The role of scavenger receptor B1 in infection with Mycobacterium tuberculosis in a murine model

Abstract

Background: The interaction between Mycobacterium tuberculosis (Mtb) and host cells is complex and far from being understood. The role of the different receptor(s) implicated in the recognition of Mtb in particular remains poorly defined, and those that have been found to have activity in vitro were subsequently shown to be redundant in vivo.

Methods and findings: To identify novel receptors involved in the recognition of Mtb, we screened a macrophage cDNA library and identified scavenger receptor B class 1 (SR-B1) as a receptor for mycobacteria. SR-B1 has been well-described as a lipoprotein receptor which mediates both the selective uptake of cholesteryl esters and the efflux of cholesterol, and has also recently been implicated in the recognition of other pathogens. We show here that mycobacteria can bind directly to SR-B1 on transfected cells, and that this interaction could be inhibited in the presence of a specific antibody to SR-B1, serum or LDL. We define a variety of macrophage populations, including alveolar macrophages, that express this receptor, however, no differences in the recognition and response to mycobacteria were observed in macrophages isolated from SR-B1(-/-) or wild type mice in vitro. Moreover, when wild type and SR-B1(-/-) animals were infected with a low dose of Mtb (100 CFU/mouse) there were no alterations in survival, bacterial burdens, granuloma formation or cytokine production in the lung. However, significant reduction in the production of TNF, IFNgamma, and IL10 were observed in SR-B1(-/-) mice following infection with a high dose of Mtb (1000 CFU/mouse), which marginally affected the size of inflammatory foci but did not influence bacterial burdens. Deficiency of SR-B1 also had no effect on resistance to disease under conditions of varying dietary cholesterol. We did observe, however, that the presence of high levels of cholesterol in the diet significantly enhanced the bacterial burdens in the lung, but this was independent of SR-B1.

Conclusion: SR-B1 is involved in mycobacterial recognition, but this receptor plays only a minor role in anti-mycobacterial immunity in vivo. Like many other receptors for these pathogens, the loss of SR-B1 can be functionally compensated for under normal conditions.

Figure 1. Identification of SR-B1 as a receptor for mycobacteria.

NIH3T3 cells stably transfected with empty vector pFBneo (negative control), SR-B1 or SIGNR1 (positive control), respectively, were incubated with BCG-GFP (A) or BCG-lux (B) and assessed for binding by fluorescence microscopy or luciferase activity, respectively. Shown is the x-fold increase of luciferase activity compared to vector control which was set as 1. Experiments were performed in triplicate and normalised to cell number by CFSE staining. (C) Western Blot showing expression of SR-B1 in various untransfected cell lines as indicated. Cellular lysates from rat liver were included as a control, and GAPDH served as loading control. (D) FACS analysis of R6F cells stably transduced with pFB (vector control), SR-B1 or SIGNR1, and stained with anti-SR-B1 or anti-SIGNRI, as indicated. (E) FACS assay showing binding/uptake of DiI-LDL to R6F cells stably expressing pFB (vector control), SR-B1 or SIGNR1. (F) Binding of FITC-labelled zymosan to R6F cells stably expressing pFB (vector control), SR-B1 or SIGNR1, as quantified by fluorometry. Shown is the x-fold increase of fluorescence compared to vector control which was set as 1. Experiments were performed in triplicate and normalised to cell number by CFSE staining.

Figure 2. Characterization of BCG and Mtb binding to SR-B1 transfected cells.

R6F cells stably transfected with the indicated constructs were incubated with either BCG-lux (left panels) or Mtb-lux (right panels), and quantified for binding by measuring luciferase activity. Experiments were performed in triplicate and normalised to cell number by CFSE staining. (A) Quantification of mycobacterial binding to SR-B1 expressing cells (white bars) or vector control cells (black bars) in the presence or absence of serum, as indicated. Shown is the x-fold increase of luciferase activity compared to vector control which was set as 1. (B) Mycobacteria (BCG-lux or Mtb-lux) were pre-incubated in medium with or without 10% serum, prior binding to R6F-SR-B1 cells also in the presence or absence of 10% serum. Shown is % of luciferase activity relative to binding to R6F-SR-B1 cells in the absence of serum, which was set at 100%. (C) Effect of LDL on binding of mycobacteria to R6F cells expressing SR-B1 or SIGNR1, as indicated. Shown is % of luciferase activity relative to R6F-SR-B1 or R6F-SIGNR1 cells, respectively, in the absence of additives. (D) Effect of BAL fluid and serum (as a control) on binding of BCG-lux or Mtb-lux to SR-B1 expressing R6F cells. Shown is % of luciferase activity relative to R6F-SR-B1 in the absence of additives. (E) Effect of anti-SR-B1 antibodies on binding of BCG-lux or Mtb-lux to SR-B1 or SIGNR1 transfected R6F cells. The white bars show % of luciferase activity relative to binding in the absence of antibody (black bars). (F) SR-B1 expressing R6F cells were incubated with mycobacteria in the presence of increasing concentrations of MβCD. Binding of BCG-lux is shown as % of luciferase activity relative to control (no MβCD).

Figure 3. The role of SR-B1 in mycobacterial recognition in primary cells.

(A) Western Blot assessing expression of SR-B1 in different macrophage populations derived from wild type C57/BL6 and SR-B1^−/− mice, as indicated. (B) Alveolar macrophages and BMDmØ isolated from wild type (black bars) and SR-B1^−/− mice (white bars) were tested for BCG-lux binding in the absence of serum. Shown is the relative luciferase activity (R.L.U.), normalised to cell number by CFSE staining. (C) TNF production at 4 hr or 24 hr after binding of BCG-lux to alveolar macrophages or BMDmØ isolated from wild type (black bars) and SR-B1^−/− mice (white bars). Shown is x-fold increase in TNF production relative to control (no BCG binding). Experiments were normalised to cell number by CFSE staining. (D) Survival of BCG-lux in infected BMDmØ isolated from wild type (solid lines) and SR-B1^−/− mice (dashed lines). After infection of macrophages, unbound mycobacteria were removed and samples taken after the indicated time points to assess luciferase activity. Experiments were performed in triplicate, normalised to cell number by CFSE staining and shown as R.L.U. relative to time point 1.

Figure 4. The role of SR-B1 in vivo following low dose infection with Mtb.

Wild type C57/BL6 and SR-B1^−/− mice were infected with 100 CFU Mycobacterium tuberculosis H37Rv by aerosol route and sacrificed after 2 and 4 months. (A) Body weight was monitored throughout the course of the experiment and is presented as % of original body weight (time of infection) with the SEM shown as solid lines for wild type animals and dashed lines for SR-B1^−/− mice. Lungs of infected animals were analysed at 2 and 4 months for histopathology by H&E staining (B). Also shown is a lung section showing the presence of clefts of accumulated cholesterol that were observed in some inflammatory foci in infected SR-B1^−/− mice. Lung sections after 4 months of infection were further anlaysed by morphometric analysis to calculate the sizes of the inflammatory lesions (C). Lung homogenates were analysed at 2 and 4 months for bacterial burden (D), as well as the production of TNF, IFNγ, IL10, IL12p70 and IL6 (E) with the black circles representing wild type and the open circles representing SR-B1^−/− animals. Shown are the data from individual mice and the median value.

Figure 5. The role of SR-B1 in vivo following high dose infection with Mtb.

Wild type C57/BL6 and SR-B1^−/− mice were infected with 1000 CFU Mycobacterium tuberculosis H37Rv by aerosol route and sacrificed after 4 months. (A) Body weight was monitored throughout the course of the experiment and is presented as % of original body weight (time of infection) with the SEM shown as solid lines for wild type animals and dashed lines for SR-B1^−/− mice. Lungs of infected animals were analysed for bacterial burden (B), TNF, IFNγ, IL10, IL12p70 and IL6 (C), histopathology (D) and inflammatory lesion size (E), with the black circles representing wild type and the open circles representing SR-B1^−/− animals. Shown are the data from individual mice and the median value. *, p<0.05.

Figure 6. The effect of cholesterol in SR-B1^−/− mice during Mtb infection.

Wild type C57/BL6 and SR-B1^−/− mice were fed either a low cholesterol diet (LC, 0.15% cholesterol) or a high cholesterol diet (HC, 1.25% cholesterol) throughout the experiment, infected with 100 CFU Mycobacterium tuberculosis H37Rv by aerosol route and sacrificed after 4 months. (A) Levels of serum cholesterol following 2 weeks on the various diets, as indicated, prior to infection (black bars representing wild type, and white bars the SR-B1^−/− animals). (B) Average mouse weight throughout the course of the experiment, presented as % of original body weight ±SEM. 4 months after infection the mice were analyzed for serum cholesterol (C), histopathology (D), bacterial burdens (E), and pulmonary TNF, IFNγ, IL10, IL12p70 and IL6 (F) with the black circles representing wild type and the open circles representing SR-B1^−/− animals. Shown are the data from individual mice and the median value. *, p<0.05.