Figure 1.
A primed innate cell population provides protection against a challenge with recombinant vaccinia virus.
Six months after intraperitoneal (ip) inoculation with vaccinia virus (primed) or PBS (naïve), IgHko mice were treated with T cell-depleting or isotype control antibody mixtures and 10 days later were challenged with 2×106 pfu rVV-luc intraperitoneally (ip). Isotype control or depleting monoclonal antibodies were administered ip every 2 weeks starting a week to ten days prior to viral challenge. (a) Flow cytometric plots of CD3 staining of peripheral blood cells from naïve control, naïve T-depleted, or primed T-depleted mice that fall within the lymphocyte gate are shown, including the percentage of total cells within the CD3+ gate at 1 week post-challenge. The TCRβ, TCRγδ, and CD8α staining profiles of the cells that fall within the CD3+ gate are also shown. (b) IVIS images of representative mice from the naïve control, naïve T cell-depleted, and primed T cell-depleted groups following the challenge with rVV-luc are shown. (c) The measurements of rVV-luc viral loads in each group over time are shown with relative viral loads based on the number of photons emitted normalized for a one minute exposure. The measurement for each individual mouse is plotted, and the center line indicates the mean with error bars representing the SEM. Statistical significance between T-depleted groups was assessed by t-tests using Mann-Whitney analysis and is indicated by the asterisk (p<0.05). Results are representative of 4 independent experiments.
Figure 2.
A Thy1+ cell population mediates innate protection against a secondary challenge with vaccinia virus.
Eight months after the initial administration of vaccinia virus (primed) or PBS (naïve), IgHko mice were treated with either isotype control or depleting monoclonal antibodies, then challenged ten days later with 2×106 pfu rVV-luc. Isotype control and depleting monoclonal antibodies were administered ip every 2 weeks beginning one week to ten days prior to viral challenge. (a) Representative flow cytometric plots of staining of all cells that fall within the CD3+ and CD3− lymphocyte gates from peripheral blood 2 weeks after the secondary challenge are shown. (b) Histograms of CD90.2 expression by the NK cell population (gated on CD3−NK1.1+ cells) isolated from peripheral blood of each of the indicated groups are shown. (c) IVIS images of representative IgHko mice from naïve isotype control, naïve and primed T cell-depleted, and naïve and primed T- and Thy1(CD90.2)+- cell depleted groups are shown at various time points following secondary challenge with rVV-luc. (d) The measurements of rVV-luc viral loads in each group over time are shown, with relative viral loads based on the number of photons emitted normalized for a one minute exposure. The measurement for each individual mouse is plotted, and the mean is indicated by the centerline with error bars representing the SEM. Statistical significance between T-depleted and T&Thy1-depleted groups was assessed by t-tests using Mann-Whitney analysis and is indicated by the asterisk (p<0.05). Results are representative of 4 independent experiments.
Figure 3.
Intracellular cytokine staining of previously primed NK cells demonstrates enhanced activation following in vitro stimulation.
Lymphocytes and NK cells isolated from collagenase-dissociated liver preparations from naïve and vaccinia virus-infected mice were cultured in vitro with the indicated stimuli for 6 hours prior to evaluation for NK cell activation (expression of CD69) and degranulation (expression of CD107) in response to the stimulation. Stimuli include plate-bound isotype control Ig, the anti-NK1.1 monoclonal antibody PK-136, formaldehyde-fixed adenovirus (2×106 vp), and formaldehyde-fixed vaccinia virus (2×106 pfu). Plots shown are gated on NK cells (NK1.1+CD3− or DX5+CD3−).
Figure 4.
Kinetic and phenotypic analysis of liver NK cell responses following primary vaccinia virus infection reveals the preferential expansion of a Thy1+ population.
Mice were administered either PBS (control) or 1×107 pfu of rVV (challenged) ip. At various time points following challenge, cells were isolated from the enzymatically-dissociated livers of control and challenged mice. (a) The absolute number of NK cells (CD3−NK1.1+), the absolute number of Thy1+ NK cells (CD3− NK1.1+ Thy1+), and the percentage of total NK cells that were Thy1+ are shown. Each data point represents an individual mouse, with the mean indicated by the horizontal bar and error bars representing the SEM. Absolute liver NK cell numbers were calculated by multiplying the total number of viable cells within the lymphocyte gate of each sample (as determined by analysis on a Guava easyCyte instrument) by the percentage of those cells determined to be CD3−NK1.1+ by flow cytometric analysis. The representation of Thy1+ cells within the total NK cell population, both as a percentage and an absolute number, was also determined based on flow cytometric analysis and the cell counts obtained by Guava analysis. Statistical analyses were performed using t tests and Mann-Whitney analysis comparing the various time points to that of the naïve control group. Statistically significant differences between virally-infected groups post-challenge and the uninfected control group are indicated by one (p<0.05) or two (p<0.01) asterisks. (b) Further phenotypic characterization of Thy1+ liver NK cells following vaccinia virus infection is shown in a series of histograms. The red traces represent cells from challenged mice and the blue traces represent cells from control mice isolated at the same time point and stained with the same antibody cocktail. (c) Quantitative analyses of the phenotypic profiles of Thy1(CD90)+ NK cells (cells within the CD3−NK1.1+CD90.2+ gate) stained with antibodies to the indicated molecules as determined by flow cytometric analysis. Statistical analyses were performed using t tests and Mann-Whitney analysis comparing the various time points to that of the naïve control group. Statistically significant differences between virally-infected groups post-challenge and the uninfected control group are indicated by one (p<0.05) or two (p<0.01) asterisks.
Figure 5.
Expansion of the NK cell population post-challenge is driven by proliferation in situ.
Groups of B6 mice were either infected with 1×107 pfu rVV ip or given a control injection of PBS. The thymidine analog 5-ethynyl-2′-deoxyuridine (EdU) was administered ip 12 hours prior to sacrifice (250 μg EdU/animal). Cell suspensions isolated from liver were stained with monoclonal antibodies to CD3, NK1.1, and Thy1 (CD90.2), fixed, permeabilized, and treated with the Click-IT EdU staining buffer with Alexa 647 azide to detect EdU incorporation. (a) The percentage of liver NK cells (CD3−NK1.1+) that incorporated EdU is plotted over time. Statistical analyses were performed using t tests and Mann-Whitney analysis comparing the various time points to that of the naïve control group. Statistically significant differences between virally-infected groups post-challenge and the uninfected control group are indicated by one (p<0.05) or two (p<0.01) asterisks. (b) The percentage of EdU+ liver NK cells expressing Thy1 (CD90.2) is plotted over time. Statistical analyses were performed using t tests and Mann-Whitney analysis comparing the various time points to that of the naïve control group. Statistically significant differences between virally-infected groups post-challenge and the uninfected control group are indicated by one (p<0.05) or two (p<0.01) asterisks. (c) Representative histograms measuring EdU incorporation by total liver NK cells is shown in the top panel, while representative histograms measuring Thy1 (CD90.2) expression by cells within the corresponding EdU+ gate are represented in the lower panel. These results are representative of two independent experimental replicates.
Figure 6.
Protection conferred on naïve RAG1ko mice by adoptive transfer of primed lymphocytes is abrogated by simultaneous administration of T cell-depleting monoclonal antibody cocktail.
Groups of RAG1ko mice received 2×106 NK-depleted hepatic lymphocytes from naïve or vaccinia virus-primed mice; a third group received both the primed lymphocytes and T cell-depleting monoclonal antibody cocktail simultaneously. All groups were challenged with 1×105 pfu rVV-luc ip 5 days later. (a) IVIS images of representative mice from the indicated recipient groups are shown at various time points following challenge with rVV-luc. (b) The measurements of rVV-luc viral loads in each group over time are shown, with relative viral loads based on the number of photons emitted, normalized for a 1 minute exposure. The measurement for each individual mouse is plotted, and the mean is indicated by the centerline with error bars representing the SEM. Statistical analyses were performed using t tests and Mann-Whitney analysis comparing the primed lymphocyte transferred & T-depleted group and the isotype-treated groups at various time points to that of the naïve control group. Statistically significant differences between the depleted group and the isotype-treated groups post-challenge are indicated by asterisks (p<0.05). Results are representative of 3 independent experiments.
Figure 7.
Innate immune protection against a vaccinia virus challenge is conferred on naïve RAG1ko mice by the adoptive transfer of primed Thy1+ liver-resident NK cells.
Six months after priming, 1×105 Thy1+ liver NK cells, 1×105 Thy1− liver NK cells, or 5×106 NK cell-depleted liver mononuclear cells from vaccinia virus-primed or naïve mice were adoptively transferred into RAG1ko mice. At the same time, mice were administered either control immunoglobulins (those mice that received NK-depleted cell transfers) or T cell-depleting monoclonal antibodies (those mice that received Thy1- and Thy1+ NK cell transfers). One week later the mice were challenged with 1×105 pfu rVV-luc intraperitoneally, and viral loads were monitored by IVIS imaging. (a) IVIS images of representative mice from each group over time following challenge are shown. (b) The ability of RAG1ko mice to resolve vaccinia virus infection by 3 weeks post-challenge is represented as the percent clearance within each group. Results were compiled from 4 separate experiments. RAG1ko mice that received primed Thy1+ liver NK cells showed enhanced clearance of vaccinia virus as compared to mice that received naïve Thy1+ NK cells, naïve Thy1-depleted NK cells, and primed Thy1-depleted NK cells (two-tailed Fischer test between groups). All statistics compare the test group relative to the group that received primed Thy1+ liver-resident NK cells. (c) The ability of RAG1ko mice to clear vaccinia virus infection throughout the course of challenge is represented as the percent clearance within each group. Results represent 4 separate experiments. RAG1ko mice that received primed Thy1+ liver NK cells showed enhanced protection against vaccinia virus challenge as compared to mice that received naïve Thy1+ NK cells, naïve Thy1− NK cells, and primed Thy1− NK cells. Statistical analysis was performed to compare Kaplan-Meier survival curves using GraphPad Prism 4 software. All statistics compare the test group relative to the group that received primed Thy1+ liver-resident NK cells. Protection provided by transferred primed Thy1+ hepatic NK cells surpassed the protection observed in mice that received hepatic naïve Thy1− NK cells (using the log rank test; p = 0.0168), primed Thy1− NK cells (p = 0.0043), and naïve Thy1+ NK (p = 0.0033). (d) Kaplan-Meier curves showing the survival of naïve RAG1ko recipients of the indicated adoptively transferred cell populations. Results represent the compilation of 4 separate experiments. RAG1ko mice that received primed Thy1+ liver NK cells showed enhanced survival of vaccinia virus challenge as compared to mice that received naïve Thy1+ NK cells, naïve Thy1− NK cells, or primed Thy1− NK cells. Statistical analysis was performed as in (c), demonstrating that host protection provided by transferred primed Thy1+ hepatic NK cells surpassed the protection observed in mice that received hepatic naïve Thy1− NK cells (log rank test; p = 0.013), primed Thy1− NK cells (p = 0.0049), or naïve Thy1+ NK (p = 0.0004). (e) Kaplan-Meier curves showing the survival of naïve RAG1ko recipients of the indicated adoptively transferred cell populations isolated from livers of naïve and MVA-primed mice. RAG1ko mice that received MVA-primed Thy1+ liver NK cells showed no evidence of enhanced survival following rVV-luc challenge as compared to mice that received naïve Thy1+ NK cells, naïve Thy1− NK cells, or MVA-primed Thy1− NK cells. Statistical analysis was performed as in (c and d). Results shown are representative of 2 independent experimental replicates.
Figure 8.
Protected RAG1ko hosts contain adoptively transferred Thy1+ NK cells, not contaminating T lymphocytes.
RAG1ko mice (CD45.2+CD90.2+) received transfers of 2×106 NK cell-depleted mononuclear cells, or 1×105 Thy1(CD90)+ NK cells from livers of naïve or vaccinia virus-primed B6.SJL mice (CD45.1+CD90.2+). Coincident with the transfers, mice received ip a mixture of isotype control (NK-depleted recipients) or T cell-depleting (purified NK cell recipients) monoclonal antibodies. All recipients were challenged with 1×105 pfu rVV-luc ip 5 days post-transfer. Shown are representative results from individual animals at 7 days (a) or 14 days (b) post-challenge. Mice were imaged via IVIS analysis through the course of the infection. At sacrifice, mononuclear cell suspensions were isolated from blood, spleens, and livers and analyzed via flow cytometry. Ovaries were harvested in 10 mM Tris buffer pH 9.0 and snap frozen in a dry ice–ethanol bath prior to processing and evaluation using a standard plaque formation assay on CV-1 cells. The recovered pfu/ovary are listed below each series of images (ND = none detected). Results are representative of two independent experiments.