Elevated temperature triggers human respiratory syncytial virus F protein six-helix bundle formation

Abstract

Human respiratory syncytial virus (RSV) is a major cause of severe lower respiratory tract infection in infants, immunocompromised patients, and the elderly. The RSV fusion (F) protein mediates fusion of the viral envelope with the target cell membrane during virus entry and is a primary target for antiviral drug and vaccine development. The F protein contains two heptad repeat regions, HR1 and HR2. Peptides corresponding to these regions form a six-helix bundle structure that is thought to play a critical role in membrane fusion. However, characterization of six-helix bundle formation in native RSV F protein has been hindered by the fact that a trigger for F protein conformational change has yet to be identified. Here we demonstrate that RSV F protein on the surface of infected cells undergoes a conformational change following exposure to elevated temperature, resulting in the formation of the six-helix bundle structure. We first generated and characterized six-helix bundle-specific antibodies raised against recombinant peptides modeling the RSV F protein six-helix bundle structure. We then used these antibodies as probes to monitor RSV F protein six-helix bundle formation in response to a diverse array of potential triggers of conformational changes. We found that exposure of 'membrane-anchored' RSV F protein to elevated temperature (45-55 degrees C) was sufficient to trigger six-helix bundle formation. Antibody binding to the six-helix bundle conformation was detected by both flow cytometry and cell-surface immunoprecipitation of the RSV F protein. None of the other treatments, including interaction with a number of potential receptors, resulted in significant binding by six-helix bundle-specific antibodies. We conclude that native, untriggered RSV F protein exists in a metastable state that can be converted in vitro to the more stable, fusogenic six-helix bundle conformation by an increase in thermal energy. These findings help to better define the mechanism of RSV F-mediated membrane fusion and have important implications for the identification of therapeutic strategies and vaccines targeting RSV F protein conformational changes.

FIG. 1. Construction of RSV Fusion (F) protein heptad repeat peptide immunogens

A) Heptad repeat (HR) domains and other structural features in the RSV Fusion protein. Key domains of the F1 subunit are shown, including the heptad repeats, HR1 (or N region) and HR2 (or C region), the fusion peptide (FP), and the transmembrane domain (TM). In the F2 subunit, the location of an additional predicted heptad repeat, HR3 is shown. Cleavage sites designate the location of furin cleavage sites in F2 and the F2-F1 junction, respectively. The location of conserved glycosylation sites are denoted by lollipop-shaped structures in F2 and F1. B) Codon optimization of HR1/HR2 cDNA in an RSV peptide expression construct. The nucleotide sequence of the codon-optimized HR1/HR2 cDNA is shown. The wild type RSV A2 strain F gene nucleotides that were replaced to permit optimal expression in E.coli are italicized. The HR1 region (nt 13-187, encoding F residues 153-209) and HR2 region (nt 232-366, encoding F residues 476-520) are joined by a linker region (underlined sequence). Also shown are the locations of restriction enzyme cloning sites and the IEGR-Factor Xa cleavage sequence. The individual HR1 and HR2 cDNA clones have sequences identical to the corresponding regions shown here. The cDNAs expressing RSV peptides NC, N, or C were cloned into the Bam HI-Hind III window of the pET-32a protein expression plasmid. C) Sequence of recombinant RSV peptides (N, C or single chain NC peptide) generated in the study. The lengths and predicted molecular weights of the peptides are shown.

FIG. 1. Construction of RSV Fusion (F) protein heptad repeat peptide immunogens

A) Heptad repeat (HR) domains and other structural features in the RSV Fusion protein. Key domains of the F1 subunit are shown, including the heptad repeats, HR1 (or N region) and HR2 (or C region), the fusion peptide (FP), and the transmembrane domain (TM). In the F2 subunit, the location of an additional predicted heptad repeat, HR3 is shown. Cleavage sites designate the location of furin cleavage sites in F2 and the F2-F1 junction, respectively. The location of conserved glycosylation sites are denoted by lollipop-shaped structures in F2 and F1. B) Codon optimization of HR1/HR2 cDNA in an RSV peptide expression construct. The nucleotide sequence of the codon-optimized HR1/HR2 cDNA is shown. The wild type RSV A2 strain F gene nucleotides that were replaced to permit optimal expression in E.coli are italicized. The HR1 region (nt 13-187, encoding F residues 153-209) and HR2 region (nt 232-366, encoding F residues 476-520) are joined by a linker region (underlined sequence). Also shown are the locations of restriction enzyme cloning sites and the IEGR-Factor Xa cleavage sequence. The individual HR1 and HR2 cDNA clones have sequences identical to the corresponding regions shown here. The cDNAs expressing RSV peptides NC, N, or C were cloned into the Bam HI-Hind III window of the pET-32a protein expression plasmid. C) Sequence of recombinant RSV peptides (N, C or single chain NC peptide) generated in the study. The lengths and predicted molecular weights of the peptides are shown.

FIG. 1. Construction of RSV Fusion (F) protein heptad repeat peptide immunogens

A) Heptad repeat (HR) domains and other structural features in the RSV Fusion protein. Key domains of the F1 subunit are shown, including the heptad repeats, HR1 (or N region) and HR2 (or C region), the fusion peptide (FP), and the transmembrane domain (TM). In the F2 subunit, the location of an additional predicted heptad repeat, HR3 is shown. Cleavage sites designate the location of furin cleavage sites in F2 and the F2-F1 junction, respectively. The location of conserved glycosylation sites are denoted by lollipop-shaped structures in F2 and F1. B) Codon optimization of HR1/HR2 cDNA in an RSV peptide expression construct. The nucleotide sequence of the codon-optimized HR1/HR2 cDNA is shown. The wild type RSV A2 strain F gene nucleotides that were replaced to permit optimal expression in E.coli are italicized. The HR1 region (nt 13-187, encoding F residues 153-209) and HR2 region (nt 232-366, encoding F residues 476-520) are joined by a linker region (underlined sequence). Also shown are the locations of restriction enzyme cloning sites and the IEGR-Factor Xa cleavage sequence. The individual HR1 and HR2 cDNA clones have sequences identical to the corresponding regions shown here. The cDNAs expressing RSV peptides NC, N, or C were cloned into the Bam HI-Hind III window of the pET-32a protein expression plasmid. C) Sequence of recombinant RSV peptides (N, C or single chain NC peptide) generated in the study. The lengths and predicted molecular weights of the peptides are shown.

FIG. 2. Characterization of the recombinant RSV HR peptides

A) Coomasie stained polyacrylamide gel showing the molecular weights of the histidine-tagged fusion peptide and purified RSV peptides (N, C, or NC) released by factor Xa cleavage of the corresponding histidine-tagged fusion proteins. A commercially synthesized HIV peptide, DP-178 (~4.5kd in size), was included for comparison. B) αRSV serum recognizes the peptides, N and NC but not the C peptide in an immunoblot. The arrow indicates the predicted migration for the C-peptide that was not recognized by αRSV serum. C) Circular Dichroism spectra of the NC single-chain peptide are consistent with a 6HB solution structure. The double minima at 222 and 208 nm are characteristic of the high level of helical secondary structure found in a 6HB.

FIG. 2. Characterization of the recombinant RSV HR peptides

A) Coomasie stained polyacrylamide gel showing the molecular weights of the histidine-tagged fusion peptide and purified RSV peptides (N, C, or NC) released by factor Xa cleavage of the corresponding histidine-tagged fusion proteins. A commercially synthesized HIV peptide, DP-178 (~4.5kd in size), was included for comparison. B) αRSV serum recognizes the peptides, N and NC but not the C peptide in an immunoblot. The arrow indicates the predicted migration for the C-peptide that was not recognized by αRSV serum. C) Circular Dichroism spectra of the NC single-chain peptide are consistent with a 6HB solution structure. The double minima at 222 and 208 nm are characteristic of the high level of helical secondary structure found in a 6HB.

FIG. 2. Characterization of the recombinant RSV HR peptides

A) Coomasie stained polyacrylamide gel showing the molecular weights of the histidine-tagged fusion peptide and purified RSV peptides (N, C, or NC) released by factor Xa cleavage of the corresponding histidine-tagged fusion proteins. A commercially synthesized HIV peptide, DP-178 (~4.5kd in size), was included for comparison. B) αRSV serum recognizes the peptides, N and NC but not the C peptide in an immunoblot. The arrow indicates the predicted migration for the C-peptide that was not recognized by αRSV serum. C) Circular Dichroism spectra of the NC single-chain peptide are consistent with a 6HB solution structure. The double minima at 222 and 208 nm are characteristic of the high level of helical secondary structure found in a 6HB.

FIG. 3. Assessment of immune response against the immunogen using a peptide ELISA

A) ELISA titers demonstrate the immunogenicity of peptide antigen NC or N+C. Antibody response of individual guinea pigs immunized with immunogen NC (guinea pig #82, #83, and #84) or N+C (guinea pig #85, #86, and #87) indicated that antisera #84(NC) and #87 (N+C) displayed the highest titers. B) Antisera αNC (#84) and αN+C (#87) are qualitatively identical in that they display the highest reactivity to peptide-complexes, NC or N+C, rather than to either of the free peptides N or C. The αNC antiserum displayed the lowest cross-reactivity to the individual N or C free peptides. Data are representative of two experiments. The dilution of αNC antibody used was 1:10,000. C) Peptide-adsorption ELISA demonstrates the specificity of αNC antiserum to RSV 6HB epitope. Adsorption of αNC antiserum with peptide complex NC or N+C abolishes the reactivity to NC (6HB) peptide. Adsorption with individual free peptide N or C does not significantly affect the reactivity to NC (6HB) peptide. The dilution of αNC antibody used was 1:10,000.

FIG. 3. Assessment of immune response against the immunogen using a peptide ELISA

A) ELISA titers demonstrate the immunogenicity of peptide antigen NC or N+C. Antibody response of individual guinea pigs immunized with immunogen NC (guinea pig #82, #83, and #84) or N+C (guinea pig #85, #86, and #87) indicated that antisera #84(NC) and #87 (N+C) displayed the highest titers. B) Antisera αNC (#84) and αN+C (#87) are qualitatively identical in that they display the highest reactivity to peptide-complexes, NC or N+C, rather than to either of the free peptides N or C. The αNC antiserum displayed the lowest cross-reactivity to the individual N or C free peptides. Data are representative of two experiments. The dilution of αNC antibody used was 1:10,000. C) Peptide-adsorption ELISA demonstrates the specificity of αNC antiserum to RSV 6HB epitope. Adsorption of αNC antiserum with peptide complex NC or N+C abolishes the reactivity to NC (6HB) peptide. Adsorption with individual free peptide N or C does not significantly affect the reactivity to NC (6HB) peptide. The dilution of αNC antibody used was 1:10,000.

FIG. 3. Assessment of immune response against the immunogen using a peptide ELISA

A) ELISA titers demonstrate the immunogenicity of peptide antigen NC or N+C. Antibody response of individual guinea pigs immunized with immunogen NC (guinea pig #82, #83, and #84) or N+C (guinea pig #85, #86, and #87) indicated that antisera #84(NC) and #87 (N+C) displayed the highest titers. B) Antisera αNC (#84) and αN+C (#87) are qualitatively identical in that they display the highest reactivity to peptide-complexes, NC or N+C, rather than to either of the free peptides N or C. The αNC antiserum displayed the lowest cross-reactivity to the individual N or C free peptides. Data are representative of two experiments. The dilution of αNC antibody used was 1:10,000. C) Peptide-adsorption ELISA demonstrates the specificity of αNC antiserum to RSV 6HB epitope. Adsorption of αNC antiserum with peptide complex NC or N+C abolishes the reactivity to NC (6HB) peptide. Adsorption with individual free peptide N or C does not significantly affect the reactivity to NC (6HB) peptide. The dilution of αNC antibody used was 1:10,000.

FIG. 4. αNC antiserum recognizes epitopes located in the F1 subunit

A) αNC antiserum recognizes RSV F protein in RSV-lysate western blot strips. The reactivity of αRSV serum is shown as a control. B) αNC antiserum selectively immunoprecipitates F1 subunit from RSV-infected cell lysate. Pre-immunization serum (PI) from guinea pig #84 is run as a negative control. RSV proteins immunoprecipitated by two RSV αF monoclonal antibodies (858-1-Sigma and 3216-Serotec) and polyclonal αRSV serum were run as controls.

FIG. 4. αNC antiserum recognizes epitopes located in the F1 subunit

A) αNC antiserum recognizes RSV F protein in RSV-lysate western blot strips. The reactivity of αRSV serum is shown as a control. B) αNC antiserum selectively immunoprecipitates F1 subunit from RSV-infected cell lysate. Pre-immunization serum (PI) from guinea pig #84 is run as a negative control. RSV proteins immunoprecipitated by two RSV αF monoclonal antibodies (858-1-Sigma and 3216-Serotec) and polyclonal αRSV serum were run as controls.

FIG. 5. Immunoprecipitation of F protein from the surface of RSV-infected cells using αNC antibodies

Cell-surface IP demonstrates the presence of 6HB epitopes on the surface of RSV-infected cells actively undergoing fusion. Cell-surface reactivity to these epitopes can be adsorbed out with NC or N+C peptide mixtures but not with individual N or C peptides.

FIG. 6. Detection of F protein 6HB epitopes on the surface of RSV-infected cells

Flow cytometry of αNC antibody-stained intact cells demonstrates: A) A relatively low amount of F protein is recognized by anti-6HB (αNC) antiserum on the surface of RSV-infected cells. B) By comparison, a large amount of total F protein is detected on RSV-infected cells using an α-F monoclonal antibody.

FIG. 7. Identification of temperature as a trigger of 6HB formational change in a flow cytometry-based triggering assay

A) Temperature course (5 min incubation) @ 37 °C, 40 °C, 42.5 °C, 45 °C, 47.5 °C, 50 °C, 52.5 °C, 55 °C, 57.5 °C, 60 °C. Temperature dependent triggering starts around 42.5 °C and plateaus between 47.5 °C and 55 °C, where a maximum signal to noise ratio of 2.4-2.5 times that observed at 37 °C (1.4) is achieved. B) Time course @ 55 °C for 5 min, 15 min, 30 min, 45 min, 60 min. 6HB formation occurs rapidly within the first 5 minutes of exposure to 55 °C. C) Loss of staining of α-NC antiserum upon adsorption with a mixture of N+C peptides but not with N or C peptide alone. Triggering was performed for 15 min at 55 °C.

FIG. 7. Identification of temperature as a trigger of 6HB formational change in a flow cytometry-based triggering assay

A) Temperature course (5 min incubation) @ 37 °C, 40 °C, 42.5 °C, 45 °C, 47.5 °C, 50 °C, 52.5 °C, 55 °C, 57.5 °C, 60 °C. Temperature dependent triggering starts around 42.5 °C and plateaus between 47.5 °C and 55 °C, where a maximum signal to noise ratio of 2.4-2.5 times that observed at 37 °C (1.4) is achieved. B) Time course @ 55 °C for 5 min, 15 min, 30 min, 45 min, 60 min. 6HB formation occurs rapidly within the first 5 minutes of exposure to 55 °C. C) Loss of staining of α-NC antiserum upon adsorption with a mixture of N+C peptides but not with N or C peptide alone. Triggering was performed for 15 min at 55 °C.

FIG. 7. Identification of temperature as a trigger of 6HB formational change in a flow cytometry-based triggering assay

A) Temperature course (5 min incubation) @ 37 °C, 40 °C, 42.5 °C, 45 °C, 47.5 °C, 50 °C, 52.5 °C, 55 °C, 57.5 °C, 60 °C. Temperature dependent triggering starts around 42.5 °C and plateaus between 47.5 °C and 55 °C, where a maximum signal to noise ratio of 2.4-2.5 times that observed at 37 °C (1.4) is achieved. B) Time course @ 55 °C for 5 min, 15 min, 30 min, 45 min, 60 min. 6HB formation occurs rapidly within the first 5 minutes of exposure to 55 °C. C) Loss of staining of α-NC antiserum upon adsorption with a mixture of N+C peptides but not with N or C peptide alone. Triggering was performed for 15 min at 55 °C.

FIG. 8. Cell-surface IP demonstrates temperature-induced triggering of six-helix bundle formation in cell-surface-expressed F protein

A) 6HB formation increases as a function of increasing temperature. A qualitative increase in F1 subunit immunoprecipitation was observed in the samples triggered at 55 or 50 °C for 15 min (lanes 6 & 7) as compared to those triggered at 37 °C (lane 5). The amount of F1 subunit immunoprecipitated following 45 °C triggering appeared to be somewhat lower than that observed at 50-55 °C, however, it still appeared to be higher than that observed at 37 °C or 40 °C (compare lane 8 with lanes 5 and 9, respectively). Lane 1 is a control showing F expression in RSV-infected cells immunoprecipitated with an anti-F monoclonal antibody (858-1). The pre-immunization serum (PI) did not immunoprecipitate any relevant proteins either in uninfected or RSV-infected cells (lanes 2 and 3). B) Specificity of the α-NC antiserum binding at elevated temperature for 6HB epitope. Increased amount of 6HB formed during elevated temperature can be adsorbed out with corresponding 6HB peptides. The α-NC antiserum was preadsorbed with 5 μg each of peptides NC, N, C, or N+C, respectively or PBS (-) overnight at 4 °C and used in triggering experiments at 55 °C for 15min as described above. The NC and N+C peptides appeared to almost completely adsorb out the reactivity of α-NC antiserum at 55 °C (compare lanes 3 and 6 with lane 2). The adsorption of α-NC antiserum with N peptide alone (lane 4) or C peptide alone (lane 5) did not appear to significantly impair their ability to immunoprecipitate F1 subunit.

FIG. 8. Cell-surface IP demonstrates temperature-induced triggering of six-helix bundle formation in cell-surface-expressed F protein

A) 6HB formation increases as a function of increasing temperature. A qualitative increase in F1 subunit immunoprecipitation was observed in the samples triggered at 55 or 50 °C for 15 min (lanes 6 & 7) as compared to those triggered at 37 °C (lane 5). The amount of F1 subunit immunoprecipitated following 45 °C triggering appeared to be somewhat lower than that observed at 50-55 °C, however, it still appeared to be higher than that observed at 37 °C or 40 °C (compare lane 8 with lanes 5 and 9, respectively). Lane 1 is a control showing F expression in RSV-infected cells immunoprecipitated with an anti-F monoclonal antibody (858-1). The pre-immunization serum (PI) did not immunoprecipitate any relevant proteins either in uninfected or RSV-infected cells (lanes 2 and 3). B) Specificity of the α-NC antiserum binding at elevated temperature for 6HB epitope. Increased amount of 6HB formed during elevated temperature can be adsorbed out with corresponding 6HB peptides. The α-NC antiserum was preadsorbed with 5 μg each of peptides NC, N, C, or N+C, respectively or PBS (-) overnight at 4 °C and used in triggering experiments at 55 °C for 15min as described above. The NC and N+C peptides appeared to almost completely adsorb out the reactivity of α-NC antiserum at 55 °C (compare lanes 3 and 6 with lane 2). The adsorption of α-NC antiserum with N peptide alone (lane 4) or C peptide alone (lane 5) did not appear to significantly impair their ability to immunoprecipitate F1 subunit.