The alphaGal epitope of the histo-blood group antigen family is a ligand for bovine norovirus Newbury2 expected to prevent cross-species transmission

Abstract

Among Caliciviridae, the norovirus genus encompasses enteric viruses that infect humans as well as several animal species, causing gastroenteritis. Porcine strains are classified together with human strains within genogroup II, whilst bovine norovirus strains represent genogroup III. Various GI and GII human strains bind to carbohydrates of the histo-blood group family which may be shared among mammalian species. Genetic relatedness of human and animal strains as well as the presence of potentially shared ligands raises the possibility of norovirus cross-species transmission. In the present study, we identified a carbohydrate ligand for the prototype bovine norovirus strain Bo/Newbury2/76/UK (NB2). Attachment of virus-like particles (VLPs) of the NB2 strain to bovine gut tissue sections showed a complete match with the staining by reagents recognizing the Galalpha1,3 motif. Alpha-galactosidase treatment confirmed involvement of a terminal alpha-linked galactose. Specific binding of VLPs to the alphaGal epitope (Galalpha3Galbeta4GlcNAcbeta-R) was observed. The binding of Galalpha3GalalphaOMe to rNB2 VLPs was characterized at atomic resolution employing saturation transfer difference (STD) NMR experiments. Transfection of human cells with an alpha1,3galactosyltransferase cDNA allowed binding of NB2 VLPs, whilst inversely, attachment to porcine vascular endothelial cells was lost when the cells originated from an alpha1,3galactosyltransferase KO animal. The alphaGal epitope is expressed in all mammalian species with the exception of the Hominidaea family due to the inactivation of the alpha1,3galactosyltransferase gene (GGTA1). Accordingly, the NB2 carbohydrate ligand is absent from human tissues. Although expressed on porcine vascular endothelial cells, we observed that unlike in cows, it is not present on gut epithelial cells, suggesting that neither man nor pig could be infected by the NB2 bovine strain.

Figure 1. Analysis of rNB2 VLPs binding and of their ligand expression by immunohistology.

VLPs at 1 µg/ml were incubated on tissue sections and detected as described in the Materials and Methods section. Binding of rNB2 VLPs to bovine duodenum surface epithelium (A), glands (B), smooth muscle (C) and negative control in absence of rNB2 VLPs (D). Effect of a 10 mM sodium periodate treatment prior to VLPs incubation (E) and serial control section pretreated in the same conditions without periodate (F). Staining of the surface epithelium (G) and vascular endothelium in connective tissue (H) of bovine duodenum by an anti-αGal mAb. Lack of binding of rNB2 VLPs on human duodenum (I). Binding of rNB2 VLPs to porcine duodenum (J). Effect of α–galactosidase treatment prior to rNB2 VLPs incubation on a porcine duodenum tissue section (K) and control section pretreated in the same conditions in absence of enzyme (L). Epithelial cells are indicated by arrows, vascular endothelium cells by small arrowheads, and smooth muscle cells by a star.

Figure 2. Expression of rNB2 VLPs ligands and of HBGAs along the bovine digestive tract.

Expression of rNB2 VLPs epithelial ligands and HBGAs on the various parts of the digestive tract was determined by immunohistochemistry (see Fig. 1). Intensity of labeling shown on the y axis was visually graded from strongly positive (grade 3), moderately positive (grade 2) to weakly positive (grade 1) and negative (grade 0). (A) NB2 VLPs ligands ; (B) αGal epitope detected using either the GS1-B4 isolectin or the 4F102c2 mAb ; (C) A blood group antigen detected using mAb 9113D10 which recognizes all types of A epitopes ; (D) Htype2/Le^y epitopes detected using either the UEA-I lectin, or the 19-0LE mAb. *In addition to epithelial cells, rNB2 VLPs, GS1-B4 and the 4F102c2 mAb stained the vascular endothelium throughout the digestive tract.

Figure 3. Red blood cell agglutination and saliva recognition.

(A) Agglutination of bovine and human erythrocytes by VLPs from the bovine NB2 strain (gray bars) and from a human GII.4 strain of the Grimsby type which binds to Le^y (black bars). Erythrocytes from 2 individual cows and human erythrocytes from A, B or O blood group individuals were used. Results are shown as hemagglutination titers (HA titers). (B) Recognition of human saliva by VLPs from the bovine NB2 strain and the human Dijon (GII.4) strain determined by ELISA. Saliva samples were grouped into nonsecretors (white bars), A secretors (black bars), B secretors (dark gray bars) and O secretors (light gray bars) with 4 samples in each group. (C) Binding of rNB2 VLPs to individual bovine saliva samples determined by ELISA. Individual samples are numbered from 1 to 17 and ranked according to increasing OD 450 nm values. (D) Expression of the A blood group antigen in individual bovine saliva samples. (E) Expression of the αGal epitope detected by human natural anti-αGal antibodies in individual bovine saliva. (F) Expression of H/Le^y epitopes detected by the UEA-I lectin in individual bovine saliva samples. In D, E and F, samples are ranked as in C.

Figure 4. Binding of rNB2 VLPs to immobilized synthetic oligosaccharides.

(A) A panel of neoglycoconjugates was coated on ELISA plates at 10 µg/ml and binding of either NB2 (white bars) or NV VLPs (black bars) was detected as described in the Materials and Methods section. OD 450 nm values on selected glycoconjugates are shown. No positive values were recorded for any of the other glycoconjugates listed in Table 1. (B) Binding of rNB2 VLPs to serially diluted structurally related PAA-conjugates. White squares: αGal trisaccharide ; black triangles: αGal-Lex ; black squares: mean of other PAA-conjugates (see Table 1 for structures). Results are shown as OD 450 nm as a function of the reciprocal of PAA-conjugate dilutions with 1/25 corresponding to 40 µg/ml.

Figure 5. Inhibition of binding to bovine saliva by α-galactosidase treatment or by human natural anti-αGal antibodies.

(A) Bovine saliva from a positive individual (#8) or PAA-conjugates (αGal trisaccharide and αGalNAc, see Table 1 for structures) were coated at a 1/1000 dilution or 10 µg/ml respectively and either treated with α–galactosidase from green coffee beans (white bars) or enzyme buffer only (black bars). (B) Bovine saliva from 2 positive individuals (#8 and #12) were coated as above and incubated with purified human anti-αGal (white bars) or blocking buffer only (black bars) for 2 hours prior to addition of rNB2 VLPs. Binding of VLPs was detected as described in the Materials and Methods section and shown as OD 450 nm values. Dashed lines indicate background level.

Figure 6. Observation of STD signals for the αGal disaccharide and the blood-group B trisaccharide.

(A) Reference ¹H NMR spectrum and (B) STD NMR spectrum of Galα3GalαOMe in the presence of bovine norovirus VLPs. (C) Reference ¹H NMR spectrum and (D) STD NMR spectrum of the methyl glycoside of the blood group B-trisaccharide in the presence of bovine norovirus VLPs. STD signals are only observed for Galα3GalαOMe. The blood group B-trisaccharide shows no STD effects, and is therefore not recognized by the VLPs. Spectra were recorded at 500 MHz at a temperature of 282 K with 256 and 816 scans, respectively.

Figure 7. Binding epitope of Galα3GalαOMe binding to rNB2 VLPs (top) as derived from STD build-up curves (bottom).

Greyscale circles encode the relative size of the saturation transfer and reflect the vicinity to protons in the binding pocket of rNB2 VLP. The larger the saturation transfer, the closer are the ligand protons to protons of the VLP binding pocket. From the Figure it is seen that the region around the glycosidic linkage is essential for recognition by the VLPs.

Figure 8. Cytofluorimetric analysis of the binding of rNB2 VLPs and the GS1-B4 lectin to rat Ggta1 transfected or control mock transfected HEK 293 cells and to porcine aortic endothelial vascular cells (PAEC) from wild type or Ggta1 KO pigs.

Cells were incubated with either the GS1-B4 lectin or rNB2 VLPs and binding was detected as described in the materials and methods section. The respective negative controls (grey plots) correspond to either cells incubated in absence of the lectin or to cells incubated in the absence of VLPs. Another negative control performed in the presence of VLPs followed by incubation with an irrelevant immune rabbit antiserum and FITC-labeled anti-rabbit IgG yielded similar results. The log of fluorescence intensities in arbitrary units is plotted against the cell number.