Identification of a missense mutation (G329A;Arg(110)--> GLN) in the human FUT7 gene.

The human FUT7 gene codes for the alpha1,3-fucosyltransferase VII (Fuc-TVII), which is involved in the biosynthesis of the sialyl Lewis x (SLe(x)) epitope on human leukocytes. The FUT7 gene has so far been considered to be monomorphic. Neutrophils isolated from patients with ulcerative colitis were examined for apparent alterations in protein glycosylation patterns by Western blot analysis using monoclonal antibodies directed against SLe(x) and SLe(x)-related epitopes. One individual showed lower levels of SLe(x) expression and an elevated expression of CD65s compared to controls. The coding regions of the FUT7 gene from this individual were cloned, and a G329A point mutation (Arg(110) --> Gln) was found in one allele, whereas the other FUT7 allele was wild type. No Fuc-TVII enzyme activity was detected in COS-7 cells transiently transfected with the mutated FUT7 construct. The FUT7 Arg(110) is conserved in all previously cloned vertebrate alpha 1,3-fucosyltransferases. Polymerase chain reaction followed by restriction enzyme cleavage was used to screen 364 unselected Caucasians for the G329A mutation, and a frequency of < or =1% for this mutation was found (3 heterozygotes). Genetic characterization of the family members of one of the additional heterozygotes identified one individual carrying the G329A mutation in both FUT7 alleles. Peripheral blood neutrophils of this homozygously mutated individual showed a lowered expression of SLe(x) and an elevated expression of CD65s when analyzed by Western blot and flow cytometry. The homozygous individual was diagnosed with ulcer disease, non-insulin-dependent diabetes, osteoporosis, spondyloarthrosis, and Sjögren's syndrome but had no history of recurrent bacterial infections or leukocytosis.

Recruitment of leukocytes to sites of inflammation or infection is initiated by interaction of leukocytes with activated vessel wall endothelium leading to "rolling" of leukocytes along endothelial cell surfaces. This interaction subsequently leads to extravasation of leukocytes into the surrounding infected or inflamed tissue (1). E-and P-selectins, which are expressed on activated endothelial cells, are involved in this interaction (2)(3)(4). The third member of the selectin family, L-selectin, is involved when lymphocytes extravasate into secondary periph-eral lymphoid organs, where it interacts with counter-receptors on the post-capillary high endothelial venules (HEV) 1 (5).
In this paper we describe for the first time a missense mutation of the FUT7 gene associated with an altered expression of SLe x and CD65s (VIM-2) epitopes on human polymorphonuclear leukocytes.

MATERIALS AND METHODS
Patients and Controls-Fifteen patients with ulcerative colitis (n ϭ 13) or proctitis (n ϭ 2) were examined. Twelve healthy controls were also studied. Restriction endonuclease analysis was performed on DNA samples from 106 plasma donors in Göteborg (Sweden) (29) and 258 unselected adult individuals from the Linköping area (Sweden). A pedigree study of FUT7 genetics was performed in one Swedish family. The study was approved by local ethical committees in Göteborg and Linköping.
Isolation of Polymorphonuclear Leukocytes-Human polymorphonuclear leukocytes (PMN) were isolated from 10 ml of freshly drawn EDTA-anticoagulated blood using density gradient centrifugation (Polymorphprep TM , Nycomed, Torshev, Norway). The cell preparations had a purity of Ͼ90% as determined by analyses on an automatic cell counter (H3 instrument, Bayer Diagnostics, Fernwald, Germany).
Immunoblot Analysis-Cell lysates were prepared by solubilizing the purified PMN in 400 l of lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 5 mM EDTA, 1% IGEPAL CA-630 (Sigma) containing 0.12 M phenylmethylsulfonyl fluoride, 2.3 M leupeptin, and 1.5 M pepstatin. Protein concentration in the supernatant was determined by the 4,4Јdicarboxy-2,2Јbiquinoline, bicinchoninic acid (BCA) method (30) (Pierce). An aliquot of the supernatant corresponding to 20 g of protein was separated by 10% SDS-polyacrylamide gel electrophoresis (31) and transferred to an Immobilon-P polyvinylidene fluoride microporous membrane (Millipore, Bedford, MA) (32). The membrane was blocked with Tris-buffered saline (TBS; 50 mM Tris-HCl, 150 mM NaCl, pH 7.5) containing 5% Tween 20 and 5% defatted milk powder (Semper, Stockholm, Sweden) at room temperature for 1 h. The membrane was then washed twice with TBS, incubated overnight at 4°C with primary antibody in TBS, 1% BSA, washed three times with TBS containing 0.3% Tween 20, and incubated with peroxidase-conjugated secondary antibody in TBS containing 1% milk powder and 0.1% Tween 20 for 1 h at room temperature. The membrane was rinsed three times with TBS containing 0.3% Tween 20, and positive bands were visualized using ECL Western blotting reagents (Amersham Pharmacia Biotech). For Western blot analysis of COS-7 cells, the membranes were blocked in 10% BSA in PBS containing 0.2% Tween 20 (PBS-T) over night at 4°C. PBS-T was used for washing, and PBS-T containing 3% BSA was used for incubations with primary and secondary antibodies.
For calculation of molecular size, prestained molecular mass standards (Bio-Rad) were used. Digital images of the blots were analyzed using a CCD camera (512 ϫ 512 pixels) in combination with a computerized imaging 8-bit system (Visage 4.6; BioImage, Ann Arbor, MI). The quantitative expression of each epitope in a lane was assessed as background-corrected optical density, integrated over all pixels in the lane and expressed as integrated optical density.
Molecular Cloning of FUT7 cDNA-Buffy coat was isolated from freshly drawn EDTA-anticoagulated blood from one heterozygously mutated patient (M. N.). Total RNA was isolated using an SV Total RNA isolation system (Promega Corp., Madison, WI) and cDNA was prepared using oligo(dT) 15 primers and a Reverse Transcription System A3500 according to the manufacturer's protocol (Promega Corp.). PCR was used to amplify the coding regions and immediately adjacent 5Јand 3Ј-flanking regions of FUT7 using 0.25 g of cDNA as template. The PCR program used an initial temperature of 8°C for 10 min, followed by 40 amplification cycles run for 15 s at 95°C, 15 s at 59°C, and 3 min at 68°C. The last extension step was kept for 10 min at 68°C. The sense primer (25 pmol), VII-3s (5Ј-gctagcgaattcCTGATCCTGG-GAGACTGTGG-3Ј), is complementary to nucleotides Ϫ20 to Ϫ1 and contains additional nucleotides (lowercase) at its 5Ј end, including an EcoRI (underlined) restriction site. The antisense primer (25 pmol), VII-4as (5Ј-gtcgactctagaGTAAGGGCCGGATGCCTGGT-3Ј), anneals to nucleotides 1107-1088, and contains additional nucleotides (lowercase) at its 5Ј end, including an XbaI (underlined) restriction site. The 1151-bp PCR product was ligated into the pCR2.1 TA-cloning vector (Invitrogen Corp., San Diego, CA). Transformation of InV␣F bacteria was performed according to the manufacturers protocol (Invitrogen Corp.). Positive clones were identified by blue-white screening, preparation of plasmids, and cleavage with XbaI and EcoRI (Life Technologies, Paisley, United Kingdom.). Twenty-four clones were analyzed with PCR-restriction fragment length polymorphism to identify TA clones with wild type or mutated alleles. Primer VII-15s (5Ј-CATCGCCCG-CTGCCACCTGAGT-3Ј), corresponding to nucleotides 216 -237, and VII-5as (5Ј-GCTGCCGCTCCTGGAAGTTGCTGAC-3Ј), corresponding to nucleotides 529 -554, were used to amplify a 338-bp fragment of FUT7 cDNA. The G329A mutation abolishes the restriction site GC2GGCCGC for NotI (Life Technologies, Inc.). When the 338-bp PCR product was treated with NotI and analyzed by gel electrophoresis, the wild type allele was digested into two products of 247 and 91 bp, whereas the mutant allele remained intact. Complete sequencing of nine TA clones was done on an Alf II-express using the Cy 5 -dye terminator kit (Amersham Pharmacia Biotech). One wild type clone and one clone containing the G329A point mutation without PCR-induced errors were chosen for subcloning of the FUT7 insert into the pSI mammalian expression vector (Promega Corp.). The resulting plasmid containing the FUT7 wild type was called pSI-wt, and the plasmid containing the mutated construct was called pSI-329.
Cloning of the FUT7 Gene from Genomic DNA-PCR was used to amplify the two coding regions and the 253-bp intron (33) as well as the immediately adjacent 5Ј-and 3Ј-flanking regions of the FUT7 gene from DNA prepared from whole blood. The PCR program used an initial temperature of 85°C for 10 min, followed by 30 amplification cycles run for 15 s at 60°C, 15 s at 59°C, and 3 min at 68°C. The last extension step was kept for 10 min at 68°C. The VII-3s sense primer (30 pmol), and the VII-4as antisense primer (30 pmol) were used. The 1404-bp PCR product was ligated into the pCR2.1 TA-cloning vector as above and sequenced with the AmpliTaq DNA polymerase FS kit (PerkinElmer Life Sciences, Foster City, CA) on an Applied Biosystems 373A DNA sequencer (PerkinElmer Life Sciences).
Transfection-COS-7 cells (ϳ10 6 cells) cultured in Dulbecco's modified Eagle's medium supplemented with 10% heat-inactivated fetal calf serum were transfected with 10 g of expression vector constructs using the DEAE-dextran method (34). The cells were transfected with pSI without insert as a negative control or with the two pSI-FUT7 constructs (pSI-wt and pSI-329). The medium was changed 24 h after transfection. Transfected cells were harvested after a 96-h growth period after transfection. Transfection efficiency was controlled using quantitative PCR analysis. Total RNA was isolated from transfected cells according to the manufacturer's instructions using a Total RNA kit (Promega Corp.) including treatment with DNase. A reverse transcription kit (Promega Corp.) was used according to the manufacturer's instructions to transcribe 1 g of total RNA. Quantitative PCR analysis was performed using the TaqMan PCR Core Reagent kit (PE Biosystems). Reactions for FUT7 quantification were performed in 30 l with 0.2 g of cDNA; 3 l of 10ϫ TaqMan Buffer A (500 mM KCl, 100 mM Tris-HCl, pH 8.3); 5 mM MgCl 2 ; 200 M each of dATP, dCTP, and dGTP; 400 M dUTP; 0.3 units of uracil-Nglucosidase; 0.75 units of AmpliTaq Gold DNA polymerase; 50 nM FUT7 probe; and 100 nM FUT7 sense and antisense primers. The following FUT7 consensus primers and probe were used: r1s, 5Ј-CTTGGCT-GACTGACTCTGG-3Ј (nucleotides Ϫ29 to Ϫ11); r2as, 5Ј-CCTCGCAGC-CTCCG-3Ј (nucleotides 28 to 41); and FUT7 probe, 5Ј-CCGTGCCCAAG-CATTATTCATCCA-3Ј (nucleotides Ϫ3 to 20). The FUT7 probe was designed to cover the sequence over the splice site in FUT7 cDNA to avoid amplification of contaminating genomic DNA sequences. As an additional control for contaminating DNA, quantitative PCR was also performed leaving out the first reverse transcription-PCR step. This did not generate any product. The PCR-program used an initial temperature of at 50°C for 2 min and then 95°C for 10 min, followed by 40 amplification cycles run for 15 s at 95°C and 1 min at 60°C. The amplifications were performed on an ABI Prism 7700 sequence detector equipped with a 96-well thermal cycler. Data were collected and analyzed with Sequence Detector version 1.6.3 software (PE Biosystems). Reactions for quantifying ␤-actin were performed exactly as described above except for using 3.5 mM MgCl 2 and 300 nM sense primer (5Ј-TCACCCACACTGTGCCCATCTACGA-3Ј), 300 nM antisense primer (5Ј-CAGCGGAACCGCTCATTGCCAATGG-3Ј), and 200 nM ␤-actin probe (5Ј-ATGCCCCCCCCATGCCATCCTGCGT-3Ј) (PE Biosystems). All analyses were performed in triplicate and with probes labeled with 6-carboxyfluorescein and 6-carboxytetramethylrhodamine.
Immunofluorescence Analysis of Lewis Antigen Expression on the Surface of Transfected COS-7 Cells-The transfected cells were trypsinized, washed, and incubated with primary antibodies against CD65s, SLe a , SLe x , and Le x . After 30 min of incubation with the primary antibody, the cells were washed with PBS without Ca 2ϩ and Mg 2ϩ and incubated another 30 min with fluorescein-conjugated rat antimouse IgG secondary antibody. After incubation with the secondary antibody, the cells were washed with PBS without Ca 2ϩ and Mg 2ϩ . The cell pellets were then fixed with Mowiol (Hoechst, Frankfurt am Main, Germany) and paraformaldehyde (4%, pH 7.3) at a ratio of 1/3 (v/v) and mounted on glass. The cells were observed under a Leitz SM-LUX epifluorescence microscope. Immunofluorescence studies were also conducted on adherent cells in eight-well tissue culture chamber slides (Nunc Inc., Naperville, IL) without using trypsin treatment.
Fucosyltransferase Assay-Enzyme activity was analyzed by measuring the incorporation of GDP-[ 14 C]fucose, 300 mCi/mmol (Amersham Pharmacia Biotech), to a sialylated type 2 acceptor substrate, NeuAc␣2-3Gal␤1-4GlcNAc␤1-sp-biotin or a sialylated type 1 acceptor substrate, NeuAc␣2-3Gal␤1-3GlcNAc␤1-sp-biotin (Syntesome, Munich, Germany). COS-7 cells transfected with pSI, pSI-wt, or pSI-329 were lysed in 50 mM MOPS buffer (pH 7.5) containing 1% Triton X-100. Apparent K m for GDP-Fuc was determined using Lineweaver-Burk plots with GDP-Fuc concentrations between 2 and 10 M and an acceptor concentration of 10 mM. Apparent K m for the sialylated type 2 acceptor was determined with acceptor concentrations between 0.25 and 10 mM and a GDP-Fuc concentration of 100 M. The assay was initiated with the addition of cell lysate (45 g of protein) to a reaction mixture containing GDP-Fuc, acceptor, 10 mM ␣-L-fucose, and 10 mM MnCl 2 in 50 mM MOPS buffer (pH 7.5). The mixture was incubated at 37°C for 2 h. The product was purified by the Sep-Pak C 18 isolation procedure (35), and analyzed by liquid scintillation counting. Product formation was also measured at 0.5, 1, and 2 h and found to be linear in this time range.
Mutation Screening by Restriction Endonuclease Analysis-Genomic DNA, isolated from 5 ml of EDTA anticoagulated blood according to Ref. 36, was amplified by primers VII-15s and VII-5as. The 338-bp product was used without prior purification for restriction enzyme analysis by NotI and electrophoresis on a 1.75% SeaKem-agarose gel (FMC), followed by ethidium bromide staining. For NotI restriction endonuclease analysis of selected family members, the primer pair VII-3s/VII-4as was used for amplification (generating a 1404-bp product) with 791-and 613-bp cleavage products.
Flow Cytometry-Flow cytometric analyses were performed on a FACScan instrument (Becton Dickinson) operating with CELLQuest software and calibrated with 6-m CaliBRITE beads with the Auto-COMP program (Becton Dickinson). One ml of EDTA-anticoagulated peripheral blood was diluted into 50 ml of lysis buffer (150 mM NH 4

Identification of a Patient with an Abnormal Expression of
SLe x and CD65s on her PMN-PMN lysates were analyzed by Western blot analysis to detect differences in expression of SLe x and CD65s. All PMN samples analyzed from healthy volunteers expressed a similar set of SLe x -carrying glycoproteins with most intensely stained bands in the molecular mass region at ϳ90 -115 kDa. A representative sample is shown in Fig. 1 (lane B). VIM-2 antibody directed against the CD65s epitope weakly stained glycoproteins with molecular masses of ϳ60 -70 kDa (Fig. 1, lane D). Western blot analyses of PMN lysates from patients with ulcerative colitis showed staining patterns comparable to the healthy population. However, one of the patients (M. N.) exhibited a different staining pattern. The Western blot analysis of PMN lysate from this individual showed a significant reduction in the staining of SLe x -bearing glycoproteins. The staining intensity was about 60% compared with control samples for identical amounts of total protein (Table I and Fig. 1 (lane A)). In addition, staining of one band in the 100-kDa region was selectively lost. This pattern was seen using two different antibodies (KM-93, CSLEX-1) both known to react with SLe x , albeit with somewhat different binding properties (Ref. 37 and data not shown). Western blot analysis of PMN lysates from this patient (M. N.) using the VIM-2 antibody directed against the CD65s epitope, showed an increased staining (480%) compared with control samples (Fig.  1, Table I). This patient was analyzed both at the time of active disease and in clinical remission at several occasions during a 2-year period. The reduced expression of SLe x and elevated expression of CD65s remained constant during this time.
Lowered SLe x Expression Correlates with a G329A Mutation in the Gene Coding for Fucosyltransferase VII-The lowered SLe x expression in PMN of patient M. N. indicated a potential defect in the Fuc-TVII enzyme. The gene coding for Fuc-TVII, FUT7, was amplified from genomic DNA by PCR and TA cloning. Plasmids were isolated from 13 bacterial clones and sequenced over the FUT7 insert. The two exons and the 253-bp intron present in the FUT7 gene were sequenced in both directions. There were no differences in the intron sequences between the two alleles from this individual. However, a G329A missense mutation was found in 7 out of the 13 bacterial clones, indicating that M. N. carried this mutation heterozygously in one allele. The G329A nucleotide change leads to an amino acid shift from arginine to glutamine at position 110. Sequence alignment (38) showed that FUT7-Arg 110 is conserved in all 16 of the ␣1,3-fucosyltransferases cloned so far from vertebrate species (13, 39 -41).
Screening for G329A by Restriction Endonuclease Analysis-A restriction fragment length polymorphism assay was  N.) (lanes A and C) and a  control sample (lanes B and D). used to screen for the G329A mutation in DNA preparations from 106 plasma donors in Göteborg and 258 unselected adults in the Linköping area. In this population, three additional individuals carrying the G329A mutation heterozygously were identified. The overall frequency of the G329A mutation in the analyzed populations was 0.82%.
Identification of an Individual Homozygous for the G329A Mutation in FUT7-DNA from another of the identified heterozygotes (M. L.) was cloned and sequenced. This confirmed the presence of the G239A mutation in one allele and no other mutations or alterations in the coding sequences or in the intron sequence. NotI restriction endonuclease analysis of M. L. and 5 of her family members are summarized in Fig. 2. Apart from the heterozygous individual M. L., her brother (R. J.) and both of her daughters (A. L. and L. L.) also showed a cleavage pattern consistent with heterozygous expression of the G329A mutation. Her husband did not carry the G329A mutation. However, the PCR product obtained from the mother of M. L. (S. J.) was not digested at all, which indicated a homozygous expression of the G329A mutation (Fig. 2). The two FUT7 exons and the intron were completely sequenced in both directions from 14 clones obtained from this individual. All clones contained the G329A mutation. No other nucleotide changes were found. This individual thus carried the isolated mutation in both of her FUT7 alleles. When PMN lysates prepared from this individual were analyzed by Western blot using antibody KM93 directed against SLe x , there was an almost complete lack of expression of SLe x -binding glycoproteins compared with control samples (Table I and Fig. 3 (lanes A and B)). When the same samples were analyzed using the VIM-2 antibody directed against CD65s, a marked increase in the expression of this epitope was found for this individual (Fig. 3, lane C). The increased staining intensity was 980% compared with control samples (Table I  and Fig. 3 (lane D)) and 205% compared with individual M. N.
Flow Cytometry Analysis of PMN from the Individual Homozygous for the G329A Mutation in FUT7-The expression of SLe x on PMN from individuals with or without the G329A mutation was investigated using flow cytometry. Most of the PMN from the homozygous individual (S. J.) showed a KM93 staining just above background. However, a subpopulation of cells from this individual showed an intermediate staining with this antibody. Antibody KM93 reacted strongly with PMN from an individual lacking the G329A mutation (Fig. 4A). Staining of PMN with the anti-SLe x antibody CSLEX-1 showed the same pattern with lower expression for the homozygous individual and a higher expression for an individual lacking the G329A mutation (Fig. 4B). In contrast to the results obtained by Western blot, there was no major differences between these individuals in staining of PMN with the anti-CD65s antibody VIM-2 (Fig. 4C). As expected, PMN from all analyzed individuals expressed a high level of Le x (Fig. 4D). Sialidase treatment of PMN prior to flow cytometry analysis reduced binding of KM93, CSLEX-1, and VIM-2 antibodies to background levels (data not shown).
Cell Surface Expression of SLe x Is Not Detected on COS-7 Cells Transfected with FUT7 G329A cDNA-The Western blot and flow cytometry analyses of PMN from hetero-and homozy-gously mutated individuals indicated that the Arg 110 3 Gln substitution affects Fuc-TVII activity. To confirm this, COS-7 cells were transiently transfected with plasmids containing either the mutated or the wild type FUT7 cDNA sequence (pSI-329 and pSI-wt, respectively). Mock transfectants using vector only (pSI) were used as negative controls. After transfection the expression of SLe x , CD65s, Le x , and SLe a was analyzed by immunofluorescence staining. Cells transfected with pSI-wt were clearly stained with anti-SLe x antibody (Fig. 5A), whereas there was no staining of cells transfected with pSI-329 with this antibody (Fig. 5B). The same pattern was seen using both KM-93 and CSLEX-1 antibodies (data not shown). This indicated that the G329A mutation significantly reduces the activity of Fuc-TVII in transfected COS-7 cells. Neither of the transfectants was stained with antibodies directed against   A and C) and a control  sample (lanes B and D). Each lane was loaded with 20 g of protein.

FIG. 3. Western blot analysis of lysed PMN isolated from the Fuc-TVII Q/Q individual (S. J.) (lanes
The blots were probed with antibodies against SLe x (KM93, lanes A and  B) and CD65s (VIM-2, lanes C and D). Molecular size standards are indicated to the left. SLe a , CD65s, or Le x .
Fucosyltransferase VII Activity Is Not Detected in COS-7 Cells Transfected with FUT7 G329A cDNA-Fuc-TVII activity was analyzed using sialylated type 1 and 2 acceptors and whole cell lysates of COS-7 cells transfected with pSI-wt, pSI-329, or pSI. The pSI-wt construct produced an active enzyme, whereas there was no detectable enzyme activity in cells transfected with the mutated construct or vector only (Table II), in accordance with the immunofluorescence results. When a sialylated type 1 chain acceptor was used, no activity was detected in either of the transfectants. As a measure of transfection efficiency, RNA was isolated from the transfected cells and a fragment of the FUT7 transcript was amplified using quantitative real-time reverse transcription-PCR analysis. There were no quantitative differences in FUT7 mRNA between pSIwt-and pSI-329-transfected cells (Table III). This pattern was seen for all transfection experiments used for Fuc-TVII activity measurements. All values were corrected for differences in total mRNA content using amplification of ␤-actin mRNA as an internal control in each experiment (Table III).
Expression of the Fuc-TVII Enzyme in Transfected COS-7 Cells-COS-7 cells transfected with pSI-wt, pSI-329, or pSI were analyzed by Western blot using purified antiserum against the Fuc-TVII enzyme. The antiserum stained a band with a molecular mass of 39 kDa with similar intensity in cells transfected with pSI-wt and pSI-329, whereas this band was not detected in mock-transfected (pSI) cells (Fig. 6). The molecular mass of the stained band corresponds to the molecular mass previously reported for FucT-VII (12). In addition, a specific band with an apparent molecular mass of 55 kDa was stained in pSI-wt-transfected cells. This band was not detected in either pSI-329-or pSI-transfected cells (Fig. 6). DISCUSSION When analyzing the expression of SLe x and SLe x -related antigens on PMN from patients with ulcerative colitis, one patient with decreased expression of SLe x was identified. The FUT7 gene of this individual was cloned and sequenced, and a single point mutation, G329A, was found in one of the alleles. This mutation gives an amino acid shift from an arginine to a glutamine at position 110 in Fuc-TVII (Arg 110 3 Gln). When the G329A mutation was screened for in two small Swedish populations, 3 out of 364 individuals were found heterozygous for this mutation (Fuc-TVII R/Q). Although a larger population must be examined to ascertain the exact overall frequency of this mutation, this indicates that it might be carried by ϳ1% of the population. FUT7 should thus be considered to be a polymorphic gene, especially since the G329A allele might be only one of several mutated alleles to be found in various populations around the world. Genetic analysis of the family members of one of the identified heterozygotes revealed an individual carrying the G329A mutation in both alleles (Fuc-TVII Q/Q). The two exons and the 253-bp intron of FUT7 (33) were fully sequenced in two of the identified heterozygotes (M. N. and M. L.) and in the homozygote. Apart from the G329A mutation, there was no other structural alteration compared with the wild type FUT7 sequence.
Western blot analysis of PMN lysates and flow cytometry of PMN showed that individuals carrying the G329A mutation had a lowered expression of SLe x , which is consistent with the hypothesis that the G329A mutation affects Fuc-TVII activity. Western blot analysis of PMN lysates from individuals carrying the G329A mutation also showed an increased staining of    CD65s. Flow cytometry analysis of PMN from the Fuc-TVII Q/Q individual also indicated an increased surface expression of CD65s compared to Fuc-TVII R/R individuals. However, the increase was not as pronounced as seen in the Western blot analysis. Since there is a possible competition among Fuc-TVII, Fuc-TIV, and Fuc-TIX for the same sialylated polylactosamine acceptor substrate, a lowered activity of Fuc-TVII would theoretically increase the substrate availability for Fuc-TIV and Fuc-TIX, which would explain the observed increase in CD65s expression (Fig. 7). Surprisingly, the major increase in CD65s antigens was detected on proteins in the 60 -70-kDa region, whereas the expression of SLe x was mainly detected on proteins migrating in the 90 -115-kDa region. This would suggest that the observed phenotypic changes are not only explained by substrate availability. Previous studies have shown a reciprocal expression of Fuc-TVII and Fuc-TIV during differentiation of HL60 cells and in HL60 cells deficient in FUT7 expression (42,43), indicating a linked transcriptional regulation of these enzymes. The possible effect of the FUT7 mutation on the transcriptional levels of fucosyltransferases in PMN must be studied further. Furthermore, there is always a possibility that the analyzed individuals in this study may have other differences in glycosyltransferase activity in addition to the lowered activity of Fuc-TVII, which would affect the glycoprotein profiles obtained in the Western blot analysis.
The phenotypic changes observed in individuals carrying the G329A mutation suggested a decreased activity of the Fuc-TVII enzyme. To study the effect of the G329A mutation in more detail, COS-7 cells were transfected with wild type and mutated FUT7 constructs. COS-7 cells transfected with the FUT7 gene containing the G329A mutation did not express SLe x on the cell surface in contrast to cells transfected with the wild type FUT7 construct. In addition, there was no detectable ␣1,3-fucosyltransferase activity in whole cell lysate of COS-7 cells transfected with the mutated construct when an ␣2,3sialylated lactosamine acceptor was used as substrate. The reported K m values of Fuc-TVII are in the low millimolar range when Neu5Ac␣2-3Gal␤1-4GlcNAc is used as acceptor (44,45). The obtained K m value for the sialylated type 2 acceptor used in the present study was 6 mM. K m for GDP-Fuc was 5 M. An acceptor concentration of 10 mM and a GDP-Fuc concentration of 100 M would ensure an individual reaction rate at saturating acceptor concentrations, but still there was no detectable activity in the cells transfected with the pSI-329 construct. Even when incubation with the cell lysate was prolonged to 18 h, the activity in the COS-7 cell transfected with the mutant construct gave the same incorporation as the mock-transfected COS-7 cells, indicating that the Arg 110 3 Gln substitution inactivates the Fuc-TVII enzyme. There was an overexpression of FUT7 transcripts in both cells transfected with the wild type and mutated constructs, and the levels of transcripts were similar for both constructs. This indicates that the decrease in enzymatic activity in the COS-7 cells was not an effect of reduced transcription efficiency for the mutated FUT7 construct. The lack of activity when an ␣2,3-sialylated type 1 chain was used as acceptor was to be expected, as only Fuc-TVII was overexpressed in the COS-7 cells and this enzyme specifically recognizes only the sialylated type 2 chain acceptor (11,12,15,44).
Western blots using the polyclonal anti-Fuc-TVII antiserum positively identified the expected 39-kDa band in pSI-wt-and pSI-329-transfected cells in about equal quantities. Interestingly, the cells transfected with pSI-wt, but not those transfected with pSI-329, showed an additional specific band at 55 kDa. This heavier band might correlate to a heterodimer or a highly glycosylated form of the enzyme and imposes an interesting question on the structural and functional consequences of the G329A mutation. The molecular explanation for the lack of this band is now under focus and will be the subject of a separate publication.
Sequence alignment showed that Fuc-TVII-Arg 110 is conserved in all 16 of the ␣1,3-fucosyltransferases cloned so far from vertebrate species (18,39,40). This amino acid is found just in between the hypervariable regions of ␣1,3-fucosyltransferases considered to be responsible for the acceptor binding domain and the peptide motifs presumed to be involved in the GDP-fucose binding (41). This arginine residue has not before been directly linked to enzymatic activity or specificity. It remains to be studied whether the Arg 110 3 Gln substitution directly affects enzyme activity or if the substitution affects other functions of the enzyme such as ER or Golgi retention and degradation. One of the naturally occurring mutations found to inactivate the Lewis enzyme (Fuc-TIII) has been found to induce susceptibility to protease digestion rather than directly affecting enzymatic binding sites (46).
The role of Fuc-TVII in the synthesis of selectin ligands has been demonstrated in vitro using antisense oligonucleotides  (19). The role of Fuc-TVII in vivo has also been clearly indicated by the generation of mice completely deficient in this enzyme (20). These mice showed blood leukocytosis, nonexistent binding of leukocytes to E-and P-selectin, impaired neutrophil trafficking in inflammation, and defects in lymphocyte recirculation. However, Fuc-TVII-deficient mice did not develop a phenotype as severe as mice deficient in E-and P-selectin. E/P-selectin-deficient mice exhibit extreme leukocytosis, systemic infections, and plasma cell proliferation (47), implying that lack of Fuc-TVII would not completely abolish all functional selectin ligands. The role of Fuc-TIV in generating selectin ligands has been debated. However, recent studies on mice deficient in Fuc-TVII and/or Fuc-TIV support a role for Fuc-TIV in selectin-dependent adhesion of leukocytes (48). Although Fuc-TVII seems to play the major role in generating selectin ligands, it is clear that inactivation of both Fuc-TVII and Fuc-TIV is needed to completely inhibit leukocyte adhesion to activated endothelium. In addition, several studies have shown that specific cell lines can synthesize selectin ligands upon transfection with only Fuc-TIV (49 -52). Studies using HL60 cells have also shown that CD65s can act as a ligand for E-selectin in in vitro flow systems (53). We are currently analyzing PMN from Fuc-TVII Q/Q and Fuc-TVII R/Q individuals in a selectin adhesion assay under dynamic flow conditions to address these questions.
The phenotype that can be related to neutrophil dysfunction in the Fuc-TVII deficient mice is similar to some of the clinical symptoms of patients with the disease called leukocyte adhesion deficiency type II (LAD II). LAD II patients and Fuc-TVIIdeficient mice both have raised leukocyte counts and impaired neutrophil rolling on E-and P-selectins. However, in contrast to Fuc-TVII-deficient mice, LAD II patients also suffer from an increased incidence of bacterial infections in early infancy. In addition, LAD II patients exhibit growth and mental retardation (54,55). The LAD II deficiency affects all fucosylated glycoconjugates including the selectin ligands. Recently, a mutation in a GDP-fucose transporter has been implied as responsible for the LAD II phenotype (56). Thus, the clinical symptoms in LAD II patients cannot be attributed to a specific deficiency in selectin ligand synthesis alone but reflect a general deficiency of fucose metabolism and transport. The individual homozygously mutated in FUT7 and presented in this paper is diagnosed with ulcer disease, non-insulin-dependent diabetes, osteoporosis, spondyloarthrosis, and Sjögren's syndrome. The latter diagnosis was confirmed by signs of keratoconjunctivitis sicca, sialoadenitis, and a positive titer for antinuclear antibodies. There was, however, no history of recurrent bacterial infections, and the white blood cell count was repeatedly within the reference range, thus excluding a phenotype similar to LAD II for this patient, nor have any consistent medical conditions associated with impaired neutrophil function been reported for the heterozygous members of this patient's family.