Macrophage migration inhibitory factor promotes clearance of pneumococcal colonization

Abstract

Human genetic polymorphisms associated with decreased expression of macrophage migration inhibitory factor (MIF) have been linked to the risk of community-acquired pneumonia. Because Streptococcus pneumoniae is the leading cause of community-acquired pneumonia and nasal carriage is a precursor to invasive disease, we explored the role of MIF in the clearance of pneumococcal colonization in a mouse model. MIF-deficient mice (Mif(-/-)) showed prolonged colonization with both avirulent (23F) and virulent (6A) pneumococcal serotypes compared with wild-type animals. Pneumococcal carriage led to both local upregulation of MIF expression and systemic increase of the cytokine. Delayed clearance in the Mif(-/-) mice was correlated with reduced numbers of macrophages in upper respiratory tract lavages as well as impaired upregulation of MCP-1/CCL2. We found that primary human monocyte-derived macrophages as well as THP-1 macrophages produced MIF upon pneumococcal infection in a pneumolysin-dependent manner. Pneumolysin-induced MIF production required its pore-forming activity and phosphorylation of p38-MAPK in macrophages, with sustained p38-MAPK phosphorylation abrogated in the setting of MIF deficiency. Challenge with pneumolysin-deficient bacteria demonstrated reduced MIF upregulation, decreased numbers of macrophages in the nasopharynx, and less effective clearance. Mif(-/-) mice also showed reduced Ab response to pneumococcal colonization and impaired ability to clear secondary carriage. Finally, local administration of MIF was able to restore bacterial clearance and macrophage accumulation in Mif(-/-) mice. Our work suggests that MIF is important for innate and adaptive immunity to pneumococcal colonization and could be a contributing factor in genetic differences in pneumococcal disease susceptibility.

Figure 1. MIF is important for the clearance of pneumococcal colonization

Wild type (WT) and MIF-deficient (Mif^-/-) mice were inoculated intranasally with 10⁷ CFU pneumococci. Lavages of the upper respiratory tract were performed thereafter to determine colonization density (CFU/ml). WT mice (black squares) display accelerated clearance of 23F pneumococci compared to Mif^-/- mice (open squares, A). The MIF-dependent defect in colonization is evident using both 23F and 6A strains of pneumococci at 7 days (B). N≥5 mice per experiment, at least two experiments. The dashed line indicates the limit of detection. Error bars represent S.D. and horizontal lines indicate mean values. Y-axis units depicting CFU/ml are on a log base 10 scale. *P < 0.05, **P < 0.01, Mann-Whitney U test.

Figure 2. Pneumococcal colonization leads to local upregulation and systemic production of MIF in WT mice

Upper respiratory tract lavages were obtained 3 days after inoculation using RNA lysis buffer. RNA was isolated and reverse transcribed, and MIF expression level was measured by quantitative RT-PCR relative to GAPDH controls. Baseline MIF expression and MIF upregulation after colonization was noted in WT (A). Serum was obtained by cardiac puncture at 3 days post-colonization and analyzed for MIF by specific ELISA. Increased circulating MIF was noted in WT mice (B). Values are relative to mock-colonized WT mice ± SD (n≥10 mice per group). *P < 0.05, **P < 0.01, unpaired t test.

Figure 3. MIF is required for the macrophage influx and MCP-1 upregulation in response to pneumococcal colonization

Upper respiratory tract lavages were obtained 3 and 7 days after inoculation and the composition of the cellular infiltrate was determined by flow cytometry. Numbers of macrophages (F4/80+, CD11b–, A) and neutrophils (Ly6G+, CD11b+, B) are shown in WT (gray bars) and Mif^-/- mice (white bars). Each bar represents the average number of events ± SD in 4 experiments, each with 5 mice. MCP-1 upregulation in the WT mice was demonstrated by quantitative RT PCR of RNA from nasal lavages and found to be reduced in the Mif^-/- mice (C). RT PCR values are relative to mock-colonized WT mice ± SD (n≥10 mice per group). *P < 0.05, **P < 0.01 Mann-Whitney U test or unpaired t test.

Figure 4. Pneumolysin mediates the effect of MIF on pneumococcal colonization

Colonization experiments were performed in WT (close squares) and Mif^-/- (open squares) mice using 10⁷ CFU of strain 23F and strain 23F^ply-. Nasal lavage was obtained at 14 days. 23F^ply- (pneumolysin-deficient) pneumococci colonized both WT and T Mif^-/- mice to a greater degree than 23F. The MIF-dependent defect in pneumococcal clearance was eliminated in the absence of pneumolysin (A). Macrophage influx, quantified by flow cytometry, was greater in WT mice colonized with strain 23F compared to 23F^ply- (B). MIF upregulation, assessed using quantitative RT PCR of nasal lavage was present in strain 23F colonization and absent in and 23F ^ply- colonization (C). Flow cytometry, mean ± SD, n=5 mice per experiment, 4 experiments. RT-PCR, relative to mock-colonized WT mice ± SD, n≥10 mice per group. Y-axis units depicting CFU/ml are on a log base 10 scale. *P < 0.05, **P < 0.01, Mann-Whitney U test or unpaired t test.

Figure 5. MIF is produced from macrophages in a process that requires the pore-forming function of pneumolysin

Human macrophages differentiated from PBMCs were infected with 23F pneumococci at an MOI of 10:1 and their culture supernatant assayed for MIF production by specific ELISA (A). THP-1 macrophages, differentiated by PMA, were infected with the indicated strain of pneumococci. MIF production after infection was diminished in the 23F^ply- and 23F^plyW433F strains compared to 23F, and restored in the revertant 23F^ply-;→ply+ strain (B). MIF production was abrogated by inhibition of p38MAPK phosphorylation by treatment with SB203580 (a specific MAPKi, C). Phosphorylation of p38MAPK was observed by western blotting after infection of cultured bone marrow derived macrophages (BMDMs) from WT mice with 23F pneumococci, and quantified by densitometry. p38 MAPK phosphorylation was diminished in BMDMs of Mif^-/- mice (D). Mean ± SD values depicted from 4 independent experiments. Representative western blot shown and densitometry performed from 4 independent experiments. *P<0.05, **P < 0.01, ***P<0.001, one-way ANOVA.

Figure 6. MIF promotes the generation of adaptive responses to pneumococcal colonization

Mice were inoculated intranasally with 10⁷ CFU of 23F pneumocci and 21 days after inoculation, were sacrificed, serum was isolated, and levels of anti-pneumococcal serum IgG determined by ELISA. Values are expressed as geometric mean titers. WT mice (black squares) had circulating higher antibody titers to pneumococcus compared to Mif^-/- mice (open squares, A). Mice were allowed to clear primary colonization for 6 weeks, and then re-challenged with 10⁷ CFU 23F pneumococci. Nasal lavage was obtained at 5 days. Mif^-/- mice had higher levels of colonization compared to WT (B). N≥5 mice per experiment, at least two experiments. . Y-axis units depicting CFU/ml are on a log base 10 scale. **P < 0.01, Mann-Whitney U test.

Figure 7. Nasopharyngeal treatment with rMIF recovers the MIF-dependent defect in macrophage recruitment and pneumococcal clearance

Mif^-/- mice were colonized with 10⁷ 23F pneumococci, and treated every other day with either 100ng of rMIF in PBS or vehicle control. Nasal lavage was obtained at 14 days. MIF treatment let to greater clearance of pneumococcal colonization compared to control (A). Macrophage influx, quantified by flow cytometry, was also greater in the rMIF treated mice compared to vehicle treated controls (B). Horizontal lines indicate mean values, dashed line is the limit of detection. Flow cytometry, mean ± SD, n=5 mice per experiment, 3 experiments. Y-axis units depicting CFU/ml are on a log base 10 scale. *P < 0.05, **P < 0.01, Mann-Whitney U test.