Verteporfin synergizes the efficacy of anti-PD-1 in cholangiocarcinoma

Abstract

Background:

Cholangiocarcinoma (CCA) is one of the primary hepatobiliary malignant neoplasms with only 10% of 5-year survival rate. Promising immunotherapy with the blockade of immune checkpoints has no clear benefit in CCA. The inhibition of YAP1 signaling by verteporfin has shown encouraging results by inhibiting cell proliferation and inducing apoptosis. This study aimed to evaluate the potential benefit of the combination of verteporfin and anti-programmed cell death 1 (PD-1) in CCA mouse model.

Methods:

We assessed the cytotoxicity of verteporfin in human CCA cell lines in vitro, including both intrahepatic CCA and extrahepatic CCA cells. We examined the in vitro effect of verteporfin on cell proliferation, apoptosis, and stemness. We evaluated the in vivo efficacy of verteporfin, anti-PD-1, and a combination of both in subcutaneous CCA mouse model.

Results:

Our study showed that verteporfin reduced tumor cell growth and enhanced apoptosis of human CCA tumor cells in vitro in a dose-dependent fashion. Nevertheless, verteporfin impaired stemness evidenced by reduced spheroid formation and colony formation, decreased numbers of cells with aldehyde dehydrogenase activity and positive cancer stem cell markers (all P < 0.05). The combination of verteporfin and anti-PD-1 reduced tumor burden in CCA subcutaneous SB1 tumor model compared to either agent alone.

Conclusions:

Verteporfin exhibits antitumor effects in both intrahepatic and extrahepatic CCA cell lines and the combination with anti-PD-1 inhibited tumor growth.

Introduction

Cholangiocarcinoma (CCA) is a heterogenous cancer derived from the biliary epithelium in the biliary tree system. It is subclassified as intrahepatic CCA (ICC) if it originates within the liver parenchyma and extrahepatic CCA (ECC) if it originates outside of liver [1]. The incidence of CCA has gradually increased across sex and racial/ethnic groups [2,3] and overall prognosis is poor [4], even with recent development of combination chemotherapy [5-7] and targeted therapies [8,9]. Immunotherapy is one of the major components of cancer therapies, especially immune checkpoint inhibitors (ICIs), which target various dysregulated immune checkpoints including programmed cell death 1 (PD-1), and execute anti-tumor effects in various types of cancers [10]. However, ICIs have exhibited limited efficacy for the treatment of CCA [11,12] both as monotherapy and in combination with locoregional therapies [13]. This underlines the urgent demand for the improvement of therapeutic efficacy of ICIs for patients with advanced CCA.

Cancer stem cells (CSCs), also referred to as tumor-initiating or -propagating cells, are tumorigenic cells that can survive cancer directed therapies and facilitate relapse and metastasis. CSCs were responsible for acquired resistance to chemotherapy, radiation therapy and immunotherapy [14-16]. Interestingly, CSCs were found to interact with surrounding immune cells and modify the tumor microenvironment to be immunosuppressive microecosystem [16,17], which results in immunotherapy resistance. Therefore, targeting CSCs has the likelihood to overcome immunotherapy resistance and improve the efficacy of immunotherapy in CCA. Verteporfin (VP) is a photodynamic drug that was approved for treating macular degeneration. VP has demonstrated promising preclinical antitumoral efficacy through the inhibition of YAP1/TAZ pathway [18]. VP was noted to impair CSC characteristics and tumor growth in vivo, including ICC [19]. Here, we described the in vitro effects of VP on ICC and ECC cell lines and showed enhanced antitumoral efficacy in combination with ICI in vivo. These data provide a new platform for translational therapeutics in CCA.

Methods

Cell lines, reagents, and mouse strain

The human ICC cell lines HuCCT-1 and SNU-1079, ECC cell lines TFK-1 and WITT cells, and murine CCA cell line SB1 [20] were used in this study. Among these, WITT cells were cultured in DMEM/F12 (Gibco, Grand Island, NY, USA) with 10% fetal bovine serum (FBS), whereas the rest were cultured in RPMI1640-GlutaMAX™-I medium (Gibco) supplemented with 10% FBS. VP (Sigma-Aldrich, St. Louis, MO, USA) was prepared according to the manufacture’s instruction. Briefly, VP was dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich) and diluted in PBS. Anti-PD-1 was purchased from BioXCell (clone 29F.1A12, Lebanon, NH, USA). C57BF/6 mice were obtained from The Charles River Laboratory. The mice with age of 8-12 weeks were used for experiments. All animals received humane care and experiments were conducted followed with Institution Guidelines and approved by the Animal Care and Use Committee of the NIH, Bethesda, Maryland, USA.

In vitro cytotoxicity, cell proliferation and apoptosis assay

For the assay of VP’s cytotoxicity, HuCCT-1, SNU-1079, TFK-1, WITT and SB1 cells (2000 cells/well) were seeded onto 96-well plates. The cells were precultured for 12 hours before the testing. Cell viability treated with different dose of VP was measured with the CellTiter-Blue Cell Viability Assay kit (Promega, Madison, WI, USA) following the manufacturer’s protocol. Median inhibitory concentration (IC50) was derived based on inhibitory results and calculated per GraphPad Prism software (GraphPad, La Jolla, CA, USA). The effects of VP on cell growth were measured by Cell Counting Kit-8 assay (Dojindo Laboratories, Kumamoto, Japan) following the manufacture’s instruction. Six replicates were performed in each group. The optical density (OD) values at 450 nm were collected using a microplate reader (SpectraMax M5, Molecular Devices, San Jose, CA, USA). Apoptosis was measured by two independent flow cytometry methods: Annexin V-fluorescein isothiocyanate (FITC) & propidium iodide apoptosis detection kit (Sigma-Aldrich) and Annexin V-PE & Apotracker™ Green (BioLegend, San Diego, CA, USA) according to the instructions of the manufacturer. The samples were run and analyzed using CytoFLEX LX platforms (Beckman-Coulter, Indianapolis, IN, USA) and results were analyzed using FlowJo software version 10.4.2 (TreeStar Inc, Ashland, Oregon, USA).

Spheroid and colony formation assay

Single cell suspension of CCA cells were cultured in 6-well ultralow attachment plates (Corning Inc., New York, NY, USA) at a density of 1000 cells/well in spheroid medium. The spheroid medium was prepared with DMEM/F12 medium supplemented in addition with 1X B27 supplement (Gibco) and human recombinant epidermal growth factor (hrEGF) (Gibco) (20 ng/mL), and bFGF (Gibco) (10 ng/mL). After incubating for 7 days, the number of spheroids was counted. The spheroid formation rate was assessed as the ratio of the number of spheroids to the number of cells cultured. For colony formation, 1000 CCA single cell suspension was seeded in a well of 6-well plate and treated with DMSO or VP for 10 days. Each group included triplicate wells. Final cell colonies were stained with crystal violet and counted manually.

Western blotting analysis

Protein samples from cells were extracted using the mammalian protein extraction reagent M-Per (Thermo Scientific, Rockford, IL, USA) supplemented with a protease inhibitor cocktail (Roche, Indianapolis, IN, USA). Antibodies against GADPH (1:5000, Santa Cruz Biotechnology, Santa Cruz, CA, USA), TAZ (1:1000, Cell Signaling), YAP1 (1:2000, Santa Cruz Biotechnology), Bcl-2 (1:2000, Cell Signaling), Bax (1:2000, Cell Signaling), pro-Caspase 3 (1:1000, Cell Signaling) and cleaved-Caspase 3 (1:1000, Cell Signaling) were used to detect individual protein expression. Western blotting imaging system was used for testing individual protein expression. Quantification was completed with ImageJ software version 1.51 (NIH, Bethesda, MD, USA).

Flow cytometry

Flow cytometry was used to detect cell populations with specific CSC surface markers. Briefly, cultured cells were treated with DMSO or VP as indicated, and were dissociated to single cell suspension and washed with cold PBS for two times before incubation with different antibodies for 30 min at 4 °C. Cells were washed one more time before flow cytometry analysis. The following antibodies were used for detecting stem cell markers by flow cytometry analysis: anti-CD24-PB human (BioLegend), anti-CD90-PE/Cy5 human (BioLegend), anti-CD133/2-APC human (Miltenyi Biotec, Milan, Italy), anti-LGR5-PE human (BioLegend), anti-CD44-PE/Cyanine7 human (BioLegend), anti-EpCAM-FITC human (Miltenyi Biotec) and anti-DCAMKL1-Alexa Fluor^® 488 (Abcam, Cambridge, MA, USA).

Aldehyde dehydrogenase (ALDH) activity analysis by flow cytometry was determined by the ALDEFLUOR™ Kit (Stem Cell Technologies, Inc., Cambridge, MA, USA) following the manufacturer’s instructions. Briefly, 1 × 10⁵ single cell suspension was incubated with activated ALDEFLUOR reagent with or without 5 μL DEAB for 1 h at 37 °C. ALDH activity was determined with flow cytometry after incubated cells were washed with ALDH assay buffer.

For mononuclear cell analysis, tumors were removed immediately after mice were sacrificed. After homogenization, debris was removed by filtering samples through nylon mesh. Tumor infiltrating cells were isolated by isotonic Percoll centrifugation (850 × g, 25 min). After red blood cells were lysed with ammonium-chloride-potassium lysing buffer, cells were incubated with indicated antibodies for 30 min at 4 °C. The following antibodies were used for detecting tumor infiltrating lymphocytes (TILs) by flow cytometry analysis: anti-CD3-FITC (clone 17A2, BD Pharmingen, San Jose, CA, USA), anti-CD4-Alexa Fluor 700 (clone GK1.5, BioLegend), anti-CD8-Pacific Blue (clone 53-6.7, BioLegend), anti-TCRbBV510 (clone H57-587, BioLegend), PBS57/CD1d-tetramer-APC (NIH core facility), anti-CD11b-Pacific Blue (clone M1/70, BioLegend), anti-Gr1-PerCP/Cy5.5 (clone RB6–8C5, BioLegend), anti-F4/80-FITC (clone BM8, BioLegend), anti-CD49-Pacific blue (clone GoH3, BioLegend). The TIL subsets were identified by markers: CD4⁺ T cells: CD3^hiCD4⁺; CD8⁺ T cells: CD3^hiCD8⁺; NK cells: TCRβ-CD49b⁺; NKT cells: TCRb⁺CD1d-Tetramer⁺; macrophage: CD11b⁺F4/80⁺; myeloid cells: CD11b⁺Gr1⁺. The absolute number of immune cells was decided by multiplying frequency by the total live cells.

All abovementioned stained cells were then examined using CytoFLEX LX platforms and results were analyzed using FlowJo software version 10.4.2 (TreeStar Inc).

Subcutaneous injection of tumor model in vivo experiments

Eight-week-old C57BL/6 male mice were used for tumor model in vivo experiments. The 1 × 10E6 SB1 cell suspension in 100 mL was injected subcutaneously in the flank. Mice were randomized before starting treatment at day 12. The mice were injected intraperitoneally with bi-weekly IgG control (alternating 10 mg/g and 5 mg/g) (n = 6), weekly anti-PD-1 (10 mg/g, clone 29F.1A12, BioXCell) [20,21] (n = 6), every-3-day VP (100mg/g) [22,23] (n = 6), the combination anti-PD-1 + VP (n = 8), respectively. Tumor size was measured by caliper as greatest diameter and performed by a blinded observer. Mice were sacrificed when tumor size reached 20 mm in diameter or tumor necrosis exceeded 50%.

Statistical analysis

Statistical analyses were performed with GraphPad Prism software (GraphPad). Data were shown as the mean of the three or more independent experiments ± standard error of mean (SEM). Student’s t-test was used. A P < 0.05 was considered statistically significant.

Results

VP inhibits CCA cell growth in a dose-dependent manner

The effect of VP on the cell growth in vitro was tested with two ICC cell lines, two ECC cell lines and mouse CCA cell line SB1 cells. There was a robust reduction in all CCA cell lines with dose-response fashion and the results showed different IC50 in different cell line (2.0 μmol/L in HuCCT-1 cells, 3.0 μmol/L in SNU-1079 cells, 1.4 μmol/L in TFK-1 cells, 1.5 μmol/L in WITT cells, 1.8 μmol/L in SB1 cells, Fig. 1A). The difference of IC50 value among different CCA cell lines may reflect the difference of tolerability and origin of individual cell lines. Similarly, we noted a reduction of tumor cell proliferation of HuCCT-1 (human ICC cell line), TFK-1 (human ECC cell line) and mouse SB1 cells after VP exposure in a dose-dependent manner measured by CCK-8 assay (Fig. 1B). Taken together, these data implied that VP inhibited human CCA cell growth in a dose-dependent fashion.

Fig. 1.

VP suppressed the cell growth of CCA cell lines. A: Cell viability and IC50 of CCA cell lines were determined by CellTiter-Blue Cell Viability Assay after exposed to different concentrations of VP for 72 h. B: Cell growth curves of HuCCT-1, TFK-1 and SB1were decided by measuring cell viability at different time point with CCK-8 assays under different concentrations of VP. Data are expressed as mean ± standard error of mean. ***P < 0.001, compared with DMSO-treated group. VP: verteporfin; CCA: cholangiocarcinoma; IC50: median inhibitory concentration.

VP facilitates apoptosis in CCA cells

VP treatment resulted in significant cell apoptosis of HuCCT-1 cells in a dose-dependent manner. The cell apoptosis rate was significantly augmented in the VP treated groups in comparison to that in the DMSO group (Fig. 2A). The similar pro-apoptosis effect of VP was also observed in TFK-1 and SB1 cells (Figs. S1A and S2A). Moreover, the expression levels of apoptosis regulatory proteins Bcl-2 and pro-Caspase 3 were decreased in HuCCT-1 cells exposed to VP treatment, whereas Bax and cleaved-Caspase 3 were increased in VP treated groups compared to those in the DMSO group (Fig. 2B). Moreover, VP decreased the protein expressions of YAP1 and downstream TAZ, determined by Western blotting (Fig. 2B). These data suggested that VP may induce CCA cell apoptosis through interfering the YAP1 signaling pathway.

Fig. 2.

VP induced apoptosis in CCA cells in vitro. A: Apoptosis was detected with Annexin V-PE/Apotracker™ green and PI/Annexin V-FITC using flow cytometry, respectively, in the HuCCT-1 cells treated with different concentration of VP (DMSO, 1 μmol/L, 2 μmol/L). B: YAP1 pathway and apoptosis-related proteins were evaluated by Western blotting and subsequent quantification. Data are expressed as mean ± standard error of mean. *P < 0.05, **P < 0.01, ***P < 0.001. VP: verteporfin; CCA: cholangiocarcinoma.

VP impairs stemness feature and expression of CSC markers in CCA cells

The VP treatments significantly decreased the number of formed spheroids in comparison to DMSO (Figs. 3A, S1B and S2B). Furthermore, we investigated whether VP affects colony formation in CCA cells. The results showed significant decrease in the number colony formation with VP exposure in comparison to that in the control group (Figs. 3B, S1C and S2C). ALDH activity has been considered an important stemness feature. We further evaluated the cell population with ALDH activity in HuCCT-1 cells after VP treatment. Consistently, we observed significant reduction of ALDH positive cell population after VP treatment compared with that in DMSO treatment (Fig. 3C). The effects of VP on cell populations with CSC marker expression were evaluated. We treated CCA cells with DMSO, VP for 24 h and cells with different CSC marker expression were evaluated with flow cytometry. We found that VP significantly reduced cell populations of different CSC markers in comparison to DMSO control group (Figs. 3D, S1D and S2D). Taken together, the results indicated that VP has potential antitumor effects by inhibiting CSCs of CCA.

Fig. 3.

VP reduced the population of cancer stem-like cells in vitro. A: Representative spheroid formation and colony formation in HuCCT-1 cells treated with different concentration of VP (left) and quantitative numbers of spheroids. B: Representative colony formation in HuCCT-1 cells treated with different concentration of VP and quantitative numbers of colonies. C: Aldefluor positive cells in HuCCT-1 cells were decided by ALDEFLUOR™ Kit and flow cytometry after treated with different concentration of VP with or without DEAB. D: Flow cytometry analysis of different CSC surface markers of HuCCT-1 under different concentration of VP (DMSO, 1 μmol/L and 2 μmol/L) treatment. All experiments were repeated at least three times. Data are expressed as mean ± standard error of mean. *P < 0.05, ***P < 0.001. VP: verteporfin; CSC: cancer stem cell.

Combination of VP and anti-PD-1 antibody impairs CCA growth in the subcutaneous tumor model

The tumor size and endpoint measurement of tumor weight of subcutaneous SB1 tumors were reduced with the treatment of combination of VP and anti-PD-1 but not monotherapy of VP or anti-PD-1 (Fig. 4B-D) which indicated the synergistic antitumor effects of VP and anti-PD-1 (Fig. 4A). The profile of TILs was analyzed to illuminate the potential mechanism of the improved therapeutic efficacy of the combination therapy. In the TILs of SB1 cells derived subcutaneous tumors, there was an increased percentage of CD4⁺ T cells and NK cells in the combination therapy group, however, the number of other TIL subsets did not change significantly. There was a slightly increased trend of CD8⁺ T cells with VP monotherapy but it did not reach significance (Fig. S3). Together, these data indicate that VP may modulate tumor immune microenvironment and enhance antitumor activity of ICI in CCA.

Fig. 4.

The combination of VP and anti-PD-1 controlled tumor growth in CCA model. A: Experiments set-up. C57BL/6 mice with subcutaneous injection of 1 × 10⁶ SB1 cells was treated with VP and/or anti-PD-1 at the indicated time points. Mice with tumors > 20 mm were scarified, but tumor size continued to be reported as 20 mm. B: Subcutaneous tumor measurements for subcutaneous SB1 CCA tumor-bearing as indicated. C: Representative tumors for SB1 tumor-bearing C57BL/6 mice treated with IgG, anti-PD-1, VP and combination therapy. D: Tumor weight of mice under different treatment described in A at day 25. Data are shown as mean ± standard error of mean. **P < 0.01. VP: verteporfin; anti-PD-1: anti-programmed cell death-1; CCA: cholangiocarcinoma; IgG: immunoglobin G.

Discussion

In the present study, we demonstrated that VP inhibited ICC and ECC cell growth and induce apoptosis. Moreover, VP impairs stemness features of CCA cells and reduces CSC population in vitro. Interestingly, VP showed synergistic antitumor effect in combination with anti-PD-1, evidenced by reduction of tumor growth and tumor weight in vivo. This synergistic antitumor effect may partially result from an immune modulation effect of VP through targeting CSCs in the tumor via YAP1 signaling pathway inhibition. These results indicate a potentially new therapeutic strategy for CCA treatment in combination with immunotherapy.

Although VP is approved for the treatment of macular degeneration, VP retains intrinsic antitumor activities by blocking YAP1/TAZ pathway and other mechanisms [18,24-26]. The antitumor effect of VP has previously been reported in various tumors [22,23,25-28]. In CCA, it has been documented that VP at the 10 or 20 μmol/L level decreased stem cell marker OCT4 and mesenchymal marker N-cadherin protein expressions and reduced in vitro spheroid formation [19]. Furthermore, VP activated the protein kinase B/mTOR signaling pathway and considerably impaired IL-6-stimulated STAT3 phosphorylation in ICC cells. It was also reported that the combination of VP and rapamycin inhibited cell proliferation and tumor growth in vivo [19].

In our study, we extended the previous observation of cytotoxicity effect of VP on ICC to ECC cells without assistance of photodynamic therapy, which is different from the previous study [26]. We showed different IC50 of VP in various CCA cells. This cytotoxicity effect was observed not only in ICC but also in ECC cell lines. The results showed lower concentration, e.g. 1 or 2 μmol/L of VP, possessed strong effects of cytotoxicity, anti-proliferation and apoptosis induction in comparison to those in a previous report in ICC [19]. Nevertheless, this dose of VP showed inhibition of stemness features including spheroid formation, colony formation, ALDH activity and CSC marker expression. These results are likely through the inhibition of the YAP1 signaling pathway, evidenced by reduction of YAP1 expression and downstream target TAZ. It has been shown that YAP1 enhanced cell proliferation, chemoresistance, and angiogenesis through downstream TEAD in ICC [19], and maintained stemness through YAP1/SOX9 in esophageal cancer [23]. There were also reports that VP-antitumor effect is independent on YAP1 inhibition, e.g. autophagy mechanism and STAT3 pathway [25], and Notch pathway [26].

ICI therapy is an encouraging therapeutic approach that has shown benefits in patients with a broad spectrum of cancer types. The efficacy of ICI in patients with CCA remains challenging. It is interesting to observe the synergistic antitumor effect of the combination of VP with anti-PD-1 in this study, despite no significant antitumor effects in the preclinical model with monotherapy. It suggests that VP should be combined with other treatment modalities to contribute its antitumor effect in vivo. Subsequent immune cell profiling from TILs did not show significant change of CD8⁺ T cells though the combination therapy dramatically increased the percentage of CD4⁺ T cells and NK cells that was not observed in the monotherapy groups with either VP or anti-PD-1. It is unclear about the increased percentage of CD4⁺ T cells and NK cells in this setting, though NK cells are critical as part of innate antitumor immunity. This may indicate that VP can modulate intratumoral immune microenvironment partially through targeting CSCs via inhibition of YAP1 and enhance anti-PD-1 efficacy in vivo. A recent study has found that VP inhibits PD-L1 expression through autophagy and the STA1/IRF1/TRIM28 signaling axis and can enhance therapeutic effects of PARP1 [29].

In conclusion, VP possessed antitumor effects through multiple cellular activities, including decreasing cell proliferation, increasing apoptosis and inhibiting CSC stemness. We also for the first time demonstrated the synergistic antitumor effects of VP and anti-PD-1 in vivo that could be a promising therapeutic strategy for CCA for which there is currently no effective immunotherapy. Future clinical studies are warranted to determine whether the combination of VP and ICI are effective on CCA progression.