Early responding dendritic cells direct the local NK response to control herpes simplex virus 1 infection within the cornea

Abstract

Dendritic cells (DCs) regulate both innate and adaptive immune responses. In this article, we exploit the unique avascularity of the cornea to examine a role for local or very early infiltrating DCs in regulating the migration of blood-derived innate immune cells toward HSV-1 lesions. A single systemic diphtheria toxin treatment 2 d before HSV-1 corneal infection transiently depleted CD11c(+) DCs from both the cornea and lymphoid organs of CD11c-DTR bone marrow chimeric mice for up to 24 h postinfection. Transient DC depletion significantly delayed HSV-1 clearance from the cornea through 6 d postinfection. No further compromise of viral clearance was observed when DCs were continuously depleted throughout the first week of infection. DC depletion did not influence extravasation of NK cells, inflammatory monocytes, or neutrophils into the peripheral cornea, but it did significantly reduce migration of NK cells and inflammatory monocytes, but not neutrophils, toward the HSV-1 lesion in the central cornea. Depletion of NK cells resulted in similar loss of viral control to transient DC ablation. Our findings demonstrate that resident corneal DCs and/or those that infiltrate the cornea during the first 24 h after HSV-1 infection contribute to the migration of NK cells and inflammatory monocytes into the central cornea, and are consistent with a role for NK cells and possibly inflammatory monocytes, but not polymorphonuclear neutrophils, in clearing HSV-1 from the infected cornea.

Figure 1. CD11c-DTR bone marrow chimeras require 6 weeks for full reconstitution of the cornea

WT BALB/C mice were lethally irradiated and reconstituted with 5×10⁵ bone marrow derived cells from CD11c-DTR donors. Whole corneas were excised 4, 6, and 9 weeks after bone marrow transfer, mounted, and analyzed via confocal microscopy observing pCD11c-EGFP⁺ cells. A) Serial images were taken starting at the corneal periphery, moving to the central cornea. Representative serial image sets of corneas of CD11c-DTR bone marrow chimera at 4, 6, and 9 weeks post-reconstitutionvisualizing pCD11c-EGFP⁺ cell recovery. B) Quantified imaging data of reconstitution; bars present mean ± SEM total number of EGFP-CD11c⁺ DCs per each of the two fields closest to the central corneal. (n=3-4 corneas/group, 2 independent experiments).

Figure 2. (Diphtheria toxin) DT treatment ablates DCs from the lymphoid organs and the cornea

CD11c-DTR chimeraswere treated ip with 175ng of DT, or mock treated with PBS. A) At 2 days post injection, whole corneas were harvested, mounted and analyzed via confocal microscopy for pCD11c-EGFP. Serial images of whole mounted corneas were taken starting at the cornea periphery, moving to the central cornea. Data arerepresentative serial image sets of DT or mock treated corneas visualizing pCD11c-EGFP+ ablation. B) At 2 days post injection, spleens and draining lymph nodes were harvested, dispersed, stained with anti-CD11c and CD11b antibodies, and analyzed via flow cytometry. Representative flow plots depicting DCs direct the NK cell response to HSV-1 infection CD11c⁺ cells within the cornea-draining lymph nodes and spleen following DT and PBS treatment.(n=4-6 mice/group, 2 independent experiments).

Figure 3. Dendritic cells are required for optimal early clearance of HSV-1 from the cornea

Wild type (WT) BALB/c mice andCD11c-DTR chimeras were infected with HSV-1-RE. CD11c-DTR chimers were treated with PBS or DT 2 days before infection (−2 dpi); or treated with DT at −2 dpi, +1 dpi, and + 4 dpi. A) At indicated time eyes were swabbed with a sterile surgical spear and assayed for live virus by standard plaque assay. The data are presented as the mean ± SEM viral plaque forming units (pfu) per cornea.B) CD11c-DTR chimeras were treated with PBS or DT 2 days before corneal infection with HSV-1 K26-RFP. At 2 dpi, whole corneas were excised, mounted, and analyzed via confocal microscopy. Representative images depicting virally infected cells (red), and pCD11c-EGFP+ DCs (green) within the viral lesion in the central cornea .C) Spleens were harvested at 4 dpi from WT mice or CD11c-DTR chimeras that received ip PBS or DT treatment 2 days before HSV-1 corneal infection; cells were dispersed, stained with antibodies to CD11c and analyzed via flow cytometry. DCs had partially recovered by 4 days after treatment (2 dpi) but remained significantly reduced relative to PBS treated controls (p<0.01). (** p<0.01, * p<0.05), n=4-12 mice per group, 2 independent experiments).

Figure 4. DC depletion does not alter extravasation of leukocytes into the peripheral cornea

Wild type (WT) and CD11c-DTR mice were infected with HSV-1 RE. CD11c-DTR chimeras were treated with PBS or DT 2 days before infection. At 3 dpicorneas were excised, and the central 2 mm dissected away from the remaining peripheral cornea with a trephine. Pools of 3 DCs direct the NK cell response to HSV-1 infection peripheral corneas were dispersed with collagenase, stained with antibodies to CD45,GL-3, CD49b, F4/80, and GR-1, and infiltrating leukocytes identified by flow cytometry (A)Representative histogram overlays depicting GL-3+ γδ T cells within the peripheral cornea. Histograms: Isotype control is grey shaded, CD11c-DTR PBS Tx is black solid, and −2 dpi DT Tx is black dotted. (B)Representative flow plots gated on CD45⁺ cells (numbers are for three corneas). (C)bars represent the mean ± SEM absolute number of neutrophils (PMN), inflammatory monocytes (IM), and NK cells per individual peripheral cornea (n=4-6 pooled corneas, 3 independent experiments).

Figure 5. NK cells in the infected cornea acquire an activation phenotype in the absence of DC

CD11c-DTR chimeras were treated with PBS or DT 2 days before corneal infection with HSV-1-RE (−2 dpi). At 3 dpi the central 2 mm of cornea was removed with a trephine and pools of 3 peripheral corneas were dispersed with collagenase, stained for the NK cell marker DX5, the activation markers CD69 and granzyme B (GrB), and analyzed by flow cytometry. A) flow plots depicting isotype controls for CD69 and Granzyme B expression on NK cells, and representative flow plots depicting peripheral NK cell CD69 and Granzyme B expression in PBS Tx; −2 dpi DT Tx; −2 and +1 DT Tx CD11c-DTR mice. B) Bars represent mean frequency ± SEM of CD69⁺, Granzyme B⁺, or CD69+GranzymeB⁺ NK cells within the corneal periphery. (n=6-8 pooled corneas, 2 independent experiments)

Figure 6. Viral replication is confined to the central cornea

CD11c-DTR mice were infected with HSV-1-K26 or HSV-1-RE. Mice were either treated with PBS or DT at −2 dpi. For mice infected with HSV-1 K26-RFP, at 2 dpi whole corneas were excised, mounted, and the entire cornea was imaged by confocal microscopy. A) Representative whole corneal images depicting virally infected cells (green) within naive, 2 dpi PBS Tx, or 2 dpi DT Tx corneas. B) In HSV-1 RE infected mice, whole corneas were excised at 3 dpiand the central and peripheral cornea were separated using a 2mm trephine. DNA was extracted from pools of 3 corneas and the amount of HSV-1 genome harbored within the central and peripheral corneas was determined by quantitative real time PCR for the gH gene. Bars represent mean fold increase ± SEM of gH copies over mock infected. (** p<0.01, n=4 mice per group, 2 independent experiments (panel A), (4 pooled corneas, 2 independent experiments, Panel B)

Figure 7. Fewer NK cells and inflammatory monocytes migrate into the central cornea following DC depletion

Wild type (WT) and CD11c-DTR mice were infected with HSV-1 RE. CD11c-DTR chimeras were treated with PBS or DT 2 days before infection (−2 dpi). At 3 dpi corneas were excised, and the central 2 mm dissected away from the remaining peripheral cornea with a trephine. Pools of 3 central corneas were dispersed with collagenase and infiltrating leukocytes were identified by flow cytometry as described in Fig. 5. (A) Representative flow plots gated on CD45+ cells (numbers are for three corneas). (B)bars represent the mean ± SEM absolute number of neutrophils (PMN), inflammatory monocytes (IM), and NK cells per individual central cornea (*p<0.05, n=4-6 pooled corneas, 3 independent experiments)

Figure 8. Depletion of NK cells results in increased viral burden within the cornea

WT BALB/C mice received ip injections of PBS or ASGM-1 antibody 1 day before corneal infection with HSV-1 RE. At indicated dpi, eyes were swabbed with a sterile surgical spear and assayed for live virus by standard plaque assay. Data are presented as the mean ± SEM viral plaque forming units (pfu) per cornea of PBS and anti-ASGM-1 treated mice. (* p<0.05, *** p<0.005,n=8-10 mice per group, 2 independent experiments)