alpha-L-Fucp-(1->2)-beta-D-Galp-(1->3)-beta-D-GlcpNAc-(1->3)-beta-D-Galp-(1->4)-D-Glcp

Abstract

The P[19] genotype belongs to the P[II] genogroup of group A rotaviruses (RVs). However, unlike the other P[II] RVs, which mainly infect humans, P[19] RVs commonly infect animals (pigs), making P[19] unique for the study of RV diversity and host ranges. Through in vitro binding assays and saturation transfer difference (STD) nuclear magnetic resonance (NMR), we found that P[19] could bind mucin cores 2, 4, and 6, as well as type 1 histo-blood group antigens (HBGAs). The common sequences of these glycans serve as minimal binding units, while additional residues, such as the A, B, H, and Lewis epitopes of the type 1 HBGAs, can further define the binding outcomes and therefore likely the host ranges for P[19] RVs. This complex binding property of P[19] is shared with the other three P[II] RVs (P[4], P[6], and P[8]) in that all of them recognized the type 1 HBGA precursor, although P[4] and P[8], but not P[6], also bind to mucin cores. Moreover, while essential for P[4] and P[8] binding, the addition of the Lewis epitope blocked P[6] and P[19] binding to type 1 HBGAs. Chemical-shift NMR of P[19] VP8* identified a ligand binding interface that has shifted away from the known RV P-genotype binding sites but is conserved among all P[II] RVs and two P[I] RVs (P[10] and P[12]), suggesting an evolutionary connection among these human and animal RVs. Taken together, these data are important for hypotheses on potential mechanisms for RV diversity, host ranges, and cross-species transmission.

Expression and purification of VP8* proteins in Escherichia coli.

The VP8* sequences (amino acids 46 to 231) from a human RV P[19] (GenBank accession number ) was expressed in E. coli BL21 and purified by glutathione S-transferase (GST) tag affinity purification, as described previously (). Recombinant VP8* proteins of other P[II] RVs (P[4], P[6], and P[8]) and a P[10] RV were also purified using the same expression system. 15N- or 15N,13C-labeled P[19] VP8* proteins were also made for the NMR studies using the same E. coli culture procedures, with the addition of 15N-minimal growth medium (22 mM KH2PO4, 50 mM Na2HPO4, 8.5 mM NaCl, 0.1 mM CaCl2, 2 mM MgSO4, 1× BME vitamins (catalog no. B6891; Sigma), 18 mM 15NH4Cl, and 22 mM glucose) or 15N,13C-minimal growth medium supplied with 22 mM [13C]glucose (). The GST tag of the labeled VP8* proteins was removed for the NMR study by cutting with thrombin (Sigma-Aldrich Co., St. Louis, MO), followed by fast protein liquid chromatography (FPLC) gel filtration to collect the VP8* protein fractions and concentration by ultracentrifugation. The purified VP8* proteins were stored at −80°C until needed.

STD NMR experiments.

All saturation transfer difference (STD) NMR spectra were acquired in Shigemi Tubes (Shigemi, USA) with a Bruker 600-MHz Avance III spectrometer at 10°C using a conventional HCN probe and the TopSpin 3.1 pulse sequence STDDIFFGG19.3. The NMR sample for mucin core 2 in complex with P[19] VP8* was prepared with 200 μM protein and 6.0 mM ligand in a final volume of 300 μl of Tris NMR buffer. The LNFPI STD sample was prepared with 130 μM P[19] VP8* and 6.6 mM LNFPI in a final volume of 300 μl of phosphate NMR buffer. This results in total protein-ligand ratios of 1:35 and 1:50 for the mucin core 2 and LNFPI samples, respectively. The protein resonances were saturated at 0 ppm with a cascade of 40 Gaussian-shaped pulses with duration of 50 ms, resulting in a total saturation time of 2 s. The off-resonance saturation was applied to 30 ppm, and 1,024 experiments were acquired for each with a 5-s recycle delay. A spin-lock filter with strength of 10 kHz and duration of 100 to 200 ms was applied to suppress the broad protein resonance signals, and Watergate 3-9-19 was used to suppress the residual water signal. The STD spectra were obtained by subtracting the on-resonance spectrum from the off-resonance reference spectrum. STD amplification factors were determined by the equation (Io − Isat)/Io, where Io and Isat are the intensities of the signals in the reference and saturated spectra, respectively, and the normalized STD amplification percentage was determined by dividing by the largest STD amplification factor for each ligand. The protons with the largest STD amplification percentage along with data from the 2D STD TOCSY spectrum, which were used to resolve overlap from the 1D spectra, were used to assign the binding epitopes.

P[19] VP8* recognizes mucin cores and the type 1 HBGA precursor.

The ligand binding specificity of the P[19] VP8* was further characterized by STD NMR. Data for a type 1 HBGA pentasaccharide, LNFPI (Fucα1-2Galβ1-3GlcNAcβ1-3Galβ1-4Glc), demonstrated that the extended type 1 chain of an internal Gal (Fucα1-2Galβ1-3GlcNAcβ1-3Gal) is also involved in the interaction with P[19] VP8*, in addition to the GlcNAc and the terminal (H) fucose ( and ). Similarly, STD NMR of P[19] VP8* with mucin core 2 showed a strong interaction with the disaccharide GlcNAcβ1-6GalNAc, in which the GlcNAc residue was highly involved ( and ).

Abstract

The P[19] genotype belongs to the P[II] genogroup of group A rotaviruses (RVs). However, unlike the other P[II] RVs, which mainly infect humans, P[19] RVs commonly infect animals (pigs), making P[19] unique for the study of RV diversity and host ranges. Through in vitro binding assays and saturation transfer difference (STD) nuclear magnetic resonance (NMR), we found that P[19] could bind mucin cores 2, 4, and 6, as well as type 1 histo-blood group antigens (HBGAs). The common sequences of these glycans serve as minimal binding units, while additional residues, such as the A, B, H, and Lewis epitopes of the type 1 HBGAs, can further define the binding outcomes and therefore likely the host ranges for P[19] RVs. This complex binding property of P[19] is shared with the other three P[II] RVs (P[4], P[6], and P[8]) in that all of them recognized the type 1 HBGA precursor, although P[4] and P[8], but not P[6], also bind to mucin cores. Moreover, while essential for P[4] and P[8] binding, the addition of the Lewis epitope blocked P[6] and P[19] binding to type 1 HBGAs. Chemical-shift NMR of P[19] VP8* identified a ligand binding interface that has shifted away from the known RV P-genotype binding sites but is conserved among all P[II] RVs and two P[I] RVs (P[10] and P[12]), suggesting an evolutionary connection among these human and animal RVs. Taken together, these data are important for hypotheses on potential mechanisms for RV diversity, host ranges, and cross-species transmission.

Expression and purification of VP8* proteins in Escherichia coli.

The VP8* sequences (amino acids 46 to 231) from a human RV P[19] (GenBank accession number ) was expressed in E. coli BL21 and purified by glutathione S-transferase (GST) tag affinity purification, as described previously (). Recombinant VP8* proteins of other P[II] RVs (P[4], P[6], and P[8]) and a P[10] RV were also purified using the same expression system. 15N- or 15N,13C-labeled P[19] VP8* proteins were also made for the NMR studies using the same E. coli culture procedures, with the addition of 15N-minimal growth medium (22 mM KH2PO4, 50 mM Na2HPO4, 8.5 mM NaCl, 0.1 mM CaCl2, 2 mM MgSO4, 1× BME vitamins (catalog no. B6891; Sigma), 18 mM 15NH4Cl, and 22 mM glucose) or 15N,13C-minimal growth medium supplied with 22 mM [13C]glucose (). The GST tag of the labeled VP8* proteins was removed for the NMR study by cutting with thrombin (Sigma-Aldrich Co., St. Louis, MO), followed by fast protein liquid chromatography (FPLC) gel filtration to collect the VP8* protein fractions and concentration by ultracentrifugation. The purified VP8* proteins were stored at −80°C until needed.

STD NMR experiments.

All saturation transfer difference (STD) NMR spectra were acquired in Shigemi Tubes (Shigemi, USA) with a Bruker 600-MHz Avance III spectrometer at 10°C using a conventional HCN probe and the TopSpin 3.1 pulse sequence STDDIFFGG19.3. The NMR sample for mucin core 2 in complex with P[19] VP8* was prepared with 200 μM protein and 6.0 mM ligand in a final volume of 300 μl of Tris NMR buffer. The LNFPI STD sample was prepared with 130 μM P[19] VP8* and 6.6 mM LNFPI in a final volume of 300 μl of phosphate NMR buffer. This results in total protein-ligand ratios of 1:35 and 1:50 for the mucin core 2 and LNFPI samples, respectively. The protein resonances were saturated at 0 ppm with a cascade of 40 Gaussian-shaped pulses with duration of 50 ms, resulting in a total saturation time of 2 s. The off-resonance saturation was applied to 30 ppm, and 1,024 experiments were acquired for each with a 5-s recycle delay. A spin-lock filter with strength of 10 kHz and duration of 100 to 200 ms was applied to suppress the broad protein resonance signals, and Watergate 3-9-19 was used to suppress the residual water signal. The STD spectra were obtained by subtracting the on-resonance spectrum from the off-resonance reference spectrum. STD amplification factors were determined by the equation (Io − Isat)/Io, where Io and Isat are the intensities of the signals in the reference and saturated spectra, respectively, and the normalized STD amplification percentage was determined by dividing by the largest STD amplification factor for each ligand. The protons with the largest STD amplification percentage along with data from the 2D STD TOCSY spectrum, which were used to resolve overlap from the 1D spectra, were used to assign the binding epitopes.

P[19] VP8* recognizes mucin cores and the type 1 HBGA precursor.

The ligand binding specificity of the P[19] VP8* was further characterized by STD NMR. Data for a type 1 HBGA pentasaccharide, LNFPI (Fucα1-2Galβ1-3GlcNAcβ1-3Galβ1-4Glc), demonstrated that the extended type 1 chain of an internal Gal (Fucα1-2Galβ1-3GlcNAcβ1-3Gal) is also involved in the interaction with P[19] VP8*, in addition to the GlcNAc and the terminal (H) fucose ( and ). Similarly, STD NMR of P[19] VP8* with mucin core 2 showed a strong interaction with the disaccharide GlcNAcβ1-6GalNAc, in which the GlcNAc residue was highly involved ( and ).