Volume 20, Issue 1, Pages 99-106 (July 2016) CRIg Functions as a Macrophage Pattern Recognition Receptor to Directly Bind and Capture Blood-Borne Gram-Positive Bacteria  Zhutian Zeng, Bas G.J. Surewaard, Connie H.Y. Wong, Joan A. Geoghegan, Craig N. Jenne, Paul Kubes  Cell Host & Microbe  Volume 20, Issue 1, Pages 99-106 (July 2016) DOI: 10.1016/j.chom.2016.06.002 Copyright © 2016 Elsevier Inc. Terms and Conditions

Cell Host & Microbe 2016 20, 99-106DOI: (10.1016/j.chom.2016.06.002) Copyright © 2016 Elsevier Inc. Terms and Conditions

Figure 1 Capture of Circulating S. aureus Is Dependent on Kupffer Cells and CRIg, but Not Complement (A) WT mice were infected i.v. with 5 × 107 CFU S. aureus (USA-300); bacterial load per organ were determined 20 min post-infection and calculated as percentage of total inoculum, and pooled results from two independent experiments were shown. For some mice, organs were also exteriorized and subjected to the whole-body imaging based on the far-red dye Syto60 on bacteria to qualitatively determine bacterial distribution. Representative images are shown. (B) WT mice were infected i.v. with 5 × 107 CFU S. aureus (USA-300-GFP), the livers were imaged by intravital microscopy showing representative images at 20 min post-infection. S. aureus-GFP (bright green), PE-anti-Ly6G (1A8)-labeled neutrophils (blue, pseudocolor), and AF 750-anti-F4/80-labeled Kupffer cells (red, pseudocolor) were shown. Scale bars: 100 μm. (C) 3D reconstitution of liver images with increasing transparency showing bacteria inside Kupffer cells. Grid length: 10 μm. (D) Mice were treated with PBS-liposomes (ctrl) or Clodronate-liposomes (CLL) 24 hr prior to S. aureus infection. The numbers of captured bacteria per FOV with the liver were counted at 20 min post-infection. (E) Age-matched WT, CR3−/−, and CRIg−/− mice were infected with GFP-S. aureus; the numbers of captured bacteria per FOV within liver was counted 20 min post-infection. (F) High-magnification intravital liver images showing bacterial capture in WT or CRIg−/− mice. AF 555-labeled anti-CRIg (red); efluor-660 anti-F4/80-labeled Kupffer cells (blue), and S. aureus-GFP (bright green) were shown. Yellow arrows indicate the captured S. aureus. Scale bar: 30 μm. (G) WT or CRIg−/− mice were i.v. infected with 5 × 107 CFU S. aureus; bacterial load in liver, blood, kidney, and spleen were determined at 20 min post-infection. (H–J) (H) Age-matched C3−/−, C4−/−, and C5−/− mice; (I) WT and Rag1−/− mice; and (J) WT and C3−/−FcγR−/− mice were i.v. infected with 5 × 107 CFU S. aureus; liver intravital images were performed. The numbers of captured bacteria per FOV were counted 20 min post-infection. Data were expressed as mean ± SEM, each dot represents one mouse; a total of six to seven FOV were counted per mouse for intravital studies. Representative results of at least two experiments were shown. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001; N.S., no significance. See also Figure S1 and Movies S1 and S2. Cell Host & Microbe 2016 20, 99-106DOI: (10.1016/j.chom.2016.06.002) Copyright © 2016 Elsevier Inc. Terms and Conditions

Figure 2 CRIg Can Directly Bind Gram-Positive Bacteria (A) Recombinant mouse (upper) or human (below) CRIg were labeled with AF-647. Their binding with S. aureus was detected by flow cytometry. AF-647-labeled α-toxin was used as a control. Blank means S. aureus only. (B) S. aureus or E. coli were incubated with indicated concentrations of AF-647 labeled mCRIg. The binding was detected by flow cytometry. (C) Representative fluorescence images of CRIg (AF647, blue) binding to S. aureus (GFP) or E. coli (RFP); scale bars: 10 μm. (D) Quantification of the binding of AF647-labeled CRIg to various bacteria strains based on the geometric mean fluorescence intensity of AF 647 acquired by flow cytometry. Pooled results from three independent experiments were shown in (D), and data were expressed as mean ± SEM. ∗∗p < 0.01; N.S., no significance. Cell Host & Microbe 2016 20, 99-106DOI: (10.1016/j.chom.2016.06.002) Copyright © 2016 Elsevier Inc. Terms and Conditions

Figure 3 CRIg Recognizes Lipoteichoic Acid (A) Binding of S. aureus-GFP or S. aureus ΔsrtA-GFP with AF647-labeled mCRIg. (B) Mice were infected with 5 × 107 CFU S. aureus-GFP or S. aureus ΔsrtA-GFP, and the captured bacteria within liver were counted over time. n = 3 to 4 per group; six to seven FOV were counted per mouse. (C) Binding of S. aureus to AF-647-labeled mCRIg in the absence or presence of purified S. aureus-derived peptidoglycan (100 μg/ml). (D) hCRIg-expressing Jurkat-T cells were incubated with indicated doses of FITC-labeled peptidoglycan. The binding was measured by flow cytometry. (E and F) (E) Binding of S. aureus with AF-647-labeled mCRIg in the presence of different concentrations of S. aureus-derived LTA or (F) in the presence of anti-LTA and ctrl Ig. (G) The binding was quantified using geometric fluorescence intensity. (H) Measurement of the binding of LTA to mCRIg- or BSA-coated plates by ELISA. (I) LTA was pre-absorbed with or without soluble mCRIg; their binding to CRIg or BSA-coated plates was then measured by ELISA. (J) AF-647-labeled CRIg binding to WT S. aureus strain (WT-LAC), LTA-negative S. aureus strains (ΔltaS-US3, ΔltaS-UN2), or WTA-negative S. aureus strain (ΔtagO). (K) Quantification of the binding between AF-647-labeled CRIg and various LTA-deficient S. aureus strains. Data were shown as mean ± SEM. Pooled results from at least three independent experiments in (G), or representative results of at least two experiments in (H)–(K) were shown. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. Cell Host & Microbe 2016 20, 99-106DOI: (10.1016/j.chom.2016.06.002) Copyright © 2016 Elsevier Inc. Terms and Conditions

Figure 4 CRIg-LTA Recognition Promotes Catching of Gram-Positive Bacteria (A) Mice were infected i.v. with wild-type (WT-LAC) or LTA-deficient S. aureus (ΔltaS-LAC) for 20 min. Bacterial load from the liver, blood, and kidney were counted as CFUs. Data were shown as mean ± SEM and were from two independent experiments; each dot represents one mouse (n = 6). (B and C) WT mice were infected with S. aureus (USA-300-GFP) and simultaneously i.v. injected with or without (Ctrl) 1 mg purified LTA. Time-lapse liver intravital imaging was performed; the numbers of (B) caught bacteria or (C) free-floating bacteria per FOV were counted every minute. n = 3 to 4 for each group. The representative of two independent experiments was shown. (D) WT mice were i.v. infected with S. aureus MW2-GFP (top, green), Syto60 labeled B. cerus (2nd row, red, pseudocolor), L. monocytogenes (3rd row, red, pseudocolor), or E. faecium (bottom, red, pseudocolor), representative intravital liver images at 10 min post-infection were shown. Scale bars: 100 μm. At the right side of the panel, the numbers of intrahepatic captured bacteria were counted 10 min post-infection and normalized to the numbers of WT mice for each bacterial strain (capture efficiency = number of captured bacteria in CRIg−/− mice/average number of captured bacteria in WT mice). Data are shown as mean ± SEM and are the representative of at least two independent experiments. n = 4 to 5 per group. ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. See also Figures S2 and S3 and Movies S3 and S4. Cell Host & Microbe 2016 20, 99-106DOI: (10.1016/j.chom.2016.06.002) Copyright © 2016 Elsevier Inc. Terms and Conditions