The chemokine receptor CCR5 plays a key role in the early memory CD8+ T cell response to respiratory virus infections

Abstract

Innate recognition of invading pathogens in peripheral tissues results in the recruitment of circulating memory CD8(+) T cells to sites of localized inflammation during the early phase of a recall response. However, the mechanisms that control the rapid recruitment of these cells to peripheral sites are poorly understood, particularly in relation to influenza and parainfluenza infections of the respiratory tract. In this study, we demonstrate a crucial role for C-C chemokine receptor 5 (CCR5) in the accelerated recruitment of memory CD8(+) T cells to the lung airways during virus challenge. Most importantly, CCR5 deficiency resulted in decreased recruitment of memory T cells expressing key effector molecules and impaired control of virus replication during the initial stages of a secondary response. These data highlight the critical importance of early memory T cell recruitment for the efficacy of cellular immunity in the lung.

Figure 1. Memory CD8⁺ T cell recruitment to the lung airways coincides with increased expression of CCR5- and CXCR3-binding chemokines

C57BL/6 mice were infected i.n. with 250 EID₅₀ Sendai virus and allowed to rest for 45 days. (A) Lung airway cells were harvested from Sendai-immune mice prior to (Control) or on days 1-5 following challenge with 3 × 10⁴ EID₅₀ influenza x31 virus. The data show staining of CD11a and Sendai NP_324-332/K^b tetramer gated on CD8⁺CD44^hi cells. The data are representative of 5 independent experiments. (B) RNA was extracted from whole lung tissue of mice following influenza x31 challenge and the expression of chemokine genes was measured by real-time RT-PCR. The data are shown as the fold induction of the indicated genes on days 1-5 post-infection relative to un-challenged mice. The data are representative of 3 independent experiments. (C) BAL fluid was harvested from Sendai-immune mice prior to (Control) or on days 1-5 following influenza x31 challenge and chemokine protein levels were measured by luminex. Each symbol represents the value from an individual mouse and the data are representative of 2 independent experiments.

Figure 2. Virus-specific memory CD8⁺ T cells in lymphoid and peripheral tissues express CXCR3 and migrate to CXCR3 ligands in vitro

Naïve C57BL/6 mice were infected with 250 EID₅₀ Sendai virus or 1500 EID₅₀ influenza PR8 virus. Mice were sacrificed on days 8, 10, 15, 21, and 35 post-infection and the expression of CCR5 and CXCR3 on antigen-specific CD8⁺ T cells from the BAL, lung parenchyma, MLN, and spleen was measured by flow cytometry. Representative staining of CCR5 and CXCR3 gated on Sendai NP_324-332/K^b+ cells is shown at day 10 (A) and day 35 (B) post-infection. (C) The frequency of Sendai NP_324-332/K^b+ (open circles) or Flu NP_366-374/D^b+ (closed squares) CD8⁺ T cells expressing CXCR3 (left column) or CCR5 (right column) was measured at the indicated times post infection. The data are graphed as the mean ± s.d. of 5 mice for each time point and are representative of 3 independent experiments. (D) Splenocytes were harvested from C57BL/6 Sendai memory mice 45-60 days post-infection. The migration of naïve (open squares) or memory (filled squares) to migrate in response to increasing concentrations of CXCL9, CXCL10, and CXCL11 was determined by using a standard in vitro chemotaxis assay. (E) The ability of Sendai-specific CD8⁺ T cells to migrate to an optimal concentration of CXCL9 (5μg/ml) was measured as described above. The plots on the left show representative staining of the frequency of Sendai NP_324-332/K^b+ cells in the input or migrated population. The chemotactic index of naïve, total memory (CD8⁺ CD44^hi), of Sendai-specific T cells is graphed on the right. (F) The ability of Sendai-specific memory CD8⁺ T cell subsets to migrate to CXCL9 was examined based on the expression of CD62L and CD27. Representative staining of CXCR3, CD62L, and CD27 gated on Sendai NP_324-332/K^b+ cells are shown on the left. The chemotactic index of Sendai NP_324-332/K^b+ cells in response to 5μg/ml CXCL9 is shown for subsets based on the expression of CD62L (top graph) and CD27 (bottom graph). The data are representative of 5 independent experiments.

Figure 3. CCR5 expression on memory CD8⁺ T cells is required for accelerated recruitment to the lung airways following virus challenge

(A) Dual adoptive transfer Sendai memory mice were generated as shown and allowed to rest for 45 days after Sendai infection prior to influenza x31 challenge. (B) Prior to influenza x31 challenge, a cohort of mice were examined to assure that Sendai NP_324-332/K^b+ memory T cells from both donor populations and the host animal were detectable in the BAL, lung, MLN, and spleen (staining for BAL gated on CD8⁺ Sendai NP_324-332/K^b+ cells is shown). (C) The expression of CD11a was examined on CD8⁺ Sendai NP_324-332/K^b+ cells in the lung airways from donor and host populations prior to or on days 1-5 following influenza virus challenge. Representative staining from an individual CXCR3-deficient (upper panels) or CCR5-deficient (lower panels) dual transfer memory mouse is shown. (D) The average frequency (mean ± s.d.) of recently recruited (CD11a^hi) Sendai NP_324-332/K^b+ cells from each population is shown for CXCR3-deficient (left graph) and CCR5-deficient (right graph) dual transfer memory mice during the course of an influenza challenge. The data are representative of 3 independent experiments.

Figure 4. Circulating memory CD8+ T cells require CCR5 to migrate to the lung airways, but not lung parenchyma, during virus challenge

(A) Splenocytes from Sendai-immune CCR5-deficient or congenic (CD90.1⁺) wild-type were harvested 2 days post-influenza challenge and mixed together so that the frequency of Sendai NP_324-332/K^b+ cells were at a 1:1 ratio. Normalized splenocytes were injected i.v. into congenic mice (CD45.1⁺) that had also been infected with influenza virus 2 days previously. (B) Recipient mice were sacrificed 18-24 hours post-transfer, and BAL, lung parenchyma, PBL, and spleen were harvested and the ratio of Sendai NP_324-332/K^b+ cells from the input populations was assessed by flow cytometry in each tissue. The ratio of Sendai NP_324-332/K^b+ cells in the input population was determined by staining an aliquot of mixed cells prior to injection. (C) Splenocytes from Sendai-immune C57BL/6 or congenic (CD90.1⁺) wild-type mice were harvested as previously described, except that B6 splenocytes were incubated in 100 μg / ml Met-RANTES prior to mixing with CD90.1⁺ splenocytes for i.v. transfer. (D) The ratio of control (CD90.1⁺) or Met-RANTES-treated (B6) Sendai NP_324-332/K^b+ cells was calculated for each tissue as described above. The data are shown as the mean ± s.d. of 5 individual mice and are representative of 3 independent experiments.

Figure 5. CCR5 is rapidly expressed on the surface of memory CD8⁺ T cells following virus challenge

C57BL/6 mice were infected i.n. with 250 EID₅₀ Sendai virus and allowed to rest for 45 days. (A) Splenocytes were harvested from Sendai-immune mice prior to or on days 1-5 post-challenge with x31 influenza virus, and the expression of CCR5 on antigen-specific memory CD8⁺ T cells was analyzed by flow cytometry. Representative CCR5 staining gated on Sendai NP_324-332/K^b+ cells is shown for days 0-3 post-challenge. (B) Cumulative data from several experiments were graphed as the frequency of CCR5⁺ cells among total Sendai NP_324-332/K^b+ cells. Each symbol represents an individual mouse and the data are combined from 3 independent experiments. (C) Splenocytes were harvested from resting Sendai memory mice (filled squares) or Sendai memory mice 2 days post-challenge with x31 influenza (open squares). Purified CD8⁺ T cells from each population were tested for their ability to migrate in vitro to increasing concentrations of CCL3. The data are displayed as the chemotactic index (mean ± s.d.) of Sendai NP_324-332/K^b+ cells from each population and are representative of 3 independent experiments. (D) Sendai-specific memory CD8⁺ T cells were analyzed for surface and intracellular expression of CCR5. Representative surface, intracellular, and isotype staining is shown gated on CD8⁺ Sendai NP_324-332/K^b+ cells. (E) The frequency of surface or intracellular CCR5⁺ cells among Sendai NP_324-332/K^b+ splenocytes was graphed for several resting Sendai memory mice. The lines shown connect the surface and intracellular results for individual mice. (F) The frequency of resident (CD11a^lo) or recently-recruited (CD11a^hi) Sendai NP_324-332/K^b+ cells in the airways positive for internal CCR5 expression in resting memory mice (left panel) or on day 3 post-challenge (right panel) was measured by flow cytometry. Each symbol represents an individual mouse. The data from parts D, E, and F are representative of 3 independent experiments.

Figure 6. CCR5-mediated recruitment is required for the expression of antiviral effector molecules by memory CD8⁺ T cells in the lung airways

(A) Mixed bone marrow chimera mice were generated from wild-type congenic and CCR5-deficient mice as shown. Following reconstitution, chimeras were infected with 250 EID₅₀ Sendai virus and rested for 45 days. Sendai-immune chimeras were harvested at this time (control) or following influenza virus challenge (day 3). Representative staining for CD11a and granzyme B by wild-type (CD45.1⁺) and CCR5-deficient Sendai NP_324-332/K^b+ cells in the airways is shown for (B) resting memory mice or (C) on day 3 following influenza virus challenge. (D) The frequency of CD11a^hi cells in the airways among wild-type and CCR5-deficient Sendai NP_324-332/K^b+ cells is graphed as the mean ± s.d. for resting control memory mice or on day 3 post-challenge. (E) The frequency of granzyme B⁺ cells among Sendai NP_324-332/K^b+ wild-type or CCR5-deficient cells in the airways of resting memory mice (BAL, control), or in the airways (BAL, day3), lung parenchyma (Lung), and spleen on day 3 post-challenge is graphed as the mean ± s.d. (F) The in vivo production of IFN-γ by Sendai-specific memory CD8⁺ T cells following virus challenge was measured by direct ICCS, and representative staining of IFN-γ and CD11a from Sendai NP_324-332/K^b+ wild-type (left panel) or CCR5-deficient (right panel) cells in the lung airways is shown. (G) The frequency of IFN-γ⁺ cells among Sendai NP_324-332/K^b+ wild-type or CCR5-deficient cells in the lung airways from resting control memory mice or on day 3 post-challenge is graphed as the mean ± s.d. for 3 individual mice. All mixed bone chimera data are representative of 3 independent experiments.

Figure 7. Defective memory CD8⁺ T cell recruitment to the lung airways in CCR5-deficient mice correlates with increased virus titers during secondary challenge

(A) C57BL/6 and CCR5-deficient mice were primed with 1500 EID₅₀ x31 influenza virus and allowed to rest for 45 days. Naïve and x31-immune age-matched mice were challenged with 6000 EID₅₀ PR8 and whole lung tissue was harvested on the indicated days post-challenge. (B) Lung virus titers were measured by plaque assay on days 2 (upper graph) and 4 (lower graph) post-PR8 challenge. The data are graphed as the number of influenza PFU per lung (mean ± s.d.) for 5 individual mice per timepoint. The data are representative of 3 independent experiments.