Glycobiology Advance Access originally published online on October 8, 2008
Glycobiology 2009 19(1):21-28; doi:10.1093/glycob/cwn098
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Association between expression of the H histo-blood group antigen, 
1,2fucosyltransferases polymorphism of wild rabbits, and sensitivity to rabbit hemorrhagic disease virus
2 INSERM, U892, Université de Nantes
3 Ecole Nationale Vétérinaire de Nantes
4 Office National de la Chasse et de la Faune Sauvage, Nantes, France
1 To whom correspondence should be addressed: Tel: +33-240-08-40-99; Fax: +33-240-08-40-82; e-mail: jlependu{at}nantes.inserm.fr
Received on August 21, 2008; revised on September 23, 2008; accepted on September 23, 2008
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
RHDV (rabbit hemorrhagic disease virus) is a highly virulent calicivirus that has become a major cause of mortality in wild rabbit populations (Oryctolagus cuniculus). It binds to the histo-blood group antigen (HBGA) H type 2 which requires an
1,2fucosyltransferase for its synthesis. In rabbit, three 1,2fucosyltransferases genes are known, Fut1, Fut2, and Sec1. Nonfunctional alleles at any of these loci could potentially confer resistance to RHDV, similar to human FUT2 alleles that determine the nonsecretor phenotype and resistance to infection by various NoV strains. In this study, we looked for the presence of H type 2 on buccal epithelial cells of wild rabbits from two geographic areas under RHDV pressure and from one RHDV-free area. Some animals with diminished H type 2 expression were found in the three populations (nonsecretor-like phenotype). Their frequency markedly increased according to the RHDV impact, suggesting that outbreaks selected survivors with low expression of the virus ligand. Polymorphisms of the Fut1, Fut2, and Sec1 coding regions were determined among animals that either died or survived outbreaks. The Fut2 and Sec1 genes presented a high polymorphism and the frequency of one Sec1 allele was significantly elevated, over 6-fold, among survivors. Sec1 enzyme variants showed either moderate, low, or undetectable catalytic activity, whereas all variant Fut2 enzymes showed strong catalytic activity. This functional analysis of the enzymes encoded by each Fut2 and Sec1 allele suggests that the association between one Sec1 allele and survival might be explained by a deficit of 1,2fucosyltransferase expression rather than by impaired catalytic activity. Key words: fucosyltransferase / histo-blood group antigens / polymorphism / resistance to pathogens / RHDV
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
RHDV (rabbit hemorrhagic disease virus) is a nonenveloped single positive strand RNA virus belonging to the lagovirus genus of the Caliciviridae family. It is an emergent virus responsible for a devastating disease of European rabbits (Oryctolagus cuniculus), called rabbit hemorrhagic disease (RHD) or rabbit calicivirus disease (RCD), characterized by a fulminating hepatitis accompanied by disseminated intravascular coagulation (Marcato et al. 1991
). Death usually occurs within 48 h postinfection. RHD was first reported in 1984 in rabbitries in China where it caused the death of millions of domestic animals. It most likely originated from animals imported from Germany. In 1987, it appeared in Italy and broke out simultaneously in several other European countries, including France in 1988. It rapidly expanded in wild rabbit populations and has become endemic (Mitro and Krauss 1993). Its impact on wild animals has been reported in Spain, France, and Australia. In this last country, RHDV was introduced to control rabbit populations since they threaten many endogenous plant and animal species and cause huge economic losses. These studies indicated that during the first epizootic, mortality rates could be extremely high, ranging from over 50% to 90% (Villafuerte et al. 1994; Villafuerte et al. 1995; Marchandeau et al. 1998; Mutze et al. 1998). At present, fighting the disease is possible in husbandry only, prophylaxy being based on systematic individual vaccination. Nonpathogenic strains of lagoviruses infecting European rabbits have also been described and evidence of serological relatedness with RHDV has been obtained, although the existence of cross-protection with virulent strains remains uncertain (Capucci et al. 1996; Moss et al. 2002; Marchandeau et al. 2005).
In wild rabbit populations, RHD is considered as a major cause of the species regression, particularly in dry areas of the south of France and the Iberian Peninsula (Marchandeau 2000, Marchandeau et al. 2000). In those areas, the drastic decrease of rabbit numbers threatens several other species that depend upon rabbit for their survival, such as a lynx species (Lynx pardinus) and two eagle species (Aquila aldaberti and Hieraaetus fasciatus). In France, the regression of a lizard species (Lacerta lepida), which uses rabbit warrens during hot summer periods, has also been noticed in areas where wild rabbits have become rare.
We observed earlier that RHDV binds to a trisaccharide present on the surface epithelial cells of the upper respiratory and digestive tracts, which are natural doors of entry for the virus (Ruvoën-clouet et al. 2000). The trisaccharide, Fuc2Galβ4GlcNAcβ-R, is a member of the histo-blood group antigen (HBGA) family called H type 2 and exists in other mammalian species, including human, although its tissue distribution may vary in a species-specific manner. Its synthesis requires a fucosyltransferase that catalyses the addition of a fucose residue in 1,2 linkage onto the galactose of the precursor unit (Marionneau et al. 2001). Three rabbit genes encode such enzymes: Fut1, Fut2, and Sec1. Only Sec1 and Fut2 are expressed in epithelial cells, Fut1 mRNA being detected in the brain only (Hitoshi et al. 1995; Hitoshi, Koikima, et al. 1996). In human, Sec1 is a pseudogene and the polymorphism of the FUT2 gene, orthologous to the rabbit Fut2 gene, controls sensitivity or resistance to several caliciviruses belonging to the norovirus genus (Tan and Jiang 2005; Le Pendu et al. 2006). It was shown that these viruses, which cause acute gastroenteritis, bind to human digestive surface epithelial cells using 1,2fucosylated structures closely related to that used by RHDV for attachment to rabbit epithelial cells. Addition of the 1,2-linked fucose residue to human gastric and duodenal surface epithelial cells is catalyzed by the FUT2 enzyme. The FUT2 gene is polymorphic with some alleles encoding a functional enzyme and other alleles being either null alleles or alleles encoding enzymes with low activity. Functional alleles confer the so-called secretor phenotype, characterized by the presence of 1,2fucosylated structures on various epithelial cell types or in secretions such as saliva. By contrast, null or weak alleles in the homozygous state confer the nonsecretor phenotype, characterized by the absence of such structures (Oriol et al. 2000). Nonsecretor individuals are resistant to infection by strains of norovirus which bind to 1,2fucosylated ligands (Lindesmith et al. 2003; Hutson et al. 2005; Thorven et al. 2005; Kindberg et al. 2007). Since RHDV binds to a similar glycan, by analogy with the observations on human-norovirus interaction, we hypothesized that rabbits devoid of H type 2 on their epithelial cells should be resistant to infection by RHDV. Given the very high impact of RHD on some European rabbit populations, if such nonsecretor-like animals exist, one expects that they should be positively selected by the virus and therefore that their frequency would increase in populations that have experienced RHD outbreaks. The 1,2fucosyltransferase polymorphism should thus provide a mechanism of protection against RHDV at the species level. To put this hypothesis to the test, the expression of the H type 2 ligand and the polymorphism of 1,2fucosyltransferases genes were determined in wild rabbit populations which have been closely monitored with respect to the RHDV impact.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
In human, the secretor or nonsecretor phenotypes have classically been defined by the presence of ABH antigens on salivary mucins. Practically, binding of the Ulex europaeus agglutinin I (UEA-I) lectin to human saliva characterizes the secretor phenotype, whereas the absence of binding characterizes the nonsecretor phenotype. For other mammals, collection of saliva is either unpractical or salivary mucins do not express ABH antigens, even though they may be expressed by various epithelial cell types. Nevetheless, the expression of these antigens can easily be determined from buccal epithelial cells collected using cotton swabs. We thus collected buccal epithelial cells from captured wild European rabbits and looked for the presence of UEA-I binding sites on these cells. Epithelial cells recovered from the swabs are generally mixed up with debris of foodstuff and sometimes blood cells due to some bleeding during the sampling process. This precluded a quantitative analysis of the UEA-I binding site expression on buccal epithelial cells by flow cytometry. However, under the microscope, such cells have a typical morphology and can be readily distinguished. We thus chose to estimate their UEA-I labeling visually in a semiquatitative manner. We observed a variable intensity of labeling from sample to sample. Some were clearly strongly positive, others completely negative, while others yet were weakly positive and classified as doubtful. Although the doubtful group was somewhat arbitrarily defined, classification into the three groups, positive, doubtful, and negative, was consistent among either two or three independent observers. The positive animals were considered to have a secretor phenotype, whereas the negative and doubtful animals were pooled into a single nonsecretor-like group. Analysis of the frequency of the two phenotypes in three populations differentially affected by RHD showed significant differences from a population to another. Less than 15% animals from the Dompierre site, where no RHD outbreak was ever recorded, presented a nonsecretor-like phenotype. None were completely UEA-I negative. By contrast, at the Cerisay site, where the epizootic killed about half of the population, almost 50% individuals had the nonsecretor-like phenotype, of which 18 and 15 were classified as doubtful and negative, respectively, and even more striking, at the Chèvreloup site, where RHD killed about 90% of the population in 1995, over 70% animals presented the nonsecretor phenotype, with 12 classified as doubtful and 11 as negative (Table I). The data are consistent with a significant association between the impact of RHD on wild European rabbits and the frequency of the nonsecretor-like phenotype.
|
Polymorphism of the rabbit
1,2fucosyltransferases genes and the relationship with survivalWith the aim to determine which genetic polymorphism was responsible for the nonsecretor-like phenotype, we sequenced the coding regions of the three rabbit
1,2fucosyltransferases genes, Fut1, Fut2, and Sec1, from animals of the Chèvreloup population which had been sampled in 1994, most of whom were alive at the start of the 1995 devastating outbreak, as well as from animals sampled soon after the 1995–1996 outbreaks. Sequences from a total of 83 animals were obtained revealing an important polymorphism quite different among the three genes. Thus, over about 1 kb sequenced for each gene, 1, 11, and 26 variable positions were detected for Fut1, Fut2, and Sec1, respectively. The unique Fut1 nucleotide change is silent and all wild rabbit Fut1 genes differed from the Fut1 reference sequence (Hitoshi et al. 1995) by one nonsilent mutation. Five and 14-amino-acid changes were detected in Fut2 and Sec1, respectively. Interestingly, the Fut2 mutations are clustered in two regions of the coding sequence while the Sec1 mutations were spread more evenly (Figure 1). The variable positions define two alleles for Fut1, five alleles for Fut2, and seven alleles for Sec1. They were denoted alphabetically for Fut1 (variants and β) and Fut2 (i.e., vA for variant A), and numerically for Sec1 (i.e., v1 for variant 1). An alignment of all sequences revealed shared polymorphisms between these three genes. Thus, seven polymorphisms localized in the region coding for the enzymes’ catalytic domains were shared between Fut2 and Sec1 (Figure 2). Considering the very high homology between these two genes in that region, the shared polymorphisms are clear signs of gene conversion events. Likewise, the single Fut1 polymorphic site is shared with Sec1 since sequence alignment indicates that the synonymous Fut1 780 C>G mutation corresponds to the Sec1 750 C>G mutation.
|
|
The frequencies of each Fut1, Fut2, and Sec1 allele among the animals who either died or survived during the epizootic were then compared (Figure 3). Allelic frequencies were quite variable with some alleles highly represented, such as Fut2 vG or Sec1v9, and some alleles, such as Fut2 vF or Sec1v1, being of low frequency. Some additional alleles found at other locations on the French territory, such as Fut2 vC, were completely absent from the Chèvreloup population, indicating that there existed some population differentiation among wild rabbits prior to the emergence of RHD in France (not shown).
|
When the animals were subdivided into deceased and survivors, some minor differences in allelic frequencies between the two groups occurred for Fut2, but none reached statistical significance (Figure 3A). Likewise, there was no difference in the frequencies of the two Fut1 alleles between survivors and nonsurvivors (not shown). By contrast, the Sec1v5 allele was significantly overrepresented among survivors to the 1995–1996 RHD outbreaks (Figure 3B). The increase was as high as 6.8-fold when considering the survivors alone, or 5.7-fold after pooling the survivors and the probable survivors’ groups.
Functional analysis of the 1,2fucosyltransferases allelic variants
Since our working hypothesis was that 1,2fucosyltransferases allelic variants would encode nonfunctional enzymes unable to synthesize the virus ligand, we analyzed the activity of each Fut2 and Sec1 enzyme variant. Fut1 is known not to be expressed in epithelial tissues (Hitoshi et al. 1995) and all animals presented the same single variation at the amino-acid level compared to the reference sequence. Thus, the wild rabbit Fut1 catalytic activity was not analyzed since it was not informative. All the Fut2 and Sec1 variants were cloned into an expression vector so as to generate N-terminal green fluorescent protein (GFP) fusion proteins. After transfection into Chinese hamster ovary (CHO) cells, which are devoid of endogenous 1,2fucosyltransferase activity, the presence of the 1,2fucosylated structures at the cell surface was measured by flow cytometry using the UEA-I lectin that detects H type 2 and a monoclonal antibody (mAb) that detects H type 3. Like most cultured cell lines, CHO cells do not express type 1 precursor and therefore the synthesis of H type 1 was not looked for. The presence of the GFP tail allowed normalization of the transfection and protein expression efficacy (Figure 4). The five Fut2 variants presented identical activities since they were able to equally synthesize cell surface H type 2 and H type 3 epitopes. Thus none of the five nonsynonymous mutations represented among these variant enzymes affected their catalytic activity. By contrast, notable differences were observed between the seven Sec1 variants. Some presented no detectable activity at all (v1, v2, v8, and v9), whereas some presented a weak activity (v4, v5, and v7) compared to Fut2 variants. Of all Sec1 variants, v5 was the most active to synthesize both H type 2 and H type 3 cell surface epitopes (Figure 5).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
Numerous studies concerning host–pathogen relationships suggest that genetic diversity of the host population is essential to prevent evolution of the pathogen toward maximal virulence. Direct proofs of a benefit conferred by this diversity have only been obtained in a restricted number of instances such as that of the major histocompatibility complex (MHC) (Woolhouse et al. 2002
). Recently, several NoV strains were shown to use 1,2fucosylated structures, as ligands and polymorphisms of the FUT2 1,2fucosyltransferase gene and of the ABO gene were shown to control sensitivity or resistance to infection (Harrington et al. 2002; Hutson et al. 2002, 2005; Marionneau et al. 2002; Lindesmith et al. 2003; Thorven et al. 2005; Kindberg et al. 2007). We suggested that this situation resulted from a coevolution between these caliciviruses and their human host, whereby the host genetic diversity would have been both maintained by the pathogen selective pressure and contributed to force the virus evolution toward its present low or moderate virulence (Le Pendu et al. 2006). Owing to their historical aspect, testing such hypotheses is particularly difficult, but the emergence of RHDV in wild European rabbit populations could provide the opportunity to validate some assumptions of the hypothesis. Indeed, RHDV binds to fucosylated glycans similar to those used by several NoV strains (Ruvoën-clouet et al. 2000; Rademacher et al. 2008). Because neither human NoV strains nor RHDV are cultivable, it has been difficult to demonstrate that HBGAs constitute functional receptors. Thus, at this stage, it has not been proven that RHDV binding to H type 2 is necessary for infection. However, the similarity with the NoV-human situation and the observation that young rabbits below 2 months of age are resistant to RHD (Morisse et al. 1991; Rodak et al. 1991) and express low levels of H type 2 make it highly plausible that it is a good candidate as a primary receptor (Ruvoën-clouet et al. 2000). If this was the case, RHDV virulence is expected to exert a very strong selective pressure on its host so as to positively select animals with diminished expression of H type 2. Our prediction is that when nonsecretor-like animals will have reached a significant frequency, they will contribute to decrease the virus transmission in rabbit populations, favoring mutant strains with decreased virulence because of a trade-off between virulence and transmission as indicated by mathematical modeling (Fouchet et al., in preparation). Such nonsecretor-like animals were thus searched in wild rabbit populations differentially affected by RHD. Owing to the difficulty to obtain tissues from these animals, hunting in heavily affected populations being now restricted, the phenotyping was performed on buccal epithelial cells. Since it is no longer allowed us to recapture animals from these populations, DNA could not be obtained from the same animals as the cellular material collected on swabs was entirely used for phenotyping. Nevertheless, for the first time, a nonsecretor-like phenotype was described in a wild mammalian species. So far a deficit of 1,2fucosylated HBGAs in secretions or on epithelial cells had been observed in human and pigs only (Oriol et al. 2000; Coddens et al. 2007). Most interestingly, the frequency of animals with this phenotype was significantly associated with the impact of RHDV outbreaks, providing indirect evidence that the H type 2 ligand is necessary for infection and that its absence or low expression should confer resistance to the disease, similar to what has been observed between NoV and human HBGAs expression. In addition, it indicates that RHDV may have selected animals of the nonsecretor-like phenotype. French wild rabbit populations show only moderate differentiation as determined from microsatellite markers (Queney et al. 2001). Nevertheless, it cannot be completely ruled out that the differential frequency of the nonsecretor-like phenotype among the three populations that we analyzed could have preexisted RHDV emergence. It was therefore important to ascertain that polymorphism at the 1,2fucosyltransferases loci was associated with survival to RHD. Although not systematically searched in rabbit breeds, we failed to observe the nonsecretor-like phenotype among 30 laboratory animals (data not shown). Fortunately, DNA samples from animals captured either before or soon after the 1995–1996 outbreaks at the Chèvreloup site were available. Monitoring data from these animals were also available, allowing us to distinguish survivors and probable survivors from nonsurvivors (in practice, animals whose tract was lost immediately after an outbreak). The coding region of the three rabbit 1,2fucosyltransferases genes were thus sequenced. We chose to sequence those regions only since the noncoding regions of the rabbit genes are unknown as yet and since in human, all known mutations that generate either null or weak alleles affect the coding regions. Surprisingly, very distinct polymorphisms were found between the three genes. Fut1 presented almost no variation whereas a large number of mutations were found in Fut2 and even more in Sec1. In addition, the distribution of the mutations was quite different between Fut2 and Sec1. Shared polymorphisms signed the existence of gene conversion which leads to convergent evolution of paralogues and increases genetic diversity (Teshima and Innan 2004). The pattern of diversity that we observed is thus consistent with a possible role for Fut2 and Sec1 expressed in epithelial cells in coping with the environment. By contrast, the near absence of variation in Fut1, which is expressed in the brain only (Hitoshi et al. 1995), suggests that it undergoes a strong negative selection to maintain functionality. Nevertheless, functional analysis of Fut2 enzyme variants revealed that all were equally active. In accordance with the first description of rabbit 1,2fucosyltransferases activity (Hitoshi, Kusunoki, et al. 1996), we observed that the reference Sec1 (v5) was functional but that its activity was lower than that of any Fut2 variant. Other variants of Sec1 showed even lower activity or no activity at all, suggesting that the Sec1 gene of rabbits is on its way to pseudogenization, similar to what happened to the apes, pig, and mice orthologues. Yet, the Sec1v5 allele was strongly associated with resistance to RHD since its frequency was largely higher among survivors than among nonsurvivors. Along with the association between the nonsecretor-like phenotype and the impact of RHD, this observation suggests that decreased expression of 1,2fucosylated ligands determines resistance to RHDV. Nevertheless, the Sec1v5 allele is not a null allele and it is always associated with a functional Fut2. It is therefore a marker of survival but cannot be the cause of the nonsecretor-like phenotype or of the resistance to RHDV. The three genes Fut1, Fut2, and Sec1 are closely located in the human, mouse, rat, cow, and pig genomes. Their genomic organization has not been determined in rabbit as yet, but it is likely similar to that of other species. The ORFs of Fut2 and Sec1 are separated by about 22–28 kb with a Fut2 untranslated 5' exon lying at near equidistance between the two genes. We therefore hypothesize that in the Chèvreloup population, the Sec1v5 allele is in linkage disequilibrium with a polymorphism affecting Fut2 expression which would be localized in its regulatory region located near Sec1. So far polymorphisms of HBGAs expression in human have been linked to mutations in the coding regions of glycosyltransferases genes. However, mutations in regulatory regions affecting gene transcription could similarly affect HBGAs expression. The Sec1v5 variant would thus be genetically linked to such a regulatory mutation in the 1,2fucosyltransferases loci rather than directly causative of the nonsecretor-like phenotype. Further work is underway to validate this hypothesis.
Regardless, the present work provides the first evidence of an 1,2fucosyltransferase gene polymorphism associated with survival to a viral disease and of a nonsecretor-like phenotype in a wild mammalian population whose frequency is also associated with the impact of the disease. The latter aspect is suggestive of a process of selection of resistant animals taking place in naturo. Although limited by the availability of the biological material (wild rabbits cannot be sampled freely and are extremely difficult to breed), these observations represent an important step in the analysis of the evolutionary relationship between HBGAs and caliciviruses.
|
Material and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
Sampling wild rabbits
Animals were caught by trapping or using ferrets (Mustela furo). Buccal epithelial cells were collected using cotton swabs from animals captured at three different locations on the French metropolitan territory in 2004. The three sites were chosen for their very different situation with regard to the RHD impact. When sampled, the Dompierre site had never been affected by the disease. In 2000 at the Cerisay site, an outbreak caused a 50% mortality (Marchandeau et al. 2005
). Finally, the wild rabbit population of the Chèvreloup site experienced two consecutive outbreaks in 1995 and 1996. In 1995, annual mortality rates at that site were estimated as high as 88% in adults and 99% in juveniles (Marchandeau et al. 1998).
In addition, DNA samples collected in 1994 and 1996, i.e., before and after the RDH outbreaks, during a study on the genetic structure of the Chèvreloup population were used (Queney et al. 2000, 2001). The Chèvreloup population being monitored for a long-term survey, the rabbits were individually marked with reflecting ear tags which allowed individual recognition from a distance and thus a follow-up of each animal during and after the outbreaks (Marchandeau et al. 2000). Three methods of recapture (ferreting, trapping, and spotlighting) were used. Thus, from that site, three groups of animals could be defined. The first group includes rabbits sampled in 1994 which survived the epizootic (N = 23). The second group includes animals sampled in 1994 and presumed to have died during either the 1995 or the 1996 outbreak (N = 24). The third group includes animals first captured soon after the 1996 outbreak that presented adult morphological features and body weight and therefore considered as probable survivors (N = 20). One may notice that in wild populations, the fate of each animal is almost impossible to record. Therefore, the classification of the rabbits into three classes imperfectly reflects their sensitivity or resistance to the disease. Some rabbits that survived the 1995–1996 outbreaks may have not been exposed to the virus and some that died may have died from causes other than RHD.
Determination of the nonsecretor-like phenotype
Cotton swabs were dipped in 1 mL phosphate-buffered saline (PBS), and then centrifuged at 1500 rpm at room temperature for 5 min. The cell pellet was collected and incubated for 1 h at room temperature with the UEA-I-FITC conjugated lectin (Sigma, St Louis, MO). The lectin specifically binds to the H type 2 motif Fuc2Galβ4GlcNAcβ-R recognized by RHDV. Cells were then washed thrice with PBS, and nuclei were stained with propidium iodide prior to observation under an epifluorescence microscope (Olympus, Hamburg, Germany). Membrane fluorescence of cells with a clear morphology of buccal epithelial cells was recorded by two or three independent observers without prior knowledge of the animals’ site of origin and qualitatively graded as strongly positive, weakly positive, or negative.
Sequencing of the rFut1, rFut2, and rSec1 open reading frames
DNA extractions were performed from ear fragments using the QIAamp DNA Mini kit (Qiagen, Courtaboeuf, France). The coding regions of the Fut1, Fut2, and Sec1 rabbit genes were amplified using primers deduced from the X80226
[GenBank]
, X91269
[GenBank]
, and X80225
[GenBank]
sequences, respectively: gen Fut2 (CCCAAGAGCT AACTAACATC CCT GCTG) and gen Fut2 inv (GCAAGAAACG GGTCTTG GCC CACGTGAC) for Fut2, gen Sec1 (CTTCAGCCTGCAC AGCCCCA GCCGTT) and gen Sec1 inv (ACTCCGTGTG GGATCCCGGG TCCATGC) for Sec1, and gen Fut1 (CT GTCTGGAA AGCCATCATC) and gen Fut1 inv (GTAGAAT CAC TCTGGCTGTG) for Fut1. Amplifications were performed in the presence of 2.5% DMSO using GoTaq (Promega, Madison, WI). Sequencing was performed on an Applied Biosystems 3730 sequencer (Ouest-Genopole/IFR26 core facility) and sequences were analyzed using the BioEdit v7.0.5.3 program (Hall T.A., North Carolina State University, Department of Microbiology).
Amplification and cloning
For cloning, amplification were performed using the AmpliTaq Gold kit with GeneAmp 10x PCR Gold Buffer and MgCl2 Solution (Applied Biosystems) with the following primers: rFut 2.3 (ATGAGCACCG CCCAGGTCCC CTTC) and rFut 2.2 inv (TCAGTGCTTG AGCAATGGGG ACAG) for Fut2, and rSec1.3 (ATGAGATTCG CCCCTGACTA TGTCC) and rSec1.2 inv (CTAGAGGCCA CTCCACAAGG C) for Sec1. The obtained amplicons are 1044 bp and 1066 bp long for Fut2 and Sec1, respectively. Cloning was performed in the pcDNA 3.1 vector using the NT-GFP Fusion TOPO TA Expression kit (Invitrogen, Paisley, UK), following the manufacturer's instructions from 2-µL fresh PCR product and a 20-min incubation at room temperature. The bacteria used for transformation were E. coli One Shot TOP10F’ (Invitrogen). Orientation of the insert was determined using the T7 Pro primer (TAATACGACT CACTATAGGG) and a primer common to the Fut2 and Sec1 inserts, rFut3.2 Inv (GTTGAGGTGG TAGTTCTGCC). All inserted sequences were verified before use in transfection experiments.
Assay of the enzyme activity of the 1,2fucosyltransferases variants
In order to study the functionality of each of the allelic variant of Fut2 and Sec1, the coding regions were cloned into a eukaryotic expression vector so as to obtain a GFP fusion protein with the GFP domain at the N-terminus. CHO-K1 cells were cultured in RPMI 1640, supplemented with 10% fetal calf serum, 2 mM L-glutamine, free nucleotides (10 µg/mL), 100 U/mL penicillin, and 100 mg/mL streptomycin (Gibco, Paisley, UK). They were cultured at confluence after dispersal with 0.025% trypsin and 0.02% EDTA. Cells were routinely checked for mycoplasma contamination using Hoëchst 33258 (Sigma, St Louis, MO) labeling. Following amplification and purification with the Miniprep Qiagen kit, the recombinant pcDNA 3.1 constructs were transfected into CHO-K1 cells using Lipofectamin 2000 (Invitrogen) according to the manufacturer's instructions. Forty-eight hours following transfection, cells were collected and the presence of 1,2-linked fucose residues was tested by flow cytometry using the UEA-I lectin (Vector Labs) and the MBr1 mAb (Alexis Biochemicals, San Diego, CA) that detect the H type 2 (Fuc2Galβ4GlcNAcβ-R) and H type 3 (Fuc2Galβ3GalNAc-R) motifs, respectively. To this aim, 2.5 x 105 viable transfected cells were incubated in the presence of either biotin-labeled lectin at 5 µg/mL (Vector Labs) or the MBr1 mAb at 5 µg/mL for 20 min at 4°C. After three washings with PBS, cells were incubated under the same conditions in the presence of either PerCP-conjugated streptavidin (BD Biosciences, San José, CA) at 0.5 µg/mL or a Cy5-conjugated antimouse IgG (BD Biosciences) at a 1/500 dilution. After three more washings with PBS, cell fluorescence was measured on a FACScalibur flow cytometer (Becton-Dickinson, Heidelberg, Germany) and analyzed using the CellQuest program (Becton-Dickinson). The transfected protein expression was quantified by the GFP fluorescence recorded on the FL1 channel. The presence of the H type 2 and H type 3 motifs was detected on the FL3 and FL4 channels, respectively.
Statistical analysis
Comparison of frequencies of the nonsecretor-like phenotype between the three rabbit populations was performed by a likelihood ratio chi-square test with one degree of freedom. Analysis of the frequencies of the Fut2 and Sec1 alleles between the Chèvreloup subgroups was performed by a likelihood ratio chi-square with one degree of freedom and correction for multiple comparisons with p corrected = 1–(1–p)n, where p is the uncorrected P value and n the number of comparisons (Sham 1998).
|
Funding |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
The BRG (Bureau des Ressources Génétiques); the FNC (Fédération Nationale des Chasseurs); and the Région des Pays de la Loire.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
None declared.
|
Acknowledgements |
|---|
The authors are thankful to Dr S. Marionneau for involvement in the initial phase of the work and to Dr G. Queney for providing DNA samples.
|
Abbreviations |
|---|
CHO, Chinese hamster ovary; GFP, green fluorescent protein, HBGA, histo-blood group antigen; mAb, monoclonal antibody; PBS, phosphate-buffered saline; RHDV, rabbit hemorrhagic disease virus; UEA-I, Ulex europaeus agglutinin I
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsFundingConflict of interest statementReferences |
|---|
Capucci L, Fusi P, Lavazza A, Pacciarini ML, Rossi C. Detection and preliminary characterization of a new rabbit calicivirus related to rabbit haemorrhagic disease virus but non pathogenic. J Virol (1996) 70:8614–8623.
Coddens A, Verdonck F, Tiels P, Rassschaert K, Goddeeris BM, Cox E. The age-dependent expression of the F18+ E. coli receptor on porcine gut epithelial cells is positively correlated with the presence of histo-blood group antigens. Vet Microbiol (2007) 122:332–341.[CrossRef][Medline]
Harrington PR, Lindensmith L, Yount B, Moe CL, Baric RS. Binding of Norwalk virus-like particles to ABH histo-blood group antigens is blocked by antisera from infected human volunteers or experimentally vaccinated mice. J Virol (2002) 76:12325–12343.
Hitoshi S, Koijima N, Kusunoki S, Inokuchi J, Kanazawa I, Tsuji S. Expression of the beta-galactoside alpha 1,2-fucosyltransferase gene suppresses axonal outgrowth of neuro2a neuroblastoma cells. J Neurochem (1996) 66:1633–1640.[Medline]
Hitoshi S, Kusunoki S, Kanazawa I, Tsuji S. Molecular cloning and expression of two types of rabbit β-galactoside 1,2fucosyltransferase. J Biol Chem (1995) 270:8844–8850.
Hitoshi S, Kusunoki S, Kanazawa I, Tsuji S. Molecular cloning and expression of a third type of rabbit GDP-l-fucose: β-D-galactoside 2--L-fucosyltransferase. J Biol Chem (1996) 271:16975–16981.
Hutson AM, Airaud F, Le Pendu J, Estes MK, Atmar RL. Norwalk virus infection associates with secretor status genotyped from sera. J Med Virol (2005) 77:116–120.[CrossRef][Web of Science][Medline]
Hutson AM, Atmar RL, Graham DY, Estes MK. Norwalk virus infection and disease is associated with ABO histo-blood group type. J Infect Dis (2002) 185:1335–1337.[CrossRef][Web of Science][Medline]
Kindberg E, Akerlind B, Johnsen C, Knudsen JD, Heltberg O, Larson G, Böttinger B, Svensson L. Host genetic resistance to symptomatic norovirus (GGII.4) infections in Denmark. J Clin Microbiol (2007) 45:2720–2722.
Le Pendu J, Ruvöen-Clouet N, Kindberg E, Svensson L. Mendelian resistance to human norovirus infections. Semin Immunol (2006) 18:375–386.[CrossRef][Web of Science][Medline]
Lindesmith L, Moe CL, Marionneau S, Ruvoën-clouet N, Jiang X, Lindblad L, Stewart PA, Le Pendu J, Baric RS. Human susceptibility and resistance to Norwalk virus infection. Nat Med (2003) 9:548–553.[CrossRef][Web of Science][Medline]
Marcato PS, Benazzi C, Vecchi G, Galeotti M, Salda LD, Sarli G, Lucidi P. Clinical and pathological features of viral haemorrhagic disease of rabbits and the European brown hare syndrome. Rev Sci Tech Off Int Epiz (1991) 10:371–392.
Marchandeau S. Enquête nationale sur les tableaux de chasse à tir saison 1998–1999. Le lapin de garenne. Faune Sauvage (2000) 251:18–25.
Marchandeau S, Chaval Y, Le Goff E. Prolonged decline in the abundance of wild European rabbits and high immunity level over three years following the arrival of rabbit haemorrhagic disease. Wildl Biol (2000) 6:141–147.
Marchandeau S, Le Gall-Recule G, Bertagnoli SJA, Botti G, Lavazza A. Serological evidence for a non-protective RHDV-like virus. Vet Res (2005) 36:53–62.[CrossRef][Medline]
Marchandeau SJC, Portejoie Y, Barraud S, Chaval Y. Impact of viral hemorrhagic disease on a wild population of European rabbits in France. J Wildl Dis (1998) 34:429–435.[Abstract]
Marionneau S, Cailleau-Thomas A, Rocher J, Le Moullac-Vaidye B, Ruvoën-clouet N, Clément M, Le Pendu J. ABH and Lewis histo-blood group antigens, a model for the meaning of oligosaccharide diversity in the face of a changing world. Biochimie (2001) 83:565–573.[CrossRef][Medline]
Marionneau S, Ruvöen-Clouet N, Le Moullac-Vaidye B, Clement M, Cailleau-Thomas A, Riuz-Palacios G, Huang PW, Jiang X, Le Pendu J. Norwalk virus binds to H types 1/3 histo-blood group antigens on gastro-duodenal epithelial cells of "secretor" individuals. Gastroenterology (2002) 122:1967–1977.[CrossRef][Web of Science][Medline]
Mitro S, Krauss H. Rabbit hemorrhagic disease: A review with special reference to its epizootiology. Eur J Epidemiol (1993) 9:70–78.[CrossRef][Medline]
Morisse J-P, Le Gall G, Boilletot E. Hepatitis of viral origin in Leporidae: Introduction and aetiological hypotheses. Rev Sci Tech Off Int Epiz (1991) 10:283–295.
Moss SR, Turner SL, Trout RC, White PJ, Hudson PJ, Desai A, Armesto M, Forrester NL, Gould EA. Molecular epidemiology of rabbit haemorrhagic disease virus. J Gen Virol (2002) 83:2461–2467.
Mutze G, Cooke B, Alexander P. The initial impact of rabbit hemorrhagic disease on European rabbit populations in South Australia. J Wildl Dis (1998) 34:221–227.[Abstract]
Oriol R, Candelier JJ, Mollicone R. Molecular genetics of H. Vox Sang (2000) 78:105–118.[Medline]
Queney G, Ferrand N, Marchandeau S, Azevodo M, Mougel F, Branco M, Monnerot M. Absence of a genetic bottleneck in a wild rabbit (Oryctolagus cuniculus) population exposed to a severe viral epizootic. Mol Ecol (2000) 9:1253–1264.[Medline]
Queney G, Ferrand N, Weiss S, Mougel F, Monnerot M. Stationary distributions of microsatellite loci between divergent population groups of the European rabbit (Oryctolagus cuniculus). Mol Biol Evol (2001) 18:2169–2178.
Rademacher C, Krishna NR, Palcic MM, Parra F, Peters T. NMR experiments reveal the molecular basis of receptor recognition by a calicivirus. J Am Chem Soc (2008) 130:3669–3675.
Rodak L, Smid B, Valicek L. Application of control measures against viral haemorrhagic disease of rabbits in the Czech and Slovak Federal Republic. Rev Sci Tech Off Int Epiz (1991) 10:513–524.
Ruvoën-clouet N, Ganière JP, André-Fontaine G, Blanchard D, Le Pendu J. Binding of rabbit hemorrhagic disease virus to antigens of the ABH histo-blood group family. J Virol (2000) 74:11950–11954.
Sham P. Statistics in Human Genetics (1998) London: Arnold.
Tan M, Jiang X. Norovirus and its histo-blood group antigen receptors: An answer to a historical puzzle. Trends Microbiol (2005) 13:285–293.[CrossRef][Web of Science][Medline]
Teshima KM, Innan H. The effect of conversion on the divergence between duplicated genes. Genetics (2004) 166:1553–1560.
Thorven M, Grahn A, Hedlund K-O, Johansson H, Wahlfrid C, Larson G, Svensson L. A homozygous nonsense mutation (428G>A) in the human FUT2 gene provides resistance to symptomatic norovirus (GGII) infections. J Virol (2005) 79:15351–15355.
Villafuerte R, Calvete C, Blanco JC, Lucientes J. Incidence of viral hemorrhagic disease in wild rabbit populations in Spain. Mammalia (1995) 59:651–659.
Villafuerte R, Calvete C, Gortazar C, Moreno S. First epizootic of rabbit hemorrhagic disease in free living populations of Oryctolagus cuniculus at Doñana national park, Spain. J Wildl Dis (1994) 30:176–179.[Abstract]
Woolhouse MEJ, Webster JP, Domingo E, Charlesworth B, Levin BR. Biological and biomedical implications of the co-evolution of pathogens and their hosts. Nat Genet (2002) 32:569–577.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||