Alveolar macrophages are early targets of mumps virus

Abstract

Formerly a common childhood pathogen, mumps virus (MuV) remains active worldwide, despite relatively high vaccine coverage. MuV is thought to infect the upper respiratory tract before disseminating to other organs; however, the early cellular targets of MuV in vivo are unknown. To address this, we generated a green fluorescent protein (GFP)-tagged vaccine strain (JL5) of MuV to infect leukocytic cell lines and found that replication was greatest in monocytes. Infection of peripheral blood mononuclear cells (PBMCs) also showed that both JL5 and a circulating strain of MuV (Iowa 2006; genotype G), preferentially infected monocytes. Further, monocyte-derived macrophages showed high susceptibility to MuV, with genotype G infecting macrophages to a much greater extent. While mice are generally resistant to MuV infection, we inoculated immunocompetent Rosa26-tdTomato mice intranasally with a GFP and Cre recombinase tagged MuV to determine whether monocytes/macrophages are important targets in vivo. We observed a small population of tdTomato⁺ cells within the lungs, which included epithelial cells; however, the vast majority were alveolar macrophages (AMs). To validate these findings, we infected murine AMs isolated from Rosa26-tdTomato mice with the GFP and Cre recombinase tagged MuV and found that while MuV could enter AMs, as determined by tdTomato positivity, only a small percentage of these expressed GFP, suggesting that inhibition in murine cells occurs postentry. To translate these findings, we infected cells from human bronchoalveolar lavage fluid with MuV and found that most infected cells were AMs. These findings highlight the high susceptibility of AMs and provide a basis for early MuV pathogenesis and subsequent dissemination.

Keywords: entry; lung; mouse; myeloid; pathogenesis.

Fig. 1.

Mumps virus readily infects monocytes in vitro and ex vivo. (A) Cell lines derived from monocytes (THP1 and U937), T cells (Jurkat and CEM), and B cells (Ramos and Daudi) were infected with MuV-JL5-GFP at an MOI of 0.1. GFP expression was measured over time for 72 h by flow cytometry. (B) Representative images showing GFP expression and syncytia formation (red arrows) are shown for each cell line at 48 hpi. (C) Supernatants from each cell line were collected at each time point to quantify viral load by plaque assay. (D) PBMCs were infected with MuV-JL5-GFP at an MOI of 1 and evaluated for GFP expression at 48 hpi by flow cytometry. (E) The percentage of GFP⁺ cells across six donors is shown. (F) Mock- or donor-matched MuV-infected PBMCs at 48 hpi were stained with surface markers to identify major cell subsets (CD14, monocytes; CD3, T cells; CD19, B cells; CD56, NK cells) for one representative donor. (G) These values were averaged for mock (n = 9), MuV-JL5-GFP-infected (n = 9), or MuV-G-GFP-infected (n = 3) cells and are represented as stacked bar plot. (H) Supernatants were collected and quantified by plaque assay to assess virus production by MuV-infected PBMCs. (Scale bars represent 300 µm.) Error bars represent the mean with SD (A and C) or SEM (E and H). A Wilcoxon matched-pairs signed rank test was utilized for statistical analysis in (E).

Fig. 2.

Macrophages are highly susceptible to MuV infection in vitro. (A) THP1 and U937 cells were differentiated into macrophages before infection with MuV-JL5-GFP (MOI of 0.1). Cells were imaged at 72 hpi for GFP expression and syncytia formation (red arrows) by microscopy. (B) Cells were further evaluated for GFP expression by flow cytometry. (C) Virus production in the supernatants was measured by plaque assay. (D) Cells were further evaluated for Annexin V binding to assess the percentage of apoptotic cells. (E) Primary monocytes isolated from PBMCs were differentiated into M1 macrophages and infected with MuV-JL5-GFP or MuV-G-GFP at an MOI of 1 (n = 3). Cells were imaged at 48 hpi to evaluate GFP expression and subsequently collected to quantify the percentage of infected cells by flow cytometry. (F) The percentage of GFP⁺ cells was quantified across 3 donors at 24 and 48 hpi. (G) The gMFI of GFP was quantified. (H) Virus production was evaluated at 24 and 48 hpi. (I) Supernatants collected at 48 hpi were evaluated for cytokine/chemokine production by Luminex. Shown is a heatmap with hierarchical clustering to visualize the Log₁₀ fold change (infected/uninfected) averaged across three donors for each analyte. The following analytes where all samples were below the level of detection were not included in the heatmap: CCL11, IL-12(p70), IL-15, IL-17F, IL-7, CXCL2. Error bars represent the mean with SD (B–D). (Scale bars represent 300 µm.)

Fig. 3.

Alveolar macrophages are early target of MuV infection in vivo. (A) A schematic of the MuV genome with GFP and Cre recombinase reporters, linked by a P2A site (MuV-JL5-GFP-Cre) along with expected outcomes following infection of Rosa26-tdTomato mice. (B) Rosa26-tdTomato mice were inoculated intranasally with MuV-JL5-GFP-Cre, and lungs were collected 2, 4, or 7 dpi and evaluated by flow cytometry to determine the tdTomato expression across time. (C) Representative flow plots are shown for mock and MuV-infected cells at 7 dpi where tdTomato expression was evaluated based on CD45 expression. (D) The percentage of tdTomato⁺ cells that were either CD45⁻ or CD45⁺ was assessed across all time points. (E) The proportion of GFP expressing was then calculated among the tdTomato⁺CD45⁺ immune cells across time. (F) Cells from the lungs of mock- or MuV-JL5-GFP-Cre-infected mice were stained with CD11b, CD11c, and CD170 (Siglec F) to immunophenotype the tdTomato⁺CD45⁺ cells. (G) Among the CD45⁺ cells, the percentage of alveolar macrophages found in the mock-infected lungs was compared to their proportion found among the tdTomato⁺ cells. (H) RNA in situ hybridization was performed using tdTomato-specific probes on mock-infected and MuV-JL5-GFP-Cre-infected mouse lungs collected 2 dpi to visualize infected cells. (Scale bars represent 100 µm.) Error bars represent the mean with SEM. A Kruskal–Wallis one-way ANOVA with Dunnett’s test (B) and a Mann–Whitney test (G) were performed for statistical analyses. ** indicates P ≤ 0.01

Fig. 4.

Mouse and human alveolar macrophage are highly susceptible targets for MuV infection. (A) BAL was performed on Rosa26-tdTomato mice to isolate alveolar macrophages, which were subsequently infected with MuV-JL5-GFP-Cre (MOI 1). Cells were imaged at 48 hpi to visualize tdTomato and GFP expression. (B) Cells were collected at each time point to quantify the kinetics of tdTomato and GFP expression with a stacked bar plot representing the total percentage of tdTomato⁺ and the proportion among these that were tdTomato⁺GFP⁺ double positive across time. (C) The gMFI calculated for both GFP and tdTomato over time by flow cytometry is shown. (D) Supernatants collected across time from mock- and MuV-infected murine AMs were evaluated for cytokine/chemokine production by multiplex ELISA. Log₁₀ fold change was quantified for each cytokine/chemokine and visualized utilizing a heatmap with hierarchical clustering. (E) Seven human BALF samples were obtained and infected with MuV-JL5-GFP at an MOI of 1. At 72 hpi, the cells were collected to assess AM purity using a combination of CD11b, CD15, HLA-DR, CD169, and CD206 as well as quantify the percentage of GFP⁺ cells. (F) A representative image of GFP expression at 48 hpi is shown. (G) The percentage of AMs and non-AMs among the GFP⁺ cells at 72 hpi was quantified for each BALF sample. (H) Supernatants collected at 3 dpi from MuV-infected cells were used to quantify virus production by plaque assay. The limit of detection is shown via a dotted line for (H). (Scale bars represent 300 µm.) Error bars represent the mean with SD. Samples with a media replacement after 24 h to remove dead cells are indicated by*.