin-Lepers, P. Delannoy, J. M. Petit, H. Leveziel, and R. Julien. 1999. Molecular cloning, expression and exon/intron organization of the bovine ß-galactoside 2,6-sialyltransferase gene. Glycobiology 9:851–863.
Miyoshi, E., N. Uozumi, K. Noda, N. Hayashi, M. Hori, and N. Taniguchi. 1997. Expression of 1,6 fucosyltransferase in rat tissues and human cancer cell lines. Int. J. Cancer 72:1117–1121.
Nguyen, A. T., E. H. Holmes, J. M. Whitaker, S. Ho, S. Shetterly, and B. A. Macher. 1998. Human 1,3/4 fucosyltransferases. Identification of amino acids involved in acceptor substrate binding by site directed mutagenesis. J. Biol. Chem. 273:25244–25249.
Noda, K., E. Miyoshi, N. Uozumi, C. X. Gao, K. Suzuki, N. Hayashi, M. Hori, and N. Taniguchi. 1998. High expression of 1,6 fucosyltransferase during rat hepatocarcinogenesis. Int. J. Cancer 75:444–450.
Norgard-Sumnicht, K. E., L. Roux, D. K. Toomre, A. Manzy, H. H. Freeze, and A. Varki. 1995. Unusual anionic N-linked oligosaccharides from bovine lung. J. Biol. Chem. 270:27634–27645.
Noronski, T., and I. Mononen. 1997. Influence of L-fucose attached 1,6 to the asparagine-linked N-acetylglucosamine on the hydrolysis of the N-glycosidic linkage by human glycosylasparaginase. Glycobiology 7:217–220.
Oriol, R., R. Mollicone, A. Cailleau, L. Balanzino, and C. Breton. 1999. Divergent evolution of fucosyltransferase genes from vertebrates, invertebrates, and bacteria. Glycobiology 9:323–334.
Oulmouden, A., A. Wierinckx, J. M. Petit, M. Costache, M. M. Palcic, R. Mollicone, R. Oriol, and R. Julien. 1997. Molecular cloning and expression of a bovine 1,3 fucosyltransferase gene homologous to a putative ancestor gene of the human FUT3-FUT5-FUT6 cluster. J. Biol. Chem. 272:8764–8773.
Paulson, J. C., and K. Colley. 1989. Glycosyltransferases. J. Biol. Chem. 264:17615–17618.
Pawson, T. 1995. Protein modules and signalling networks. Nature 373:573–580.
Pearson, W. R., and D. J. Lipman. 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA 85:2444–2448.
Shimizu, H., K. Ochiai, K. Ikenaka, K. Mikoshiba, and S. Hase. 1993. Structures of N-linked sugar chains expressed mainly in mouse brain. J. Biochem. 114:334–338.
Staudacher, E., and L. Marz. 1998. Strict order of (Fuc to Asn-linked GlcNAc) fucosyltransferases forming core-difucosylated structures. Glycoconj. J. 15:355–360.[Web of Science][Medline]
Stubbs, H. J., H. J. Lih, T. L. Gustafson, and K. G. Rice. 1996. Influence of core fucosylation on the flexibility of a biantennary N-linked oligosaccharide. Biochemistry 35:937–947.
Uozumi, N., S. Yanagidani, E. Miyoshi, Y. Ihara, T. Sakuma, C. X. Gao, T. Teshima, S. Fujii, T. Shiba, and N. Taniguchi. 1996. Purification and cDNA cloning of porcine brain GDP-L-Fuc: N-acetyl-ß-D-glucosaminide 1,6 fucosyltransferase. J. Biol. Chem. 271:27810–27817.
Vo, L., S. Lee, M. C. Marcinko, E. H. Holmes, and B. A. Macher. 1998. Human 1,3/4 fucosyltransferases. A single amino acid at the COOH terminus of FucT III and V alters their kinetic properties. J. Biol. Chem. 273:25250–25255.
Wierinckx, A., D. Mercier, A. Oulmouden, J. M. Petit, and R. Julien. 1999. Complete genomic organization of futb encoding a bovine 3-fucosyltransferase: exons in human orthologous genes emerged from ancestral intronic sequences. Mol. Biol. Evol. 16:1535–1547.[Abstract]
Wilson, R., R. Ainscough, K. Anderson et al. (53 co-authors). 1994. 2.2 Mb of contiguous nucleotide sequence from chromosome III of C. elegans. Nature 368:32–38.
Xu, Z., V. Loc, and A. Macher. 1996. Structure-function analysis of human 1,3-fucosyltransferase. J. Biol. Chem. 271:8818–8823.
Yamaguchi, N., J. Fujii, S. Inoue, N. Uozumi, S. Yanagidani, Y. Ikeda, M. Egashira, O. Miyoshi, N. Niikawa, and N. Taniguchi. 1999. Mapping of the -1,6-fucosyltransferase gene, FUT8, to human chromosome 14q24.3. Cytogenet. Cell Genet. 84:58–60.[Web of Science][Medline]
Yanagidani, S., N. Uozumi, Y. Ihara, E. Miyoshi, N. Yamaguchi, and N. Taniguchi. 1997. Purification and cDNA cloning of GDP-L-Fuc:N-acetyl-ß-D-glucosaminide: 1–6 fucosyltransferase (1–6 FucT) from human gastric cancer MKN45 cells. J. Biochem. 121:626–632.
Yazawa, S., N. Kochibe, T. Nishimura, C. Shima, I. Takai, M. Adachi, T. Asao, T. Hada, Y. Enoki, and L. R. Juneja. 1998. A novel method for determination of (1,6)fucosyltransferase activity using a reducing oligosaccharide from egg yolk as a specific acceptor. Glycoconj. J. 15:863–871.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. K. Qasba and B. Ramakrishnan Letter to the Glyco-Forum: Catalytic domains of glycosyltransferases with 'add-on' domains Glycobiology, May 1, 2007; 17(5): 7G - 9G. [Full Text] [PDF]
|
|
|
|
|
|
H. Ihara, Y. Ikeda, S. Toma, X. Wang, T. Suzuki, J. Gu, E. Miyoshi, T. Tsukihara, K. Honke, A. Matsumoto, et al. Crystal structure of mammalian {alpha}1,6-fucosyltransferase, FUT8 Glycobiology, May 1, 2007; 17(5): 455 - 466. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Ihara, Y. Ikeda, and N. Taniguchi Reaction mechanism and substrate specificity for nucleotide sugar of mammalian {alpha}1,6-fucosyltransferase--a large-scale preparation and characterization of recombinant human FUT8 Glycobiology, April 1, 2006; 16(4): 333 - 342. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Paschinger, E. Staudacher, U. Stemmer, G. Fabini, and I. B. H. Wilson Fucosyltransferase substrate specificity and the order of fucosylation in invertebrates Glycobiology, May 1, 2005; 15(5): 463 - 474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Martinez-Duncker, J.-C. Michalski, C. Bauvy, J.-J. Candelier, B. Mennesson, P. Codogno, R. Oriol, and R. Mollicone Activity and tissue distribution of splice variants of {alpha}6-fucosyltransferase in human embryogenesis Glycobiology, January 1, 2004; 14(1): 13 - 25. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Saunier, J.-P. Barreaud, A. Eggen, R. Oriol, H. Leveziel, R. Julien, and J.-M. Petit Organization of the Bovine {alpha}2-Fucosyltransferase Gene Cluster Suggests that the Sec1 Gene Might Have Been Shaped Through a Nonautonomous L1-Retrotransposition Event Within the Same Locus Mol. Biol. Evol., November 1, 2001; 18(11): 2083 - 2091. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||