Evasion by stealth: inefficient immune activation underlies poor T cell response and severe disease in SARS-CoV-infected mice

Abstract

Severe Acute Respiratory Syndrome caused substantial morbidity and mortality during the 2002-2003 epidemic. Many of the features of the human disease are duplicated in BALB/c mice infected with a mouse-adapted version of the virus (MA15), which develop respiratory disease with high morbidity and mortality. Here, we show that severe disease is correlated with slow kinetics of virus clearance and delayed activation and transit of respiratory dendritic cells (rDC) to the draining lymph nodes (DLN) with a consequent deficient virus-specific T cell response. All of these defects are corrected when mice are treated with liposomes containing clodronate, which deplete alveolar macrophages (AM). Inhibitory AMs are believed to prevent the development of immune responses to environmental antigens and allergic responses by interacting with lung dendritic cells and T cells. The inhibitory effects of AM can also be nullified if mice or AMs are pretreated with poly I:C, which directly activate AMs and rDCs through toll-like receptors 3 (TLR3). Further, adoptive transfer of activated but not resting bone marrow-derived dendritic cells (BMDC) protect mice from lethal MA15 infection. These results may be relevant for SARS in humans, which is also characterized by prolonged virus persistence and delayed development of a SARS-CoV-specific immune response in individuals with severe disease.

Figure 1. Effect of CL treatment on weight loss, mortality, histological changes and virus titers in MA15-infected BALB/c mice.

(A) BALB/c mice (6–8 weeks old) were treated with 75 µl CL or PBS at before or after intranasal infection with 3×10⁴ PFU MA15 virus in 25 µl DMEM. Weight loss and mortality were monitored daily. n = 12 mice in PBS group; 20 mice in CL group. (B) For virus titers, lungs were homogenized and titeted on Vero E6 cells. Viral titers are expressed as PFU/g tissue. n = 4 mice/group/time point. *P values of <0.05. (C) BALB/c mice were treated with CL or PBS 18–24 h prior to infection with 3×10⁴ PFU MA15 virus. Lungs were removed at the indicated time points p.i.. Lungs were fixed in zinc formalin, and paraffin embedded. Sections were stained with hematoxylin and eosin.

Figure 2. Respiratory dendritic cell recruitment, migration and activation in MA15-infected mice after CL or PBS treatment.

Mice were treated with CL or PBS 18–24 h prior to infection with 3×10⁴ PFU MA15. Lungs were harvested at the indicated time points, and after enzyme digestion, single cell suspensions were acquired. Cells were stained for CD11c, MHC class II, CD11b, CD86 and CD40 expression. Total numbers of inflammatory cells and of CD11c⁺MHC II⁺ rDC in the lung are shown (A). CD86 and CD40 expression was measured on aDCs (CD11c⁺CD11b⁻MHC II⁺) and iDCs (CD11c⁺CD11b⁺MHC II⁺). An example of CD86 expression at day 6 p.i. (B) and a summary of MHC^highCD86⁺ or MHC^highCD40⁺ expression frequencies (C) are shown. Data are representative of two independent experiments and are the mean values±SEM (n = 7–8 mice/group/time point). (D) Mice were treated with CL or PBS 18–24 h before i.n. inoculation of 50 µl 8 mM CFSE. 6 h after CFSE instillation, mice were infected with 3×10⁴ PFU MA15 virus or were mock infected. At the indicated time points p.i., single cell suspensions were prepared from lung DLNs and gated for CD11c expression by flow cytometry. The values represent the percentage of CFSE⁺ cells within the CD11c⁺ DC population per LN. n = 4 mice/group/time point. *P values of <0.05.

Figure 3. MA15-specific T cell responses in the lungs after CL treatment.

Mice were treated with CL or PBS, 18–24 h prior to infection with 3×10⁴ PFU MA15 virus. At the indicated time points, single cell suspension were prepared from lungs, and stimulated with SARS-CoV CD8 (S366, S521 and S1061) or CD4 (N353) T cell peptides for 6 h in the presence of brefeldin A. Frequencies (A) and numbers (B) of total and MA15-specific T cells (determined by IFN-γ intracellular staining) are shown. Data are representative of two to four independent experiments n = 5–8 mice/group/time point. (C) In vivo cytotoxicity assays were performed on day 6 p.i.. Target cells were co-stained with PKH26 and different concentrations of CFSE (0.1 µM or 1 µM) and then incubated with SARS-CoV specific CD8 T cell peptides (0.1 µM CFSE) or in the absence of added peptides (1 µM CFSE) at 37°C for 1 h. 5×10⁵ target cells from each group were mixed together (1×10⁶ in total) and transferred i.n. to infected mice. 12 h after transfer, single cell suspensions were prepared from the lung and examined by flow cytometry. n = 3–4 mice/group. Data are representative of two independent experiments. *P values of <0.05.

Figure 4. Phenotype and numbers of AM in MA15-infected lungs after treatment with CL or PBS and AM-mediated inhibition of aDC activation and T cell proliferation in vitro.

Mice were treated with CL or PBS at day −1 prior to infection with 3×10⁴ PFU MA15. CD86, CD40, F4/80 and CD200R expression on CD11c⁺CD11b⁻Siglec F⁺ AM (A) and numbers and frequency (B) of AM were determined by flow cytometry. Black, isotype control; green, naive; blue day 2; red, day 4. yellow, D6. Data are representative of two independent experiments and are the mean values±SEM (n = 7–8 mice/group/time point). *P values of <0.05. (C) To assess the ability of AM to suppress aDC activation in vitro. AMs were harvested from BAL (bronchoalveolar) fluid and cultured at 4×10⁴/well in 96-well dishes for 48 h before use. aDCs were purified from naïve mice lungs by FACS sorting. aDCs were cultured in the presence or absence of AMs together at a 1∶1 ratio for 24 h at 37°C and subjected to flow cytometry. Data are representative of four independent experiments. (D) To assess the ability of AM to inhibit T cell proliferation in vitro, single cell suspensions were prepared from the lungs of naïve mice or CL-treated MA15-infected mice at day 8 p.i. Cells were incubated on plastic dishes for 2 h at 37°C, to remove AMs. Naïve lung cells (4×10⁵/96-well) were stained with 1 µM CFSE and stimulated with either 2.5 µg/ml Con A or 1 µg/ml soluble CD3 antibody with or without AMs (4×10⁴ / 96-well) for 72 h. Solid line, without AMs; gray, with AMs. (E) Total CD8 T cells were purified by microbeads from lung cells of AM-depleted MA15-infected mice at day 8, stained with 1 µM CFSE and stimulated with splenocytes from naïve mice or CD8 T cell-depleted infected lung cells (4×10⁵/96-well) that were pulsed with SARS-CoV CD8 T cell peptides with or without AM (4×10⁴ / 96-well) for 72 h. Cells were then subjected to flow cytometry. Solid line, without AMs; gray, with AMs. Data are repesentative of two independent experiments.

Figure 5. Protective effects of poly I:C treatment.

(A) Mice were treated with 20 µg poly I:C or 5 µg LPS 18–24 h before infection with MA15. Weight loss and mortality were monitored daily. n = 18 in LPS group; 14 mice in Poly I:C group. (B) Lungs were harvested and homogenized and virus was titered on Vero E6 cells. Viral titers are expressed as PFU/g tissue. (n = 4 mice/group/time point). (C) Single cell suspensions were prepared from lungs of naïve and treated mice. CD86, CD40. F4/80 and CD200R expression by CD11c⁺CD11b⁻Siglec F⁺ AMs after poly I:C, LPS or no treatment was determined by flow cytometry. The frequencies of MHC II⁺ CD86⁺ cell populations are shown. (D and E) Mice were treated with poly I:C or LPS 18–24 h prior to MA15 infection. At day 7 p.i., single cell suspensions were prepared from lungs, and stimulated with SARS-CoV CD8 (S366, S521 and S1061) or CD4 (N353) T cell peptides for 6 h in the presence of brefeldin A. Cells were analyzed for IFN-γ expression. Frequency (D) and numbers (E) of virus specific T cells are shown. Data are representative of two independent experiments and are the mean values±SEM (n = 5–8 mice/group/time point). (F) In vivo cytotoxicity assays were performed on day 6 p.i. Target cells were co-stained with PKH26 and different concentrations of CSFE, then pulsed with/without SARS-CoV specific CD8 T cell peptides, mixed together (1×10⁶ in total) and transferred i.n. to mice. 12 h after transfer, lung cells were examined by flow cytometry. n = 3–4 mice/group. Data are representative of two independent experiments.

Figure 6. Poly I:C treatment partially reverses AM inhibition of T cell proliferation in vitro.

(A) AMs were harvested from BAL fluid, and cultured at 2.5×10⁵/well in 24-well dishes for 48 h in the presence of 20 µg/ml poly I:C or 1 µg/ml LPS. Cells were detached and subjected to flow cytometry. (B) Single cell suspension were prepared from spleens of naïve mice, stained with 1 µM CFSE and stimulated with either 2.5 µg/ml Con A or 1 µg/ml soluble CD3 antibody for 72 h in the presence or absence of poly I:C-stimulated AMs from (A). Samples were then subjected to flow cytometry. Data are representative of two independent experiments.

Figure 7. Activation of BMDCs and protective effect of adoptive transfer of activated but not resting BMDCs.

(A) LPS (1 µg/ml) activated BMDCs were co-cultured with AMs harvested from BAL of naïve mice for 24 h. Phenotype changes were assessed by flow cytometry. AM co-culture did not inhibit costimulatory molecule expression by previously activated BMDCs. (B) BMDCs were stimulated with 20 µg/ml poly I:C or 1 µg/ml LPS or MA15 virus (m.o.i. = 5) and assayed for CD86 expression. Both poly I:C and LPS activated AM, as measured by CD86 expression. (C) 3×10⁵ activated or resting BMDCs were transferred by i.n. inoculation 18 h before MA15 infection (3×10⁴ PFU/mouse). Weight loss and mortality were monitored daily. n = 12 mice in resting BMDC group; 15 mice in activated BMDC group. (D) Lungs were homogenized and virus titered on Vero E6 cells. Viral titers are expressed as PFU/g tissue. (n = 4 mice/group).

Figure 8. Enhanced DC migration to DLN and MA15-specific T cell response after. transfer of activated but not resting BMDCs.

(A) Activated or resting BMDCs were stained with 1 µM CFSE, and adoptively transferred to mice. After 18 h, mice were infected with 3×10⁴ PFU MA15. Single cell suspensions were prepared from DLNs and CFSE⁺ cells were identified by flow cytometry. Total CFSE⁺ cells and LN cells numbers are shown in (B). Activation of BMDC enhanced migration to DLN and also increased total DLN cellularity. Data are representative of two independent experiments and are the mean values±SEM (n = 6–8 mice/group/time point). (C and D) Activated or resting BMDCs were transferred 18–24 h prior to infection with MA15. At day 7 p.i., single cell suspensions were prepared from lungs, and stimulated with SARS-CoV CD8 (S366, S521 and S1061) or CD4 (N353) T cell peptides for 6 h in the presence of brefeldin. Cells were analyzed for IFN-γ expression A. Frequency (C) and numbers (D) of MA15-specific T cells are shown. Data are representative of two independent experiments and are the mean values±SEM (n = 6–7 mice/group/time point). (E) In vivo cytotoxicity assays were performed on day 6 p.i. Target cells were co-stained with PKH26 and different concentrations of CSFE, pulsed with/without SARS-CoV specific CD8 T cell peptides, mixed together (1×10⁶ in total) and transferred i.n. to mice. 12 h after transfer, lung cells were examined by flow cytometry. n = 3–4 mice/group. Data are representative of two independent experiments.