Simultaneously targeting cancer-associated fibroblasts and angiogenic vessel as a treatment for TNBC

Abstract

Fibrotic tumor stroma plays an important role in facilitating triple-negative breast cancer (TNBC) progression and chemotherapeutic resistance. We previously reported a rationally designed protein (ProAgio) that targets integrin αvβ3 at a novel site. ProAgio induces apoptosis via the integrin. Cancer-associated fibroblasts (CAFs) and angiogenic endothelial cells (aECs) in TNBC tumor express high levels of integrin αvβ3. ProAgio effectively induces apoptosis in CAFs and aECs. The depletion of CAFs by ProAgio reduces intratumoral collagen and decreases growth factors released from CAFs in the tumor, resulting in decreased cancer cell proliferation and apoptotic resistance. ProAgio also eliminates leaky tumor angiogenic vessels, which consequently reduces tumor hypoxia and improves drug delivery. The depletion of CAFs and reduction in hypoxia by ProAgio decreases lysyl oxidase (LOX) secretion, which may play a role in the reduction of metastasis. ProAgio stand-alone or in combination with a chemotherapeutic agent provides survival benefit in TNBC murine models, highlighting the therapeutic potential of ProAgio as a treatment strategy.

Graphical abstract

Figure 1.

Abundant stroma rich in integrin α_Vβ₃-expressing CAFs and collagen is associated with poor TNBC patient survival. (A) Representative images of H&E staining, Sirius red staining, α-SMA, and FAP IHC staining of low stroma (upper panel) and high stroma (lower panel) in the breast tumors of TNBC patients (n = 80). Scale bars, 200 µm. (B and C) Kaplan-Meier overall survival (B) and DFS (C) analyses of low and high COL1A1 expression (COL1A1 low, n = 44; COL1A1 high, n = 38) in the tumors of TNBC patients. (D) Representative images of IHC staining of integrin β₃ in the breast tumor of TNBC patients (lower panel, n = 80; scale bar, 500 µm) compared with adjacent normal breast tissue (upper panel, n = 3; scale bar, 100 µm). (E) Levels of integrin α_V (IB: Integrin α_V), integrin β₃ (IB: Integrin β₃), and α-SMA (IB: α-SMA) in the nonactivated (−TGF-β) and activated (+TGF-β) hMFs were analyzed by immunoblot. hMFs were activated by culturing for 48 h in the presence of 5 ng/ml TGF-β. Nonactivated hMFs are the cells that are cultured for 1 d without TGF-β. (F) Representative Co-IF images of integrin β₃ (green) and α-SMA (red) in activated hMFs (upper panel) and murine CAFs (lower panel). Nuclei were counterstained with Hoechst (blue). Scale bars, 100 µm. (G and H) Kaplan-Meier overall survival analyses (G) of ITGB3 (integrin β₃: low, n = 50; high, n = 32) and DFS analyses (H) of ITGB3 (integrin β₃: low, n = 45; high, n = 30) in the tumors of TNBC patients. (I) Co-immunoprecipitation (coIP) of caspase 8 with integrin β₃ (IP: Integrin β₃) was analyzed by immunoblot (IB: caspase 8). Activated hMFs were treated with 5 µM ProAgio for 3 h. Immunoblot of integrin β₃ (IB: Integrin β₃) indicates the amount of β₃ that was coprecipitated down in the coIP. Input represents 10% of the total protein used for IP. (J) Apoptosis in murine CAFs (left panel) and activated hMFs (right panel) that were treated with 5 µM ProAgio (red bar) compared with vehicle (PBS, gray bar) was analyzed by apoptosis kit. TCGA dataset of TNBC patients was obtained from cbioportal (https://www.cbioportal.org). HR, hazard ratio. Immunoblot of β-actin (IB: β-actin) in E and I is a loading control. Experiments were performed at least in triplicate. Error bars in J represent mean ± SEM. *, P < 0.05; **, P < 0.01; ****, P < 0.0001 by log-rank test (B, C, G, and H) or unpaired Student’s t test (J).

Figure S1.

High stroma is associated with poor overall survival and DFS; breast cancer CAFs express high levels of integrin α_vβ₃; there is a close correlation between integrin β₃, α-SMA, and FAP levels in TNBC patient tissue samples; and ProAgio induces apoptosis in breast cancer CAFs. (A–D) Kaplan-Meier overall survival analyses of low and high COL1A2 (low, n = 34; high, n = 48; A) and FAP (low, n = 39; high, n = 43; C) and DFS of low and high COL1A2 (low, n = 40; high, n = 35; B) and FAP (low, n = 46; high, n = 29; D) gene expression in the tumors of TNBC patients. HR, hazard ratio. TCGA dataset of TNBC patients was obtained from cbioportal (https://www.cbioportal.org). (E) Level of integrin β₃ (IB: Integrin β₃) in hMFs and hbCAFs was analyzed by immunoblot. GAPDH is a loading control. (F) Representative images of IF costaining of α-SMA (red) and integrin β₃ (green) in the tumor sections of 4T1 mice. Nuclei were counterstained with DAPI (blue). Scale bar, 100 µm. (G) Schematic illustration of the MACS technique. (H and I) The mRNA levels (H) and quantification of integrin α_v (ITGAV) and integrin β₃ (ITGB3; I) in the indicated magnetically sorted cells from MMTV-PyMT breast tumor tissue were analyzed by RT-PCR (n = 2). β-actin is a loading control. (J) Correlation of integrin β₃ protein expression with α-SMA/FAP protein expression. Regression analysis for expression of the integrin β₃ (IHC) versus α-SMA/FAP (IHC) in tumor sections of TNBC patients (n = 18). (K) Co-immunoprecipitation (coIP) of caspase 8 with integrin β₃ (IP: Integrin β₃) was analyzed by immunoblot (IB: Caspase 8). Nonactivated hMFs (NA) and activeted hMFs (A) were treated with 5 µM ProAgio for 3 h. Immunoblot of integrin β₃ (IB: Integrin β₃) indicates the amount of β₃ that was coprecipitated down in the coIP. Input represents 10% of the total protein used for IP. (L) Levels of CC8 (IB: C-Caspase 8) in hbCAFs upon treatment with vehicle or 10 µM ProAgio for 3 h were analyzed by immunoblot. Immunoblot of GAPDH is a loading control. (M) Apoptosis in hbCAFs that were treated with 10 µM ProAgio (red bar) compared with vehicle (PBS, gray bar) was analyzed by apoptosis kit. (N) Representative images of Co-IF staining of vinculin (green) and rhodamine phalloidin (actin, red) of activated hMFs treated with vehicle or ProAgio. Nuclei were counterstained with DAPI (blue). Scale bars, 100 µm. All data are representative of at least three independent biological replicates. Error bars in M represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001 by log-rank test (A–D) or unpaired Student’s t test (M).

Figure S2.

ProAgio does not induce apoptosis in TNBC cell lines; ProAgio decreases cancer cell proliferation in mouse models; and ProAgio reduces levels of α-SMA by inducing apoptosis in breast cancer CAFs and mediates apoptosis in cancer cells. (A) Levels of integrin α_V (IB: Integrin α_V) and integrin β₃ (IB: Integrin β₃) in the cell lysates of indicated TNBC cell lines were analyzed by immunoblot. β-actin is a loading control (n = 2). (B) Cell viability of indicated cells upon treatment with indicated concentrations of ProAgio for 24 h was analyzed by MTT assay. Experiments were performed at least in triplicate. (C–F) Representative images of IF staining (C, green) or IHC staining of Ki67 (E) and quantification of percentage of Ki67-positive cells in the sections of 4T1 tumor (D) or Ki67⁺ area in the sections of MDA-MB-231 tumor (F) upon treatment with vehicle or ProAgio. Insets, ×1.5. n = 6–8/group. Scale bars, 100 µm. (G and H) Representative images of IF costaining of pancytokeratin (red) and Ki67 (green; G) and quantification of both Ki67⁺pancytokeratin⁺ area (H) in the tumor sections of 4T1 mice treated with vehicle or ProAgio. Nuclei were counterstained with DAPI (blue); n = 5/group. Scale bar, 100 µm. (I) mRNA levels of α-SMA (left panel) and FAP (right panel) measured by qRT-PCR in the tumors of 4T1 mice treated with vehicle (gray bar) or ProAgio (red bar; n = 4/group). (J and K) Representative images of three-color IF staining of pancytokeratin (PanCK, white), CC3 (green), and α-SMA (red; J) and quantification of CC3⁺pancytokeratin⁺ and CC3⁺α-SMA⁺ area (K) in tumor sections of 4T1 mice treated with vehicle or ProAgio. Nuclei were counterstained with DAPI (blue); n = 5–7/group. Scale bar, 50 µm. (L) Levels of integrin α_V (IB: Integrin α_V), integrin β₃ (IB: Integrin β₃), CC3 (IB: CC3), and CC8 (IB: C Caspase 8) in the 4T1 tumor tissue lysates of mice treated with vehicle or ProAgio were analyzed by immunoblot (n = 5). GAPDH is the loading control. Error bars in B and I represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns denotes nonsignificant by unpaired Student’s t test.

Figure 2.

ProAgio reduces breast tumor growth and prolongs survival in tumor-bearing mice. (A and B) Therapy regimen of ProAgio (i.p.; indicated by red arrows) in 4T1 (A) and MDA-MB-231 (B) mice. (C–F) Kaplan-Meier survival analysis (n = 18; median survival: vehicle, 22 d; ProAgio, 26 d; C), mean tumor volume (D), mean tumor weight (E), and representative images of tumor (F) of 4T1 tumor-bearing mice treated with vehicle or ProAgio (vehicle, n = 15; ProAgio, n = 30). (G–J) Kaplan-Meier survival analysis (n = 8; median survival: vehicle, 29 d, ProAgio, 41 d; G), mean tumor volume (n = 10; H), mean tumor weight (n = 6; I), and gross images of breast tumor (J) of MDA-MB-231 female nude mice treated with vehicle or ProAgio. (K) Therapy regimen of 20 mg/kg ProAgio (i.p.; indicated by red arrows) in MMTV-PyMT mice. (L–O) Kaplan-Meier survival analysis (n = 12; median survival: vehicle, 27 d; ProAgio, 42.5 d; L), mean tumor burden (n = 12; M), average tumor weight (n = 6; N), and representative images showing gross appearance of tumors where dotted lines demarcate tumor masses (O) of MMTV-PyMT mice treated with vehicle or ProAgio. Error bars represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired Student’s t test.

Figure 3.

ProAgio decreases CAFs and collagen in TNBC mouse models by inducing apoptosis in breast cancer CAFs. (A and B) Representative images of IF staining of α-SMA (A) and quantification of α-SMA⁺ area (B) in the tumors of 4T1 mice treated with vehicle or ProAgio; n = 6/group. Scale bars, 100 µm. (C and D) Representative images of IHC staining of α-SMA (C) and quantification of α-SMA⁺ area (D) in the tumors of MDA-MB-231 mice treated with vehicle or ProAgio; n = 6/group. Scale bar, 100 µm. (E and F) Representative images of Sirius red staining (E) and quantification of collagen area (F) in the tumors of 4T1 mice treated with vehicle or ProAgio; n = 7/group. Scale bars, 100 µm. (G and H) Representative images of Sirius red staining (G) and quantification of collagen area (H) in the tumors of MDA-MB-231 mice treated with vehicle or ProAgio; n = 6/group. Scale bars, 100 µm. (I–K) Representative images of IF costaining of α-SMA (red) and CC3 (green; I), quantification of apoptotic cells (CC3⁺ area; J), and apoptotic breast cancer CAFs (α-SMA⁺ CC3⁺ area; K) in tumor sections of 4T1 mice treated with vehicle or ProAgio. Nuclei were counterstained with DAPI (blue); n = 5–6/group. Scale bars, 100 µm. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired Student’s t test.

Figure S3.

ProAgio reduces levels of GFs and signaling downstream; the addition of GFs or CAF-conditioned media mediates apoptotic resistance in cancer cells; and ProAgio induces apoptosis in breast CAFs and therefore sensitizes cancer cells to apoptotic induction by a chemotherapeutic agent. (A and B) Representative images of IHC staining of PDGF-BB, EGF, and IGF1 (A) and quantitative analysis of PDGF-BB, EGF, and IGF1-positive area (B) in the tumor sections of 4T1 mice treated with vehicle or ProAgio. Scale bars, 100 µm; n = 3–4/group. (C) Intratumoral levels of PDGF-BB, EGF, and IGF1 were determined by ELISA assay in the tumor extracts of 4T1 mice (n = 4/group). (D and E) Representative images of IF staining of pPDGFRβ, pEGFR, IHC staining of pIGF1R, pAKT, and pERK (D), and quantitative analysis of pPDGFRβ, pEGFR, pIGF1R, pAKT, and pERK-positive cells (E) in the tumor sections of 4T1 mice treated with vehicle or ProAgio. In IF images, nuclei were counterstained with DAPI (blue). The quantifications are presented as fold change using the ProAgio treatment group as a reference. Scale bars, 100 µm; n = 4–5/group. (F–K) Cell viability of TNBC cells, including MDA-MB-231 (F and G), 4T1 (H and I), BT549 (J), and HCC1806 (K) upon treatment with indicated concentrations of indicated drugs for 48 h in the culture media containing the vehicle (i.e., PBS [black line] or indicated GFs [red line]). (L and M) Cell viability of MDA-MB-231 (L) and 4T1 (M) cells upon treatment with indicated concentrations of DOX for 48 h in the culture media containing the conditioned medium from activated breast fibroblasts (blue line) or the indicated GFs (red line) compared with the vehicle (black line). (N) Schematic of the experimental setup for transwell for co-culture assay. (O) Levels of pERK and pAKT in MDA-MB-231 cells co-cultured with activated breast fibroblasts (CAFs) treated with vehicle or ProAgio were analyzed by immunoblot. β-actin is the loading control. (P) Cell viability of MDA-MB-231 cells upon treatment with indicated concentrations of DOX for 48 h in the activated human breast fibroblasts (CAFs) and MDA-MB-231 cells (red line) or the pretreated activated human breast fibroblasts (CAFs) and MDA-MB-231 cells with ProAgio (blue line) compared with vehicle (black line). ProAgio (5 µM) was added to the activated human breast fibroblasts (CAFs) for 6 h followed by DOX treatment for 48 h. Cell viability was analyzed by MTT assay. All data are representative of at least three independent biological replicates. Error bars represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns denotes nonsignificant by unpaired Student’s t test.

Figure 4.

ProAgio reduces metastasis to the lungs by modulating fibrotic stroma at the lung metastatic site. (A and B) Representative images of H&E staining of the lungs (A) and quantification of number of lung metastatic nodules (B) in tumor sections of 4T1 mice treated with vehicle or ProAgio (vehicle, n = 13; ProAgio, n = 22).Scale bar, 100 µm. (C and D) Representative images of H&E staining of the lungs (C) and quantification of the number of lung metastatic nodules (D) in tumor sections of MMTV-PyMT mice treated with vehicle or ProAgio. Scale bar, 4 mm. (E–G) Representative images of IHC staining of α-SMA (upper panel) or Sirius red staining (lower panel; E) and quantification of α-SMA⁺ area (F) and collagen area (G) in the sections of metastatic lungs of 4T1 mice treated with vehicle (gray bar) or ProAgio (red bar). Normal represents the lung sections from mice without a tumor and treatment (open bar); n = 8/group. Scale bars, 100 µm. (H and I) Representative images of Co-IF of α-SMA (red) and CC3 (C-caspase 3; green; H) and quantification of apoptotic α-SMA⁺ CAFs (α-SMA⁺ CC3⁺ area; I) in the sections of metastatic lungs of 4T1 mice treated with vehicle or ProAgio. Nuclei were counterstained with DAPI (blue). White arrowheads indicate colocalization; n = 5/group. Scale bar, 100 µm. (J) Levels of integrin α_V (IB: Integrin α_V), integrin β₃ (IB: Integrin β₃), and α-SMA (IB: α-SMA) in the nonactivated (−TGF-β) and activated (+TGF-β) hLFs were analyzed by immunoblot. hLFs were activated by culturing for 48 h in the presence of 5 ng/ml TGF-β. Nonactivated hLFs are the cells that are cultured for 1 d without TGF-β. β-actin is a loading control. Data are representative of at least three independent biological replicates. (K) Therapy regimen of the 4T1 spontaneous model. On day 15, the mice were sacrificed and the lungs were collected. (L and M) Representative images of H&E staining (L) and quantification of tumor nodules (M) in lung sections of the vehicle- or ProAgio-treated animals. Scale bars, top panel: 1,000 µm; lower panel: 100 µm. Error bars represent mean ± SEM. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired Student’s t test.

Figure S4.

ProAgio treatment reduces the number and size of metastatic lung nodules in mouse models, and ProAgio treatment induces apoptosis in integrin α_vβ₃–expressing activated lung fibroblasts and decreases α-SMA and collagen levels in metastatic lungs in mouse models. (A–C) Representative images of H&E staining of lungs (A) and quantification of metastatic lung nodule number (B) and lung metastatic nodule diameter (C) of MDA-MB-231 mice treated with vehicle or ProAgio (n = 6/group). (D and E) Quantification of the lung metastatic nodule diameter of 4T1 orthotopic (D) and MMTV-PyMT (E) mice treated with vehicle or ProAgio (n = 5 or 6/group). (F) Levels of integrin α_V (IB: Integrin α_V), integrin β₃ (IB: Integrin β₃), CC3 (IB: C Caspase 3), CC8 (IB: C Caspase 8), and CK19 (IB: CK19) in the lung tissue lysates of 4T1 mice treated with vehicle or ProAgio were analyzed by immunoblot (n = 4). GAPDH is the loading control. (G) Representative images of IF staining of integrin α_Vβ₃ (green), α-SMA (red), and vimentin (green) in the activated hLFs. Nuclei were counterstained with DAPI (blue). Scale bar, 100 µm. (H) Apoptosis in activated lung fibroblasts that were treated with 5 µM ProAgio was analyzed by Annexin V kit. (I) Co-immunoprecipitation (coIP) of caspase 8 with integrin β₃ (IP: Integrin β₃) was analyzed by immunoblot. hLFs were activated with 5 ng/ml TGF-β for 48 h and subsequently treated with 5 µM ProAgio for 4 h before preparation of the extract. Immunoblot of integrin β₃ (IB: Integrin β₃) indicates the amount of integrin β₃ that was precipitated down in coIP. IgG is the loading control. Input represents 5% of the total protein used for IP, and β-actin is the loading control. (J) coIP of caspase 8 with integrin β₃ (IP: Integrin β₃) was analyzed by immunoblot (IB: Caspase 8). Nonactivated hLFs (NA) and activated hLFs (A) were treated with 5 µM ProAgio for 3 h. Immunoblot of integrin β₃ (IB: Integrin β₃) indicates the amount of β₃ that was coprecipitated down in the coIP. Input represents 10% of the total protein used for IP. Experiments were performed in triplicate. (K) Quantification of the lung metastatic nodule diameter from the lung sections of the 4T1 spontaneous mouse model (n = 5/group). (L–O) Representative images of IHC staining of α-SMA (L), Sirius red staining of collagen (N), and quantification of α-SMA–positive area (M) and collagen-positive area (O) in the lung sections of vehicle- or ProAgio-treated 4T1 spontaneous mice (tail vein model) compared with normal mice. Scale bars, 100 µm; scale bars in N, 1,000 µm (left panel); n = 5–6/group. Error bars represent mean ± SEM. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired Student’s t test.

Figure 5.

ProAgio eliminates angiogenic vessels, enhances blood perfusion into the tumor, and consequentially reduces hypoxia. (A–D) Representative images of IF CD31 staining (A), quantification of vessel number (B), number of branch points (C), and vessel length (D) per view field in tumor sections of 4T1 mice treated with vehicle or ProAgio. Nuclei were counterstained with DAPI (blue). Scale bars, 100 µm (n = 6–7/group). (E and F) Representative IHC images of CD31 staining (E) and quantification of CD31⁺ area (F) in the breast tumor sections of MMTV-PyMT mice treated with vehicle or ProAgio. Insets, ×1.5.Scale bars, 100 µm (n = 5/group). (G and H) Representative fluorescence images of lectin perfusion in tumor vessels (G) and quantification of lectin-positive area (red) per total CD31-positive area (H; green, represented by white arrows) in tumor sections of 4T1 mice treated with vehicle or ProAgio.Scale bars, 100 µm (n = 4/group). (I and J) Representative IF images of dextran leakage of tumor vessels (J) and quantification of dextran⁺ area (I) in the breast tumor sections of 4T1 orthotopic mice. Dextran⁺ area is presented as a percentage per total section of the CD31⁺ area. Scale bar, 100 µm (n = 4/group). (K) Intratumoral levels of PTX in the extracts of tumors of 4T1 mice treated with vehicle or ProAgio. PTX levels were presented as nanograms of PTX per milligrams of tumor tissue (n = 4/group). (L and M) Representative fluorescence images of FITC-conjugated PTX (∼1 kD; L), and quantification of fluorescence signals (M) in the tumor sections of 4T1 mice treated with vehicle or ProAgio. Tumors were harvested at the end of the experiment (n = 3 or 4/group). Scale bars, 100 µm. (N and O) Representative images of IHC staining of hypoxyprobe-1 (N, left panels; scale bar, 500 µm; n = 4/group) and IF staining of Hif-1α (N, right panels; scale bars, 200 µm; n = 4/group) and quantification of hypoxic area (O) in the tumor sections of 4T1 mice treated with vehicle or ProAgio. (P and Q) Representative images of IHC staining of HIF-1α (P) and quantitative analysis of HIF-1α (Q) in the breast tumors of MMTV-PyMT mice treated with vehicle or ProAgio. Scale bar, 100 µm. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by unpaired Student’s t test.

Figure 6.

ProAgio decreases hypoxia-induced LOX secretion in the breast tumor in mouse models. (A–C) Cellular levels of LOX mRNA in 4T1 cells (A), MDA-MB-231 cells (B) in normoxia (gray bar) or hypoxia (blue bar) condition, or hMFs nonactivated (−TGF-β, gray bar) or activated (+TGF-β, red bar) by TGF-β (C) were analyzed by qRT-PCR. All data are representative of at least three independent biological replicates. (D and E) Representative images of IF staining of LOX (D) and quantification of LOX staining (E) in the tumors of 4T1 mice treated with vehicle or ProAgio. Nuclei were counterstained with DAPI (blue). Scale bar, 100 µm (n = 8/group). (F and G) Representative images of IHC staining of LOX (F) and quantification of LOX staining (G) in the breast tumors of MMTV-PyMT mice treated with vehicle or ProAgio. Scale bars, 100 µm (n = 6/group). (H and I) mRNA analysis of LOX (H), LOXL2 (left panel, I), and LOXL4 (right panel, I) measured by qRT-PCR in the tumors of 4T1 mice treated with vehicle (gray bar) or ProAgio (red bar). n = 4/group. (J and K) Quantification of LOX activity in the tumor extracts of 4T1 mice (J) and MMTV-PyMT mice (K) treated with vehicle or ProAgio (n = 5–6/group). (L–O) Representative images of Sirius red staining (left panels) and the corresponding cross-linked collagen images by polarized microscopy on Sirius red staining (right panels; L) and quantitative analysis of cross-linked collagen (M and N) in the breast tumors of 4T1 orthotopic (upper panels; n = 7/group) and MMTV-PyMT (lower panels; n = 5/group) mice treated with vehicle or ProAgio. Scale bars, 100 µm. (O) Quantification of cross-linked collagen per total collagen area in the breast tumor sections of 4T1 (left panel) and MMTV-PyMT mice (right panel) treated with vehicle (gray bar) or ProAgio (red bar); n = 4/group. Error bars in A, B, C, H, I, and K represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns denotes nonsignificant by unpaired Student’s t test.

Figure 7.

ProAgio decreases hypoxia-induced LOX-mediated metastasis to the lungs in mouse models. (A and B) Quantification of LOX activity in the serum of 4T1 orthotopic mice (n = 5–6/group; A) and MMTV-PyMT mice (n = 6–8/group; B) treated with vehicle or ProAgio. (C–F) Representative images of IHC staining of LOX (C and E) and quantification of LOX (D and F) in the lung metastatic tumors of 4T1 mice (C and D) and MMTV-PyMT mice (E and F) treated with vehicle or ProAgio (n = 6/group). Scale bars, 100 µm. (G and H) Representative images of Sirius red staining visualized under brightfield (upper panel) or polarized light microscopy (lower panel; G) and quantitative analysis of cross-linked collagen (under polarized light; H) in the lung metastatic tumors of 4T1 mice treated with vehicle or ProAgio. Scale bar, 100 µm (n = 7/group). (I–K) Representative images of Sirius red staining for collagen (left panel) and its corresponding cross-linked images by polarized microscopy on Sirius red staining (right panel; I) and quantification of collagen (J) and cross-linked collagen (K) in the metastatic lungs of MMTV-PyMT mice treated with vehicle or ProAgio (n = 7/group). (L–O) Representative flow cytometric plots of CD11b (L and N) and the population of CD11b-positive cells (M and O) in the metastatic lungs of 4T1 mice (n = 4–5/group; M) and MMTV-PyMT mice (n = 5 or 6/group; O) treated with vehicle or ProAgio. SSC, side scatter. (P) Representative IHC images of low and high LOX staining in the breast tumors of TNBC patients (n = 80). Scale bars, 100 µm. (Q) Kaplan-Meier survival analysis of low and high LOX (low, n = 34; high, n = 48) gene expression in the breast tumors of TNBC patients. HR, hazard ratio. TCGA dataset of TNBC patients was obtained from cbioportal (https://www.cbioportal.org). Error bars in B, M, and O represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 by log-rank test (Q) or by unpaired Student’s t test.

Figure 8.

ProAgio enhances chemotherapeutic efficacy in 4T1 orthotopic and MMTV-PyMT mouse models. (A) Therapy regimen of ProAgio (red arrows) and PTX (green arrows, i.p. 3 mg/kg) in 4T1 mice. (B) Representative images of tumors of 4T1 mice treated with indicated agents. (C–E) Kaplan-Meier survival plot (n = 8–11; median survival [days]: vehicle = 17, ProAgio = 21, PTX = 20, ProAgio + PTX = 26.5; C), mean tumor volume (n = 8–12; D), and mean tumor weight (E) of 4T1 mice treated with indicated agents (n = 5–8). (F) Therapy regimen of ProAgio (red arrows) and PTX (green arrows, i.p. 5 mg/kg) in MMTV-PyMT mice. (G) Representative images showing appearance of tumors indicated by dotted circles on MMTV-PyMT mice treated with indicated agents. (H and I) Kaplan-Meier survival analysis (n = 18–21; median survival [days]: vehicle = 32, ProAgio = 43, PTX = 33, ProAgio + PTX = 48; H), mean tumor burden (total volumes of all tumors; I) of MMTV-PyMT mice treated with indicated agents (n = 6–8). (J and K) Representative images of IHC staining and quantitative analyses of α-SMA (J) and Sirius red staining and quantitative analyses of collagen (K) in the breast tumors of MMTV-PyMT mice treated with indicated agents. Scale bars, 100 µm (n = 4–6/group). Error bars in D, E and I represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****; P < 0.0001; ns denotes nonsignificant by log-rank test (C and H) or unpaired Student’s t test.

Figure S5.

ProAgio enhances the effect of chemotherapeutics in mouse models. (A) Therapy regimen of 10 mg/kg ProAgio (i.p.) and 3 mg/kg DOX (i.p.) in 4T1 orthotopic mice. (B) Representative gross images of breast tumors of orthotopic 4T1 mice treated with indicated agents. (C–E) Kaplan-Meier survival analyses (n = 11–12; median survival [days]: vehicle = 21, ProAgio = 31, DOX = 23, ProAgio + DOX = 36; C), mean tumor volume (D), and mean tumor weight (E) of orthotopic 4T1 mice treated with indicated agents. (F and G) Mean tumor volume of each nodule (F) and average tumor weight (G) of MMTV-PyMT mice treated with indicated agents (n = 6–8). (H and I) Representative IHC images of CC3 staining (H) and quantification of CC3-positive cells (I) in 4T1 mice treated with indicated agents. Scale bars, 100 µm; n = 8/group. (J and K) Representative H&E staining of lungs (J), and quantification of lung nodules number (K) in 4T1 mice upon treatment with indicated agents; n = 5 or 6/group. Scale bars, 1,000 µm. Error bars represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns denotes nonsignificant by unpaired Student’s t test.

Figure 9.

ProAgio enhances the efficacy of chemotherapeutics. (A and B) Representative images of IHC staining and quantitative analyses of Ki67 (A) and CC3 (B) in the breast tumors of MMTV-PyMT mice treated with indicated agents.Insets, ×3;Scale bars, 100 µm (n = 4–6/group). (C–E) Representative H&E images of lung sections (C), number of metastatic lung nodules by histology (D), and number of metastatic lung nodules on the lung surface (E) of MMTV-PyMT mice treated with indicated agents. Black arrows indicate metastatic nodules in lungs. Scale bars, 1,000 µm(n = 5–6/group). (F) Schematic illustration of action of ProAgio in TNBC. Error bars in D and E represent mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.001; ns denotes nonsignificant by unpaired Student’s t test.