- Email: [email protected]
Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes
Cellular Immunology xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Cellular Immunology journal homepage: www.elsevier.com/locate/ycimm
Short communication
Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes Hideya Onishi a,⇑, Akiko Fujimura a, Yasuhiro Oyama a, Akio Yamasaki a, Akira Imaizumi a,b, Makoto Kawamoto a, Mitsuo Katano a, Masayo Umebayashi c, Takashi Morisaki c a b c
Department of Cancer Therapy and Research, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Shukoukai Inc., Tokyo, Japan Fukuoka General Cancer Clinic, Fukuoka, Japan
a r t i c l e
i n f o
Article history: Received 28 June 2016 Revised 3 August 2016 Accepted 6 August 2016 Available online xxxx Keywords: Hedgehog signaling PDL-1 Hypoxia Activated lymphocytes
a b s t r a c t We investigated whether hypoxia-induced activation of Hh signaling contributes to PDL-1 expression in cancer and whether it affects the anti-tumor function of activated lymphocytes. Hypoxia augmented PDL-1 expression and inhibition of Hh signaling reduced PDL-1 expression under hypoxia. When activated lymphocytes were cocultured with cancers treated with a Hh inhibitor, activated lymphocyte cell numbers increased under hypoxia. In contrast, this increase was abrogated when cancer cells were treated with a PDL-1 neutralizing antibody. These results suggest that Hh signaling is one of regulatory pathways of PDL-1 expression under hypoxia and that inhibiting Hh signaling may induce lymphocyte anti-tumor activity. Ó 2016 Elsevier Inc. All rights reserved.
1. Introduction Hedgehog (Hh) signaling plays a pivotal role in growth and morphogenic patterning in a wide variety of tissues during embryonic development [1]. Recent studies have revealed that Hh signaling is reactivated in various types of cancers and contributes to cellular invasion, progression, proliferation, and cancer initiation [2–5]. Hh signaling therefore warrants evaluation as a therapeutic target and promising results have been observed in vitro. However, good clinical endpoints have not yet been reported in clinical trials using Hh inhibitors [6]. We considered this discrepancy between in vitro results and clinical trials and suspect one reason may be the cancer microenvironment and hypoxic conditions. Almost all in vitro results were conducted under normoxic conditions, and tumor hypoxia had not been taken into consideration. Indeed, we have already shown that biological functions in cancerous and immune cells, such
as activated lymphocytes and dendritic cells (DCs), are modulated differently under hypoxic conditions than under normoxic conditions [7–9]. Notably, it was recently reported that Hh signaling is activated under hypoxic conditions [10,11]. Antibodymediated blockade of programmed death ligand-1 (PDL-1) as well as programmed death-1 (PD-1) revealed sustained tumor suppression and prolonged stabilization of disease in patients with advanced cancer [12–14]. Regulation of PDL-1 may thus be a promising therapeutic strategy. However, the mechanism (s) regulating PDL-1 expression remains unclear. Recently, it has been shown that hypoxia induces PDL-1 expression in cancer cells through hypoxia inducible factor (HIF)-1a [14]. To assess the potential effectiveness of new cancer therapies, we investigated whether hypoxia-induced activation of Hh signaling contributes to PDL-1 expression in cancer and whether it modulates the anti-tumor function of activated lymphocytes. 2. Materials and methods
Abbreviations: Hh, Hedgehog; MAML3, mastermind-like3; RBPJ, recombination signal binding protein for immunoglobulin-kappa-J region; PD-1, programmed death-1; PDL-1, programmed death ligand-1; Smo, Smoothened; PSK, proteinbound polysaccharide; NKG2D, natural-killer group 2, member D. ⇑ Corresponding author at: Cancer Therapy and Research, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan. E-mail address: [email protected] (H. Onishi).
2.1. Induction of activated lymphocytes Three different human peripheral blood mononuclear cells (PBMCs) were maintained in RPMI-1640 (Nacalai Tesque, Kyoto, Japan) supplemented with 0.5% human serum, 2000 units/ml penicillin (Meijiseika, Tokyo, Japan), 10 lg/ml streptomycin
http://dx.doi.org/10.1016/j.cellimm.2016.08.003 0008-8749/Ó 2016 Elsevier Inc. All rights reserved.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
2
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
(Meijiseika) and 200 U/ml IL-2 (Primmune) in 2.5 lg/ml anti-CD3 monoclonal antibody (OKT3, JANSSEN PHARMACEUTICAL K.K., Tokyo, Japan)-coated T-75 flask for 7 days. Thereafter the lymphocytes (non-adherent fracture of culture) were transferred to oxygen permeable culture bags and were cultured in RPMI medium with 200 U/ml IL-2 for an additional 7 days. Then, the lymphocytes were collected as activated lymphocytes and were used experimentally. 2.2. Cell culture and reagents Human pancreatic ductal adenocarcinoma cells lines, Panc-1, gallbladder cancer cell lines, GBd15 and NOZ, and small cell lung
cancer cell line, SBC-5 were maintained in RPMI 1640 medium supplemented with 10% fetal calf serum (FCS; Life Technologies, Grand Island, NY) and antibiotics (2000 units/ml of penicillin and 10 lg/ml of streptomycin). For normoxic conditions, cells were cultured in 5% CO2 and 95% air. For hypoxic conditions, cells were cultured in 1% O2, 5% CO2, and 94% N2 using a multigas incubator (Sanyo, Tokyo, Japan). Cyclopamine (Wako, Osaka, Japan), an Smoothened (Smo) inhibitor was diluted in 99% ethanol. Cancer cells were treated with 10 lM cyclopamine over night to inhibit Hh signaling. To block PDL-1 expressed on cancer cells, 10 lg/ml of anti-human CD274 neutralizing Ab (Biolegend, San Diego, CA, USA) was treated for 30 min. Mouse IgG2a isotype control (eBioscience, San Diego, CA, USA) was used as control.
Fig. 1. Hh signaling contributes to PDL-1 expression on cancerous cells under hypoxic conditions. (A) PDL-1 expressions on Panc-1 (a), NOZ (b), GBd (c) and SBC-5 (d) cells under normoxia and hypoxia were investigated by FACS. Dotted line; control IgG, solid line; PDL-1 expression under normoxia, filled hystogram; PDL-1 expression under hypoxia. (B) FACS analyses of PDL-1 expression on Panc-1 cells. Filled histogram shows Gli-si RNA (a, h), Smo-siRNA (b, i), MAML-3 siRNA (c, j), RBPJ siRNA (d, k), and RBPJ plasmid transfected cells (e, l), respectively, and PSK-treated (f, m), and cyclopamine-treated cells (g, n). Vacant lines represent control cells. Cells were cultured under normoxic (a-g) and hypoxic (h-n) conditions for 2 days. (C) Normoxic cancer cells were pretreated with 10 lM cyclopamine for 24 h, then cultured under hypoxic conditions for 2 days. PDL-1 expression on NOZ (a), GBd (b), and SBC-5 (c) cells were examined by FACS. Filled histogram shows cyclopamine-treated cells and vacant histogram shows control cells.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
2.3. RNA interference Small interfering RNA (siRNA) for Gli1 (ON-TARGETplus SMART pool, L-003896), siRNA for Smo (ON-TARGETplus SMART pool, L-005726), siRNA for MAML3 (ON-TARGET plus SMART pool, L-013813), siRNA for RBPJ (ON-TARGET plus SMART pool, L-007772) and negative control siRNA (ON-TARGET plus si CONTROL non-targeting pool, D-001810) were purchased from Dharmacon RNA Technologies (Chicago, IL). Cells (0.2 106 cells/well) seeded in 6-well plates were transfected with 100 nM siRNA under normoxia using Lipofectamine RNAiMAX Reagent (Invitrogen) according to the manufacturer’s instructions. Cells were used for experiments at 2 days after transfection. Then, cells were cultured additionally under normoxia and hypoxia in each experiments. 2.4. Plasmids Plasmids pFN21A HaloTag CMV Flexi-RBPJ vector and pFN21AB5901 control empty vector were purchased from Promega (Madison, WI, USA). Cells (0.2 106 cells/well) seeded in 6-well plates were transfected with 2.5 lg plasmids under normoxia using Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions. 2.5. Fluorescence activated cell sorting (FACS) To analyze the expression of natural-killer group 2, member D (NKG2D), CD3, CD4, CD8 on activated lymphocytes and PDL-1 on cancer cells, the cells were incubated for 1 h with fluorescein isothiocyanate (FITC)-conjugated anti-CD3 and CD4 monoclonal antibodies (mAbs), or phycoerythrin (PE)-conjugated anti-NKG2D, CD8, and PDL-1 mAbs (BD Pharmingen, San Diego, CA, USA). Mouse IgG was used as an isotype control (BD Pharmingen). For staining, cells were washed twice with PBS and incubated in PBS containing 3% bovine serum albumin (Sigma, St. Louis, MO, USA) and 0.1% NaN3 (FACS buffer; Sigma) and the appropriate concentration of labeled mAb for 1 h at 4 °C. After washing with FACS buffer, the fluorescence intensity or percent positive cells was measured using a FACSCalibur flow cytometer (BD Pharmingen) and analyzed with CELLQuest software (BD Pharmingen). 2.6. Lymphocytes proliferation assay Panc-1 cells (1 105 cells/well) treated with or without 10 lM of cyclopamine or 10 lg/ml of anti-PDL1 Ab were seeded in a 12-well plate and incubated for 24 h to allow adherence. Effector cells (activated lymphocytes) at an effector:target (E:T) ratio of 5:1 were added to the culture and were incubated under normoxic condition and hypoxic condition for 48 h. Then number of lymphocytes were counted by light microscope. 2.7. Statistical analysis An unpaired two-tailed Student’s t-test was used for statistical analysis. A P-value of <0.05 was considered significant. 3. Results 3.1. Hh signaling contributes to PDL-1 expression on cancerous cells under hypoxic conditions It has recently been reported that PDL-1 expression increases under hypoxic compared with normoxic conditions [14]. To confirm this result using cell lines of refractory cancer types, we first compared PDL-1 expression under normoxic and hypoxic
3
conditions using Panc-1, NOZ, GBd, and SBC-5 cell lines. Hypoxic PDL-1 expression was elevated relative to normoxic conditions in all four cell lines (Fig. 1A). Previous reports have shown that HIF1a contributes to increased PDL-1 under hypoxic conditions. In contrast, we have shown that Smo transcription is upregulated under hypoxia independent of HIF-1a [10]. Therefore, we next evaluated if Hh signaling augments PDL-1 expression under hypoxic conditions. Our results indicate that MAML-3, RBPJ, and PSK modulate Smo expression under such conditions [15,16]. We therefore used siRNAs for Gli1, Smo, MAML3 and RBPJ, PSK, RBPJ plasmid and cyclopamine to inhibit expression. We did not observe any significant changes in PDL-1 expression under normoxic conditions (Fig. 1B). Unexpectedly, PDL-1 expression increased in normoxic Smo siRNA transfected cells. In contrast, PDL-1 expression in Gli1 siRNA or MAML3 siRNA transfected and cyclopaminetreated cells decreased under hypoxic conditions, whereas RBPJ plasmid transfection had no effect (Fig. 1B). This suggests that endogenous RBPJ expression may be sufficient to induce PDL-1 expression. We confirmed decreased expressions of PDL-1 in cyclopamine-treated other cell lines, such as NOZ, GBd, and SBC5 cell lines grown under hypoxic conditions (Fig. 1C). Collectively, these results suggest that Hh signaling modulates PDL-1 expression in cancer cells under hypoxic conditions. 3.2. Suppression of PDL-1 expression in cancer cells by cyclopamine augments lymphocyte activation We next sought to understand how a change in PDL-1 expression affects lymphocyte function. We therefore evaluated activated lymphocyte proliferation following inhibition of PDL-1 expression by blocking Hh signaling. Because Panc-1 cells showed the strongest response to cyclopamine-induced PDL-1 suppression, Panc-1 cells with or without cyclopamine treatment were used as target cells. Most activated lymphocytes used in this study consisted of CD8+ cells rather than CD4+ cells (Fig. 2A), and CD3+NKG2D+ T cells rather than CD3-NKG2D+ natural killer cells (Fig. 2B). Activated lymphocytes from these three individuals were used as effector cells, which proliferated when cocultured with cyclopaminetreated cancers under hypoxia. In contrast, treatment with a PDL1 neutralizing antibody abrogated this effect, suggesting that Hh inhibitor-reduced PDL-1 expression may play a pivotal role in regulating the proliferation of activated lymphocytes under hypoxia (Fig. 2C). Interestingly, the percent of CD8+ lymphocytes decreased in both the cyclopamine- and anti-PDL-1-treated groups under hypoxic conditions in all three cases (Fig. 2D), while there was no significant change in CD3 expression (Fig. 2E). It may be that CD8 lymphocytes were used to kill cancer cells more efficiently following cyclopamine and anti-PDL-1 treatment. Consistent with these results, NKG2D expression increased on activated lymphocytes in anti PDL-1Ab-treated group (Fig. 2F). These results suggest that cyclopamine suppression of PDL-1 expression may augment lymphocyte activation. 4. Discussion PDL-1 expression on Smo siRNA transfected- or MAML3 siRNA transfected Panc-1 cells increased compared to control under normoxia (Fig. 1B). Because normoxic condition such as 20% O2 does not exist in vivo, other signaling pathway or other unknown factors may contribute as a mechanism. We should consider tumor hypoxic condition that Hh signaling is highly activated, but not normoxic condition that Hh signaling is activated weakly. On the other hand, unexpectedly, PDL-1 expression on RBPJ siRNA transfected- and PSK treated Panc-1 cells did not decrease (Fig. 1B). Although we have shown that RBPJ inhibition and PSK
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
4
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
Fig. 2. Suppression of PDL-1 expression in cancer cells by cyclopamine augments lymphocyte activation. (A, B) Expressions of CD4/CD8 (A) and CD3/NKG2D (B) on activated lymphocytes were estimated by FACS. (C-F) Normoxic Panc-1 cells were pretreated with 10 lM cyclopamine for 24 h. Thereafter, Panc-1 cells were pretreated again with 10 lg/ml of anti-human PDL-1 neutralizing Ab or control antibody (IgG) for 30 min. Panc-1 (target cells) were then cocultured with activated lymphocytes (effector cells) under normoxic and hypoxic conditions for 48 h. The effector/target ratio was 5. Cell number (C), expressions of CD8 (D), CD3 (E), and NKG2D (F), were examined by light microscopy and FACS, respectively. *p < 0.05. Bars show SD. NS; not significant.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
5
Fig. 3. Schematic findings of this study based on previous results. Gray field shows findings from this study. Hh inhibition may be a promising drug target, not only as a direct tumor suppressor agent, but also as a PDL-1 inhibitor.
treatment decreased Smo expression under hypoxia [16], some questions such as whether Smo inhibition by RBPJ suppression is direct or indirect, and which molecules of PSK contributed to the regulation of Smo transcription remains unclear. The research toward these directions may answer these questions. In the experiment of coculture with CD3+CD8+ NKG2D+ dominant activated lymphocytes and cancer cells, CD8 expression decreased and NKG2D expression increased when anti PDL-1 Ab was treated with cancers under hypoxia (Fig. 2D and F). We think that activated lymphocytes may tend to work well by blocking of hypoxia-induced PDL-1expression using anti PDL-1 Ab. Moreover, our previous study revealed that NKG2D expression on activated lymphocytes increased under hypoxia compared to normoxia [9]. In cancer microenvironment, various kinds of phenotypic change may occur in activated lymphocytes. Recently, the effect of Nivolmab against refractory cancers is taking much attention, however, it is difficult to prove directly the kinetics of lymphocytes against cancer cells in vivo after the administration of Nivolmab. Our results may aid understanding these issues. Previously, we and other researchers have shown that Hh signaling induces cell cycle-dependent tumor proliferation and invasion by stimulating metalloproteinase expression, and that inhibiting Hh signaling suppresses invasion and proliferation under hypoxic conditions [10,11,17]. Considering these results together, Hh inhibition appears to be a promising drug target, not only as a direct tumor suppressor agent, but also as a PDL-1 inhibitor. However, we must note that inhibiting Hh signaling may negatively affect activated lymphocytes and DCs [7,9]. Clinical administration of a Hh inhibitor may therefore require a brief intermission prior to treating with immune checkpoint inhibitors to modulate activated lymphocytes. Fig. 3 shows a summary of our previous results. In this study, we have shown that Hh signaling may modulate PDL-1 expression under hypoxic conditions. Additionally, we found that Hh inactivation and/or the blockade of PDL-1 induces lymphocyte anti-tumor activity. We plan to further examine tumor PDL-1 regulation under hypoxic conditions and to explore cancer therapeutics using a Hh inhibitor, as clinical trials using Hh inhibitors have not yet met anticipated clinical endpoints [6].
Conflicts of interest The authors declare no conflict of interest. Acknowledgments We thank Ms Kaori Nomiyama for skillful technical assistance. This study was supported by JSPS KAKENHI Grant Number 15K10055. References [1] P.W. Ingham, A.P. McMahon, Hedgehog signaling in animal development: paradigms and principles, Genes Dev. 15 (2001) 3059–3087. [2] M. Souzaki, M. Kubo, M. Kai, et al., Hedgehog signaling pathway mediates the progression of non-invasive breast cancer to invasive breast cancer, Cancer Sci. 102 (2011) 373–381. [3] G. Feldmann, V. Fendrich, K. McGovern, et al., An orally bioavailable smallmolecule inhibitor of Hedgehog signaling inhibits tumor initiation and metastasis in pancreatic cancer, Mol. Cancer Ther. 7 (2008) 2725–2735. [4] O. Bermudez, E. Hennen, I. Koch, M. Lindner, O. Eickelberg, Gli1 mediates lung cancer cell proliferation and Sonic Hedgehog-dependent mesenchymal cell activation, PLoS One 8 (2013) e63226. [5] H.X. Fan, S. Wang, H. Zhao, et al., Sonic hedgehog signaling may promote invasion and metastasis of oral squamous cell carcinoma by activating MMP-9 and E-cadherin expression, Med. Oncol. 31 (2014) 41. [6] D.V. Catenacci, M.R. Junttila, T. Karrison, et al., Randomized phase Ib/II study of gemcitabine plus placebo or vismodegib, a hedgehog pathway inhibitor, in patients with metastatic pancreatic cancer, J. Clin. Oncol. 33 (2015) 4284– 4292. [7] H. Onishi, T. Morisaki, A. Kiyota, et al., The Hedgehog inhibitor suppresses the function of monocyte-derived dendritic cells from patients with advanced cancer under hypoxia, Biochem. Biophys. Res. Commun. 436 (2013) 53–59. [8] T. Ogino, H. Onishi, H. Suzuki, T. Morisaki, M. Tanaka, M. Katano, Inclusive estimation of complex antigen presentation functions of monocyte-derived dendritic cells differentiated under normoxia and hypoxia conditions, Cancer Immunol. Immunother. 61 (2012) 409–424. [9] H. Onishi, T. Morisaki, A. Kiyota, et al., The Hedgehog inhibitor cyclopamine impairs the benefits of immunotherapy with activated T and NK lymphocytes derived from patients with advanced cancer, Cancer Immunol. Immunother. 62 (2013) 1029–1039. [10] H. Onishi, M. Kai, S. Odate, et al., Hypoxia activates the hedgehog signaling pathway in a ligand-independent manner by upregulation of Smo transcription in pancreatic cancer, Cancer Sci. 102 (2011) 1144–1150. [11] J. Lei, J. Ma, Q. Ma, et al., Hedgehog signaling regulates hypoxia induced epithelial to mesenchymal transition and invasion in pancreatic cancer cells via a ligand-independent manner, Mol. Cancer 12 (2013) 66.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
6
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
[12] J.R. Brahmer, S.S. Tykodi, L.Q. Chow, et al., Safety and activity of anti-PD-L1 antibody in patients with advanced cancer, N. Engl. J. Med. 366 (2012) 2455–2465. [13] S.L. Topalian, F.S. Hodi, J.R. Brahmer, et al., Safety, activity, and immune correlates of anti-PD-1 antibody in cancer, N. Engl. J. Med. 366 (2012) 2443–2454. [14] M.Z. Noman, G. Desantis, B. Janji, et al., PD-L1 is a novel direct target of HIF-1a, and its blockade under hypoxia enhanced MDSC-mediated T cell activation, J. Exp. Med. 211 (2014) 781–790. [15] H. Onishi, A. Yamasaki, M. Kawamoto, A. Imaizumi, M. Katano, Hypoxia but not normoxia promotes smoothened transcription through upregulation of RBPJ and mastermind-like 3 in pancreatic cancer, Cancer Lett. 371 (2016) 143–150.
[16] H. Onishi, T. Morisaki, F. Nakao, S. Odate, T. Morisaki, M. Katano, Protein-bound polysaccharide decreases invasiveness and proliferation in pancreatic cancer by inhibition of hedgehog signaling and HIF-1a pathways under hypoxia, Cancer Lett. 335 (2013) 289–298. [17] H. Onishi, Y. Morifuji, M. Kai, K. Suyama, H. Iwasaki, M. Katano, Hedgehog inhibitor decreases chemosensitivity to 5-fluorouracil and gemcitabine under hypoxic conditions in pancreatic cancer, Cancer Sci. 103 (2012) 1272–1279.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
Contents lists available at ScienceDirect
Cellular Immunology journal homepage: www.elsevier.com/locate/ycimm
Short communication
Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes Hideya Onishi a,⇑, Akiko Fujimura a, Yasuhiro Oyama a, Akio Yamasaki a, Akira Imaizumi a,b, Makoto Kawamoto a, Mitsuo Katano a, Masayo Umebayashi c, Takashi Morisaki c a b c
Department of Cancer Therapy and Research, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Shukoukai Inc., Tokyo, Japan Fukuoka General Cancer Clinic, Fukuoka, Japan
a r t i c l e
i n f o
Article history: Received 28 June 2016 Revised 3 August 2016 Accepted 6 August 2016 Available online xxxx Keywords: Hedgehog signaling PDL-1 Hypoxia Activated lymphocytes
a b s t r a c t We investigated whether hypoxia-induced activation of Hh signaling contributes to PDL-1 expression in cancer and whether it affects the anti-tumor function of activated lymphocytes. Hypoxia augmented PDL-1 expression and inhibition of Hh signaling reduced PDL-1 expression under hypoxia. When activated lymphocytes were cocultured with cancers treated with a Hh inhibitor, activated lymphocyte cell numbers increased under hypoxia. In contrast, this increase was abrogated when cancer cells were treated with a PDL-1 neutralizing antibody. These results suggest that Hh signaling is one of regulatory pathways of PDL-1 expression under hypoxia and that inhibiting Hh signaling may induce lymphocyte anti-tumor activity. Ó 2016 Elsevier Inc. All rights reserved.
1. Introduction Hedgehog (Hh) signaling plays a pivotal role in growth and morphogenic patterning in a wide variety of tissues during embryonic development [1]. Recent studies have revealed that Hh signaling is reactivated in various types of cancers and contributes to cellular invasion, progression, proliferation, and cancer initiation [2–5]. Hh signaling therefore warrants evaluation as a therapeutic target and promising results have been observed in vitro. However, good clinical endpoints have not yet been reported in clinical trials using Hh inhibitors [6]. We considered this discrepancy between in vitro results and clinical trials and suspect one reason may be the cancer microenvironment and hypoxic conditions. Almost all in vitro results were conducted under normoxic conditions, and tumor hypoxia had not been taken into consideration. Indeed, we have already shown that biological functions in cancerous and immune cells, such
as activated lymphocytes and dendritic cells (DCs), are modulated differently under hypoxic conditions than under normoxic conditions [7–9]. Notably, it was recently reported that Hh signaling is activated under hypoxic conditions [10,11]. Antibodymediated blockade of programmed death ligand-1 (PDL-1) as well as programmed death-1 (PD-1) revealed sustained tumor suppression and prolonged stabilization of disease in patients with advanced cancer [12–14]. Regulation of PDL-1 may thus be a promising therapeutic strategy. However, the mechanism (s) regulating PDL-1 expression remains unclear. Recently, it has been shown that hypoxia induces PDL-1 expression in cancer cells through hypoxia inducible factor (HIF)-1a [14]. To assess the potential effectiveness of new cancer therapies, we investigated whether hypoxia-induced activation of Hh signaling contributes to PDL-1 expression in cancer and whether it modulates the anti-tumor function of activated lymphocytes. 2. Materials and methods
Abbreviations: Hh, Hedgehog; MAML3, mastermind-like3; RBPJ, recombination signal binding protein for immunoglobulin-kappa-J region; PD-1, programmed death-1; PDL-1, programmed death ligand-1; Smo, Smoothened; PSK, proteinbound polysaccharide; NKG2D, natural-killer group 2, member D. ⇑ Corresponding author at: Cancer Therapy and Research, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan. E-mail address: [email protected] (H. Onishi).
2.1. Induction of activated lymphocytes Three different human peripheral blood mononuclear cells (PBMCs) were maintained in RPMI-1640 (Nacalai Tesque, Kyoto, Japan) supplemented with 0.5% human serum, 2000 units/ml penicillin (Meijiseika, Tokyo, Japan), 10 lg/ml streptomycin
http://dx.doi.org/10.1016/j.cellimm.2016.08.003 0008-8749/Ó 2016 Elsevier Inc. All rights reserved.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
2
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
(Meijiseika) and 200 U/ml IL-2 (Primmune) in 2.5 lg/ml anti-CD3 monoclonal antibody (OKT3, JANSSEN PHARMACEUTICAL K.K., Tokyo, Japan)-coated T-75 flask for 7 days. Thereafter the lymphocytes (non-adherent fracture of culture) were transferred to oxygen permeable culture bags and were cultured in RPMI medium with 200 U/ml IL-2 for an additional 7 days. Then, the lymphocytes were collected as activated lymphocytes and were used experimentally. 2.2. Cell culture and reagents Human pancreatic ductal adenocarcinoma cells lines, Panc-1, gallbladder cancer cell lines, GBd15 and NOZ, and small cell lung
cancer cell line, SBC-5 were maintained in RPMI 1640 medium supplemented with 10% fetal calf serum (FCS; Life Technologies, Grand Island, NY) and antibiotics (2000 units/ml of penicillin and 10 lg/ml of streptomycin). For normoxic conditions, cells were cultured in 5% CO2 and 95% air. For hypoxic conditions, cells were cultured in 1% O2, 5% CO2, and 94% N2 using a multigas incubator (Sanyo, Tokyo, Japan). Cyclopamine (Wako, Osaka, Japan), an Smoothened (Smo) inhibitor was diluted in 99% ethanol. Cancer cells were treated with 10 lM cyclopamine over night to inhibit Hh signaling. To block PDL-1 expressed on cancer cells, 10 lg/ml of anti-human CD274 neutralizing Ab (Biolegend, San Diego, CA, USA) was treated for 30 min. Mouse IgG2a isotype control (eBioscience, San Diego, CA, USA) was used as control.
Fig. 1. Hh signaling contributes to PDL-1 expression on cancerous cells under hypoxic conditions. (A) PDL-1 expressions on Panc-1 (a), NOZ (b), GBd (c) and SBC-5 (d) cells under normoxia and hypoxia were investigated by FACS. Dotted line; control IgG, solid line; PDL-1 expression under normoxia, filled hystogram; PDL-1 expression under hypoxia. (B) FACS analyses of PDL-1 expression on Panc-1 cells. Filled histogram shows Gli-si RNA (a, h), Smo-siRNA (b, i), MAML-3 siRNA (c, j), RBPJ siRNA (d, k), and RBPJ plasmid transfected cells (e, l), respectively, and PSK-treated (f, m), and cyclopamine-treated cells (g, n). Vacant lines represent control cells. Cells were cultured under normoxic (a-g) and hypoxic (h-n) conditions for 2 days. (C) Normoxic cancer cells were pretreated with 10 lM cyclopamine for 24 h, then cultured under hypoxic conditions for 2 days. PDL-1 expression on NOZ (a), GBd (b), and SBC-5 (c) cells were examined by FACS. Filled histogram shows cyclopamine-treated cells and vacant histogram shows control cells.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
2.3. RNA interference Small interfering RNA (siRNA) for Gli1 (ON-TARGETplus SMART pool, L-003896), siRNA for Smo (ON-TARGETplus SMART pool, L-005726), siRNA for MAML3 (ON-TARGET plus SMART pool, L-013813), siRNA for RBPJ (ON-TARGET plus SMART pool, L-007772) and negative control siRNA (ON-TARGET plus si CONTROL non-targeting pool, D-001810) were purchased from Dharmacon RNA Technologies (Chicago, IL). Cells (0.2 106 cells/well) seeded in 6-well plates were transfected with 100 nM siRNA under normoxia using Lipofectamine RNAiMAX Reagent (Invitrogen) according to the manufacturer’s instructions. Cells were used for experiments at 2 days after transfection. Then, cells were cultured additionally under normoxia and hypoxia in each experiments. 2.4. Plasmids Plasmids pFN21A HaloTag CMV Flexi-RBPJ vector and pFN21AB5901 control empty vector were purchased from Promega (Madison, WI, USA). Cells (0.2 106 cells/well) seeded in 6-well plates were transfected with 2.5 lg plasmids under normoxia using Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions. 2.5. Fluorescence activated cell sorting (FACS) To analyze the expression of natural-killer group 2, member D (NKG2D), CD3, CD4, CD8 on activated lymphocytes and PDL-1 on cancer cells, the cells were incubated for 1 h with fluorescein isothiocyanate (FITC)-conjugated anti-CD3 and CD4 monoclonal antibodies (mAbs), or phycoerythrin (PE)-conjugated anti-NKG2D, CD8, and PDL-1 mAbs (BD Pharmingen, San Diego, CA, USA). Mouse IgG was used as an isotype control (BD Pharmingen). For staining, cells were washed twice with PBS and incubated in PBS containing 3% bovine serum albumin (Sigma, St. Louis, MO, USA) and 0.1% NaN3 (FACS buffer; Sigma) and the appropriate concentration of labeled mAb for 1 h at 4 °C. After washing with FACS buffer, the fluorescence intensity or percent positive cells was measured using a FACSCalibur flow cytometer (BD Pharmingen) and analyzed with CELLQuest software (BD Pharmingen). 2.6. Lymphocytes proliferation assay Panc-1 cells (1 105 cells/well) treated with or without 10 lM of cyclopamine or 10 lg/ml of anti-PDL1 Ab were seeded in a 12-well plate and incubated for 24 h to allow adherence. Effector cells (activated lymphocytes) at an effector:target (E:T) ratio of 5:1 were added to the culture and were incubated under normoxic condition and hypoxic condition for 48 h. Then number of lymphocytes were counted by light microscope. 2.7. Statistical analysis An unpaired two-tailed Student’s t-test was used for statistical analysis. A P-value of <0.05 was considered significant. 3. Results 3.1. Hh signaling contributes to PDL-1 expression on cancerous cells under hypoxic conditions It has recently been reported that PDL-1 expression increases under hypoxic compared with normoxic conditions [14]. To confirm this result using cell lines of refractory cancer types, we first compared PDL-1 expression under normoxic and hypoxic
3
conditions using Panc-1, NOZ, GBd, and SBC-5 cell lines. Hypoxic PDL-1 expression was elevated relative to normoxic conditions in all four cell lines (Fig. 1A). Previous reports have shown that HIF1a contributes to increased PDL-1 under hypoxic conditions. In contrast, we have shown that Smo transcription is upregulated under hypoxia independent of HIF-1a [10]. Therefore, we next evaluated if Hh signaling augments PDL-1 expression under hypoxic conditions. Our results indicate that MAML-3, RBPJ, and PSK modulate Smo expression under such conditions [15,16]. We therefore used siRNAs for Gli1, Smo, MAML3 and RBPJ, PSK, RBPJ plasmid and cyclopamine to inhibit expression. We did not observe any significant changes in PDL-1 expression under normoxic conditions (Fig. 1B). Unexpectedly, PDL-1 expression increased in normoxic Smo siRNA transfected cells. In contrast, PDL-1 expression in Gli1 siRNA or MAML3 siRNA transfected and cyclopaminetreated cells decreased under hypoxic conditions, whereas RBPJ plasmid transfection had no effect (Fig. 1B). This suggests that endogenous RBPJ expression may be sufficient to induce PDL-1 expression. We confirmed decreased expressions of PDL-1 in cyclopamine-treated other cell lines, such as NOZ, GBd, and SBC5 cell lines grown under hypoxic conditions (Fig. 1C). Collectively, these results suggest that Hh signaling modulates PDL-1 expression in cancer cells under hypoxic conditions. 3.2. Suppression of PDL-1 expression in cancer cells by cyclopamine augments lymphocyte activation We next sought to understand how a change in PDL-1 expression affects lymphocyte function. We therefore evaluated activated lymphocyte proliferation following inhibition of PDL-1 expression by blocking Hh signaling. Because Panc-1 cells showed the strongest response to cyclopamine-induced PDL-1 suppression, Panc-1 cells with or without cyclopamine treatment were used as target cells. Most activated lymphocytes used in this study consisted of CD8+ cells rather than CD4+ cells (Fig. 2A), and CD3+NKG2D+ T cells rather than CD3-NKG2D+ natural killer cells (Fig. 2B). Activated lymphocytes from these three individuals were used as effector cells, which proliferated when cocultured with cyclopaminetreated cancers under hypoxia. In contrast, treatment with a PDL1 neutralizing antibody abrogated this effect, suggesting that Hh inhibitor-reduced PDL-1 expression may play a pivotal role in regulating the proliferation of activated lymphocytes under hypoxia (Fig. 2C). Interestingly, the percent of CD8+ lymphocytes decreased in both the cyclopamine- and anti-PDL-1-treated groups under hypoxic conditions in all three cases (Fig. 2D), while there was no significant change in CD3 expression (Fig. 2E). It may be that CD8 lymphocytes were used to kill cancer cells more efficiently following cyclopamine and anti-PDL-1 treatment. Consistent with these results, NKG2D expression increased on activated lymphocytes in anti PDL-1Ab-treated group (Fig. 2F). These results suggest that cyclopamine suppression of PDL-1 expression may augment lymphocyte activation. 4. Discussion PDL-1 expression on Smo siRNA transfected- or MAML3 siRNA transfected Panc-1 cells increased compared to control under normoxia (Fig. 1B). Because normoxic condition such as 20% O2 does not exist in vivo, other signaling pathway or other unknown factors may contribute as a mechanism. We should consider tumor hypoxic condition that Hh signaling is highly activated, but not normoxic condition that Hh signaling is activated weakly. On the other hand, unexpectedly, PDL-1 expression on RBPJ siRNA transfected- and PSK treated Panc-1 cells did not decrease (Fig. 1B). Although we have shown that RBPJ inhibition and PSK
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
4
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
Fig. 2. Suppression of PDL-1 expression in cancer cells by cyclopamine augments lymphocyte activation. (A, B) Expressions of CD4/CD8 (A) and CD3/NKG2D (B) on activated lymphocytes were estimated by FACS. (C-F) Normoxic Panc-1 cells were pretreated with 10 lM cyclopamine for 24 h. Thereafter, Panc-1 cells were pretreated again with 10 lg/ml of anti-human PDL-1 neutralizing Ab or control antibody (IgG) for 30 min. Panc-1 (target cells) were then cocultured with activated lymphocytes (effector cells) under normoxic and hypoxic conditions for 48 h. The effector/target ratio was 5. Cell number (C), expressions of CD8 (D), CD3 (E), and NKG2D (F), were examined by light microscopy and FACS, respectively. *p < 0.05. Bars show SD. NS; not significant.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
5
Fig. 3. Schematic findings of this study based on previous results. Gray field shows findings from this study. Hh inhibition may be a promising drug target, not only as a direct tumor suppressor agent, but also as a PDL-1 inhibitor.
treatment decreased Smo expression under hypoxia [16], some questions such as whether Smo inhibition by RBPJ suppression is direct or indirect, and which molecules of PSK contributed to the regulation of Smo transcription remains unclear. The research toward these directions may answer these questions. In the experiment of coculture with CD3+CD8+ NKG2D+ dominant activated lymphocytes and cancer cells, CD8 expression decreased and NKG2D expression increased when anti PDL-1 Ab was treated with cancers under hypoxia (Fig. 2D and F). We think that activated lymphocytes may tend to work well by blocking of hypoxia-induced PDL-1expression using anti PDL-1 Ab. Moreover, our previous study revealed that NKG2D expression on activated lymphocytes increased under hypoxia compared to normoxia [9]. In cancer microenvironment, various kinds of phenotypic change may occur in activated lymphocytes. Recently, the effect of Nivolmab against refractory cancers is taking much attention, however, it is difficult to prove directly the kinetics of lymphocytes against cancer cells in vivo after the administration of Nivolmab. Our results may aid understanding these issues. Previously, we and other researchers have shown that Hh signaling induces cell cycle-dependent tumor proliferation and invasion by stimulating metalloproteinase expression, and that inhibiting Hh signaling suppresses invasion and proliferation under hypoxic conditions [10,11,17]. Considering these results together, Hh inhibition appears to be a promising drug target, not only as a direct tumor suppressor agent, but also as a PDL-1 inhibitor. However, we must note that inhibiting Hh signaling may negatively affect activated lymphocytes and DCs [7,9]. Clinical administration of a Hh inhibitor may therefore require a brief intermission prior to treating with immune checkpoint inhibitors to modulate activated lymphocytes. Fig. 3 shows a summary of our previous results. In this study, we have shown that Hh signaling may modulate PDL-1 expression under hypoxic conditions. Additionally, we found that Hh inactivation and/or the blockade of PDL-1 induces lymphocyte anti-tumor activity. We plan to further examine tumor PDL-1 regulation under hypoxic conditions and to explore cancer therapeutics using a Hh inhibitor, as clinical trials using Hh inhibitors have not yet met anticipated clinical endpoints [6].
Conflicts of interest The authors declare no conflict of interest. Acknowledgments We thank Ms Kaori Nomiyama for skillful technical assistance. This study was supported by JSPS KAKENHI Grant Number 15K10055. References [1] P.W. Ingham, A.P. McMahon, Hedgehog signaling in animal development: paradigms and principles, Genes Dev. 15 (2001) 3059–3087. [2] M. Souzaki, M. Kubo, M. Kai, et al., Hedgehog signaling pathway mediates the progression of non-invasive breast cancer to invasive breast cancer, Cancer Sci. 102 (2011) 373–381. [3] G. Feldmann, V. Fendrich, K. McGovern, et al., An orally bioavailable smallmolecule inhibitor of Hedgehog signaling inhibits tumor initiation and metastasis in pancreatic cancer, Mol. Cancer Ther. 7 (2008) 2725–2735. [4] O. Bermudez, E. Hennen, I. Koch, M. Lindner, O. Eickelberg, Gli1 mediates lung cancer cell proliferation and Sonic Hedgehog-dependent mesenchymal cell activation, PLoS One 8 (2013) e63226. [5] H.X. Fan, S. Wang, H. Zhao, et al., Sonic hedgehog signaling may promote invasion and metastasis of oral squamous cell carcinoma by activating MMP-9 and E-cadherin expression, Med. Oncol. 31 (2014) 41. [6] D.V. Catenacci, M.R. Junttila, T. Karrison, et al., Randomized phase Ib/II study of gemcitabine plus placebo or vismodegib, a hedgehog pathway inhibitor, in patients with metastatic pancreatic cancer, J. Clin. Oncol. 33 (2015) 4284– 4292. [7] H. Onishi, T. Morisaki, A. Kiyota, et al., The Hedgehog inhibitor suppresses the function of monocyte-derived dendritic cells from patients with advanced cancer under hypoxia, Biochem. Biophys. Res. Commun. 436 (2013) 53–59. [8] T. Ogino, H. Onishi, H. Suzuki, T. Morisaki, M. Tanaka, M. Katano, Inclusive estimation of complex antigen presentation functions of monocyte-derived dendritic cells differentiated under normoxia and hypoxia conditions, Cancer Immunol. Immunother. 61 (2012) 409–424. [9] H. Onishi, T. Morisaki, A. Kiyota, et al., The Hedgehog inhibitor cyclopamine impairs the benefits of immunotherapy with activated T and NK lymphocytes derived from patients with advanced cancer, Cancer Immunol. Immunother. 62 (2013) 1029–1039. [10] H. Onishi, M. Kai, S. Odate, et al., Hypoxia activates the hedgehog signaling pathway in a ligand-independent manner by upregulation of Smo transcription in pancreatic cancer, Cancer Sci. 102 (2011) 1144–1150. [11] J. Lei, J. Ma, Q. Ma, et al., Hedgehog signaling regulates hypoxia induced epithelial to mesenchymal transition and invasion in pancreatic cancer cells via a ligand-independent manner, Mol. Cancer 12 (2013) 66.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003
6
H. Onishi et al. / Cellular Immunology xxx (2016) xxx–xxx
[12] J.R. Brahmer, S.S. Tykodi, L.Q. Chow, et al., Safety and activity of anti-PD-L1 antibody in patients with advanced cancer, N. Engl. J. Med. 366 (2012) 2455–2465. [13] S.L. Topalian, F.S. Hodi, J.R. Brahmer, et al., Safety, activity, and immune correlates of anti-PD-1 antibody in cancer, N. Engl. J. Med. 366 (2012) 2443–2454. [14] M.Z. Noman, G. Desantis, B. Janji, et al., PD-L1 is a novel direct target of HIF-1a, and its blockade under hypoxia enhanced MDSC-mediated T cell activation, J. Exp. Med. 211 (2014) 781–790. [15] H. Onishi, A. Yamasaki, M. Kawamoto, A. Imaizumi, M. Katano, Hypoxia but not normoxia promotes smoothened transcription through upregulation of RBPJ and mastermind-like 3 in pancreatic cancer, Cancer Lett. 371 (2016) 143–150.
[16] H. Onishi, T. Morisaki, F. Nakao, S. Odate, T. Morisaki, M. Katano, Protein-bound polysaccharide decreases invasiveness and proliferation in pancreatic cancer by inhibition of hedgehog signaling and HIF-1a pathways under hypoxia, Cancer Lett. 335 (2013) 289–298. [17] H. Onishi, Y. Morifuji, M. Kai, K. Suyama, H. Iwasaki, M. Katano, Hedgehog inhibitor decreases chemosensitivity to 5-fluorouracil and gemcitabine under hypoxic conditions in pancreatic cancer, Cancer Sci. 103 (2012) 1272–1279.
Please cite this article in press as: H. Onishi et al., Hedgehog signaling regulates PDL-1 expression in cancer cells to induce anti-tumor activity by activated lymphocytes, Cell. Immunol. (2016), http://dx.doi.org/10.1016/j.cellimm.2016.08.003