The dual role of asporin in breast cancer progression

Abstract

Asporin has been reported as a tumor suppressor in breast cancer, while asporin-activated invasion has been described in gastric cancer. According to our in silico search, high asporin expresion associates with significantly better relapse free survival (RFS) in patients with low-grade tumors but RFS is significantly worse in patients with grade 3 tumors. In line with other studies, we have confirmed asporin expression by RNA scope in situ hybridization in cancer associated fibroblasts. We have also found asporin expression in the Hs578T breast cancer cell line which we confirmed by quantitative RT-PCR and western blotting. From multiple testing, we found that asporin can be downregulated by bone morphogenetic protein 4 while upregulation may be facilited by serum-free cultivation or by three dimensional growth in stiff Alvetex scaffold. Downregulation by shRNA inhibited invasion of Hs578T as well as of CAFs and T47D cells. Invasion of asporin-negative MDA-MB-231 and BT549 breast cancer cells through collagen type I was enhanced by recombinant asporin. Besides other investigations, large scale analysis of aspartic acid repeat polymorphism will be needed for clarification of the asporin dual role in progression of breast cancer.

Keywords: 3D cultivation; asporin; breast cancer; grade; stiffness.

Figure 1. Asporin expression has different prognostic value in low-grade and advanced breast cancer

A. High asporin expresion associates with significantly better relapse free survival (RFS) in patients with low-grade tumors but RFS is significantly worse in patients with grade 3 tumors. B. High asporin expression is associated with significantly worse RFS both in estrogen receptor positive and negative grade 3 tumors, in particular with metastasis to lymph nodes. HR, hazard ratio; N, number of patients in the KMPLOT analysis.

Figure 2. Validation of asporin antibody (Proteinatlas #HPA008435) in breast cancer cells, gingival fibroblasts and human dental pulp stem cells (hDPSCs)

Black arrow indicates strong upregulation of asporin in hDPSCs after odontogenic differentiation. BT-549-ASPN with stable transfected asporin open reading frame sequence display a band with higher molecular weight (a similar band was observed also in MDA-MB-231-ASPN cells, data not shown). Validation of other antibodies is provided in Supplementary Figure S2.

Figure 3. Glycoprotein asporin in Hs578T breast cancer cells

A. Asporin is consistently upregulated by serum starvation (SFM-ctrl, serum-free medium control) and downregulated by two-day treatment with 100 ng/ml bone morphogenetic protein 4 (BMP4). Mild asporin downregulation by TGF-beta was not reproduced in other experiments. BMP4 also decreased phosphorylation of focal adhesion at tyrosine 397 (FAKy397). Inhibition of translation by cotreatment with 10 μg/ml cycloheximide led to downregulation of asporin but not of FAK. All treatments were performed in serum free media for 48 hours. B. Expression of asporin is upregulated by 12-day cultivation of Hs578T cells in Alvetex polystyrene scaffold (3D) in comparison to normal 2D conditions. C. Formallin-fixed paraffin-embedded Hs578T cells (stained with hematoxylin-eosin) in polystyrene Alvetex scaffold (transparent stellar shapes; magnification 1000x). D. Four-day cultivation of Hs578T cells in hypoxic conditions does not consistently change asporin expression. E. Expression of asporin is successfully downregulated with three different shRNAs. F. Invasion of Hs578T cells (parental cells are displayed in green, control cells with scrambled shRNA are in red) through collagen matrix is inhibited by shRNAs 17 and 19 (blue and purple, respectively). Similar inhibition was observed also for the shRNA 18 (data not shown). Reading of cell index was every 10 minutes up to 42 hours with the real time cell monitoring xCELLigence instrument (error bars indicate standard deviations from quadruplicate measurement). All experiments were performed at least three times and representative blots/chart are shown.

Figure 4. Recombinant asporin enhances invasion of MDA-MB-231 A. and BT-549 B. breast cancer cells through collagen matrix

Both models were seeded at a density of 75000 cells per well into CIM plates coated by collagen stiffened in presence (green) or absence (red) of 10 nM recombinant asporin. The bottom chambers were filled with complete medium containing 10% FCS (fetal calf serum) as a chemoattractant. Reading of cell index was every 10 minutes up to 42 hours with the real time cell monitoring xCELLigence instrument (error bars indicate standard deviations from quadruplicate measurement). All experiments were performed at least three times and representative charts are shown.

Figure 5. Asporin is detected by RNA scope in situ hybridization in invasive breast cancer

Human dental pulp stem cells before A. and after B. odontogenic differentiation, as well as endometrium C., were used as positive controls. Specific dot-like positivity was found in cancer associated fibroblasts both in ductal D. and lobular E-F. invasive carcinomas (black arrows). Asporin positivity was rarely observed also in cancer cells (F, white arrow). Magnification is 1000x while lower magnification (400x) of the same samples is provided in the Supplementary Figure S5.

Figure 6. Invasion of breast cancer associated fibroblasts and T47D cells is attenuated by asporin downregulation

A. Transwell experiment indicate coordinated invasion of cancer associated fibroblasts with T47D cells through extracellular matrix composed of collagen, matrigel and crosslinking ribose (70000 CAFs, green; 70000 T47D, red; CAFs + T47D, 35000 each, blue; 35000 CAFs, purple). Western blot analysis shows downregulation of asporin by shRNA B. and its upregulation upon 3D cultivation of CAFs in Alvetex polystyrene scaffold C. D. Downregulation of asporin attenuated invasion of CAFs and T47D cells (70000 CAFs shRNA ctrl, green; 70000 CAFs shRNA18, red; CAFs shRNA ctrl + T47D, 35000 each, purple; CAFs shRNA 18 + T47D, 35000 each, blue). All experiments were performed three times and representative blots/charts are shown.