A Single-Amino-Acid Substitution in the P2 Domain of VP1 of Murine Norovirus Is Sufficient for Escape from Antibody Neutralization

Cells, virus, and antibodies.MNV-1 CW.1 P3 (hereafter MNV-1) and anti-MNV-1 MAb A6.2.1 (immunoglobulin G2a [IgG2a]) were kindly provided by H. W. Virgin and C. E. Wobus. MAb A6.2.1 was described previously (40). Murine RAW 264.7 macrophages (ATCC TIB-71) were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 4.5 g/liter glucose, 10% fetal bovine serum (Atlanta Biologicals), 10 mM HEPES, 4 mM l-glutamine, 1.5 g/liter sodium bicarbonate, and 1% penicillin-streptomycin. MNV-1 was propagated in RAW 264.7 cells as previously described (40). MAb 61.21 to SMV was described previously (24) and was used as an IgG2a isotype-matched irrelevant antibody.

Phage-displayed oligopeptide library screens.Phage library screens were performed as described previously (24). Briefly, 10¹² PFU of library J404 (4) in 75 μl Tris-buffered saline (TBS) (50 mM Tris-Cl [pH 7.5], 150 mM NaCl) were combined with 100 μg MAb A6.2.1 in phage buffer consisting of TBS containing 1 mg/ml bovine serum albumin (BSA) and 0.5% (vol/vol) Tween 20. The phage-MAb mixture was then combined with 100 μl Gammabind Plus Sepharose beads (GE Healthcare) and incubated for 1 h at room temperature. The mixture was loaded onto a 10-ml Poly-Prep chromatography column (Bio-Rad) and washed with phage buffer. Antibody-phage complexes were eluted with 0.1 M glycine (pH 2.2). First-round eluate was amplified by infecting K91 bacterial cells in late log phase in soft agar. Phages were precipitated with 0.15 volumes of a solution containing 16% polyethylene glycol 8000 (PEG 8000) and 0.5 M NaCl. Second- and third-round selections were performed in the same way, and third-round titers were determined by plaque assay.

Phage characterization.Two assays were used to identify third-round phages that reacted with the MAb. First, plaque lifts were performed on the third-round population. Nitrocellulose membranes were probed with MAb A6.2.1 at a final concentration of 4 μg/ml in phosphate-buffered saline (PBS)-0.5% BSA. Positive plaques were detected with a 1:1,000 dilution of alkaline phosphatase-conjugated goat anti-mouse secondary antibody. Positive plaques were picked from the original plate and amplified.

In a second assay, third-round plaques from the plaque lifts were transferred into 2 ml of 2× yeast-tryptone medium supplemented with 75 μg/ml kanamycin and cultured for 12 h at 37°C. A 1.5-ml aliquot of each culture was centrifuged for 5 min in a microcentrifuge. Equal volumes of 16% PEG 8000 and 5 M NaCl were added to the supernatants, and precipitated phages were pelleted and then suspended in 50 μl of TBS. The phage concentrate was adhered to the wells of a 96-well Immulon-1B plate overnight at 4°C. The wells were blocked with PBS-5% BSA for 45 min at room temperature. MAb A6.2.1 was added to each well at a final concentration of 1.5 μg/ml, and plates were incubated for 2 h at room temperature. The wells were washed three times with PBS-0.05% Tween 20, and 50 μl of a 1:4,000 dilution of peroxidase-conjugated goat anti-mouse IgG was then added to each well. Plates were incubated for 45 min at room temperature. After washing three times with PBS-Tween 20, the reactions were developed by the addition of 100 μl/well citrate buffer with o-phenylenediamine dihydrochloride (OPD) and H₂O₂ (10 ml 50 mM citrate buffer, 10 mg OPD, and 5 μl 35% H₂O₂). The reaction was stopped with 50 μl of 2.5 M sulfuric acid, and the absorbance at 490 nm was measured on a Molecular Devices Versamax microplate reader. Phages that showed reactivity above an irrelevant phage control were selected for sequencing.

Phage DNA was purified using the QIAprep Spin M13 kit (QIAGEN), and the nonapeptide insertion region was sequenced with primer J534.3 (5′-GTTTTGTCGTCTTTCCAGACG-3′) using the ABI 310 genetic analyzer and Big Dye v3.1 chemistry.

Peptide enzyme-linked immunosorbent assays (ELISAs).The synthetic peptide used in these studies contained the phage-derived consensus sequence flanked by five (N terminus) and seven (C terminus) residues of M13 phage coat protein pIII (CFYSHSGWWEDHGQLGPPVKVLC). The peptide was constrained by the disulfide linkage of terminal cysteines. Disulfide bond formation was confirmed by mass spectrometry (Global Peptide Services). A constrained peptide derived from SMV (CFYSHSWLPAPIDKLGPPVKVLC) was described previously (24) and was used as an irrelevant control peptide.

ELISAs were performed with both constrained and linearized peptides. To convert the constrained peptide to a linear form, 100 μg of peptide was added to 80 μl sterile water, followed by the addition of the reducing agent Tris(2-carboxyethyl)-phosphine HCl to a final concentration of 30 mM. Reaction mixtures were incubated for 30 min at room temperature. Twofold dilutions of constrained or linearized peptide starting at 2 μg/μl in PBS were added to a MaxiSorp 96-well plate (Nunc) along with the indicated positive and negative controls. The plate was incubated overnight at 4°C, and the wells were then blocked with 5% BSA in PBS for 4 h at room temperature. The wells were probed with 50 μl of a 1-μg/ml concentration of A6.2.1. The plate was washed with PBS and then probed with peroxidase-conjugated goat anti-mouse secondary antibody (Jackson Scientific) for 45 min at room temperature. The plate was washed with PBS and developed with OPD substrate, and the absorbance was then measured at 490 nm.

Generation of neutralization escape mutants.Neutralization escape mutants were generated as previously described (30). A total of 1 × 10⁷ PFU of MNV-1 were incubated for 30 min with a 10-fold molar excess of MAb A6.2.1 (∼40 μg) in 1 ml of complete DMEM. This mixture was then used to infect RAW 264.7 cells. The medium was removed, and 1 ml of medium containing 1/10 the original amount of antibody was added. The cells and medium were harvested at 36 h postinfection. Following three freeze-thaw cycles, cellular debris was pelleted by centrifugation for 5 min at 5,000 × g. Virus in the supernatants was amplified by inoculating a 10-cm dish of RAW 264.7 cells with 500 μl of supernatant diluted in 5 ml of complete DMEM. The cells and medium were collected and subjected to three freeze-thaw cycles, and the supernatant was clarified by centrifugation. Second- and third-round selections were performed by combining 20 μg A6.2.1 antibody with 2.5 ml of the first-round virus for the second round and 30 μg antibody with 2 ml of the second round virus for the third round of selection. The infections and amplifications were carried out as described above. Neutralization assays were performed at each amplification step to monitor the resistant phenotype.

Plaque reduction neutralization assay.Plaque reduction neutralization assays were performed with wild-type MNV-1 and third-round escape mutants as previously described (40). Virus populations were diluted in DMEM to 700, 70, and 7 PFU/ml, and 20 μg A6.2.1 antibody was added to each dilution in a total volume of 800 μl. Similar dilutions without antibody were prepared as controls. Virus-MAb mixtures were adsorbed to RAW 264.7 cells for 30 min at 37°C. Plaques were counted 2 to 3 days postinfection.

Viral RNA extraction and sequencing.Viral RNA for reverse transcription-PCR was extracted as previously described (23). Plaques were picked with 9-in. borosilicate glass Pasteur pipettes, transferred into 500 μl complete DMEM, and used to inoculate RAW 264.7 cells. Cultures were harvested by scraping and collecting the cells and medium, performing three freeze-thaw cycles, and then pelleting the cell debris for 5 min at 5,000 × g. Five hundred microliters of supernatant was combined with an equal volume of 16% PEG 8000-0.8 M NaCl. Virus was precipitated for 30 min at 4°C and then centrifuged for 15 min at maximum speed in a microcentrifuge. The pellet was suspended in 300 μl proteinase K digestion buffer (0.1 M Tris-Cl [pH 7.5], 12.5 mM EDTA, 0.15 M NaCl, 1% [wt/vol] sodium dodecyl sulfate). Fifteen micrograms of proteinase K was added to each sample and incubated for 30 min at 37°C. Equal volumes of 10% cetyltrimethylammonium bromide and 4 M NaCl were added, and the samples were mixed by vortexing for 10 s and then incubated for another 30 min at 56°C. The samples were extracted with an equal volume of 1:1 phenol-chloroform (pH 4.3). RNA was precipitated in 2.5 volumes ethanol and 0.2 M sodium acetate and then suspended in 25 μl of RNase-free water.

Reverse transcription was performed with 60 ng of viral RNA template and 2 pmol of the negative-sense oligonucleotide mnv6753(−) (5′-GAACATTTGAGATGGAATTGCC-3′) that anneals to MNV-1 RNA at nucleotide 6753. The reaction mixture was heated for 5 min at 65°C and then cooled on ice. Four microliters of 5× first-strand buffer (Invitrogen), 2.0 μl of 0.1 M dithiothreitol, and 1 μl Superscript II reverse transcriptase (Invitrogen) were added to the template and incubated for 20 min at 50°C and then for 30 min at 42°C. The cDNA was amplified by PCR using 2 μl of the reverse transcription reaction mixture in standard Pfu Turbo DNA polymerase (Stratagene) PCR using the negative-sense primer mnv6753(−) and forward primer mnv5007(+) that anneals to MNV-1 RNA at nucleotide 5007 (5′-GGAACAATGGATGCTGAGACCC-3′). VP1 cDNA fragments were sequenced using an ABI 310 genetic analyzer and Big Dye v3.1 chemistry starting with primers mnv6753(−) and mnv5007(+). Additional primers were designed from this sequence to obtain the complete VP1 sequence. The internal sequencing primers designed were mnv5545(+), a positive-sense primer that anneals at nucleotide 5545 (5′-GTGCGTCGAGTCCTTTGGCATG-3′), and negative-sense primer mnv6201(−) that anneals at nucleotide 6201 (5′-GGCCGCAGCAGTGACGCTGGCG-3′). Sequences were aligned and analyzed using Seqman II DNA sequence analysis software (DNASTAR Inc.). Each mutation was verified by at least two independent sequence determinations.

A peptide selected by MAb A6.2.1 aligns with residues 327 to 335 in VP1.MAb A6.2.1 recognizes VP1 and neutralizes MNV-1 infection in RAW 264.7 cells (40). The sequences of 20 third-round phage clones selected by MAb A6.2.1 revealed a consensus peptide sequence, GWWEDHGQL (Fig. 1). Notable conserved residues in the selected sequences include a glutamic acid at position 4 that was present in 70% of the clones, an aspartic acid at position 5 in 80% of the clones, a histidine at position 6 in 60% of the clones, and a leucine at position 9, conserved in 90% of the clones. Four residues in the A6.2.1 peptide aligned with amino acids 327 to 335 in the hypervariable P2 domain of VP1: G327 and then, sequentially, G333, Q334, and L335. This sequence lies immediately adjacent to an epitope on SMV VP1 at positions 318 to 327 that is reactive with a MAb that inhibits SMV VLP-mediated hemagglutination. This epitope is part of a consensus sequence, AP(L/I)GXPD, conserved in all norovirus strains sequenced so far, including MNV-1 (24) (see Fig. 4). Sequence alignments suggest that G327 corresponds to the conserved G residue in the consensus peptide sequence.

• Open in new tab
• Download powerpoint

FIG. 1.

Nonapeptides selected from the J404 phage library by anti-MNV-1 MAb. Shown are peptide sequences obtained from 20 third-round phage clones and the consensus sequence derived from the clones. Shaded residues indicate the most conserved residues, and the derived consensus sequence is shown below the alignment.

MAb A6.2.1 recognizes a structurally constrained synthetic peptide.MAb A6.2.1 recognizes VP1 in an ELISA but not by immunoblot with denatured protein, suggesting the interaction between the MAb and VP1 is conformation dependent (40). To confirm the reactivity of the phage epitope with the MAb, a synthetic peptide containing the peptide flanked by M13 pIII residues was synthesized to yield the sequence CFYSHSGWWEDHGQLGPPVKVLC. The peptide was constrained by disulfide linkage through the N- and C-terminal cysteine residues. A6.2.1 reacted strongly with this peptide by ELISA in a dose-dependent manner (Fig. 2). Antibody reactivity was dependent upon the conformational presentation of the peptide because the reactivity was abolished when the constrained peptide was linearized by reducing the disulfide bond. The pIII residues in the peptide were not responsible for binding because an irrelevant constrained peptide of the same length with the same flanking residues (CFYSHSWLPAPIDKLGPPVKVLC) did not react with the MAb. These data all are consistent with the reactivity of MAb A6.2.1 with MNV-1 being conformation dependent and suggest that residues in VP1 surrounding the selected amino acid residues contribute significantly to the structural presentation of the epitope.

• Open in new tab
• Download powerpoint

FIG. 2.

MAb A6.2.1 reacts with constrained but not linear consensus peptide by ELISA. (A) One hundred micrograms of constrained (black bars) or Tris(2-carboxyethyl)-phosphine-reduced (vertical stippled bars) peptide containing the consensus phage epitope was adsorbed onto microtiter plates in serial twofold dilutions and then probed with MAb A6.2.1 as described in Materials and Methods. Gray bars are wells containing serial dilutions of MNV-1. (B) Thirty-three micrograms of constrained phage epitope probed with A6.2.1 (black bar) in the ELISA format described in Materials and Methods. Irrelevant isotype control MAb 61.21 (vertical stippled bar), 33 μg of a constrained irrelevant peptide probed with A6.2.1 (diagonal stippled bar), MNV-1 probed with A6.2.1 (spotted bar), and primary and secondary antibody only (horizontal stippled bar) are shown.

Neutralization escape mutants reveal a single-amino-acid change in the P2 domain of VP1.Neutralization escape mutants were generated to compare the epitope selected by phage display to changes in VP1 associated with resistance to the MAb. After three rounds of selection, infectivity was no longer inhibited (Fig. 3). The sequence of the cDNA amplified from the third-round virus population revealed two nucleotide changes relative to the sequence of the starting virus. One mutation was a G-to-U change at nucleotide 1145. This mutation translated to a change from L to F at position 386. Sequence was also obtained from 18 plaques from the third-round population that had been amplified in the presence of MAb. The L386F mutation was present in all of the clones sequenced, whether propagated in the presence of blocking concentrations of MAb A6.2.1 or in 1/10 of the inhibitory concentration, suggesting that this mutation is stable in culture.

• Open in new tab
• Download powerpoint

FIG. 3.

MNV-1 passaged three times in the presence of MAb A6.2.1 is resistant to neutralization. (A) Titers of wild-type MNV-1 in the absence (left) and presence (right) of 20 μg of A6.2.1. (B) Mutant MNV-1 (MNV-1m) titers after three rounds of amplification in the presence of 20 μg A6.2.1 MAb. Plaque assays shown are in the absence (left) and presence (right) of the MAb.

A second change of G to A at nucleotide 873 that translated to an E-to-K substitution at amino acid residue 296 was identified. The population sequenced contained a mixture of these two residues. A K at position 296 was reported in the sequence of the original MNV-1 isolate (22). Serial passage of MNV-1 in RAW 264.7 cells resulted in a mixture of K and E in the passaged population and then an E in the final passage (P3) (40). It is unlikely that this residue contributes to resistance to neutralization in this study because MAb A6.1.2 neutralizes both the original isolate (22) and the RAW 264.7-adapted CW.1 P3 clone (40).

Predicted location of the MNV-1-neutralizing epitope on VP1.MNV-1 and NV VP1 are 40.6% identical over the full-length of the protein (15), with the most significant variation being predictably present in the P2 domain (Fig. 4). NV is the only norovirus strain for which crystallographic data have been reported. Therefore, the sequence alignments shown in Fig. 4 were used to gain insight into a potential relationship between the peptide derived from phage display and the mutation in the neutralization-resistant viruses. Regions of the NV VP1 sequence that aligned with the relevant MNV-1 residues were identified and are highlighted on the VP1 structure (Fig. 4 and 5). The L386F mutation was in close proximity to W375-S378 of NV VP1 on the surface of the molecule. Residues G333, Q334, and L335 of the phage library-derived epitope aligned near G319-D327 of VP1 sequence. These two regions in the NV VP1 structure that corresponded to the escape mutant and phage peptide were in close proximity to each other, and a large portion is surface exposed.

• Open in new tab
• Download powerpoint

FIG. 4.

Sequence alignments of the P2 domain of VP1 of MNV-1 and select human norovirus strains. Amino acid sequence alignment incorporating the variable P2 domain of VP1 of human noroviruses NV, SMV, VA387, and MNV-1 is shown. Conserved residues are indicated with asterisks. Red lines below the alignment indicate residues in MNV-1 VP1 defined by the phage peptide isolated with MAb A6.2.1 (G327, G333, Q334, and L335). The red dot below the alignment marks the L386F mutation in the escape mutants. Blue lines above the alignment mark SMV and putative NV epitopes described previously (24). Black dots mark residues suggested to be important for carbohydrate binding of gII human norovirus strain VA387 (5).

• Open in new tab
• Download powerpoint

FIG. 5.

Putative location of MNV-1-neutralizing epitope. Shown is the VP1 monomer illustrating the location of amino acid residues of NV that align with the MNV-1 peptide selected from the phage library (red) and the escape mutant (yellow). The residues in yellow are NV amino acids W375, S377, and P378. The residues in red are D327, W328, and H329. The structure was drawn and analyzed with NCBI CN3D4.1.