Protection against prolonged pneumococcal infection involves structural changes in C-reactive protein and subsequent binding to both phosphocholine and amyloids on the bacterial surface

C-reactive protein (CRP) protects mice during the initial stages of <i>Streptococcus pneumoniae</i> infection. In order to be protective against all stages of infection, we hypothesize that CRP binds to two different ligands on pneumococci. In its native form, CRP binds to phosphocholine residues of C-polysaccharide to activate complement. In its altered form, CRP binds to amyloid-like structures (amyloids) formed on complement inhibitors recruited by pneumococci. We employed CRP knockout mice to test this hypothesis. In one approach, both wild-type CRP and E42Q/F66A/T76Y/E81A mutant CRP (E-CRP-1) were administered together. E-CRP-1 does not bind to phosphocholine but binds to amyloids. In another approach, Y40F/E42Q mutant CRP (E-CRP-2) was administered. E-CRP-2 binds to both phosphocholine and amyloids. When CRP was administered to mice 12 h after inoculation, then unlike wild-type CRP by itself, the combination of wild-type CRP and E-CRP-1 was protective and E-CRP-2 alone was protective. We also detected amyloids on pneumococci. The serum levels of the amyloid-binding protein, serum amyloid P component (SAP), were higher in CRP knockout mice than in wild-type mice. Also, the basal SAP levels were higher in female than in male mice and, conversely, male mice were more susceptible than female mice to severe infection. We conclude that the protection against prolonged pneumococcal infection requires structural changes in CRP and binding to both phosphocholine and amyloids on pneumococci. The sources of amyloids can be virulence factors or recruited complement inhibitors or both. Combined data also raise the possibility that SAP cooperates with CRP in reducing bacteremia and bacterial load.