Glycobiology vol. 8 no. 1 pp. 87–94, 1998
Conserved structural features in eukaryotic and prokaryotic fucosyltransferases
Christelle Breton2, Rafael Oriol1 and Anne Imberty
CERMAV-CNRS (affiliated to the University Joseph Fourier), BP 53, F-38041
Grenoble Cedex 9, France and 1INSERM U178, University of Paris-Sud XI,
F-94807 Villejuif cedex, France
Received on June 23, 1997; revised on July 23, 1997; accepted on July 31, 1997
2To
whom correspondence should be addressed
Fucosyltransferases are the enzymes transferring fucose
from GDP-Fuc to Gal in an α1,2-linkage and to GlcNAc in
α1,3-, α1,4-, or α1,6-linkages. Since all fucosyltransferases
utilize the same nucleotide sugar, their specificity will
probably reside in the recognition of the acceptor and in the
type of linkage formed. A search of nucleotide and protein
databases yielded more than 30 sequences of fucosyltransferases originating from mammals, chicken, nematode, and
bacteria. On the basis of protein sequence similarities, these
enzymes can be classified into four distinct families: (1) the
α-2-fucosyltransferases, (2) the α-3-fucosyltransferases, (3)
the mammalian α-6-fucosyltransferases, and (4) the bacterial
α-6-fucosyltransferases. Nevertheless, using the sensitive
hydrophobic cluster analysis (HCA) method, conserved
structural features as well as a consensus peptide motif have
been clearly identified in the catalytic domains of all α-2 and
α-6-fucosyltranferases, from prokaryotic and eukaryotic
origin, that allowed the grouping of these enzymes into one
superfamily. In addition, a few amino acids were found
strictly conserved in this family, and two of these residues
have been reported to be essential for enzyme activity for a
human α-2-fucosyltransferase. The α-3-fucosyltransferases
constitute a distinct family as they lack the consensus peptide,
but some regions display similarities with the α-2 and
α-6-fucosyltranferases. All these observations strongly
suggest that the fucosyltransferases share some common
structural and catalytic features.
Key words: fucosyltransferases/hydrophobic cluster analysis/
protein sequence analysis
Introduction
Fucosyltransferases are a group of enzymes catalyzing the
transfer of fucose from GDP-Fuc, in α1,2-, α1,3-, α1,4-, and
α1–6-linkages, with inversion of anomeric configuration, to
various oligosaccharide acceptors. Fucosylated glycoconjugates
may play important roles in physiological and pathological
processes such as fertilization, embryogenesis, lymphocyte
trafficking, immune response, and cancer metastasis (SchenkelBrunner, 1995; Staudacher, 1996). All eukaryotic fucosyltransferases cloned to date have the typical domain structure of type
II transmembrane proteins, consisting of a short NH2-terminal
cytoplasmic tail, a transmembrane domain and a stem region
followed by a large globular COOH-terminal catalytic domain
(Joziasse, 1992; Kleene and Berger 1993). Since all fucosyltrans 1998 Oxford University Press
ferases utilize the same nucleotide sugar as donor, their specificity
will probably reside in the recognition of the acceptor substrate
and in the type of linkage formed.
The α-2- and α-3/4-fucosyltransferases have attracted a large
interest in recent years as they are involved in the last steps of the
biosynthesis of ABH and Lewis related carbohydrate antigens
(Henry et al., 1995; Watkins, 1995; Costache et al., 1997). Seven
human FUT genes (named FUT1 to FUT7 according to the
chronology of cloning) and 14 animal genes (named according to
their homology with the human counterparts) have been cloned
and sequenced. FUT1 and FUT2 genes encode for the α-2-fucosyltransferases H and Se (secretor), respectively, while FUT3 to
FUT7 genes code for the α-3-fucosyltransferases Fuc-TIII to
Fuc-TVII. The enzymes Fuc-TIII and Fuc-TV also display
α-4-fucosyltransferase activity (Weston et al., 1992a; Wong et
al., 1992). A third group of mammalian fucosyltransferases has
emerged with the recent cloning of the pig and human α-6-fucosyltransferases which are implied in the synthesis of N-glycans
(Uozumi et al., 1996; Yanagidani et al., 1997). These two
enzymes exhibit no significant homology with other cloned
fucosyltransferases.
The presence of fucosyltransferase activities has also been
demonstrated in plants, insects, snails, parasites, and bacteria (for
a review, see Staudacher, 1996). At the present time, two bacterial
enzymes (NodZ), from rhizobia species (Azorhizobium caulinodans and Bradyrhizobium japonicum) have been cloned (Stacey
et al., 1994; Mergaert et al., 1996). These enzymes catalyze the
transfer of fucose from GDP-Fuc in α1,6 linkage to the GlcNAc
at the reducing end of Nod factors. Other protein sequences
originating from the random genomic sequencing of Caenorhabditis elegans (Wilson et al., 1994) and the bacterium Yersinia
enterocolitica (Zhang et al., 1997) have been considered as
putative fucosyltransferases since they display a low but
significant homology with some mammalian fucosyltransferases,
but their functionality has not yet been demonstrated.
The sensitive hydrophobic cluster analysis (HCA) method
(Gaboriaud et al., 1987; Lemesle-Varloot et al., 1990) has been
successfully used for the grouping of proteins exhibiting low
sequence identity and for prediction of catalytic residues in a
number of glycosyl hydrolases (Henrissat et al., 1995). Using
HCA and fold recognition methods, we have previously
suggested that mammalian α-2- and α-3-fucosyltransferases,
despite their lack of sequence identity, are related proteins sharing
the same type of fold (Breton et al., 1996). We have also predicted
the occurrence of a nucleotide binding domain of the Rossmantype in the catalytic domain of these enzymes. The aim of the
present study was to extend the protein sequence analysis to all
the prokaryotic and eukaryotic fucosyltransferase sequences
available in data banks, to provide insight into structure–function
relationships of this class of enzymes.
Results and discussion
A search of nucleotide and protein databases yielded 37 sequences
of fucosyltransferases originating from mammals, chicken, nema87
C.Breton, R.Oriol and A.Imberty
tode, and bacteria. Among them, 13 are partial sequences,
hypothetical proteins, or pseudogenes. Only the complete sequences, listed in Table I, were considered in this study. The protein
sizes are extremely variable ranging from 283 amino acids for the
bacterial Yersinia enterocolitica sequence to 575 residues for the
mammalian α-6-fucosyltransferases. On the basis of protein
sequence similarities, these sequences can be classified into four
families (1) the α-2-fucosyltransferases, (2) the α-3-fucosyltransferases, (3) the mammalian α-6-fucosyltransferases, and (4) the
bacterial α-6-fucosyltransferases. Large homologies in protein
sequences were found for human and animal fucosyltransferases
within each family, ranging from 50% to more than 80% overall
identity for members of the α-2-fucosyltransferase family and
from 40% to 99% for members of the α-3-fucosyltransferase
family. The Y.enterocolitica sequence (an hypothetical fucosyltransferase) exhibits a low but significant homology with other
sequences of α-2-fucosyltransferase (22–27%) and was included
in this family. The three Caenorhabditis elegans sequences are also
hypothetical fucosyltransferases which can be grouped into the
α-3-fucosyltransferase family as they share 25–31% of identity
with the human and animal α-3-fucosyltransferases. The two
recently cloned mammalian α-6-fucosyltransferases display high
homology (96% of identity) and appear to be unrelated with the
two bacterial enzymes NodZ.
Table I. Origin of the fucosyltransferase sequences
Accession
number
Origin
Gene
Enzyme
Size (aa)
References
Man
FUT1
H
365
Larsen et al., 1990
α-2-Fucosyltransferases
M35531
L50534
Pig
FUT1
H
365
Cohney et al., 1996
X80226
Rabbit
FUT1
H-I
373
Hitoshi et al., 1995
Y09883
Mouse
FUT1
H
377
Hitoshi et al., 1996a
U17894
Man
FUT2
Se
343
Kelly et al., 1995
X91269
Rabbit
FUT2
Se-III
354
Hitoshi et al., 1996b
X80225
Rabbit
Sec1
Se-II
347
Hitoshi et al., 1995
Y09882
Mouse
Sec1
Sec1
368
Hitoshi et al., 1996a
U46859
Y.enterocolitica
ORF11.8
yeFuc-Ta
283
Zhang et al., 1997
X53578
Man
FUT3
Fuc-TIIIc
361
Kukowska-Latallo et al.,1990
Y14033
Chimpanzee
FUT3
Fuc-TIIIc
372
Costache et al., in press
374
Weston et al., 1992a
α-3-Fucosyltransferases
M81485
Man
FUT5
Fuc-TVc
Y14034
Chimpanzee
FUT5
Fuc-TVc
374
Costache et al., in press
L01698
Man
FUT6
Fuc-TVI
359
Weston et al. 1992b
Y14035
Chimpanzee
FUT6
Fuc-TVI
359
Costache et al., in press
X87810
Bovine
FUTb
Fuc-Tb
365
Oulmouden et al., 1997
M58596
Man
FUT4
Fuc-TIV
405
Goelz et al., 1990
U33457
Mouse
FUT4
Fuc-TIV
433
Gersten et al., 1995
U58860
Rat
FUT4
Fuc-TIV
433
Sajdel-Sulkowska et al., 1997
U73678
Chicken
FUT4
Fuc-TIV
356
Lee et al., 1996
X78031
Man
FUT7
Fuc-TVII
341
Sasaki et al., 1994
U45980
Mouse
FUT7
Fuc-TVII
389
Smith et al., 1996
Z66497
C.elegans
K08F8.3
ceFUTAa
451
Wilson et al., 1994
331(C-t)b
Wilson et al., 1994
450(N-t)b
Wilson et al., 1994
U40028
C.elegans
T05A7.5
ceFUTBa
U40028
C.elegans
T05A7.5
ceFUTCa
Mammalian α-6-fucosyltransferases
D86723
Pig
FUT8
Fuc-TVIII
575
Uozumi et al., 1996
D89289
Man
FUT8
Fuc-TVIII
575
Yanagidani et al., 1997
Bacterial α-6-fucosyltransferases
L18897
A.caulinodans
nodZ
NodZ
328
Mergaert et al., 1996
L22756
B.japonicum
nodZ
NodZ
324
Stacey et al. 1994
Only the complete cds were analyzed in the present study.
aHypothetical proteins. The ORF T05A7.5 encodes a large protein (1653 aa) which exhibits significant homology with α-3-fucosyltransferases in its N-term and
C-term parts. The three C.elegans sequences were named for this study ceFUTA, ceFUTB, and ceFUTC.
bEstimated size of the domain sharing homology with α-3-fucosyltransferases.
cFuc-TIII and Fuc-TV also express α-4-fucosyltransferase activity.
88
Conserved structural features in fucosyltransferases
Fig. 1. Protein sequence alignments of the two highly conserved regions I and II of α-3-fucosyltransferases. Highly conserved amino acids are represented in
boldface characters. The asterisks indicate identical residues. Protein designations are as in Table I.
The degree of identity between proteins belonging to the
different families is too low (typically less than 20%) to obtain a
reliable global alignment using the classical 1D sequence
alignment methods (Bestfit, Gap, or Pileup of the GCG package).
Because of its known sensitivity at very low sequence identity, the
HCA method was used for the comparison of fucosyltransferase
sequences. The effectiveness of HCA comes from its ability to
significantly detect the regular secondary structure elements
constituting the hydrophobic core of globular proteins (Woodcock et al., 1992). This method uses a bidimensional representation of the protein sequence and the clusters of contiguous
hydrophobic residues are drawn. The analysis involves the visual
comparison of hydrophobic cluster shapes and their distribution
in order to find similarities between protein sequences.
The α-3-fucosyltransferases constitute a homogenous family
since these enzymes display between 30% and more than 90% of
identity. The HCA alignment is straightforward and similarities
were clearly observed in the catalytic domains of the three
C.elegans sequences and the mammalian α-3-fucosyltransferases. However, we have to keep in mind that the C.elegans
proteins are putative fucosyltransferases and that their functionality still needs to be proved. Since the catalytic amino acids are
submitted to intense conservation pressure, they are generally
located in the best conserved regions. The divergence brought in
by the invertebrate sequences can help to identify such regions but
the number of conserved amino acids is still too high to predict
those which could be important for activity. Nevertheless, two
highly homologous regions, named I and II, were detected and
they are represented in Figure 1 as a multiple linear sequence
alignment deduced from the HCA comparison of all the
sequences of this family. Seven residues in region I and 12 in
region II are found to be identical. Other highly conserved basic,
aromatic, and aliphatic amino acids are also observed in these two
regions.
In contrast, the α-2-fucosyltransferases, the mammalian
α-6-fucosyltransferases, and the bacterial α-6-fucosyltransferases constitute a more heterogenous family. As indicated
above, these three groups of fucosyltransferases display no
significant homology. However, by HCA, a region spanning
approximately 100 residues was found conserved in the catalytic
domain of all the enzymes belonging to this large family. Figure
2 displays the partial HCA plots of four selected Fuc-Ts. These
four protein sequences exhibit 16–25% overall identity. The
vertical lines delineate hydrophobic clusters which were found
conserved in the C-terminal part of all the α-2 and α-6-fucosyltransferases. Homologies were also observed in the N-terminal
part of the catalytic domains of the human H enzyme and the
putative Y.enterocolitica fucosyltransferase (dotted vertical
lines). In addition, a consensus peptide sequence corresponding
to the boxed region in Figure 1, has been clearly identified. This
peptide motif (V/I - G - V/I - H - V/I - R - R/H - G/T - D/N) is
detailed in Figure 3 with all protein members exhibiting α-2 and
α-6-fucosyltransferase activities. At first sight, the B.japonicum
NodZ sequence appeared to be different in the C-terminal end
from its counterpart from A.caulinodans, and we were unable to
find similarities by HCA in this region. However, a shift in the
reading frame in this region restored the similarity with the other
sequences and extended the sequence of B.japonicum NodZ by
approximately 40 residues. These observations allow the grouping of these enzymes in one superfamily, and one can expect that
these proteins share common structural and catalytic features in
the concerned region.
What could be the significance of such a conserved motif
present in prokaryotes and eukaryotes and more strikingly in
enzymes catalyzing two very different reactions (i.e,. α-2-linkage
on galactose and α-6 linkage on N-acetylglucosamine)? The
common feature of all the fucosyltransferases is the use of the
same nucleotide sugar and one can argue that the consensus
89
C.Breton, R.Oriol and A.Imberty
Fig. 2. HCA comparison of eukaryotic and prokaryotic α-2- and α-6-fucosyltransferases. The one-letter code for amino acid is used except for Gly, Pro, Ser, and
Thr, which are represented by diamonds, stars, squares with solid dots, and open squares, respectively. The vertical lines indicate the correspondences found
between the protein sequences. The best conserved residues are indicated on a black background.
Fig. 3. Sequence alignment of the best conserved region in α-2- and
α-6-fucosyltransferases. The consensus peptide motif is indicated in boldface
characters.
peptide sequence is part of the nucleotide binding domain.
Although the peptide motif has not been detected in the
90
α-3-fucosyltransferase family, a closer examination of the HCA
plots suggests that the region containing the peptide motif in α-2
and α-6-fucosyltransferases could correspond to the region I
defined for the α-3-fucosyltransferases in Figure 1. These two
regions display similar hydrophobic cluster shapes as well as
conserved basic and acidic residues (Figure 4). Similarly, the
region II of the α-3-fucosyltransferases (i.e., Fuc-TIII230–260)
could match with another well conserved region in α-2 and
α-6-fucosyltransferases (i.e., H300–330) as they share common
structural features in the HCA representation. Nevertheless, the
correspondences indicated in Figure 4 are still highly speculative
especially for the region II.
Some important structural and functional information have
been recently reported for several fucosyltransferases. These
studies were aimed to identify the amino acids that are possibly
involved in nucleotide sugar and acceptor recognition and also to
get new insights in the mechanism of these enzymes. Sixty-one
and 75 amino acids can be eliminated from the N-terminus of
Fuc-TIII and Fuc-TV, respectively, without significant loss of
enzyme activity (Xu et al., 1996). These truncation experiments
allowed the delineation of the stem region and the catalytic
domain. Thus the catalytic domains of these two enzymes begin
with a large hydrophobic region in which a conserved Trp (Trp68
in Fuc-TIII) was shown to be essential for activity (Elmgren et al.,
1997). In contrast the removal of one or more amino acids from
Conserved structural features in fucosyltransferases
Fig. 4. Partial HCA plots comparison of the peptide motif region of human
H enzyme and the region I of human Fuc-TIII. The conserved hydrophobic
clusters are shaded and the conserved basic and acidic residues are indicated
on a black background.
the C-terminus resulted in a dramatic or total loss of enzyme
activity (Xu et al., 1996). Domain swapping experiments were
also performed to localize the region that account for the
difference observed in acceptor specificity between Fuc-TIII,
Fuc-TV, and Fuc-TVI (Legault et al., 1995; Xu et al., 1996). As
few as 11 nonidentical amino acids, found within a hypervariable
peptide segment positioned at the N-terminus of the catalytic
domain (region 103–153 in Fuc-TIII), determine whether or not
these enzymes can add fucose in α1,4-linkage to type I acceptor
(Legault et al., 1995). Another study has also demonstrated the
importance of a cysteine residue probably located in the GDP-Fuc
binding pocket of Fuc-TIII, Fuc-TV, and Fuc-TVI enzymes
(Holmes et al., 1995). This GDP-fucose-protectable cysteine
(Cys143 in Fuc-TIII) is located at the end of the region identified
as being involved in acceptor recognition. In a very recent report,
the amino acids essential for the activity of Fuc-TVI were
investigated through chemical modification of the enzyme with
group-selective reagents (Britten and Bird, 1997). Cysteine and
histidine residues were shown to be essential for activity, but no
inactivation was observed with reagents selective for arginine and
lysine. The authors have identified five histidine residues
conserved within the human α-3-fucosyltransferases. From our
HCA alignment, His108 and His109 of Fuc-TVI appear to be the
most conserved ones. However, a GDP-fucose-protectable lysine
residue has been shown to be present in an α-3-fucosyltransferase
isolated from human lung small cell carcinoma NCI-H69 cells
(Holmes, 1992).
The recognition of complex carbohydrates by proteins occurred through hydrogen bond interactions with hydroxyl groups
on carbohydrate and through stacking of the sugar ring with
aromatic amino acids (Vyas, 1991). A series of deoxygenated
type I and II acceptors were used to identify key polar groups on
acceptors that are essential in enzyme binding (Hindsgaul et al.,
1991; deVries et al., 1995; Du et al., 1996). In addition to the
reactive hydroxyl at C-3 or C-4 of GlcNAc, Fuc-TIII, FucT-IV,
and FucT-V were shown to have an absolute requirement for a
hydroxyl group at carbon C-6 of galactose (de Vries et al., 1995;
Du et al., 1996). In a previous study, Hindsgaul et al. (1991) have
suggested that, for the enzymes Fuc-TIII and Fuc-TVI, the
reactive acceptor hydroxyl group is involved in a critical
hydrogen bond donor interaction with a base-acting group on the
enzyme which removes the developing proton during the
glycosyl transfer. This group has not yet been identified and could
vary from one enzyme to another, but typical hydrogen bond
acceptors include histidines and carboxylates. However, with
regard to the mechanism of action of Fuc-TV, it has been recently
proposed that the glycosidic bond cleavage occurs prior to the
nucleophilic attack (Murray et al., 1997), and this enzyme has
been shown to have a catalytic residue with pKa = 4.1, which is
presumably a carboxylic acid residue (Murray et al., 1996).
Product inhibition studies have also demonstrated that Fuc-TV
has an ordered, sequential mechanism with GDP-Fuc binding
first and the product GDP releasing last (Qiao et al., 1996).
The HCA method also permits the identification of strictly
conserved amino acids which are of interest because they can be
involved in nucleotide donor and acceptor recognition and in
catalysis. Invariant residues such as aspartic or glutamic acid
could be of particular importance for catalysis (Murray et al.,
1996), whereas basic residues could interact with the phosphate
groups of the nucleotide sugar donor (Vrielink et al., 1994), and
aromatic and polar residues are frequently found in protein–
carbohydrate interactions (Vyas et al., 1991). Among the
α-3-fucosyltransferases, more than 20 basic, acidic, and polar
residues are found conserved in the catalytic domains of these
enzymes, and thus it is not possible to predict which ones are the
most important for activity. Only two cysteine residues both
located in the N-terminal part of the catalytic domain were found
highly conserved in all the α-3-fucosyltransferases (Cys81 and
Cys91 in Fuc-TIII). In contrast, only a few amino acids are strictly
conserved in the α-2 and α-6-fucosyltransferase superfamily, and
they are displayed in Figure 2 on a black background. The peptide
motif contains three conserved basic (Arg, His) and one aspartic
acid (or Asn in NodZ sequences) residues. Other invariant amino
acids, including two acidic residues were also found. Two
cysteine residues are found conserved among all the α-2-fucosyltransferases including the Y.enterocolitica sequence
(corresponding to Cys169 and Cys268 in the human H enzyme).
More than 20 sporadic nonprevalent inactivating mutations in
FUT1 have been reported among H deficient individuals
worldwide (Kelly et al., 1994; Kaneko et al., 1997; Wagner and
Flegel 1997). Three types of mutation can give the H deficient
phenotype: missense, nonsense, or microdeletions resulting in
frameshifts. The single point missense mutations occurring in the
best conserved part of the catalytic domain of the human H
enzyme are listed in Table II. All the amino acids concerned by
these mutations are strictly conserved among the mammalian
α-2-fucosyltransferases. The corresponding amino acids in the
Y.enterocolitica protein, pig α-6-Fuc-T and A.caulinodans NodZ
sequences were deduced from the HCA alignment. Identical or
conservative residues were found at the same positions in the
Y.enterocolitica sequence. Two of these mutations (Arg220Cys
and Asp278Asn) are indeed interesting as they involve amino
acids which are found conserved in all the α-2 and α-6-fucosyltransferases. The residue Arg220 is located in the consensus
peptide motif which we propose to be part of the nucleotide
binding domain. The isosteric replacement Asp278Asn
completely abolishes enzyme activity. These observations strongly
suggest a role for these two amino acids in the catalytic process.
91
C.Breton, R.Oriol and A.Imberty
Table II. Summary of some known human H inactivating mutations and the proposed corresponding amino acids in the Y.enterocolitica protein, pig
α-6-fucosyltransferase, and A.caulinodans NodZ sequences, deduced from the HCA alignment
Human H
mutation
yeFuc-T
pα6Fuc-T
NodZ
References
Asp148→Tyr
Asn
–
–
Kaneko et al., 1997
Tyr154→His/Cys
Tyr
–
–
Kaneko et al., 1997; Wang et al., 1997;
Yu et al., 1997, Wagner et al., 1997
Leu164→His
Ile
–
–
Kelly et al., 1994
Trp171→Cys
Tyr
–
–
Wagner et al., 1997
Arg220→Cys
Arg
Arg
Arg
Yu et al., 1997
Tyr241→His
Tyr
Leu
Leu
Kaneko et al., 1997
Val259→Glu
Val
Leu
Leu
Wagner et al., 1997
Trp267→Cys
Trp
Leu
Val
Johnson et al., 1994
Asp278→Asn
Asp
Glu
Glu
Johnson et al.,1994
Ala315→Val
Ala
–
–
Wagner et al., 1997
Asn327→Thr
Asn
–
–
Yu et al., 1997
(-), No homology found in this region.
Combining HCA and fold recognition methods, we have
previously suggested that the mammalian α-2 and α-3-fucosyltransferases have a similar fold consisting of alternating
α-helices and β-strands (Breton et al., 1996). One of the most
probable folds could be that of the β-glucosyltransferase from
phage T4, an enzyme catalyzing the transfer of glucose from
UDP-Glc to hydroxymethylcytosine in DNA (Vrielink et al.,
1994). In the crystal structure, the fold comprises two α/β
domains separated by a wide cleft constituting the binding pocket
and the C-terminal domain is a typical nucleotide binding motif,
the so-called Rossman fold (Rossman et al., 1975). In the present
study, the HCA analysis was extended to all the eukaryotic and
prokaryotic fucosyltransferase sequences and it allowed us to
refine our previous predictions. Based on the assumption that
fucosyltransferases possess a nucleotide binding domain, it is
highly probable that it is located in the C-terminus part of these
enzymes, starting approximately in the region preceding the
peptide motif in α-2- and α-6-fucosyltransferases and in the
corresponding region I in α-3-fucosyltransferases, with a minimal size estimated to approximately 100 residues. Consequently,
we can assign the acceptor binding region in the N-terminus part
of the catalytic domain. These observations are in agreement with
the proposed location of the acceptor region of Fuc-TIII, Fuc-TV,
and Fuc-TVI reported by Legault et al. (1995) and Xu et
al.(1996).
the β-glucosyltransferase of phage T4. All these observations
suggest that the fucosyltransferases have evolved from a common
ancestor even if we cannot totally exclude a case of convergent
evolution.
Conclusion
During the review of the manuscript, the sequence of a new
α-3-fucosyltransferase from Helicobacter pylori was published
in GenBank under accession number AF006039. The nucleotidededuced protein sequence is composed of 333 amino acids. This
enzyme appears to have a different domain organization and
display very low overall sequence identity with the other
members of the α-3-fucosyltransferase family. However, we can
properly align by HCA the region 1–220 of the H.pylori sequence
with the region 100–361 of Fuc-TIII. Interestingly, this sequence
contains almost all the highly conserved residues of the region II
and four of the conserved residues of the region I. No cysteine
residue was found conserved between the eukaryotic α-3-fucosyltransferase and the H. pylori sequence.
The present work demonstrates the presence of conserved
structural features as well as a peptide motif in prokaryotic and
eukaryotic α-2 and α-6-fucosyltransferases. We have also
identified a few invariant amino acids which are probably
involved in the catalytic process of these enzymes and which
constitute potential targets for site-directed mutagenesis studies.
Despite the lack of the peptide motif in the α-3-fucosyltransferases, some regions could be aligned by HCA with the α-2 and
α-6-fucosyltranferases. These similarities occur in a region of the
catalytic domain of these enzymes that could correspond to the
nucleotide binding domain, by analogy to the crystal structure of
92
Materials and methods
Protein sequences were retrieved from GenBank or Swiss-Prot
databanks. Pairwise sequence comparison were done using
Bestfit or Gap programs (GCG package) and sequence similarity
searches using Blast (Altschul et al., 1990).
HCA is a graphical protein sequence comparison method
based on the detection and comparison of hydrophobic clusters
which are presumed to correspond to the regular secondary
structure elements constituting the hydrophobic core of globular
proteins (Gaboriaud et al., 1987; Lemesle-Varloot et al., 1990).
The protein sequences are represented on a duplicated α-helical
net, and the clusters of contiguous hydrophobic (V, I, L, F, M, W,
Y) residues are drawn. The HCA plots were obtained using the
program HCA-Plot V2.0 (Doriane S.A., Le Chesnay, France).
Graphical manipulation of the HCA plots were carried out using
IslandDraw V3.0 (Island Graphics Corp., Hoofddorp, The
Netherlands).
Note added in proof
Conserved structural features in fucosyltransferases
Acknowledgments
The work was supported in part by Grant 9514111 Action
Concertée Coordonnée des Sciences du Vivant (ACCSV14) from
Ministère de l’Education Nationale de l’Enseignement Supérieur
et de la Recherche (MENESR, France), and the Immunology
concerted action 3026PL950004 of the Biotechnology program
from the European Union (EU). C.B. is a staff member of Institut
National de la Recherche Agronomique.
Abbreviations
Fuc-T, fucosyltransferase; GDP-Fuc, guanosine diphosphate-β-l-fucose; Gal, galactose; GlcNAc, N-acetylglucosamine;
HCA, hydrophobic cluster analysis; UDP-Glc, uridine diphosphate-α-d-glucose.
References
Altschul,S.F., Gish,W., Miller,W., Myers,E.W. and Lipman,D.J. (1990) Basic
local alignment search tool. J. Mol. Biol., 215, 403–410.
Breton,C., Oriol,R. and Imberty,A. (1996) Sequence alignment and fold
recognition of fucosyltransferases. Glycobiology, 6, vii–xii.
Britten,C.J. and Bird,M.I. (1997) Chemical modification of an α3-fucosyltransferase; definition of amino acid residues essential for enzyme activity.
Biochim. Biophys. Acta, 1334, 57–64.
Cohney,S., Mouhtouris,E., McKenzie,I.F. and Sandrin,M.S. (1996) Molecular
cloning of the gene coding for pig α(1,2)fucosyltransferase. Immunogenetics,
44, 76–79.
Costache,M., Apoil,P.A., Cailleau,A., Elmgren.,A., Larson,G., Henry,S.,
Blancher,A., Iordachescu,D., Oriol,R. and Mollicone,R. (in press) Evolution
of fucosyltransferase genes in vertebrates. J. Biol. Chem.
Costache,M., Cailleau,A., Fernandez-Mateos,P., Oriol,R. and Mollicone,R.
(1997) Advances in molecular genetics of α-2- and α-3/4-fucosyltransferases.
Transf. Clin. Biol., 4, 367–382.
de Vries,T., Srnka,C.A., Palcic,M.M., Swiedler,S.J., van den Eijnden,D. and
Macher,B.A. (1995) Acceptor specificity of different length constructs of
human recombinant α1,3/4-fucosyltransferase. J. Biol. Chem., 270,
8712–8722.
Du,M. and Hindsgaul,O. (1996) Recognition of β-D-Galp-(1–3)-β-D-GlcpNAcOR acceptor analogues by the Lewis α–(1,3/4)-fucosyltransferase from human
milk. Carbohydr. Res., 286, 87–105.
Elmgren,A., Mollicone,R., Costache,M., Börjeson,C., Oriol,R., Harrington,J. and
Larson,G. (1997) Significance of individual point mutations, T202C and
C314T, in the human Lewis (FUT3) gene for expression of Lewis antigens by
the human α(1,3/1,4)fucosyltransferase, Fuc-TIII. J. Biol. Chem., 272,
21994–21998.
Gaboriaud,C., Bissery,V.L., Benchetrit,T. and Mornon,J.P. (1987) Hydrophobic
cluster analysis: an efficient new way to compare and analyse amino acid
sequences. FEBS Lett., 224, 149–155.
Gersten,K.M., Natsuka,S., Trinchera,M., Petryniak,B., Kelly,R.J., Hiraiwa,N.,
Jenkins,N.A., Gilbert,D.J., Copeland,N.G. and Lowe,J.B. (1995) Molecular
cloning, expression, chromosomal assignment and tissue-specific expression
of a murine α-(1,3)-fucosyltransferase locus corresponding to the human
ELAM-1 ligand fucosyltransferase. J. Biol. Chem., 270, 25047–25056.
Goelz,S.E., Hession,C., Goff,D., Griffiths,B., Tizard,R., Newman,B., Chi-Rosso,G.
and Lobb,R. (1990) ELFT: a gene that directs the expression of an ELAM-1
ligand. Cell, 63, 1349–1356.
Henrissat,B., Callebaut,I., Fabrega,S., Lehn,P., Mornon,J.P. and Davies,G. (1995)
Conserved catalytic machinery and the prediction of a common fold for
several families of glycosyl hydrolases. Proc. Natl. Acad. Sci. USA, 92,
7090–7094.
Henry,S., Oriol,R. and Samuelsson,B.E. (1995) Lewis histo-blood group system
and associated secretory phenotypes. Vox Sang., 69, 166–182.
Hindsgaul,O., Kaur,K.J., Srivastava,G., Blaszczyk-Thurin,M., Crawley,S.C.,
Heerze,L.D. and Palcic,M.M. (1991) Evaluation of deoxygenated oligosaccharide acceptor analogs as specific inhibitors of glycosyltransferases. J.
Biol. Chem., 266, 17858–17862.
Hitoshi,S., Kusunoki,S., Kanazawa,I. and Tsuji,S. (1995) Molecular cloning and
expression of two types of rabbit β-galactoside α1,2-fucosyltransferase. J.
Biol. Chem., 270, 8844–8850.
Hitoshi,S., Kusunoki,S., Kanazawa,I. and Tsuji,S (1996a) Molecular cloning and
expression of a mouse GDP-l-fucose: β-D-galactoside 2-α-L-fucosyltransferase. GenBank/EMBL accession number Y09882 (Sec1) and Y09883 (H).
Hitoshi,S., Kusunoki,S., Kazanawa,I. and Tsuji,S. (1996b) Molecular cloning and
expression of a third type of rabbit GDP-L-fucose: β-D-galactoside
2-α-L-fucosyltransferase. J. Biol. Chem., 271, 16975–16981.
Holmes,E.H. (1992) Presence of an essential lysine residue in a GDP-fucose
protected site of the α1,3fucosyltransferase from human small cell lung
carcinoma NCI-H69 cells. Arch. Biochem. Biophys., 296, 562–568.
Holmes,E.H., Xu,Z., Sherwood,A.L. and Macher,B.A. (1995) Structure–function
analysis of human α1–3fucosyltransferases. A GDP-fucose protected,
N-ethylmaleimide-sensitive site in FucT-III and FucT-V corresponds to
Ser178 in Fuct-IV. J. Biol. Chem., 270, 8145–8151.
Johnson,P.H., Mak,M.K., Leong,S., Broadberry,R., Duraisamy,G., Gooch,A.,
Lin,C.M., Makar,I., Okubo,Y., Smart,E., Koepsall,E. and Ewers,M. (1994)
Analysis of mutations in the blood group H gene in donors with H-deficient
phenotypes (abstract). Vox Sang., 67(Suppl. 2), 25.
Joziasse,D.H. (1992) Mammalian glycosyltransferases: genomic organization
and protein structure. Glycobiology, 2, 271–277.
Kaneko,M., Nishihara,S., Shinya,N., Kudo,T., Iwasaki,H., Seno,T., Okubo,Y. and
Narimatsu,H. (1997) Wide variety of point mutations in the H gene of Bombay
and para-Bombay individuals which inactivate H enzyme. Blood, 90,
839–849.
Kelly,R.J., Ernst,L.K., Larsen,R.D., Bryant,J.G., Robinson,J.S. and Lowe,J.B.
(1994) Molecular basis for H blood group deficiency in Bombay (Oh) and
para-Bombay individuals. Proc. Natl. Acad. Sci. USA, 91, 5843–5847.
Kelly,R.J., Rouquier,S., Giorgi,D., Lennon,G.G. and Lowe,J.B. (1995) Sequence
and expression of a candidate for the human secretor blood group
α(1,2)fucosyltransferase gene (FUT2). J. Biol. Chem., 270, 4640–4649.
Kleene,R. and Berger,E.G. (1993) The molecular and cell biology of glycosyltransferases. Biochim. Biophys. Acta, 1154, 283–325.
Kukowska-Latallo,J.F., Larson,R.D., Nair,R.P. and Lowe,J.B. (1990) A cloned
human cDNA determines expression of a mouse stage-specific embryonic
antigen and the Lewis blood group α(1,3/1,4)fucosyltransferase. Genes Dev.,
4, 1288–1303.
Larsen,R.D., Ernst,L.K., Nair,R.P. and Lowe,J.B. (1990) Molecular cloning,
sequence and expression of a human GDP-L-fucose: β-D-galactoside
α2-L-fucosyltransferase cDNA that can form the H blood group antigen. Proc.
Natl. Acad. Sci. USA, 87, 6674–6678.
Lee,K.P., Carlson,L.M., Woodcock,J.B., Ramachandra,N., Schultz,T.L.,
Davis,T.A., Lowe,J.B., Thomson,C.B. and Larsen,R.D. (1996) Molecular
cloning and characterization of CFT1, a developmentally regulated avian
α-1,3-fucosyltransferase gene. J. Biol. Chem., 271, 32960–32967.
Legault,D.J., Kelly,R.J., Natsuka,Y.and Lowe,J.B. (1995) Human α(1,3/1,4)fucosyltransferases discriminate between different oligosaccharide acceptor
substrates through a discrete peptide fragment. J. Biol. Chem., 270,
20987–20996.
Lemesle-Varloot,L., Henrissat,B., Gaboriaud,C., Bissery,V., Morgat,A. and
Mornon,J.P. (1990) Hydrophobic cluster analysis: procedures to derive
structural and functional information from 2-D representation of protein
sequences. Biochimie, 72, 555–574.
Mergaert,P., D’Haeze,W., Fernandez-Lopez,M., Geelen,D., Goethals,K.,
Prome,J.C., Van Montagu,M. and Holsters,M. (1996) Fucosylation and
arabinosylation of Nod factors in Azorhizobium caulinodans: involvement of
nolK, nodZ, as well as noeC and/or downstream genes. Mol. Microbiol., 21,
409–419.
Murray,B.W., Takayama,S., Schultz,J. and Wong,C.H. (1996) Mechanism and
specificity of human α1,3-fucosyltransferase V. Biochemistry, 35,
11183–11195.
Murray,B.W., Wittmann,V., Burkart,M.D., Hung,S.C. and Wong,C.H. (1997)
Mechanism of human α-1,3-fucosyltransferase V: glycosidic cleavage occurs
prior to nucleophilic attack. Biochemistry, 36, 823–831.
Oulmouden,A., Wierinckx,A., Petit,J.M., Costache,M., Palcic,M.M., Mollicone,R.,
Oriol,R. and Julien,R. (1997) Molecular cloning and expression of a bovine
α(1,3)fucosyltransferase gene homologous to the ancestor gene of the human
FUT3-FUT5-FUT6 cluster. J. Biol. Chem., 272, 8764–8773.
Qiao,L., Murray,B.W., Shimazaki,M., Schultz,J. and Wong,C.H. (1996) Synergistic inhibition of human α-1,3-fucosyltransferase V. J. Am. Chem. Soc., 118,
7653–7662.
Rossmann,M.G., Liljas,A., Bränden,C.I. and Banaszak,L.J. (1975) Evolutionary
and structural relationships among dehydrogenases. In Boyer,P.D. (ed.), The
Enzymes, Vol. 11. Academic Press, New York, pp. 61–102.
Sajdel-Sulkowska,E.M., Smith,F.I., Wiederschain,G. and McCluer,R.H. (1997)
Cloning of rat α1,3-fucosyltransferase gene: a new member of the fucosyltransferase IV family. Glycoconj. J., 14, 249–258.
93
C.Breton, R.Oriol and A.Imberty
Sasaki,K., Kurata,K., Funayama,K., Nagata,M., Watanabe,E., Ohta,S., Hanai,N.
and Nishi,T. (1994) Expression cloning of a novel α(1,3)fucosylytransferase
that is involved in biosynthesis of the sialyl Lewisx carbohydrate determinants
in leukocytes. J. Biol. Chem., 269, 14730–14737.
Schenkel-Brunner,H. (1995) Human blood groups. Springer-Verlag, Vienna.
Smith,P.L., Gersten,K.M., Petryniak,B., Kelly,R.J., Rogers,C., Natsuka,Y.,
Alford,J.A.,3rd, Scheidegger,E.P., Natsuka,S. and Lowe,J.B. (1996) Expression
of the α(1,3)fucosyltransferase Fuc-TVII in lymphoid aggregate high endothelial
venules correlates with expression of L-selectin ligands. J. Biol. Chem., 271,
8250–8259.
Stacey,G., Luka,S., Sanjuan,J., Banfalvi,Z., Nieuwkoop,A.J., Chun,J.Y.,
Forsberg,L.S. and Carlson,R. (1994) NodZ, a unique host-specific nodulation
gene, is involved in the fucosylation of the lipo-oligosaccharide nodulation
signal of Bradyrhizobium japonicum. J. Bacteriol., 176, 620–633.
Staudacher,E. (1996) α1,3-Fucosyltransferases. Trends Glycosci. Glycotechnol.,
8, 391–408.
Uozumi,N., Yanagidani,S., Miyoshi,E., Ihara,Y., Sakuma,T., Gao,C.X.,
Teshima,T., Fujii,S., Shiba,T. and Taniguchi,N. (1996) Purification and cDNA
cloning of porcine brain GDP-L-Fuc:N-acetyl-β-D-glucosaminide α1–6fucosyltransferase. J. Biol. Chem., 271, 27810–27817.
Vrielink,A., Rüger,W., Driessen,H.P.C. and Freemont,P.S. (1994) Crystal structure of the DNA modifying enzyme β-glucosyltransferase in the presence and
absence of the substrate uridine diphosphoglucose. EMBO J., 13, 3413–3422.
Vyas,N.K. (1991) Atomic features of protein–carbohydrate interactions. Curr.
Opinion Struct. Biol., 1, 732–740.
Wagner,F.F. and Flegel,W.A. (1997) Polymorphism of the h allele and the
population frequency of sporadic non-functional alleles. Transfusion, 37,
284–290.
Wang,B., Koda,Y., Soejima,M. and Kimura,H. (1997) Two missense mutations of
H type α(1,2)fucosyltransferase gene (FUT1) responsible for para-Bombay
phenotype. Vox Sang., 72, 31–35.
Watkins,W.M. (1995) Biosynthesis. Molecular basis of antigenic specificity in the
ABO, H and Lewis blood group systems. In Montreuil,J., Schachter,H. and
Vliegenthart,J.F.G. (eds.), Glycoproteins. Elsevier, Amsterdam, pp. 313–390.
Weston,B.W., Nair,R.P., Larsen,R.D. and Lowe,J.B. (1992a) Isolation of a novel
human α(1,3)fucosyltransferase gene and molecular comparison to the
human Lewis blood group α(1,3/1,4)fucosyltransferase gene. J. Biol. Chem.,
267, 4152–4160.
94
Weston,B.W., Smith,P.L., Kelly,R.J. and Lowe,J.B. (1992b) Molecular cloning of
a fourth member of a human α(1,3)fucosyltransferase gene family. J. Biol.
Chem., 267, 24575–24584.
Wilson,R., Ainscough,R., Anderson,K., Baynes,C., Berks,M., Bonfield,J.,
Burton,J., Connell,M., Copsey,T., Cooper,J., Coulson,A., Craxton,M., Dear,S.,
Du,Z., Durbin,R., Favello,A., Fulton,L., Gardner,A., Green,P., Hawkins,T.,
Hillier,L., Jier,M., Johnston,L., Jones,M., Kershaw,J., Kirsten,J., Laister,N.,
Latreille,P., Lightning,J., Lloyd,C., McMurray,A., Mortimore,B.,
O’Callaghan,M., Parsons,J., Percy,C., Rifken,L., Roopra,A., Saunders,D.,
Shownkeen,R., Smaldon,N., Smith,A., Sonnhammer,E., Staden,R., Sulston,J.,
Thierry-Mieg,J., Thomas,K., Vaudin,M., Vaughan,K., Waterston,R.,
Watson,A., Weinstock,L., Wilkinson-Sproat,J. and Wohldman,P. (1994) 2.2
Mb of contiguous nucleotide sequence from chromosome III of C.elegans.
Nature, 368, 32–38.
Wong,C.H., Dumas,D.P., Ichikawa,Y., Koseki,K., Danishefsky,S.J., Weston,B.W.
and Lowe,J.B. (1992) Specificity, inhibition and synthetic utility of a
recombinant human α-1,3-fucosyltransferase. J. Am. Chem. Soc., 114,
7321–7322.
Woodcock,S., Mornon,J.P. and Henrissat,B. (1992) Detection of secondary
structure elements in proteins by hydrophobic cluster analysis. Prot. Eng., 5,
629–635.
Xu,Z., Vo,L. and Macher,B.A. (1996) Structure-function analysis of human
α1–3fucosyltransferases. Amino acids involved in acceptor substrate
specificity. J. Biol. Chem., 271, 8818–8823.
Yanagidani,S., Uozumi,N., Ihara,Y., Miyoshi,E., Yamaguchi,N. and Taniguchi,N.
(1997) Purification and cDNA cloning of GDP-l-Fuc:N-acetyl-β-d-glucosaminyl fucosyltransferase (α1–6FucT) from human gastric cancer MKN45
cells. J. Biochem., 121, 626–632.
Yu,L.C., Yang,Y.H., Broadberry,R.E., Chen,Y.H. and Lin,M. (1997) Heterogeneity of the human H blood group α(1,2)fucosyltransferase gene among
para-Bombay individuals. Vox Sang., 72, 36–40.
Zhang,L., Radziejewska-Lebrecht,J., Krajewska-Pietrasik,D., Toivanen,P. and
Skurnik,M. (1997) Molecular and chemical characterization of the lipopolysaccharide O-antigen and its role in the virulence of Yersinia enterocolitica
serotype O:8. Mol. Microbiol., 23, 63–76.