Serum amyloid P aids complement-mediated immunity to Streptococcus pneumoniae

Abstract

The physiological functions of the acute phase protein serum amyloid P (SAP) component are not well defined, although they are likely to be important, as no natural state of SAP deficiency has been reported. We have investigated the role of SAP for innate immunity to the important human pathogen Streptococcus pneumoniae. Using flow cytometry assays, we show that SAP binds to S. pneumoniae, increases classical pathway-dependent deposition of complement on the bacteria, and improves the efficiency of phagocytosis. As a consequence, in mouse models of infection, mice genetically engineered to be SAP-deficient had an impaired early inflammatory response to S. pneumoniae pneumonia and were unable to control bacterial replication, leading to the rapid development of fatal infection. Complement deposition, phagocytosis, and control of S. pneumoniae pneumonia were all improved by complementation with human SAP. These results demonstrate a novel and physiologically significant role for SAP for complement-mediated immunity against an important bacterial pathogen, and provide further evidence for the importance of the classical complement pathway for innate immunity.

Figure 1. Binding of hSAP to ST2, ST4, and ST23F Strains of S. pneumoniae

(A) Results of whole-cell ELISAs presented as maximum ODs after incubation with different concentrations of hSAP (given below each column as μg/ml), using the E. coli O111:B4 strain as a negative control. Error bars represent SDs, and asterisks mark significant p-values for comparisons of results for S. pneumoniae in 20 μg/ml hSAP versus medium alone (2-tailed t tests, *p < 0.01, ***p < 0.0001). (B) Examples of flow cytometry histograms demonstrating hSAP binding to the surface of the three different S. pneumoniae strains, the E. coli O111:B4 strain, and the S. pyogenes H372 strain (positive control) after incubation in human serum. (C) Effect of different concentrations of EDTA on SAP binding to the ST2 S. pneumoniae strain in human serum. Error bars represent SDs, and asterisks mark significant p-values for comparisons of results for EDTA versus serum alone (2-tailed t tests, ***p < 0.0001). (D) Effect of addition of 100 mM PC on SAP and CRP binding to the ST2 S. pneumoniae strain in human serum. Grey columns, results for human serum; white columns, results for human serum in the presence of PC. Addition of 100 mM bovine serum albumin had no effect on SAP binding (unpublished data). Error bars represent SDs, and p-values are indicated above the columns. (E) Examples of flow cytometry histograms demonstrating inhibition of hSAP binding to the surface of the ST2 S. pneumoniae strains by addition of 100 mM PC to human serum.

Figure 2. Effects of SAP on C3b Deposition on S. pneumoniae Measured Using Flow Cytometry

(A) Time course of the proportion of ST2 S. pneumoniae bacteria positive for C3b after incubation in serum from wild-type and Apcs^−/− mice. For the comparison of results for wild-type versus Apcs^−/− mice, p < 0.001 at all time points from 1 to 20 min. (B and C) Proportion of ST4 (B) and ST23F (C) S. pneumoniae bacteria positive for C3b after incubation for 20 min in serum from wild-type and Apcs^−/− mice. (D and E) Examples of flow cytometry histograms of C3b deposition on ST2 (D) and ST23F (E) S. pneumoniae strains after incubation in PBS or serum from wild-type or Apcs^−/− mice. (F) Effect on the proportion of ST2 S. pneumoniae positive for C3b of addition of 10 μg/ml hSAP to serum from Apcs^−/− mice. (G) Effect on the proportion of ST2 S. pneumoniae positive for C3b of addition of an equal volume of serum from wild-type mice to serum from Apcs^−/− mice. (H) Effect of addition of hSAP (50 μg/ml, white column) on C3b deposition on the ST2 S. pneumoniae strain in serum from Apcs ^−/− .C1qa ^−/− mice. Grey columns, results for Apcs ^−/− .C1qa ^−/− serum; white columns, results for Apcs ^−/− .C1qa ^−/− serum in the presence of hSAP. For panels (A–C) and (F–H), error bars represent SDs, and in (A) when not visible are too small to be seen outside of the symbol. p-Values are calculated using 2-tailed t tests.

Figure 3. Effects of SAP on C1q Binding to S. pneumoniae Measured Using Flow Cytometry

(A–C) Proportion of ST2 (A), ST4 (B), and ST23F (C) S. pneumoniae bacteria positive for C1q after incubation in 90 μg/ml human C1q with or without addition of 10 or 50 μg/ml hSAP. (D) Example of flow cytometry histograms of C1q binding to ST4 S. pneumoniae after incubation in 90 μg/ml human C1q with or without addition of 50 μg/ml hSAP. (E) Proportion of S. pneumoniae positive for C1q after incubation in human serum with (black shading) or without (grey shading) addition of 50 μg/ml hSAP, using C1q-depleted serum (white shading) as a negative control. (F) Example of flow cytometry histograms of C1q binding to ST2 S. pneumoniae after incubation in human serum with or without addition of 50 μg/ml hSAP. For panels (A–C) and (E), error bars represent SDs, and p-values are calculated using 2-tailed t tests.

Figure 4. Effect of SAP on Phagocytosis of S. pneumoniae

(A–C) Phagocytosis (presented as proportion of HL60 cells associated with fluorescent bacteria) of (A) ST2, (B) ST4, and (C) ST23F after incubation in different dilutions of serum from wild-type (circles) or Apcs^−/− (squares) mice. Results for incubation in HBSS are shown by the triangle symbol, and for the ST2 strain the results for a 50:50 mix of serum from wild-type and Apcs^−/− mice (diamonds) are also included. (D) Example of a flow cytometry histogram of phagocytosis of ST2 S. pneumoniae by HL60 cells after incubation in HBSS or serum from wild-type or Apcs^−/− mice. (E) Effect of addition of 5 or 50 μg/ml exogenous hSAP on phagocytosis of the ST2 S. pneumoniae strain in serum from Apcs^−/− mice. (F) Effect of addition of hSAP (50 μg/ml) on phagocytosis of the ST2 S. pneumoniae strain in serum from Apcs ^−/− .C1qa ^−/− mice. Grey column, results for Apcs ^−/− .C1qa ^−/− serum; white column, results for Apcs ^−/− .C1qa ^−/− serum in the presence of hSAP. For panels (A–C), (E), and (F), asterisks mark significant p-values for comparisons of results for wild-type or mixed serum to Apcs^−/− serum (2-tailed t tests, *p < 0.01, **p < 0.001, ***p < 0.0001). All error bars represent SDs and when not visible are too small to be seen outside the symbol.

Figure 5. Clearance of the ST2 S. pneumoniae Strain from Wild-Type and Apcs ^−/− Mice Inoculated Intravenously with 1.0 × 10⁶ cfu

Each data point represents log₁₀ cfu/ml results for a single mouse, with the bar showing the median for each group. (A) Results for blood 2 h after inoculation. (B) Results for blood 4 h after inoculation. (C) Results for spleen homogenates 4 h after inoculation. Data is obtained from one experiment that is representative of two separate experiments. p-Values for Mann–Whitney U comparisons between wild-type and Apcs ^−/− mice are given below the title for each panel.

Figure 6. Susceptibility of Apcs ^−/− Mice to S. pneumoniae Pneumonia

(A) Progression to lethal disease in groups of nine wild-type (circles) and Apcs ^−/− (squares) mice inoculated i.n. with 1.0 × 10⁶ cfu of ST2 S. pneumoniae. For the comparison between wild-type and Apcs^−/− mice, p = 0.004 (log rank test). (B) S. pneumoniae cfu in blood recovered from wild-type (circles) and Apcs^−/− (squares) mice 48 h after i.n. with 1.0 × 10⁶ cfu of ST2 S. pneumoniae. Bars represent the median cfu recovered for each group, and the p-value for the comparison between wild-type and Apcs ^−/− mice was 0.041 (Mann–Whitney U test).

Figure 7. Cytokine Levels in BALF Recovered from Wild-Type or Apcs ^−/− Mice 4 h or 24 h after i.n. Inoculation with 1.0 × 10⁶ cfu of ST2 S. pneumoniae

Error bars represent SDs, and the p-values (2-tailed t tests) for comparison between wild-type and Apcs ^−/− mice are given below the title for each panel. (A) Cytokine levels 4 h after inoculation. (B) Cytokine levels 24 h after inoculation.