A tyrosine-to-histidine switch at position 18 of the Ross River virus E2 glycoprotein is a determinant of virus fitness in disparate hosts

Abstract

Arthritogenic alphaviruses are human pathogens maintained in nature through alternating replication in vertebrates and mosquitoes. Using chimeric viruses, we previously reported that replacement of the PE2 coding region of the T48 strain of Ross River virus (RRV-T48) with that from the attenuated DC5692 strain, which differ by 7 amino acids, resulted in an attenuated disease phenotype in a mouse model of RRV-induced rheumatic disease. Here, we demonstrate that introduction of one of these amino acid differences, a tyrosine (Y)-to-histidine (H) change at position 18 of the E2 glycoprotein (E2 Y18H), into the RRV-T48 genetic background was sufficient to generate a virus that caused dramatically less severe musculoskeletal disease in mice. The attenuated phenotype of RRV-T48 E2 Y18H was associated with reduced viral loads in musculoskeletal tissues, reduced viremia, and less efficient virus spread. Consistent with these findings, RRV-T48 E2 Y18H replicated less well in mammalian cells in vitro due to significantly reduced PFU released per infected cell. In contrast, RRV-T48 E2 Y18H replicated more efficiently than RRV-T48 in C6/36 mosquito cells. Competition studies confirmed that RRV-T48 E2 Y18H had a fitness advantage in mosquito cells and a fitness disadvantage in mammalian cells. Interestingly, all sequenced Ross River viruses encode either a tyrosine or a histidine at E2 position 18, and this holds true for other alphaviruses in the Semliki Forest antigenic complex. Taken together, these findings suggest that a tyrosine-to-histidine switch at E2 position 18 functions as a regulator of RRV fitness in vertebrate and invertebrate cells.

Fig 1

RRV E2 Y18H is an attenuating mutation in a mouse model. Three- to four-week old C57BL/6J mice were inoculated with PBS or 10³ PFU of RRV-T48, E3 R59G, E2 Y18H, E2 I67M, E2 H94R, E2 R251K, E2 H256Q, or E2 E302V in the left rear footpad (n = 3 per group). (A and B) At 24-h intervals, mice were assessed for weight gain (A) and scored for the development of disease signs, including loss of gripping ability, hind limb weakness, and altered gait (B). Each data point represents the arithmetic mean ± standard deviation. Data were evaluated for statistically significant differences by two-way ANOVA followed by Bonferroni's test. ***, P < 0.001. (C) At 10 dpi, mice were sacrificed and perfused with 4% paraformaldehyde. Five-micrometer-thick paraffin-embedded sections generated from the quadriceps muscle were stained with H&E. Images are representative of 3 mice per group. (D and E) Cumulative percent weight gain and disease scores for PBS-inoculated mice (n = 3) and mice inoculated with 10³ PFU RRV-T48 (n = 10), 10³ PFU RRV-T48 E2 Y18H (n = 14), or 10⁵ PFU E2 Y18H (n = 3). Data are combined from 3 to 4 independent experiments. Each data point represents the arithmetic mean ± standard deviation. Data were evaluated for statistically significant differences by two-way ANOVA followed by Bonferroni's test. **, P < 0.01; ***, P < 0.001.

Fig 2

The E2 Y18H mutation reduces RRV tissue titers and spread. Three- to four-week old C57BL/6J mice were inoculated with 10³ PFU of RRV-T48 or E2 Y18H mutant virus by injection in the left rear footpad. At 1, 3, 5, and 7 dpi, mice (n = 4 to 7) were sacrificed, blood was collected via cardiac puncture, and mice were perfused via intracardial injection with 1× PBS. Tissues were dissected, weighed, and homogenized, and the amounts of infectious virus present were quantified via plaque assays. Dashed lines indicate the limit of detection. Data were evaluated for statistically significant differences by two-way ANOVA followed by Bonferroni's test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig 3

The E2 Y18H mutation affects RRV replication in a cell-type-dependent manner. (A to F) C2C12 murine muscle cells (A and B), primary human synovial fibroblasts (C and D), or C6/36 mosquito cells (E and F) were inoculated with RRV-T48 or RRV-T48 E2 Y18H at an MOI of 0.01 (A, C, and E) or 5 (B, D, and F). At 0 hpi (input) and 1, 6, 12, 18, and 24 hpi, the amounts of infectious virus present in culture supernatants were quantified by plaque assays. Each data point represents the arithmetic mean ± standard deviation. Data were evaluated for statistically significant differences by two-way ANOVA followed by Bonferroni's test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. (G) C2C12 murine muscle cells were inoculated with RRV-T48 or RRV-T48 E2 Y18H at an MOI of 0.01 and incubated at 37°C or 30°C. At 0 hpi (input) and 1, 12, and 24 hpi, the amounts of infectious virus present in culture supernatants were quantified by plaque assays. Each data point represents the arithmetic mean ± standard deviation. Data were evaluated for statistically significant differences by two-way ANOVA followed by Bonferroni's test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Fig 4

The amino acid at RRV position 18 toggles RRV fitness in disparate host cells. (A to F) C6/36 cells were inoculated at an MOI of 0.01 with a 1:1 ratio (A and B), a 1:10 ratio (C and D), or a 10:1 ratio (E and F) of either T48 to E2 Y18HΔRsrII (A, C, and E) or T48ΔRsrII to E2 Y18H (B, D, and F). At 24 hpi, culture supernatants were collected, and titers were determined on BHK-21 cells. New C6/36 cells were then inoculated at an MOI of 0.01. This process was repeated a total of 5 to 10 times. At the passages shown, total RNA was isolated from culture supernatants, and a portion of the E2 gene was amplified by RT-PCR. PCR amplicons were digested with RsrII, and band intensities of undigested and digested products were quantified to determine the relative ratios of each virus genome. Data were evaluated for statistically significant differences by two-way ANOVA followed by Bonferroni's test. **, P < 0.01; ***, P < 0.001. (G and H) C2C12 cells were infected at an MOI of 1 with a 1:1, 1:10, or 10:1 ratio of T48 to E2 Y18HΔRsrII (G) or T48ΔRsrII to E2 Y18H (H). At 24 hpi, culture supernatants were collected, and titers were determined on BHK-21 cells. New C2C12 cells were then inoculated at an MOI of 1. This process was repeated a total of three times, and the relative amounts of each genotype present in culture supernatants were quantified as described above. Data were analyzed for statistically significant differences by a two-tailed t test.

Fig 5

The E2 Y18H mutation does not affect the pH of RRV entry. (A) Surface view of the alphavirus E1 (green) and E2 heterodimer showing the location and structural orientation of E2 domain A (red), E2 domain B (magenta), E2 domain C (maroon), E2 arches (tan), and E2 position 18 (blue). The E1-E2 heterodimer was modeled based on data reported under Protein Data Bank accession number 3N42 and was displayed by using PyMol (http://www.pymol.org/). (B) Expanded view of E2 position 18 (blue) and potential interacting residues in E1 (green). (C) RRV-T48 and RRV-T48 E2 Y18H were adsorbed onto C2C12 cells on ice at an MOI of 5. Following adsorption, cells were treated with fusion medium at the indicated pH for 1 min and then incubated in the presence of 20 mM NH₄Cl. At 18 h postfusion, the percentage of GFP-positive cells was quantified by flow cytometry. No statistically significant differences were detected.

Fig 6

The E2 Y18H mutation does not affect the kinetics or magnitude of RRV structural gene expression. C2C12 murine muscle cells or C6/36 mosquito cells were inoculated at an MOI of 5. (A) At 6, 9, 12, 18, and 24 hpi, capsid protein expression was analyzed by Western blotting. (B) Relative capsid band intensities (n = 3) were quantified by using ImageLab 4.0 software. No statistically significant differences were detected. (C and D) C2C12 cells were inoculated at an MOI of 5. At 6 and 18 hpi, cells were harvested and stained for E2 surface expression with an anti-E2 monoclonal antibody followed by a PE-conjugated secondary antibody. (C) Representative histogram displaying the anti-E2-specific fluorescence intensity of mock-, RRV-T48-, and RRV-T48 E2 Y18H-infected cells at 18 hpi. (D) The mean fluorescence intensity (MFI) of E2 staining at 6 and 18 hpi was determined by flow cytometry (n = 3). Each bar represents the arithmetic mean ± standard deviation. No significant differences were detected.

Fig 7

The E2 Y18H mutation impacts a late stage of the RRV replication cycle. C2C12 murine muscle cells or C6/36 mosquito cells were inoculated with GFP-expressing RRV-T48 or RRV-T48 E2 Y18H at an MOI of 5. (A and B) At 18 hpi, the percentage of GFP-positive cells was quantified by flow cytometry. Each data point represents an independent culture of cells. Data were analyzed for statistically significant differences by two-tailed t tests with Welch's correction. (C and D) At 18 hpi, the amounts of infectious virus present in culture supernatants were quantified by plaque assays. Each data point represents an independent culture of cells. Data were analyzed for statistically significant differences by two-tailed t tests with Welch's correction. (E and F) The PFU released per infected cell were calculated based on the total number of PFU present in culture supernatants and the percent GFP-positive cells. Each data point represents an independent culture of cells. Data were analyzed for statistically significant differences by two-tailed Student's t test with Welch's correction.

Fig 8

The E2 Y18H mutation reduces the infectivity of RRV in a cell-type-dependent manner. C2C12 murine muscle cells were inoculated with GFP-expressing RRV-T48 or RRV-T48 E2 Y18H at an MOI of 5. (A and B) RRV RNA copies (A) and PFU present in clarified C2C12 culture supernatants (B) were quantified by qRT-PCR and plaque assays, respectively. Each data point represents an individual cell culture. Data were analyzed for statistically significant differences by two-tailed t tests with Welch's correction. (C to H) RRV-T48 and RRV-T48 E2 Y18H were grown in C2C12 cells and BHK-21 cells. Clarified cell culture supernatants were analyzed directly (C to E), or virions were purified through a 60% to 20% discontinuous sucrose gradient prior to analysis (F to H). PFU (C and F) and RRV RNA copies (D and G) were quantified by plaque assays and qRT-PCR, respectively. The genome-to-PFU ratios were calculated based on the number of genomes/ml and the number of PFU/ml present in culture supernatants of gradient-purified virus stocks (E and H).