Skip to main content
NCBI home page
Infection and Immunity logoLink to Infection and Immunity
. 2015 Apr 15;83(5):1845–1852. doi: 10.1128/IAI.03058-14

C-Reactive Protein Protects Mice against Pneumococcal Infection via Both Phosphocholine-Dependent and Phosphocholine-Independent Mechanisms

Toh B Gang a, Gregory A Hanley b, Alok Agrawal a,c,✉
Editor: L Pirofski
PMCID: PMC4399050  PMID: 25690104

Abstract

The mechanism of action of C-reactive protein (CRP) in protecting mice against lethal Streptococcus pneumoniae infection is unknown. The involvement of the phosphocholine (PCh)-binding property of CRP in its antipneumococcal function previously has been explored twice, with conflicting results. In this study, using three different intravenous sepsis mouse models, we investigated the role of the PCh-binding property of CRP by employing a CRP mutant incapable of binding to PCh. The ability of wild-type CRP to protect mice against infection was found to differ in the three models; the protective ability of wild-type CRP decreased when the severity of infection was increased, as determined by measuring mortality and bacteremia. In the first animal model, in which we used 25 μg of CRP and 107 CFU of pneumococci, both wild-type and mutant CRP protected mice against infection, suggesting that the protection was independent of the PCh-binding activity of CRP. In the second model, in which we used 25 μg of CRP and 5 × 107 CFU of pneumococci, mutant CRP was not protective while wild-type CRP was, suggesting that the protection was dependent on the PCh-binding activity of CRP. In the third model, in which we used 150 μg of CRP and 107 CFU of pneumococci, mutant CRP was as protective as wild-type CRP, again indicating that the protection was independent of the PCh-binding activity of CRP. We conclude that both PCh-dependent and PCh-independent mechanisms are involved in the CRP-mediated decrease in bacteremia and the resulting protection of mice against pneumococcal infection.

INTRODUCTION

Infection with Streptococcus pneumoniae is one of the most common causes of community-acquired pneumonia and septicemia worldwide (reviewed in references 1–3). C-reactive protein (CRP) is a plasma protein whose level in the blood is dramatically increased in patients with S. pneumoniae infection (reviewed in references 4–8). In experiments using animal models, passively administered human CRP, transgenic human CRP, and murine CRP all have been shown to protect mice against lethal infection with S. pneumoniae, as determined by decreased bacteremia and increased survival (9–12). In murine models of infection involving passively administered human CRP, CRP exerted its protective effect only when injected 6 h before to 2 h after the infection but not when mice received CRP 24 h or 36 h postinfection (13, 14). Thus, the CRP-mediated protection of mice requires the presence of CRP in the early stages of infection. However, the mechanism of the protective action of CRP in mice during the early stages of infection is not known.

CRP binds to pneumococci in both serum and Ca2+-containing buffers (14). The binding of CRP to pneumococci is mediated via a Ca2+-dependent interaction between CRP and phosphocholine (PCh) residues present on C-polysaccharide (PnC) of the cell wall of pneumococci (15). Because PnC-complexed CRP activates the complement system in both human and mouse sera (4, 16; reviewed in reference 17), it has been proposed that the mechanism of the antipneumococcal function of CRP involves the ability of CRP to bind to pneumococci through PCh groups present on PnC and the subsequent activation of the complement system by pneumococcus-bound CRP (18). Previously, we have investigated whether the binding of CRP to PCh on pneumococci was required for the protection of mice against pneumococcal infection (14, 19). Employing site-directed mutagenesis, we generated two CRP mutants, F66A/E81A and F66A/T76Y/E81A, both incapable of binding to PCh, and employed them in mouse protection experiments. We hypothesized that if the binding of CRP to PCh were required for the protection of mice against pneumococcal infection, then mutant CRP should not be protective. Two papers were published on these investigations, and the results reported in these papers were contradictory to each other (14, 19). According to one report (14), the CRP-mediated protection was independent of the binding of CRP to PCh. According to another report (19), the CRP-mediated protection was dependent on the binding of CRP to PCh.

In the first report (14), we used 150 μg of CRP and 108 CFU of pneumococci and the F66A/E81A CRP mutant (20). We found that F66A/E81A CRP mutant was as capable as wild-type (WT) CRP in protecting mice against infection. Mutant CRP decreased mortality and prolonged survival of infected mice as well as WT CRP did. The results indicated that the CRP-mediated protection was independent of the binding of CRP to PCh. However, two features of this animal model were not consistent with the other previously published models (9, 10); in our model, less than 108 CFU of pneumococci did not provide suitable survival and mortality data, and less than 150 μg CRP was not protective against infection induced by 108 CFU of pneumococci.

In the second report (19), the animal model was created by using either 25 μg or 150 μg of CRP and 5 × 107 CFU of pneumococci; in this study, we used the F66A/T76Y/E81A CRP mutant. We found that, unlike WT CRP, the F66A/T76Y/E81A CRP mutant was not protective at both 25-μg and 150-μg doses. Mutant CRP did not decrease mortality and did not prolong the survival of infected mice. The results indicated that the CRP-mediated protection was dependent on the binding of CRP to PCh. However, although not statistically significant, a slight protective effect of mutant CRP was noted at the 150-μg dose (19). One feature of this animal model also was not consistent with the other published models; in our model, although both doses of WT CRP were protective, the protection was not as much as that seen in other published models (9, 10).

The purpose of the current study was to generate data to explain the findings of the previous two reports (14, 19). In this study, we used different animal models which were consistent with the other previously published models (9, 10) and an F66A/T76Y/E81A CRP mutant that is incapable of binding to PCh. The results indicated that both of our previously published reports were correct and that CRP protects mice during the early stages of infection via both PCh-dependent and PCh-independent mechanisms.

MATERIALS AND METHODS

CRP.

The method for constructing the cDNA for CRP mutant F66A/T76Y/E81A has been reported previously (19). Expression of mutant cDNA in CHO cells, isolation of a CHO cell line expressing mutant CRP, and purification of F66A/T76Y/E81A mutant CRP from the cell culture supernatant were performed as described previously (19). Native WT CRP was purified from discarded human pleural fluid as described previously (19). For mouse protection experiments, both purified WT and mutant CRP were treated with the Detoxi-Gel endotoxin-removing gel according to the manufacturer's instructions (Thermo). The concentration of endotoxin in CRP was determined by using the Limulus amebocyte lysate kit (QCL-1000) according to the manufacturer's instructions (Lonza). Binding activity of CRP for PCh was evaluated by using PCh-conjugated bovine serum albumin (PCh-BSA), PnC, and pneumococci as the ligands, as described previously (19), except that CRP was used at concentrations of up to 10 μg/ml.

Pneumococci.

Pneumococci (S. pneumoniae type 3, strain WU2) were maintained virulent, stored, and used as described previously (19). The concentration, purity, and viability of pneumococci were confirmed by plating on sheep blood agar.

Mouse protection experiments.

Male C57BL/6J mice (Jackson ImmunoResearch Laboratories) were maintained according to protocols approved by the University Committee on Animal Care. Mice were 8 to 10 weeks old when used in experiments. Two separate mouse protection experiments were performed using two batches of purified WT and mutant CRP. Mice first were injected intravenously (i.v.) with either 25 μg or 150 μg of WT or mutant CRP in 150 μl Tris-buffered saline (TBS) containing 2 mM CaCl2. The endotoxin content in 25 μg and 150 μg WT CRP was 0.77 ± 0.37 endotoxin units (EU) and 4.64 ± 2.19 EU, respectively. The endotoxin content in 25 μg and 150 μg mutant CRP was 0.53 ± 0.28 EU and 3.22 ± 1.65 EU, respectively. After 30 min, mice were injected i.v. with either 107 CFU or 5 × 107 CFU of pneumococci in 100 μl of saline. Survival of mice was recorded three times per day for 10 days. Survival curves were generated using GraphPad Prism 4 software. To determine P values for the differences in the survival curves among various groups, the survival curves were compared using the software's log-rank test. To determine bacteremia (CFU/ml) in the surviving mice, blood was collected daily for 5 days from the tip of the tail vein, diluted in normal saline, and plated on sheep blood agar for colony counting. The bacteremia value for dead mice was taken as >108 CFU/ml, because mice died when the bacteremia exceeded 108 CFU/ml. The plotting and statistical analyses of the bacteremia data were performed by using GraphPad Prism 4 software and Mann-Whitney nonparametric two-sample rank test.

PCh-binding inhibition assays.

Microtiter wells were coated with either 10 μg/ml of PCh-BSA, 10 μg/ml of PnC, or 107 CFU pneumococci in TBS overnight at 4°C, as described previously (19). The unreacted sites in the wells were blocked with TBS containing 0.5% gelatin for 45 min at room temperature. CRP diluted in TBS containing 5 mM CaCl2, 0.1% gelatin, and 0.02% Tween 20 (TBS-Ca) was added to the wells. To determine the effects of PCh (Sigma-Aldrich) and dAMP (D6250; Sigma-Aldrich) on the binding of CRP to PCh-BSA, PnC, and pneumococci, CRP was added to the wells in the presence of either 10 mM PCh or 10 mM dAMP. To determine the requirement of Ca2+ for the binding of CRP to PCh, CRP was diluted in TBS containing 5 mM EDTA, 0.1% gelatin, and 0.02% Tween 20. After incubating the plates for 2 h at 37°C, unbound CRP was aspirated, followed by washing the wells with TBS-Ca. Rabbit polyclonal anti-CRP antibody (Sigma-Aldrich) and HRP-conjugated donkey anti-rabbit IgG (Thermo) were used to detect bound CRP. Color was developed and the A405 was read in a microtiter plate reader, as described previously (19).

RESULTS

At the 25-μg dose, both WT and mutant CRP protect mice against infection with 107 CFU of pneumococci.

Figure 1 (left) shows the combined results of two separate protection experiments using 25 μg of CRP, 107 CFU of pneumococci, and six mice in each of the three groups in each experiment. Median survival time (the time taken for the death of 50% of mice) for mice injected with bacteria alone (control group A) was 80 h. The median survival time for mice injected with bacteria and WT CRP (group B) was more than 10 days. The median survival time for mice injected with bacteria and mutant CRP (group C) was 176 h. The P values for the difference in the survival curves between groups A and B, A and C, and B and C were 0.001, 0.02 and 0.24, respectively.

FIG 1.

FIG 1

Open in a new tab

(Left) Survival of mice infected with 107 CFU of S. pneumoniae with or without 25 μg of either WT or mutant CRP. CRP was injected first; pneumococci were injected 30 min later. Deaths were recorded three times a day for 10 days. The data are combined from two separate experiments with six mice in each group in both experiments. The P values for the difference in the survival curves between groups A and B, A and C, and B and C are 0.001, 0.02, and 0.24, respectively. (Right) Bacteremia in mice infected with 107 CFU of S. pneumoniae with or without 25 μg of either WT or mutant CRP. Blood was collected from each surviving mouse for the first 5 days after infection. Bacteremia was determined by plating. Each dot represents one mouse. The horizontal line in each group of mice represents the median value of bacteremia in that group. A bacteremia value of >108 indicates a dead mouse. On day 1, the P values for the difference between groups A and B, A and C, and B and C are 0.03, 0.41, and 0.17, respectively. On day 2, the P values for the difference between groups A and B, A and C, and B and C are 0.002, 0.09, and 0.20, respectively.

Bacteremia values, determined every day for 5 days in each mouse from the protection experiment, are plotted as scatter diagrams (Fig. 1, right). In mice injected with bacteria alone (group A), median bacteremia was approximately 2.5 × 103 CFU/ml of blood 1 day postinfection. In mice injected with bacteria and either WT CRP (group B) or mutant CRP (group C), median bacteremia was <103 CFU/ml 1 day postinfection in both groups. In group A, bacteremia increased dramatically after day 1, and mice died once bacteremia exceeded 108 CFU/ml. In mice administered WT CRP, there was no increase in bacteremia past day 1. In mice administered mutant CRP, there was an increase in bacteremia past day 1, but it reached only approximately 107 CFU/ml even after 5 days. On day 1, the P values for the difference between groups A and B, A and C, and B and C were 0.03, 0.41, and 0.17, respectively. On day 2, the P values for the difference between groups A and B, A and C, and B and C were 0.002, 0.09, and 0.20, respectively. Thus, although the bacteremia values for mutant CRP were not significantly different from that of the control group for both days 1 and 2, the bacteremia values for mutant CRP also were not significantly different from that of WT CRP for both days.

We concluded that both WT and mutant CRP decreased mortality and prolonged survival of mice injected with 107 CFU of pneumococci, and that the increased resistance to infection was associated with the maintenance of a reduced bacteremia or a slower increase in bacteremia. The results obtained from this animal model (25 μg of CRP and 107 CFU of pneumococci) suggest that the CRP-mediated protection of mice from infection is independent of the PCh-binding activity of CRP.

With 5 × 107 CFU of pneumococci and CRP at the 25-μg dose, mutant CRP was different from WT CRP in protecting mice against infection.

We next changed the animal model in which we used the same dose of CRP as that in the experiment shown in Fig. 1, but we increased the number of pneumococci 5-fold. Figure 2 (left) shows the combined results of two separate protection experiments using 25 μg of CRP, 5 × 107 CFU of pneumococci, and six mice in each group in each experiment. The median survival time for mice injected with bacteria alone (control group A) was 40 h, for mice injected with bacteria and WT CRP (group B) it was 78 h, and for mice injected with bacteria and mutant CRP (group C) it was 46 h. In all three groups, mice began dying at 40 h. In the control group, all mice died by 64 h. In the mutant CRP-treated group, all mice died by 116 h. However, in the WT CRP-treated group, only 75% of mice died in 10 days. The P values for the difference in the survival curves between groups A and B, A and C, and B and C are <0.001, 0.03, and 0.003, respectively. Survival in both WT CRP-treated and mutant CRP-treated groups was significantly different from that for the control group, but the results for the WT CRP-treated group also were significantly different from those for the mutant CRP-treated group.

FIG 2.

FIG 2

Open in a new tab

(Left) Survival of mice infected with 5 × 107 CFU of S. pneumoniae with or without 25 μg of either WT or mutant CRP. CRP was injected first; pneumococci were injected 30 min later. Deaths were recorded three times a day for 10 days. The data are combined from two separate experiments with six mice in each group in both experiments. The P values for the difference in the survival curves between groups A and B, A and C, and B and C are <0.001, 0.03, and 0.003, respectively. (Right) Bacteremia in mice infected with 5 × 107 CFU of S. pneumoniae with or without 25 μg of either WT or mutant CRP. Blood was collected from each surviving mouse for the first 5 days after infection. Bacteremia was determined by plating. Each dot represents one mouse. The horizontal line in each group of mice represents the median value of bacteremia in that group. A bacteremia value of >108 indicates a dead mouse. On day 1, the P values for the difference between groups A and B, A and C, and B and C are <0.005, 0.16, and 0.02, respectively. On day 2, the P values for the difference between groups A and B, A and C, and B and C are <0.005, 0.25, and 0.03, respectively.

Bacteremia values, determined every day for 5 days in each mouse from the protection experiment, are plotted as scatter diagrams (Fig. 2, right). In mice injected with bacteria alone (group A), 1 day postinfection, median bacteremia was approximately 1.9 × 105 CFU/ml of blood. In mice injected with bacteria and WT CRP (group B), median bacteremia was 1.5 × 103 CFU/ml 1 day postinfection. In mice injected with bacteria and mutant CRP (group C), 1 day postinfection, median bacteremia was 2.9 × 104 CFU/ml. In group A, bacteremia increased dramatically by day 1, and bacteremia exceeded 108 CFU/ml by day 2. In mice administered WT CRP, there was an increase in bacteremia past day 1, but it took 3 days to exceed 108 CFU/ml when those mice died. Similar to the control group, in mice administered mutant CRP, bacteremia increased dramatically after day 1, and bacteremia exceeded 108 CFU/ml by day 2. On day 1, the P values for the difference between groups A and B, A and C, and B and C are <0.005, 0.16, and 0.02, respectively. On day 2, the P values for the difference between groups A and B, A and C, and B and C are <0.005, 0.25, and 0.03, respectively. Thus, statistically significant differences in bacteremia were observed between the control (group A) and WT CRP-treated groups (group B) for both days but not between the control and the mutant CRP-treated groups (group C) for either day. Also, the bacteremia values for WT CRP were significantly different from that of mutant CRP.

We concluded that, in contrast to WT CRP, mutant CRP did not decrease mortality and did not prolong survival of infected mice due to failure in reducing bacteremia or maintaining reduced bacteremia. The results obtained from this animal model subjected to a markedly greater load of pneumococci indicate that CRP-mediated protection of mice against infection is dependent on the PCh-binding activity of CRP under these circumstances.

Mutant CRP is as capable as WT CRP in protecting mice against infection when large doses of CRP are employed.

We changed the animal model again: we used the same dose of pneumococci as that in the experiment shown in Fig. 1, but we increased the dose of CRP by 6-fold. Figure 3 (left) shows the combined results of two separate protection experiments using 150 μg of CRP, 107 CFU of pneumococci, and six mice in each group in each experiment. The median survival time for mice injected with bacteria alone (control group A) was 80 h. During this period, survival was 92% in the infected mice treated with either WT or mutant CRP. The median survival time for mice injected with bacteria and either WT or mutant CRP (groups B and C) was more than 10 days in both groups. In the control group, mice began dying at 40 h, and only 9% of mice survived up to 10 days. However, in the WT CRP-treated and mutant CRP-treated groups, no deaths occurred in 64 h, and 75% and 89% of mice survived up to 10 days, respectively. The P values for the difference in the survival curves between groups A and B, A and C, and B and C are <0.005, <0.005, and 0.33, respectively.

FIG 3.

FIG 3

Open in a new tab

(Left) Survival of mice infected with 107 CFU of S. pneumoniae with or without 150 μg of either WT or mutant CRP. CRP was injected first; pneumococci were injected 30 min later. Deaths were recorded three times a day for 10 days. The data are combined from two separate experiments with six mice in each group in both experiments. The P values for the difference in the survival curves between groups A and B, A and C, and B and C are <0.005, <0.005, and 0.33, respectively. (Right) Bacteremia in mice infected with 107 CFU of S. pneumoniae with or without 150 μg of either WT or mutant CRP. Blood was collected from each surviving mouse for the first 5 days after infection. Bacteremia was determined by plating. Each dot represents one mouse. The horizontal line in each group of mice represents the median value of bacteremia in that group. A bacteremia value of >108 indicates a dead mouse. On day 1, the P values for the difference between groups A and B, A and C, and B and C are 0.09, 0.005, and 0.31, respectively. On day 2, the P values for the difference between groups A and B, A and C, and B and C are 0.02, <0.005, and 0.85, respectively.

Bacteremia values, determined every day for 5 days in each mouse from the protection experiment, are plotted as scatter diagrams (Fig. 3, right). In mice injected with bacteria alone (group A), 1 day postinfection, median bacteremia was approximately 3.1 × 103 CFU/ml of blood. In mice injected with bacteria and WT CRP (group B), median bacteremia was approximately 200 CFU/ml 1 day postinfection. In mice injected with bacteria and mutant CRP (group C), 1 day postinfection, median bacteremia was 100 CFU/ml. In group A, bacteremia increased dramatically by day 1, and mice died once bacteremia exceeded 108 CFU/ml. In mice administered either WT or mutant CRP, there was no significant increase in bacteremia past day 1. On day 1, the P values for the difference between groups A and B, A and C, and B and C are 0.09, 0.005, and 0.31, respectively. On day 2, the P values for the difference between groups A and B, A and C, and B and C are 0.02, <0.005, and 0.85, respectively. Thus, the bacteremia values for both WT and mutant CRP were significantly different from that of the control group, and the bacteremia value for mutant CRP was not significantly different from that of WT CRP.

We concluded that mutant CRP was as effective as WT CRP in decreasing mortality and prolonging the survival of mice, and that the increased resistance to infection was associated with the maintenance of a reduced bacteremia. The results obtained from this animal model (150 μg of CRP and 107 CFU of pneumococci) suggest that the CRP-mediated protection of mice against infection is independent of the PCh-binding activity of CRP when very large doses of CRP are administered.

The ability of CRP to protect mice against infection depends on the animal model.

The combined results of the protection experiments using WT CRP (25 μg in Fig. 1 and 150 μg in Fig. 3) are shown in Fig. 4. The median survival time for mice injected with 107 CFU of pneumococci (group A) was 80 h. The median survival time for mice injected with 107 CFU of pneumococci and 25 μg of WT CRP (group B) was more than 10 days. The median survival time for mice injected with 5 × 107 CFU of pneumococci (group C) was only 40 h. The median survival time for mice injected with 5 × 107 CFU of pneumococci and 25 μg of WT CRP (group D) was 78 h. The P values for the difference in the survival curves between groups A and B, C and D, A and C, and B and D are 0.001, <0.005, <0.005, and 0.03, respectively. Thus, the protective ability of CRP decreased when the severity of infection was increased, suggesting that the protective ability of even WT CRP decreases with increasing doses of pneumococci.

FIG 4.

FIG 4

Open in a new tab

Survival of mice infected with 107 CFU and 5 × 107 CFU of S. pneumoniae with or without 25 μg of WT CRP. CRP was injected first; pneumococci were injected 30 min later. Deaths were recorded three times a day for 10 days. The data are combined from two separate experiments with six mice in each group in both experiments. The P values for the difference in the survival curves between groups A and B, C and D, A and C, and B and D are 0.001, <0.005, <0.005, and 0.03, respectively.

Comparison of binding of WT and mutant CRP to PCh-containing ligands.

We have reported previously that the F66A/T76Y/E81A CRP mutant did not bind to PCh-containing ligands (19). If mutant CRP does not bind to PCh, and if the PCh-binding property of CRP is required for CRP to protect mice against pneumococcal infection, then mutant CRP should not have protected mice against infection regardless of the doses of CRP and pneumococci to set up the animal model. The finding that mutant CRP did protect mice against infection in one defined animal model (150 μg mutant CRP and 107 CFU pneumococci) warranted the reassessment of the PCh-binding activity of mutant CRP. Previously, our PCh-binding assays utilized CRP in the 5-ng/ml to 50-ng/ml range, because this range provided the linear part of the binding curve. We, however, used 10 μg/ml of mutant CRP to reassess the PCh-binding ability of mutant CRP, using two different PCh-containing ligands, PCh-BSA (Fig. 5A) and PnC (Fig. 5B), and whole live pneumococci (Fig. 5C). As shown, for equivalent binding (A405) of CRP to any of the three ligands, approximately 100 times more mutant CRP was required than WT CRP, indicating that the PCh-binding ability of mutant CRP was approximately 99% less than that of WT CRP.

FIG 5.

FIG 5

Open in a new tab

Binding of CRP to PCh-containing ligands. Microtiter wells were coated with PCh-BSA (A), PnC (B), and pneumococci (C). The unreacted sites in the wells were blocked with gelatin. CRP diluted in TBS-Ca then was added to the wells. Bound CRP was detected by using rabbit polyclonal anti-CRP antibody and HRP-conjugated donkey anti-rabbit IgG. Color was developed, and the absorbance was read at 405 nm. A representative of three experiments is shown.

Because the dose of mutant CRP we used in the mouse protection experiments was only 6 times higher than that of WT CRP and not 100 times higher, it is not possible that the PCh-independent protection was due to the residual PCh-binding activity of mutant CRP in cases where a 6-times-higher dose of mutant CRP was used. Although we do not know if there is an indirect, PCh-independent interaction between mutant CRP and pneumococci in vivo, the data obtained from these assays suggest that the protection of mice seen with mutant CRP did not involve PCh-mediated interaction between CRP and pneumococci.

Inhibition of binding of WT and mutant CRP to PCh-containing ligands.

We next determined the molecular basis of the residual binding of mutant CRP to PCh-containing ligands. We used inhibition assays in which we used 10 mM PCh and 10 mM dAMP as fluid-phase inhibitors of binding of CRP to immobilized PCh-containing ligands. We also performed assays in EDTA-containing buffers to confirm that the binding of any CRP to PCh-containing ligands was Ca2+ dependent, because the binding of CRP to PCh requires Ca2+. As shown in Fig. 6A, the binding of WT CRP to PCh-BSA was completely inhibited by PCh at all concentrations of CRP. The binding also was inhibited by dAMP, although dAMP was a less effective inhibitor than PCh, as reported previously (15). In contrast to WT CRP, the residual binding of mutant CRP to PCh-BSA was completely inhibited by both PCh and dAMP (Fig. 6B), indicating that the binding of mutant CRP to PCh was mediated through the phosphate group of PCh. The weak binding and strong inhibitory power of dAMP suggested that mutant CRP retained the ability to bind to phosphate. Similar results were obtained with PnC and pneumococci for both WT and mutant CRP (Fig. 6C to F). In all assays (Fig. 6A to F), no binding was observed in EDTA, indicating that the binding of mutant CRP was Ca2+ dependent and was through the phosphate groups on the ligands.

FIG 6.

FIG 6

Open in a new tab

Inhibition of binding of WT (left) and mutant (right) CRP to PCh-containing ligands. (A and B) Inhibition of binding of CRP to PCh-BSA by PCh and dAMP. Microtiter wells were coated with PCh-BSA. CRP at increasing concentrations in TBS-Ca, with and without 10 mM PCh or 10 mM dAMP, was added to the wells. (C and D) Inhibition of binding of CRP to PnC by PCh and dAMP. Microtiter wells were coated with PnC. CRP at increasing concentrations in TBS-Ca, with and without 10 mM PCh or 10 mM dAMP, was added to the wells. (E and F) Inhibition of binding of CRP to pneumococci by PCh and dAMP. Microtiter wells were coated with pneumococci. CRP at increasing concentrations in TBS-Ca, with and without 10 mM PCh or 10 mM dAMP added to the wells. In all experiments, the requirement of Ca2+ for the binding of CRP to PCh-containing ligands also was evaluated by using CRP at increasing concentrations in TBS containing 5 mM EDTA. In all experiments, bound CRP was detected by using rabbit polyclonal anti-CRP antibody and HRP-conjugated donkey anti-rabbit IgG. Color was developed, and the absorbance was read at 405 nm. A representative of three experiments is shown.

DISCUSSION

We tested the hypothesis that the mechanism of CRP-mediated protection of mice against S. pneumoniae infection during the early stages of infection involves binding of CRP to PCh groups present on pneumococci. We compared the protective ability of WT CRP that binds to PCh and pneumococci with that of a CRP mutant that does not bind to either PCh or pneumococci in mice infected with lethal doses of pneumococci. We used three different animal model setups. Our major findings were the following: (i) in the animal model using 25 μg of CRP and 107 CFU of pneumococci, both WT and mutant CRP protected mice against infection; (ii) in the model using 25 μg of CRP and 5 × 107 CFU of pneumococci, WT CRP protected but mutant CRP did not protect mice against infection; and (iii) in the animal model using 150 μg of CRP and 107 CFU of pneumococci, mutant CRP was as capable as WT CRP in protecting mice against infection. Thus, by using different animal model setups, it was revealed that CRP works through both PCh-dependent and PCh-independent mechanisms in protecting mice against early-stage pneumococcal infection.

Previously, we initially reported that CRP-mediated protection of mice was not dependent on the PCh-binding property of CRP (14) but subsequently reported that the CRP-mediated protection of mice was dependent on the PCh-binding property of CRP (19). We provided an explanation for these contradictory reports by indicating that the CRP mutants used in the two studies were different (19), namely, F66A/E81A in the first study and F66A/T76Y/E81A in the second study. We also postulated that the contradictory findings were due to differences in the purity of the two CRP preparations employed, frozen and thawed mutant CRP in the first study and freshly purified mutant CRP in the second study (19). Our current findings, however, indicate that the contradictory findings were due to the differences in the setup of the animal model; in terms of median survival time, WT CRP was much less protective in the second study than in the first study. It was important that an animal model be used in which WT CRP showed prolonged protection in terms of median survival time, which was needed to evaluate the protective ability of mutant CRP.

There are two different CRP mutants available, F66A/E81A and F66A/T76Y/E81A, which do not bind to either PCh or pneumococci. However, the F66A/T76Y/E81A CRP mutant binds to phosphoethanolamine much more avidly than the F66A/E81A CRP mutant (19). To purify F66A/E81A CRP, immunoaffinity chromatography is required, which results in poor recovery of pentameric CRP due to the use of an acidic pH buffer (14). We chose the F66A/T76Y/E81A CRP mutant for this study, because it was easier to purify this mutant by affinity chromatography using a phosphoethanolamine-agarose column with higher recovery of the purified protein. It has been shown previously that the clearance rate of F66A/T76Y/E81A CRP mutant from mouse circulation is similar to that of WT CRP (19). Complete biochemical characteristics of the F66A/T76Y/E81A CRP mutant have been described previously (19).

We found that the binding of mutant CRP to PCh ligands, including pneumococci, was approximately only 1% of that of WT CRP and that the residual binding was mediated only through the phosphate group of PCh. The PCh-binding site of CRP consists of a hydrophobic pocket formed mainly by Phe66, Thr76, and Glu81 and two Ca2+ ions which are bound to CRP by interactions with seven amino acids from other parts of the protein. The phosphate group of PCh directly coordinates with the two Ca2+ ions. The choline group of PCh lies within the hydrophobic pocket. Phe66 provides hydrophobic interactions with the three methyl groups of choline. Thr76 is critical for creating the appropriately sized pocket to accommodate PCh. Glu81 interacts with the positively charged nitrogen atom of choline (21–23). Mutant CRP retained its Ca2+-dependent binding to phosphate because the mutations in the F66A/T76Y/E81A CRP mutant were not in the Ca2+-binding site. Our conclusion that CRP used a PCh-binding-independent mechanism to protect mice against infection also was based on the finding that the amount of mutant CRP required to protect mice against infection was the same as that of WT CRP, despite the drastic difference in the avidity of binding of WT and mutant CRP to pneumococci.

The significance of the PCh-dependent mechanism is the initial and immediate clearance of pneumococci, as has been shown recently with endogenous mouse CRP (12). The definition of the PCh-independent mechanism is not known. We hypothesize that the PCh-independent mechanism involves first a structural change in CRP, which then is followed by the interaction between structurally altered CRP and factor H-bound pneumococci, as we have discussed previously (4, 14, 19). This hypothesis is based on three findings. First, pneumococci are able to recruit factor H, a regulator of complement activation, on their surface, and factor H-coated pneumococci are resistant to attack by the complement system (4, 24). Second, it has been shown that the pentameric structure of CRP is altered in a variety of conditions (4, 25–31), and that structurally altered CRP gains the ability to bind to immobilized factor H (4, 25, 26, 31–35). Third, based on the bacteremia data, 150 μg of mutant CRP was more protective than 25 μg, unlike WT CRP, for which there was no difference between the use of 25 and 150 μg CRP. The half-life (t1/2) of CRP in mice is only approximately 4 h (36). Administering a higher dose of CRP allows CRP to stay longer in the host. We hypothesize that the CRP mutant, when given at the beginning of the experiment, undergoes structural changes during its stay in the host to express the factor H-binding property, regardless of the mutations at the PCh-binding site. Once factor H on pneumococci is bound to structurally altered CRP, such pneumococci may not be resistant to complement attack any longer.

In order to protect mice from infection, CRP must bind to pneumococci and that mutant CRP must be doing so in vivo, although not through the PCh groups present on the pneumococcal surface. Besides the structurally altered CRP-factor H hypothesis described in the previous paragraph, several other possible mechanisms, including one based on the interaction between CRP and ficolins, have been reported (4, 16, 37–39). As proposed previously (19), the development of infection models involving passively administered human WT and mutant CRP in CRP knockout (KO) mice and SAP KO mice (12, 40–45), the development of mice transgenic for the F66A/T76Y/E81A CRP mutant, and also the use of a strain of S. pneumoniae incapable of recruiting factor H in mouse protection experiments may provide more information on the PCh-independent mechanism of the antipneumococcal function of CRP.

ACKNOWLEDGMENTS

This work was supported by National Institutes of Health grant HL71233 and an ETSU Research Development Committee Major Grant.

We thank David E. Briles (University of Alabama at Birmingham, Birmingham, AL) for the gift of S. pneumoniae type 3, strain WU2. We thank Irving Kushner for critical reading of the manuscript.

REFERENCES

  • 1.Tuomanen EI, Austrian R, Masure HR. 1995. Pathogenesis of pneumococcal infection. N Engl J Med 332:1280–1284. doi: 10.1056/NEJM199505113321907. [DOI] [PubMed] [Google Scholar]
  • 2.Malley R, Anderson PW. 2012. Serotype-independent pneumococcal experimental vaccines that induce cellular as well as humoral immunity. Proc Natl Acad Sci U S A 109:3623–3627. doi: 10.1073/pnas.1121383109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Van der Poll T, Opal SM. 2009. Pathogenesis, treatment, and prevention of pneumococcal pneumonia. Lancet 374:1543–1556. doi: 10.1016/S0140-6736(09)61114-4. [DOI] [PubMed] [Google Scholar]
  • 4.Agrawal A, Suresh MV, Singh SK, Ferguson DA Jr. 2008. The protective function of human C-reactive protein in mouse models of Streptococcus pneumoniae infection. Endocr Metab Immune Disord Drug Targets 8:231–237. doi: 10.2174/187153008786848321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kushner I, Samols D. 2011. Oswald Avery and the pneumococcus. Pharos Alpha Omega Alpha Honor Med Soc 74:14–18. [PubMed] [Google Scholar]
  • 6.Mook-Kanamori BB, Geldhoff M, van der Poll T, van de Beek D. 2011. Pathogenesis and pathophysiology of pneumococcal meningitis. Clin Microbiol Rev 24:557–591. doi: 10.1128/CMR.00008-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Young NM, Foote SJ, Wakarchuk WW. 2013. Review of phosphocholine substituents on bacterial pathogen glycans: synthesis, structures and interactions with host proteins. Mol Immunol 56:563–573. doi: 10.1016/j.molimm.2013.05.237. [DOI] [PubMed] [Google Scholar]
  • 8.Song JY, Eun BW, Nahm MH. 2013. Diagnosis of pneumococcal pneumonia: current pitfalls and the way forward. Infect Chemother 45:351–366. doi: 10.3947/ic.2013.45.4.351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Yother J, Volanakis JE, Briles DE. 1982. Human C-reactive protein is protective against fatal Streptococcus pneumoniae infection in mice. J Immunol 128:2374–2376. [PubMed] [Google Scholar]
  • 10.Mold C, Nakayama S, Holzer TJ, Gewurz H, Du Clos TW. 1981. C-reactive protein is protective against Streptococcus pneumoniae infection in mice. J Exp Med 154:1703–1708. doi: 10.1084/jem.154.5.1703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Szalai AJ, Briles DE, Volanakis JE. 1995. Human C-reactive protein is protective against fatal Streptococcus pneumoniae infection in transgenic mice. J Immunol 155:2557–2563. [PubMed] [Google Scholar]
  • 12.Simons JP, Loeffler JM, Al-Shawi R, Ellmerich S, Hutchinson WL, Tennent GA, Petrie A, Raynes JG, De Souza JB, Lawrence RA, Read KD, Pepys MB. 2014. C-reactive protein is essential for innate resistance to pneumococcal infection. Immunology 142:414–420. doi: 10.1111/imm.12266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Nakayama S, Gewurz H, Holzer T, Du Clos TW, Mold C. 1983. The role of the spleen in the protective effect of C-reactive protein in Streptococcus pneumoniae infection. Clin Exp Immunol 54:319–326. [PMC free article] [PubMed] [Google Scholar]
  • 14.Suresh MV, Singh SK, Ferguson DA Jr, Agrawal A. 2007. Human C-reactive protein protects mice from Streptococcus pneumoniae infection without binding to pneumococcal C-polysaccharide. J Immunol 178:1158–1163. doi: 10.4049/jimmunol.178.2.1158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Volanakis JE, Kaplan MH. 1971. Specificity of C-reactive protein for choline phosphate residues of pneumococcal C-polysaccharide. Proc Soc Exp Biol Med 136:612–614. doi: 10.3181/00379727-136-35323. [DOI] [PubMed] [Google Scholar]
  • 16.Suresh MV, Singh SK, Ferguson DA Jr, Agrawal A. 2006. Role of the property of C-reactive protein to activate the classical pathway of complement in protecting mice from Streptococcus pneumoniae infection. J Immunol 176:4369–4374. doi: 10.4049/jimmunol.176.7.4369. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Agrawal A, Gang TB, Rusiñol AE. 2014. Recognition functions of pentameric C-reactive protein in cardiovascular disease. Mediators Inflamm 2014:1–6. doi: 10.1155/2014/319215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Volanakis JE. 1982. Complement activation by C-reactive protein complexes. Ann N Y Acad Sci 389:235–250. doi: 10.1111/j.1749-6632.1982.tb22140.x. [DOI] [PubMed] [Google Scholar]
  • 19.Gang TB, Hammond DJ Jr, Singh SK, Ferguson DA Jr, Mishra VK, Agrawal A. 2012. The phosphocholine-binding pocket on C-reactive protein is necessary for initial protection of mice against pneumococcal infection. J Biol Chem 287:43116–43125. doi: 10.1074/jbc.M112.427310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Agrawal A, Simpson MJ, Black S, Carey MP, Samols D. 2002. A C-reactive protein mutant that does not bind to phosphocholine and pneumococcal C-polysaccharide. J Immunol 169:3217–3222. doi: 10.4049/jimmunol.169.6.3217. [DOI] [PubMed] [Google Scholar]
  • 21.Shrive AK, Cheetham GMT, Holden D, Myles DAA, Turnell WG, Volanakis JE, Pepys MB, Bloomer AC, Greenhough TJ. 1996. Three dimensional structure of human C-reactive protein. Nat Struct Biol 3:346–354. doi: 10.1038/nsb0496-346. [DOI] [PubMed] [Google Scholar]
  • 22.Thompson D, Pepys MB, Wood SP. 1999. The physiological structure of human C-reactive protein and its complex with phosphocholine. Structure 7:169–177. doi: 10.1016/S0969-2126(99)80023-9. [DOI] [PubMed] [Google Scholar]
  • 23.Mullenix MC, Mortensen RF. 1994. Calcium ion binding regions in C-reactive protein: location and regulation of conformational changes. Mol Immunol 31:615–622. doi: 10.1016/0161-5890(94)90169-4. [DOI] [PubMed] [Google Scholar]
  • 24.Jarva H, Janullczyk R, Hellwage J, Zipfel PF, Björck L, Meri S. 2002. Streptococcus pneumoniae evades complement attack and opsonophagocytosis by expressing the pspC locus-encoded Hic protein that binds to short consensus repeats 8-11 of factor H. J Immunol 168:1886–1894. doi: 10.4049/jimmunol.168.4.1886. [DOI] [PubMed] [Google Scholar]
  • 25.Hammond DJ Jr, Singh SK, Thompson JA, Beeler BW, Rusiñol AE, Pangburn MK, Potempa LA, Agrawal A. 2010. Identification of acidic pH-dependent ligands of pentameric C-reactive protein. J Biol Chem 285:36235–36244. doi: 10.1074/jbc.M110.142026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Singh SK, Thirumalai A, Hammond DJ Jr, Pangburn MK, Mishra VK, Johnson DA, Rusiñol AE, Agrawal A. 2012. Exposing a hidden functional site of C-reactive protein by site-directed mutagenesis. J Biol Chem 287:3550–3558. doi: 10.1074/jbc.M111.310011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Xiao MA, Ji SR, Wu Y. 2013. Regulated conformational changes in C-reactive protein orchestrate its role in atherogenesis. Chin Sci Bull 58:1642–1649. doi: 10.1007/s11434-012-5591-3. [DOI] [Google Scholar]
  • 28.Wang MY, Ji SR, Bai CJ, El Kebir D, Li HY, Shi JM, Zhu W, Costantino S, Zhou HH, Potempa LA, Zhao J, Filep JG, Wu Y. 2011. A redox switch in C-reactive protein modulates activation of endothelial cells. FASEB J 25:3186–3196. doi: 10.1096/fj.11-182741. [DOI] [PubMed] [Google Scholar]
  • 29.Zhang J, Yang L, Anand GS, Ho B, Ding JL. 2011. Pathophysiological condition changes the conformation of a flexible FBG-related protein, switching it from pathogen-recognition to host-interaction. Biochimie 93:1710–1719. doi: 10.1016/j.biochi.2011.06.003. [DOI] [PubMed] [Google Scholar]
  • 30.Okemefuna AI, Nan R, Miller A, Gor J, Perkins SJ. 2010. Complement factor H binds at two independent sites to C-reactive protein in acute phase concentrations. J Biol Chem 285:1053–1065. doi: 10.1074/jbc.M109.044529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Hakobyan S, Harris CL, Van den Berg CW, Fernandez-Alonso MC, De Jorge EG, De Cordoba SR, Rivas G, Mangione P, Pepys MB, Morgan BP. 2008. Complement factor H binds to denatured rather than to native pentameric C-reactive protein. J Biol Chem 283:30451–30460. doi: 10.1074/jbc.M803648200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Sjoberg AP, Trouw LA, Clark SJ, Sjölander J, Heinegård D, Sim RB, Day AJ, Blom AM. 2007. The factor H variant associated with age-related macular degeneration (His-384) and the non-disease-associated form bind differentially to C-reactive protein, fibromodulin, DNA and necrotic cells. J Biol Chem 282:10894–10900. doi: 10.1074/jbc.M610256200. [DOI] [PubMed] [Google Scholar]
  • 33.Mihlan M, Stippa S, Józsi M, Zipfel PF. 2009. Monomeric CRP contributes to complement control in fluid phase and on cellular surfaces and increases phagocytosis by recruiting factor H. Cell Death Differ 16:1630–1640. doi: 10.1038/cdd.2009.103. [DOI] [PubMed] [Google Scholar]
  • 34.Ji SR, Wu Y, Potempa LA, Liang YH, Zhao J. 2006. Effect of modified C-reactive protein on complement activation: a possible complement regulatory role of modified or monomeric C-reactive protein in atherosclerotic lesions. Arterioscler Thromb Vasc Biol 26:935–941. doi: 10.1161/01.ATV.0000206211.21895.73. [DOI] [PubMed] [Google Scholar]
  • 35.Deban L, Jarva H, Lehtinen MJ, Bottazzi B, Bastone A, Doni A, Jokiranta TS, Mantovani A, Meri S. 2008. Binding of the long pentraxin PTX3 to factor H: interacting domains and function in the regulation of complement activation. J Immunol 181:8433–8440. doi: 10.4049/jimmunol.181.12.8433. [DOI] [PubMed] [Google Scholar]
  • 36.Motie M, Schaul KW, Potempa LA. 1998. Biodistribution and clearance of 125I-labeled C-reactive protein and 125I-labeled modified C-reactive protein in CD-1 mice. Drug Metab Dispos 26:977–981. [PubMed] [Google Scholar]
  • 37.Ng PML, Le Saux A, Lee CM, Tan NS, Lu J, Thiel S, Ho B, Ding JL. 2007. C-reactive protein collaborates with plasma lectins to boost immune response against bacteria. EMBO J 26:3431–3440. doi: 10.1038/sj.emboj.7601762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zhang J, Yang L, Ang Z, Yoong SL, Tran TTT, Anand GS, Tan NS, Ho B, Ding JL. 2010. Secreted M-ficolin anchors onto monocyte transmembrane G protein-coupled receptor 43 and cross talks with plasma C-reactive protein to mediate immune signaling and regulate host defense. J Immunol 185:6899–6910. doi: 10.4049/jimmunol.1001225. [DOI] [PubMed] [Google Scholar]
  • 39.Vassal-Stermann E, Lacroix M, Gout E, Laffly E, Pedersen CM, Martin L, Amoroso A, Schmidt RR, Zähringer U, Gaboriaud C, Di Guilmi AM, Thielens NM. 2014. Human L-ficolin recognizes phosphocholine moieties of pneumococcal teichoic acid. J Immunol 193:5699–5708. doi: 10.4049/jimmunol.1400127. [DOI] [PubMed] [Google Scholar]
  • 40.Jones NR, Pegues MA, McCrory MA, Kerr SW, Jiang H, Sellati R, Berger V, Villalona J, Parikh R, McFarland M, Pantages L, Madwed JB, Szalai AJ. 2011. Collagen-induced arthritis is exacerbated in C-reactive protein-deficient mice. Arthritis Rheum 63:2641–2650. doi: 10.1002/art.30444. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Teupser D, Weber O, Rao TN, Sass K, Thiery J, Fehling HJ. 2011. No reduction of atherosclerosis in C-reactive protein (CRP)-deficient mice. J Biol Chem 286:6272–6279. doi: 10.1074/jbc.M110.161414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Rutemark C, Alicot E, Bergman A, Ma M, Getahun A, Ellmerich S, Carroll MC, Heyman B. 2011. Requirement for complement in antibody responses is not explained by the classic pathway activator IgM. Proc Natl Acad Sci U S A 108:E934–E942. doi: 10.1073/pnas.1109831108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Togashi S, Lim SK, Kawano H, Ito S, Ishihara T, Okada Y, Nakano S, Kinoshita T, Horie K, Episkopou V, Gottesman ME, Costantini F, Shimada K, Maeda S. 1997. Serum amyloid P component enhances induction of murine amyloidosis. Lab Investig 77:525–531. [PubMed] [Google Scholar]
  • 44.Yuste J, Botto M, Bottoms SE, Brown JS. 2007. Serum amyloid P aids complement-mediated immunity to Streptococcus pneumoniae. PLoS Pathog 3:1208–1219. doi: 10.1371/journal.ppat.0030120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Noursadeghi M, Bickerstaff MCM, Gallimore JR, Herbert J, Cohen J, Pepys MB. 2000. Role of serum amyloid P component in bacterial infection: protection of the host or protection of the pathogen. Proc Natl Acad Sci U S A 97:14584–14589. doi: 10.1073/pnas.97.26.14584. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES