Stabilin-1 is expressed in human breast cancer and supports tumor growth in mammary adenocarcinoma mouse model

Abstract

Stabilin-1 is a multifunctional scavenger receptor expressed on alternatively-activated macrophages. Stabilin-1 mediates phagocytosis of "unwanted-self" components, intracellular sorting, and endocytic clearance of extracellular ligands including SPARC that modulates breast cancer growth. The expression of stabilin-1 was found on tumor-associated macrophages (TAM) in mouse and human cancers including melanoma, lymphoma, glioblastoma, and pancreatic insulinoma. Despite its tumor-promoting role in mouse models of melanoma and lymphoma the expression and functional role of stabilin-1 in breast cancer was unknown. Here, we demonstrate that stabilin-1 is expressed on TAM in human breast cancer, and its expression is most pronounced on stage I disease. Using stabilin-1 knockout (ko) mice we show that stabilin-1 facilitates growth of mouse TS/A mammary adenocarcinoma. Endocytosis assay on stabilin-1 ko TAM demonstrated impaired clearance of stabilin-1 ligands including SPARC that was capable of inducing cell death in TS/A cells. Affymetrix microarray analysis on purified TAM and reporter assays in stabilin-1 expressing cell lines demonstrated no influence of stabilin-1 expression on intracellular signalling. Our results suggest stabilin-1 mediated silent clearance of extracellular tumor growth-inhibiting factors (e.g. SPARC) as a mechanism of stabilin-1 induced tumor growth. Silent clearance function of stabilin-1 makes it an attractive candidate for delivery of immunomodulatory anti-cancer therapeutic drugs to TAM.

Keywords: SPARC; breast cancer; scavenger receptor; stabilin-1; tumor-associated macrophages.

Figure 1. Stabilin-1 expression in human breast cancer

Patient breast cancer samples were analysed by IHC for stabilin-1 (n=31) and CD68 (n=30) expression. A. Representative images from the same tumor regions are shown for TNM stages I, II, III and IV disease. Scale bars: 100 μm. Quantitative IHC analysis for the intensity of stabilin-1 B. and CD68 C. expression for TNM stages I, II, III and IV is shown as a fold change. *p<0.05, Kruskal-Wallis test with Dunn's multiple comparisons.

Figure 2. Heterogeneity of stabilin-1 expression by TAM in human breast cancer

Immunofluorescent staining for co-expression of stabilin-1 with other cellular markers was performed on human breast cancer biopsies and analysed by confocal microscopy. Panel A. shows the area with low stabilin-1 expression on CD68high TAM; panel B. shows region with stabilin-1 expression on CD68low/negative cells; panel C. shows co-expression of stabilin-1 on CD68+ TAM; panel D. demonstrates the area with heterogeneous CD68 and stabilin-1 expression. Co-expression of stabilin-1, CD68, and CD163 is shown on panel E. yellow arrow indicates stabilin-1+CD163+CD68+ triple positive cell, white arrow indicates stabilin-1+CD163-CD68- cell; co-expression of stabilin-1 with CD34, CD31 and pan-cytokeratin is shown on panels F, G and H. respectively.

Figure 3. Stabilin-1 expression in TS/A mouse mammary adenocarcinoma tumor

Wild type (wt) and stabilin-1 knockout (ko) mice were injected s.c. with 5×10⁶ TS/A tumor cells for 21 day. A. IHC analysis of stabilin-1 expression in TS/A tumors from wt and stabilin-1 ko mice. Representative images from 5 tumors are shown. B. Immunofluorescent/confocal microscopy analysis of CD68, CD31, and stabilin-1 expression in wt TS/A tumors 21 day after tumor cell inoculation. Representative images from 3 tumors are shown. Scale bars: 47, 62 μm.

Figure 4. The effect of stabilin-1 on tumor growth and infiltration of TS/A tumors by immune cells

Wild type (wt) and stabilin-1 knockout (ko) mice were injected s.c. with 5×10⁶ TS/A tumor cells. A. Tumor volume (mm³) in wt (n=12) and stabilin-1 ko (n=12) mice was measured 21 days after tumor cell inoculation. The data are presented as box plots with medians indicated, **p<0.01, Mann-Whitney U test. B. IHC analysis of TS/A tumors for the expression of CD31 (vascular marker), F4/80 (macrophage marker), CD3 (T cell marker) and Gr-1 (neutrophil marker) on day 21 after tumor cell inoculation. Representative images from 5 tumors are shown. Scale bars: 100 μm.

Figure 5. Endocytic clearance function is impaired in stabilin-1 knockout TAM

TAM from wt and stabilin-1 ko animals were cultured overnight (15 h) followed by addition of acLDL-Alexa488 A. and SPARC-FITC B. at a final concentration of 5 μg/ml and 10 μg/ml respectively, and incubated for 30 min at 37°C. Quantification of surface bound/internalized ligands was performed using BD FACS Canto II flow cytometer. The data of one representative experiment out of three are presented as mean ± SD, *p<0.05 (n=3 for wt and ko), **p<0.01 (n=8 for wt, n=7 for ko), Student's t-test. C. TAM from wt (n=3) and stabilin-1 ko (n=3) mice were cultured overnight (15 h) followed by addition of SPARC-FITC at a final concentration of 10 μg/ml, and incubated for 30 min at 37°C. Cells were fixed in PFA and subjected to confocal microscopy analysis. Representative images of single cells without nuclear staining are shown. Scale bars: 8,86 μm (left panel) and 8,22 μm (right panel). D. TS/A cells were cultured in vitro and analyzed for SPARC expression using Western Blotting. Normal goat IgG was used as a negative control. E. TS/A tumor sections from 3 wt mice (21 days post-injection) were immunofluorescently stained using goat anti-mouse SPARC antibodies and scanned using Leica TCS SP2 confocal microscope. Representative images are shown. Scale bars: 22,5 μm. F. TS/A cells were cultured without stimulation or in the presence of recombinant SPARC (10 μg/ml) for 48h followed by assessment of percentage of late apoptotic/dead cells (SYTOX green+ and SYTOX Green+/annexin V+ cells) by flow cytometry. The data are mean ± SD of triplicates for one out of two representative experiments, **p<0.01, Student's t-test.

Figure 6. Assessment of stabilin-1 role in the activation of PKCβ gene expression

A. TAM were isolated from wt (n=6) and stabilin-1 ko (n=6) mice and analyzed for PKCβ gene expression on mRNA level using Real-time PCR. Data are mean ± SD for one out of two experiments, **** p<0.0001, Student's t-test. B. TAM were isolated from wt (n=5) and stabilin-1 ko (n=5) mice and analyzed for PKCβ protein expression by Western blotting. Representative image and protein loading controls (right panel) are shown. C. The expression of stabilin-1 in HEK293 cells transfected with empty vector (HEK293-EV) and full-length stabilin-1 (HEK293-Stab1) was demonstrated by flow cytometry. Red histograms indicate isotype controls. D. The endocytosis of acLDL-Alexa488 by two HEK293-EV clones (1,2) and two HEK293-Stab1 clones (3,4) is presented. Data are mean ± SD of triplicates. E. HEK293-EV (n=2) or HEK293-Stab1 (n=2) clones were transfected with luciferase reporter construct containing human PKCβ promoter and stimulated with acLDL (5μg/ml) or left untreated. Luciferase activity was measured 48h after acLDL stimulation. The experiment was repeated 3 times. Data are mean ± SD; ns - not significant. F. HEK293 clones transfected with EV (n=3) or stabilin-1 (n=3) were stimulated with acLDL (5μg/ml) for 48h or left untreated and assessed for expression of endogenous PKCβ by Real-time PCR. Data expressed as mean ± SD, ns-not significant.