br Corresponding author at National
Corresponding author at: National Institute on Deafness and Other Communication Disorders, National Institutes of Health, 10 Center Drive, Room 7N240B, Bethesda, MD 20892, United States. E-mail addresses: [email protected], [email protected] (N.C. Schmitt).
Fig. 1. Schema of immunogenic cell death. Immunogenic cell death (ICD) is the process by which stressed, dying CP-456773 induce translocation of damage-associated molecular patterns (DAMPs). The cell surface expression and the release of such molecular patterns can stimulate tumor antigen presentation, thereby activating adaptive immunity. CRT, calreti-culin; HSP70/90, heat shock protein 70/90; ATP, adenosine triphosphate; HMGB1, high mobility group box 1; IFN, interferon; CXCL10, C-X-C motif chemokine 10; TCR, T-cell receptor; MHC, major histocompatibility complex.
ICD and/or upregulate MHC class I expression [2,3]. Our study focused on platinum chemotherapy drugs. Cisplatin is a standard treatment for head and neck squamous cell carcinoma (HNSCC). Carboplatin is a cisplatin analog with less toxicity that is administered to HNSCC pa-tients who cannot cisplatin . Oxaliplatin is a third-generation pla-tinum compound used for colorectal cancer . Preclinical studies involving non-HNSCC cancer types suggest that oxaliplatin is a bona fide ICD inducer, but cisplatin is not [6,9]. The widely-held idea that oxaliplatin is more immunogenic than cisplatin is based on a few stu-dies suggesting that cisplatin fails to induce calreticulin exposure on the tumor cell surface, which can be restored by enhancing ER stress [6,9,12,13]. We aimed to investigate whether oxaliplatin might be a better inducer of ICD and MHC class I/PD-L1 expression in HNSCC models, which would provide a rationale for using oxaliplatin, instead of cisplatin, in future clinical paradigms combining platinum che-motherapy with immunotherapies .
We used human HNSCC cell lines to compare the eﬀects of cisplatin versus oxaliplatin on the release and/or expression of calreticulin (CRT), heat shock proteins (HSPs), HMGB1, ATP, MHC class 1 and PD-L1 in vitro. A syngeneic mouse model of oral cancer was used to assess the ability of tumor cells killed by these agents to promote ICD in vivo, and to compare the eﬀects of these chemotherapy drugs with or without PD-1 blockade on tumor growth.
Materials and methods
Human UMSCC-46 and UMSCC-74A cell lines (HPV-negative) were obtained from Dr. T. Carey at the University of Michigan and were authenticated and maintained as described . UPCI SCC90 cells (HPV-positive) were obtained from Dr. R. Ferris and Dr. S. Gollin at the University of Pittsburgh and were validated as described [16,17]. Human cell lines were maintained in MEM or DMEM with 1% peni-cillin/streptomycin, 1% glutamine and 10% FBS. Mouse oral cancer 1 (MOC1) cells were obtained from Dr. R. Uppaluri in 2014 and were authenticated and maintained as previously described [18,19]. Cell lines were regularly tested for Mycoplasma and cultured for no more than 6 months or 20 passages before use.
Antibodies and reagents
Recombinant human interferon-gamma (IFN-γ) and antibodies to human MHC class I and PD-L1 were obtained from Biolegend. Antibodies to calreticulin (CRT), HSP70 and HSP90 were obtained from Abcam. The anti-PD-1 antibody for in vivo mouse treatments (clone RMP1-14) was from BioXCell. Fluorescent-conjugated flow cytometry antibodies for mouse tumor experiments were obtained from eBioscience (CD137/41BB) or Biolegend (CD8, CD45, CD80, CD11b, CD11c, CD107a, Ly6G, Ly6C, H-2Kb/H-2Db). Pharmaceutical grade chemotherapy was obtained from the veterinary pharmacy at NIH.
In vitro drug treatments and cell death assays
Cells were plated at 50,000–100,000 cells per well in 6-well plates and allowed to adhere overnight prior to drug treatments. Cell lines were treated with a range of platinum chemotherapy doses for 72 h, then assessed for cell death by annexin/7AAD flow cytometry assays (BD Biosciences) according to manufacturer instructions. Cell death curves were generated using GraphPad Prism software and used to es-timate the dose required to kill 40% of the cells in 72 h (LD40).