Ly6C+ monocytes in the skin promote systemic alphavirus dissemination

Abstract

Alphaviruses are mosquito-transmitted pathogens that induce high levels of viremia, which facilitates dissemination and vector transmission. One prevailing paradigm is that, after skin inoculation, alphavirus-infected resident dendritic cells migrate to the draining lymph node (DLN), facilitating further rounds of infection and dissemination. Here, we assess the contribution of infiltrating myeloid cells to alphavirus spread. We observe two phases of virus transport to the DLN, one that occurs starting at 1 h post infection and precedes viral replication, and a second that requires replication in the skin, enabling transit to the bloodstream. Depletion of Ly6C⁺ monocytes reduces local chikungunya (CHIKV) or Ross River virus (RRV) infection in the skin, diminishes the second phase of virus transport to the DLN, and delays spread to distal sites. Our data suggest that infiltrating monocytes facilitate alphavirus infection at the initial infection site, which promotes more rapid spread into circulation.

Keywords: CP: Immunology; CP: Microbiology; alphavirus; dissemination; immunity; lymph node; monocytes; mouse models; pathogenesis; viremia.

Figure 1.. Depletion of myeloid cells reduces CHIKV dissemination

(A and B) Cohorts of 4-week-old male C57BL/6J mice were inoculated with 10³ FFU of CHIKV (strain La Reunion [LR]-2006) in the left rear footpad. Animals were sacrificed at either 1, 4, 8, or 18 hpi, and viral RNA was quantified by qRT-PCR from serum and spleen (n = 6, two experiments). (C–J) Mice were administered anti-GR-1 or isotype control antibody via intraperitoneal injection 24 h prior to inoculation with CHIKV-LR (C–F) or RRV (strain T48) (G–J) in the left rear footpad (ipsilateral foot). Serum and indicated tissues were harvested at 18 hpi, and viral RNA was quantified (n = 6–10, two or three experiments). Values in each panel are shown as FFU equivalents (eq) normalized to volume of serum or tissue weight. Bars indicate median values. Statistical analysis: (C–J) Mann-Whitney test; ns, not significant, *p < 0.05, **p < 0.01. See also Figures S1–S4.

Figure 2.. CHIKV replication in the skin and accumulation in the DLN is impaired by anti-GR-1 antibody

(A and B) Cohorts of 4-week-old male C57BL/6J mice were inoculated with CHIKV-LR in the left rear footpad and animals were sacrificed at 1, 4, or 8 hpi. Ipsilateral foot skin and DLNs were collected at each time point, and viral RNA was quantified by qRT-PCR (n = 6, two experiments). (C and D) Three cohorts of mice were inoculated with CHIKV-LR and sacrificed at 1, 4, and 8 hpi foot skin and DLN were harvested and the levels of subgenomic (E1 probe) and genomic (nsp1 probe) RNA were measured and a (E1:nsp1) ratio was generated (n = 6, two experiments). (E) Mice pre-treated (day −1) with anti-GR-1 or isotype control antibody were inoculated with CHIKV and sacrificed at 18 hpi, and E1:nsP1 RNA ratios were generated for ipsilateral foot skin and DLN (n = 7, two experiments). (F and G) Mice were treated with anti-GR1 or isotype control mAb and inoculated with CHIKV as described in (E). At 1, 4, 8, or 18 hpi, ipsilateral foot skin and DLN were harvested, and viral RNA levels (E1 probe) were determined (n = 6, two experiments). Bars indicate median values. Statistical analysis: (C–E) Mann-Whitney test (comparisons made between anti-GR1 and isotype controls), ns, not significant, *p < 0.05, **p < 0.01. See also Figure S1.

Figure 3.. Virus accumulation in the medullary sinus and fluid phase of the DLN is reduced after anti-GR-1 treatment

(Four-week-old male C57BL/6J mice pre-treated (day −1) with anti-GR-1 or isotype control antibody were inoculated in the footpad with 10³ FFU of CHIKV-LR. (A) DLNs were harvested, and lymphatic fluid was collected. After clarification by centrifugation, a plaque assay was used to quantify infectious virus (n = 6–7, two experiments). (B and C) DLNs were harvested, and tissue homogenates were assessed for infectious virus by plaque assay (B) and viral RNA (C) (n = 6–7, two experiments). (D) Popliteal DLNs were harvested at 18 hpi, fixed, incubated in 30% sucrose for 48 h, cryo-sectioned, and then stained with the following antibodies: anti-CHIKV E1 (CHK-166 N297Q, white), anti-ERTR7 (red), and anti-CD169 (green). Sections were imaged using confocal microscopy and representative DLNs are shown (n = 6, two experiments). Scale bars, 200 μm. Boxed insets denote areas of CHIKV-positive cells. Images are representative of two experiments (n = 6). (E and F) Microscopic images of entire DLN were quantitated using automated image processing software to determine the total volume with detectable CHIKV fluorescence (after anti-CHIKV E1 staining) and the total fluorescent signal per section (n = 6–7, two experiments). Bars indicate median values. Statistical analysis: (A–C, E, and F) Mann-Whitney test, *p < 0.05, **p < 0.01. See also Figure S1.

Figure 4.. Anti-GR-1 mAb administration does not affect trafficking to the DLN of non-replicating alphaviruses

(A) Four-week-old male C57BL/6J mice were inoculated in the footpad with 10⁵ FFU of AURV and harvested at 1 or 18 hpi. Ipsilateral foot skin was collected and assessed for levels of viral RNA. (B and C) Mice were inoculated in the footpad with 10⁵ FFU of AURV and harvested at 1, 4, 8, or 18 hpi. Ipsilateral foot skin (B) and DLN (C) were collected at each time point, and the levels of subgenomic (E1 probe) (shown alone in A) and genomic (nsp3 probe) RNA were measured, and a ratio (E1:nsp3) was generated (n = 6, two experiments). (D–G) Mice pre-treated (day −1) with anti-GR-1 or isotype control antibody were inoculated with 10⁵ FFU of AURV (D and E) or 10⁵ FFU of EILV-CHIKV (CHIKV strain 99659) (F and G), and tissues were harvested at 18 hpi. The levels of viral RNA (E1 probe) were measured and normalized per weight of tissue or per lymph node (D and E: n = 7 or 9, two experiments. F and G: n = 11, three experiments). (H and I) Anti-GR-1 or isotype control-treated mice were inoculated with 10³ FFU of CHIKV-LR (n = 6, two experiments). At 8 hpi, 40 ng of 2,000-kDa FITC-dextran was injected into the ipsilateral footpad. Four hours later, DLNs were harvested and imaged using confocal microscopy. Scale bars, 200 μm. (H) The amount of FITC-dextran present was also determined after tissue homogenization by measuring the fluorescence signal (I) (n = 6, two experiments). Bars indicate median values. Statistical analysis: (D–G, and I) Mann-Whitney test; ns, not significant. See also Figure S1.

Figure 5.. Ly6C/Ly6G cell depletion affects CHIKV replication in the CD45⁻non-hematopoietic cells in the skin

(A–C) Four-week-old male C57BL/6J mice were administered anti-GR-1 or isotype control antibody (day −1) and then inoculated in the footpad with 10⁵ FFU of CHIKV AF15561-Venus reporter virus. Ipsilateral foot skin was harvested at 18 hpi and digested enzymatically to release stromal cells. Cell suspensions were incubated with anti-CD45 and anti-CHIKV E1, and flow cytometry was performed to assess the percentage (A) and number (B and C) of CD45⁻ cells expressing CHIKV antigen (Ag, A and B) or Venus (YFP, A and C) (n = 6, two experiments). Gating strategies are shown in Figure S4. (D–H) Mice were administered anti-GR-1 or isotype control antibody and then inoculated with CHIKV-AF15561-Venus reporter virus. Ipsilateral foot skin was harvested at 18 hpi and digested enzymatically. Cell suspensions were incubated with anti-CD45, anti-CD31, anti-E-cadherin, anti-EpCAM, anti-Sca-1, and anti-CD29 antibodies, and flow cytometry was performed to assess the percentage (D) and number of YFP⁺ endothelial cells (E), epithelial cells (F), fibroblasts (G), and keratinocytes (H). (n = 9 or 10, two experiments) Naive controls were used to set staining gates for all samples. Bars indicate median values. Statistical analysis: (B, C, and E–H) Mann-Whitney test, *p < 0.05, **p < 0.01. See also Figures S1 and S5.

Figure 6.. Monocytes in the skin support CHIKV infection

(A–D) Cohorts of 4-week-old male C57BL/6J mice were inoculated with 10³ FFU of CHIKV-LR and sacrificed at 0, 8, or 18 hpi. Ipsilateral foot skin was harvested and processed. Cell suspensions were incubated with antibodies against CHIKV E1 (Ag), CD45, B220, CD3, NK1.1, Ly6G, Ly6C, CD115, and CD11b, and flow cytometry was performed to assess the percentages (A and C) and numbers (B and D) of Ly6C^hi (monocytes, top panels) and Ly6G⁺ (neutrophils, bottom panels) of cells expressing CHIKV antigen (n = 6, two experiments). (E and F) Four-week-old C57BL/6J mice were inoculated in the footpad with 10⁵ FFU of CHIKV AF15561-Venus reporter virus. At 0, 18, or 36 hpi, ipsilateral foot skin and DLN were harvested, and cell staining and flow cytometry were performed. The percentages (E) and numbers (F) of Ly6C^hi monocytes expressing YFP are shown (n = 6, two experiments). Cells from naive mice were used to set staining gates. Gating strategies for this figure are shown in Figure S5. See also Figures S6 and S7. Bars indicate median values.

Figure 7.. Depletion with anti-Ly6C but not anti-Ly6G antibodies reduces infection in the skin and delays CHIKV dissemination

(A) Representative flow plots showing monocyte and neutrophil depletions. Left, isotype control; middle, anti-Ly6G depletion; right, anti-Ly6C depletion. Populations are gated as follows: monocytes, CD45⁺B220⁻CD3⁻NK1.1⁻CD11b⁺Ly6B^hi, side scatter^lo cells; neutrophils, CD45⁺B220⁻CD3⁻NK1.1⁻CD11b⁺, side scatter^hi cells. (B–M) Twenty-four hours prior to footpad inoculation with 10³ FFU of CHIKV-LR, cohorts of 4-week-old male C57BL/6J mice were administered anti-Ly6G (B–G), anti-Ly6C (H–M), or isotype control antibodies. At 18 hpi, ipsilateral foot skin (B and H), DLN (C and I), serum (D and J), spleen (E and K), ipsilateral feet (F and L), and contralateral feet (G and M) were harvested, and viral RNA was quantified by qRT-PCR (n = 6–9, two or three experiments). Bars indicate median values. Statistical analysis: Mann-Whitney test, ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001. See also Figure S1.