Targeting Macrophage Migration Inhibitory Factor to inhibit T cell immunosuppression in the tumor microenvironment and improve cancer outcomes in Head and Neck Squamous Cell Carcinoma

Abstract

Head and neck squamous cell carcinoma (HNSCC) is the 7^th most common cancer globally with a 40–50% survival rate. Although macrophage migration inhibitory factor (MIF) is overexpressed in most solid tumors and promotes tumor growth and invasion, the therapeutic potential of MIF inhibition in HNSCC is yet to be explored. In this study, we investigated the efficacy of CPSI-1306, a small-molecule MIF inhibitor, on HNSCC cell growth and cancer associated signaling pathways in vitro, as well as its impact on T cells in the HNSCC tumor microenvironment in vivo. CPSI-1306 did not reduce HNSCC cell proliferation in vitro, and mildly decreased VEGF and EGFR expression. However, CPSI-1306 significantly reduced tumor development in two orthotopic mouse oral cancer (MOC-2 and MOC-1) HNSCC models. Interestingly, CPSI-1306 treatment increased T cell infiltration to the tumor microenvironment and completely abrogated immunosuppressive checkpoint markers TIGIT, TIM3, and CTLA-4, but not PD-1 on tumor infiltrating CD8⁺ T cells. This was accompanied by increased CD8⁺ T cell expression of antitumoral cytokines IFN-γ and TNFα in the draining lymph nodes and Granzyme B in the tumor microenvironment of CPSI-1306 treated tumor bearing mice. Our studies demonstrate that the small molecule MIF inhibitor CPSI-1306 potently inhibits T cell immunosuppression in the tumor microenvironment and reduces tumor growth in HNSCC. These studies open a novel therapeutic option for modulating anti-tumoral T cell immunity to improve HNSCC outcomes by targeting MIF.

1. Introduction

Head and Neck Squamous Cell Carcinoma (HNSCC) is the 7th most common cancer worldwide with a 5-year survival rate of 40–50% [1, 2]. Human Papilloma Virus (HPV) negative HNSCC, partly attributable to alcohol and tobacco consumption, is associated with a worse prognosis [3]. Treatment for HPV negative HNSCC consists of radiation and highly invasive surgery which causes disfigurement. Chemoradiotherapy, with cisplatin/docetaxel followed by radiation has increased progression free survival in recent clinical studies [4]. However, chemoradiotherapy related toxicities have become a growing concern with neutropenia, thrombocytopenia, and dysphagia, ruling out elderly patients and those with comorbidities due to treatment related deaths [5]. Recently, molecular and immune therapies have emerged as viable treatment options, with multi-modal approaches including anti-PD-1 antibody, VEGF and mTOR inhibitors, and EGFR inhibitors such as cetuximab demonstrating modest increases in median overall survival when combined with chemotherapy [6]. Neoadjuvant immunotherapy such as PD-1 checkpoint blockade is emerging as a promising strategy for improving anti-tumor immune responses, particularly in head and neck squamous cell carcinoma (HNSCC). Although PD-1 checkpoint blockade has been approved as first line treatment for advanced HNSCC patients, underlying mechanisms of enhanced treatment responses, as well as optimal therapeutic regimens for improved disease and progression free survival are still incompletely understood [7].

Studies investigating factors affecting efficacy of neoadjuvant therapy demonstrate that systemic immune responses play a central role in tumor eradication. For example, Spitzer et al showed that peripheral immune activation, particularly the expansion of CD4⁺ T cells, is essential for the long-term protection against tumor recurrence and for generating immune memory [8]. More recent studies highlight the critical role of lymph nodes in orchestrating immune responses during PD-L1 immunotherapy in HNSCC patients. One such study demonstrated that progenitor exhausted CD8⁺ T cells in uninvolved lymph nodes are clonally related to exhausted T cells in the tumor, and that anti-PD-L1 therapy can promote their activation and differentiation [9]. Together, these studies highlight that effective neoadjuvant immunotherapy not only targets the tumor microenvironment but also induces broad systemic immune activation, with lymph nodes acting as critical hubs for immune activation and coordination across tissues. These insights suggest that combining local and systemic immune responses could significantly enhance therapeutic outcomes in HNSCC and other cancers. However, there remains a great need for further research into therapeutic options that target the HNSCC tumor microenvironment to enhance the responses to these immunotherapy treatments and improve anti-tumoral efficacy.

Macrophage migration inhibitory factor (MIF) is a pleiotropic cytokine produced by immune and non-immune cells, originally discovered in the 1960s to inhibit the random migration of macrophages [10]. MIF is involved in regulating innate immunity, as it can augment expression of various cytokines including TNF-α, IL-1, IL-6, and IL-8. MIF also counteracts the immunosuppressive effects of glucocorticoids [11]. Numerous studies since then have identified MIF as an important biomarker of cancer progression, being overexpressed in multiple solid tumors. MIF has been associated with osteosarcoma, gastric cancer, lung squamous cell carcinoma, pancreatic cancer, and various forms of HNSCC including nasopharyngeal, oral squamous cell, and esophageal squamous cell carcinomas [12]. MIF binds to CD74/CD44, CXCR2, or CXCR4, and promotes activation of cancer associated MAPK and PI3K-AKT pathways related to chronic inflammation, proliferation, and apoptotic evasion [13, 14]. MIF also promotes the expression of biomarkers of cell invasion and metastases, and MIF knockdown downregulates expression of matrix metalloproteins MMP2/9 [15]. Importantly, MIF expression is correlated with cancer aggressiveness [16]. As such, targeting MIF potentially offers a viable approach in the treatment of solid tumors.

The mechanistic role of MIF in HNSCC is still under active investigation. Early reports demonstrated that exogenous MIF promotes the invasive phenotype of HNSCC cells [17]. High MIF expression in HNSCC tumor tissue is linked to recurrence, metastasis, and poor survival, and MIF knockdown in HNSCC cells decreases cell proliferation and migration, potentially mediated by modulation of MAPK pathways [18]. Previous research by our group further demonstrated that genetic MIF deletion results in improved HNSCC outcomes, associated with decreased inflammatory markers, including IL-1β, TNF-α, and chemokines CCL3 and CXCL1 [19]. This is further supported by Kindt et al, where mice with decreased MIF expression are reported to have better survival, delayed onset of tumors, and better responses to chemotherapy [20]. Recently, Chen et al. demonstrated that MIF knockdown can be combined with UV radiation therapy to promote HNSCC anti-cancer effects [21]. These studies provide insights on the mechanistic effects of MIF on cancer associated pathways, and immunosuppression in the HNSCC tumor micro-environment. Further, these studies highlight the need for in-depth investigation of non-toxic, molecular MIF inhibitors in HNSCC treatment.

MIF inhibition via anti-MIF antibodies was a preclinical option for targeting MIF in the late 1990s and 2000s [22–24]. However, in the past few decades targeting MIF via small molecule inhibitors has proven to be a more effective, better tolerated, and cost-effective approach to cancer treatment [25]. Small molecule MIF inhibition offers a direct way of blocking MIF binding to its corresponding ligands, via interaction with its tautomerase binding site. Various classes of small inhibitors have shown effectiveness in targeting MIF in thyroid cancers, squamous cell carcinoma, glioblastoma, and colon cancers. 4-IPP, a phenylpyrimidine class of inhibitors, has shown significant pharmacological inhibition of MIF in squamous cell carcinoma, decreasing cell proliferation on a dose-dependent basis [26]. ISO-1, a newer generation isooxazoline class inhibitor, has been implicated in inhibiting MIF and reducing cell growth in melanoma, prostate, and colon cancers [27]. Interestingly, Wang et al reported that CPSI-1306 may be up to 100 times more potent when compared to ISO-1, providing a greater indication of using this newer line of inhibitors as a potential therapeutic for targeting cancer. CPSI-1306, is an isoxazoline class of MIF inhibitor that has shown significant results in murine models of triple negative breast cancer and UVB induced squamous cell carcinoma but has not been studied in head and neck cancer [28, 29]. CPSI-1306 inhibits the trimerization of MIF secreted by cancer cells, decreasing its ability to bind to its respective ligands and receptors and decreasing its pro-cancer effects. Given the greater potency of MIF inhibition by CPSI-1306 compared to previous generation MIF inhibitors, combined with its reported minimal toxicity to normal cells, CPSI-1306 offers a potentially viable approach to targeting MIF associated pathways in HNSCC treatment.

In this study we determined the therapeutic efficacy of MIF inhibition with CPSI-1306 in HNSCC treatment using in vitro and two in vivo models (MOC-2 and MOC-1). We further describe mechanisms underlying CPSI-1306 mediated inhibition of oral carcinogenesis, specifically via inhibition of T cell immunosuppressive pathways in the HNSCC tumor microenvironment. Our results reveal a novel therapeutic option for modulating anti-tumoral T cell immunity to improve HNSCC outcomes by targeting MIF.

2. Materials and Methods

2.1. Cell culture and treatments

CPSI 1306 was generously provided by L2 Diagnostics, LLC New Haven, CT. HNSCC cell lines CAL-27 SCC-83 and CA-83, were cultured in DMEM, 10% FBS, and 1% Pen-strep glutamine with supplemental 1% non-essential amino acids. MOC2 cells were cultured in IMDM/F12, 2:1, with 5% FBS, 1% Pen-strep glutamine, 5 ug/mL insulin, 40 ng/mL hydrocortisone, and 5 ng/mL human recombinant EGF.

2.2. Cell Viability assay

Cell viability assay was performed using alamarBlue dye (Thermo Fisher Scientific). 2 × 10³ cells were seeded per well in 100ul of medium into 96-well plates. Cells were allowed to incubate overnight and serially treated with 100 μM with CPSI-1306 for 72 hours. After incubation, 10 ul alamarBlue dye was added and incubated for 6 hours. Absorbance was measured at 570 nm and 600 nm, and percent viability was calculated as described.

2.3. Histology

Tumors were formalin-fixed and paraffin-embedded as previously described[30]. For immunohistochemistry (IHC), Tissue sections (5 μm) were stained for MIF (Proteintech, 20415–1-AP) with secondary biotinylated antibody for goat anti-rabbit (Vector Labs, BA-1000) along with a hematoxylin counterstain. For immunofluorescence (IF), tissue sections (5 μm) were stained for CD8 (Invitrogen, A21434) and Granzyme B (Abcam, AB4059) with the secondary antibodies Alexa Fluor 488-conjugated goat anti-rabbit (Invitrogen, A11034) and Alexa Fluor 555 goat anti-rat (Invitrogen, A21434). Each section was then counterstained with DAPI (BioLegend, San Diego, CA). Confocal imaging was performed sing a Zeiss LSM 700 confocal microscope (CarlZeiss, Munich, Germany). Regions of positive stains were quantified using ImageJ for both IHC and IF.

2.4. Real time quantitative PCR

Human and mouse primer sequences for BCL2, VEGFA, EGFR and CCND1 were obtained using PRIMER BANK (http://pga.mgh.harvard.edu/primerbank/index.html). PCR amplification was performed using the PowerUp SYBR Green Master Mix (Thermofisher Scientific, Foster Coty, CA, USA) for detection. Data normalization was performed to housekeeping genes GAPDH and β-actin.

2.5. Western blotting

Proteins were extracted from in vitro cultures of CAL27, SCC83, CA83, and MOC2 treated with CPSI-1306 at 1 and 10 μM concentrations for 24 h. 30 μg of protein was loaded onto a 10% Tris-HCL gel. Proteins were transferred to a PVDF membrane, blocked using 5% non-fat dry milk and incubated with the rabbit phospho-p44/42 MAPK (Erk1/2) (Cell Signaling, 4370S), p44/42 MAPK (Erk1/2) (Cell Signaling, 4695S) and Rabbit GAPDH (Cell Signaling, 2118S) primary antibody overnight. Blots were incubated with goat anti-rabbit HRP linked secondary antibody (Thermo Fisher Scientific, Rockland, IL). Chemiluminescence was detected by ECL western blotting substrate (Thermo Scientific, Waltman, MA). Quantification of images was performed using the ImageJ FIJI package (Ver. 2.1.0) in ImageJ.

2.6. Animals and Ethics

Experiments were approved by the Institutional Animal Care and Use Committee of the Ohio State University (IACUC) and animals housed and cared for by University Laboratory Animal Resources guidelines. Female C57BL/6 mice were purchased from Jackson Laboratory (Bar Harbor, ME) and were injected orthotopically with MOC-2 (3 × 10⁴ cells) or MOC-1 (5 × 10⁵ cells) into the right buccal cavity. Mice were randomly assigned into vehicle control (n=10) and CPSI-1306 (n=10) groups after injection for each orthotopic cell line model. On days 5 and 8 post-injection, mice began treatment via oral gavage with CPSI-1306 (20 mg/kg in 15% DMSO, 0.5% methylcellulose in water) or vehicle 5 times per week for 2 weeks in MOC-2 tumor-bearing mice and 6 weeks for MOC-1 tumor-bearing mice. Mice weight and tumor volume measurements were recorded twice a week. Tumors were analyzed for volume via equation V= L × S² × 0.5, where L is longest diameter and S is short diameter. At terminal sacrifice, spleen, tumors, and lymph nodes were harvested for flow cytometry and gene expression analysis. Tumors were imaged and analyzed for volume at sacrifice.

2.7. Flow Cytometry

For in vitro analysis of MIF expression, TE1177, SCC83, CA83, and CAL27 cells were incubated with MIF conjugated antibody (Invitrogen, 367401) and MOC2 cells were incubated with MIF fluorochrome conjugated antibody (ProteinTech, 20415–1-AP). Single cell suspensions were generated from spleens, draining lymph nodes and tumors of experimental mice, which were stained extracellularly with fluorochrome conjugated antibodies for CD4, CD8, CD45, CTLA4, TIM3, TIGIT, LAG3, CD69, and PD1 (BioLegend. San Jose, CA, USA). Cells were also incubated with antibodies targeting IL-2, Il-10, IFN- γ, Perforin, Granzyme B, and TNF- α to identify intracellularly expressed markers (BioLegend. San Jose, CA, USA). For myeloid subsets, cells were incubated with fluorochrome conjugated antibodies for Cd11b, CD11c, IA/IE, CD80, F4/80, Ly6C, Ly6G, and PD-L1 (BioLegend. San Jose, CA, USA). Samples were analyzed using FACS Celesta flow cytometer (BD Biosciences, San Jose, CA). Flow cytometric analysis was completed using FlowJo (Tree Star Inc., Ashland, OR, USA).

2.8. Enzyme-Linked Immunosorbent Assay (ELISA)

Serum was collected from blood of MOC-2 and MOC-1 tumor bearing mice treated with CPSI-1306 or vehicle control. The production of cytokines IL-6 and TNF-α was measured using ELISA. All the capture and detection antibodies were purchased from BioLegend (San Diego, CA, USA).

2.8. Statistical Analysis

Statistical analyses were conducted using GraphPad Prism software v9.2.0 (GraphPad Software, San Diego, CA, USA). Student’s t test (two-sided) was used to determine the statistical significance between the groups.

3. Results

3.1. – CPSI-1306 reduces tumor progression in orthotopic in vivo models of HNSCC.

We determined the therapeutic potential of CPSI-1306 in orthotopic models of HNSCC in vivo using the well characterized MOC-2 and MOC-1 HNSCC cell lines [31, 32]. MOC-2 tumor bearing mice were treated with either CPSI-1306 (20mg/kg/day for 2 weeks) or vehicle via oral gavage, 5 days following tumor injections. MOC-1 tumor bearing mice were treated with either CPSI-1306 (20mg/kg/day for 6 weeks) or vehicle via oral gavage, 8 days following tumor injections. Mouse tumors were measured, with reduced tumor volumes noted for MOC-2 and MOC-1 tumors at Day 17 and beginning at Day 36 post injection respectively (Figure 1A). Mouse weights were recorded, with no significant weight differences but on average being higher with CPSI-1306 treatment (Supplementary Figure 1). Mouse tumors were also measured and quantified using ImageJ post sacrifice, with volumes similarly being significantly reduced with CPSI-1306 treatment in both tumor models (Figure 1B-D). This data suggests that CPSI-1306 can reduce head and neck cancer cell growth in vivo. We further characterized expression of MIF in mice bearing tumors after injection with MOC2 HNSCC cells. As expected, MOC2 tumor bearing mice expressed high MIF levels, in tumors of both CPSI-1306 and vehicle treated mice (Figure 1E). MIF expression was also detected in spleens of tumor bearing mice, but not on MIF knockout mice (Figure 1F).

Figure 1:

A) Tumor volumes of MOC2 and MOC1injected mice treated with CPSI-1306 or vehicle control, measured in mm³ B) Tumor sizes of tumors from vehicle control and CPSI-1306 treated MOC2 and MOC1 injected mouse groups, as determined by image J analysis C) Images of tumors from MOC2 injected mice treated with vehicle control and CPSI-1306. D) Images of tumors from MOC1 injected mice treated with vehicle control and CPSI-1306.. E) Representative immunohistochemistry (IHC) images of tumors of mice injected with MOC2 cancer cells stained with MIF antibody. F) Representative IHC images of spleen tissue of wild type C57BL/6 or Mif knock-out mice stained with MIF antibody. Data reported as mean ± SEM, n>=5, *P<0.05.

3.2. – CPSI-1306 enhances T-cell infiltration to the HNSCC tumor microenvironment.

We determined potential mechanisms underlying CPSI-1306 mediated anticancer effects. First, we analyzed CD4⁺ and CD8⁺ T cell infiltration into the tumor microenvironment of MOC2 and MOC1 tumor bearing mice (Figure 2A). CD4⁺ T-cell infiltration was significantly increased in tumors of CPSI treated MOC-2 tumor bearing mice compared to vehicle controls, while a modest but not significant increase in CD8⁺ T cells was observed in CPSI-1306 treated MOC-2 tumor-bearing mice (Figure 2B-C). Both CD4⁺ and CD8⁺ T cells were significantly increased in tumors of CPSI-1306 treated MOC-1 tumor bearing mice compared to vehicle controls (Figure 2D-E). These findings correspond to previous studies showing that MIF inhibition may promote anticancer effects through improved immune activation and function [33]. To determine potential mechanisms underlying the increased T cell infiltration of CPSI-1306 treated tumor bearing mice, we analyzed gene expression levels of chemokines associated with enhanced T cell recruitment. We observed increased expression levels of CCL5 and CXCL9 in the tumor microenvironment of CPSI-1306 treated MOC-2 tumor bearing mice compared to vehicle control group (Figure 2C). Taken together, our data indicates that reduction in tumor progression in CPSI-1306-treated mice is associated with the upregulation of antitumoral T cell attracting chemokines and increased recruitment of T cells to the tumor microenvironment.

Figure 2:

A) Flow cytometry gating strategy of T cell from single cell suspensions derived from tumors of mice treated with CPSI-1306 or vehicle control. Gating for CD4+ and CD8+ cells was performed on CD45+ CD3+ cells. B) Representative flow cytometry plot for CD4+ and CD8+ T cells in MOC2 tumors of mice treated with vehicle control and CPSI-1306. C) Graphical representation of CD4+ and CD8+ T cell subsets expressed as a percentage of CD45+ cells and total live cells in MOC2 tumors of CPSI-1306 and vehicle control treated groups. D) Representative flow cytometry plot for CD4+ and CD8+ T cells in MOC1 tumors of mice treated with vehicle control and CPSI-1306. (E) Graphical representation of CD4+ and CD8+ T cell subsets expressed as a percentage of CD45+ cells and total live cells in MOC1 tumors of CPSI-1306 and vehicle control treated groups. (F) Gene expression profile of chemokines Ccl5, Cxcl9, Cxcl10, Cxcl11, and Ccl21 in tumors obtained from tumor bearing mice treated with CPSI-1306 or vehicle control, normalized to β-actin. * p value < 0.05, ** p value < 0.01

3.3. – CPSI-1306 completely abrogates immunosuppressive checkpoint markers TIGIT, TIM3, and CTLA-4, but not PD-1 on tumor infiltrating CD8⁺ T cells.

Next, we analyzed functional states of T cell subsets recruited to the HNSCC tumor microenvironment following CPSI-1306 treatment. In HNSCC, checkpoint inhibitor markers TIM-3, CTLA-4, TIGIT and PD-1 are overexpressed in T cells and is linked to T-cell exhaustion and dysfunction, dampening effector T-cell responses. Our analysis of tumor infiltrating T cells in MOC-2 tumor bearing mice revealed that CPSI-1306 treatment completely abrogated expression of CTLA-4 (Figure 3A-B), TIM3 (Figure 3D-E) and TIGIT (Figure 3G-H) in CD8⁺ tumor infiltrating lymphocytes compared to vehicle controls, with no significant changes in the spleens or draining lymph nodes. No significant change in PD1 expression was noted in CD8⁺ T cells in the spleen, draining lymph nodes, or tumors (Figure 3J-K). In MOC-1 tumor bearing mice, although levels of the checkpoint inhibitor markers CTLA-4, TIM-3 and TIGIT were not as highly expressed in tumor infiltrating CD8⁺ T cells compared to MOC-2 tumor bearing mice, CPSI-1306 treatment significantly reduced the expression of CTLA-4 (Figure 3C) and TIM-3 (Figure 3F), but not TIGIT (Figure 3I) and PD-1 (Figure 3L) in tumor infiltrating CD8⁺ T cells. We also analyzed the expression of these immunosuppressive markers on CD4⁺ T cells in the spleen, lymph node, and tumors of MOC-2 and MOC1 tumor injected mice treated with vehicle and CPSI-1306. CPSI-1306 treatment did not reduce expression of these checkpoint markers CTLA-4, TIM3, TIGIT, and PD1 in tumor infiltrating CD4+ T cells of MOC-2 and MOC-1 tumor bearing mice, compared to vehicle treated controls (Figure 3A-L). Our data suggests that CPSI-1306 selectively targets tumor infiltrating CD8⁺ T cells to improve anti-tumor immunity in the HNSCC tumor microenvironment.

Figure 3:

Representative flow cytometry plots and summary bar graphs showing percentage of CD4+ and CD8+ T cells expressing A-C) CTLA-4, D-F) TIM3, G-I) TIGIT and J-L) PD-1 in spleens (SP), draining lymph nodes (LN), and tumors (TM) of MOC2 and MOC1 injected mice treated with a vehicle control or CPSI-1306.. Data are presented as mean ± SE. ^*** P value < 0.001, ^**P value < 0.01, * P value < 0.05.

3.4. – CPSI-1306 enhances T cell effector functions in tumors and draining lymph nodes of HNSCC tumor-bearing mice.

We analyzed the effect of CPSI-1306 treatment on T cell effector function in HNSCC tumor bearing mice. To do this, we examined intracellular expression of IFN-γ, TNFα, and Granzyme b (Gzmb) in CD4⁺ and CD8⁺ T cells in spleens, draining lymph nodes and tumors of tumor bearing mice treated with CPSI-1306 or vehicle control. We observed significantly increased levels of IFN-γ producing CD8⁺ T cells in draining lymph nodes compared to vehicle control MOC-2 tumor bearing mice (Figure 4A). Both CD4⁺ and CD8⁺ T cells in the lymph nodes displayed increased levels of TNF-α compared to control treated mice (Figure 4B). Although was slightly but not significantly increased in CD4⁺ and CD8⁺ T cells in the lymph nodes with CPSI-1306 treatment (Figure 4C), Gzmb producing CD8⁺ T cells were significantly increased in the tumor microenvironment of CPSI-1306 treated mice compared to vehicle controls (Figure 4D), demonstrating higher CD8⁺ T cell cytotoxic activity in the primary tumor. Our analysis of systemic immune responses showed slightly, but not significant, increase in serum TNFα, and reduced levels of IL-6 in CPSI-1306 treated MOC-2 and MOC-1 tumor bearing mice compared to vehicle controls (Supplementary Figure 1). Taken together, in addition to modulating CD8⁺ tumor cell checkpoint markers, CPSI-1306 may enhance anti-tumoral immune responses in the primary tumors and draining lymph nodes of metastatic MOC-2 tumor bearing mice, suggesting an immunomodulatory mechanism of anticancer activity by the MIF inhibitor CPSI-1306.

Figure 4:

Representative flow cytometry plots with associated summary bar graphs showing percentage of CD4+ and CD8+ T cells expressing A) IFN-γ, (B) TNF-α and (C) Gzmb in Lymph nodes and Spleens of MOC2 and MOC1 injected tumor bearing mice treated with vehicle control or CPSI-1306. D) Representative immunofluorescence staining of primary tumor tissue from vehicle control and CPSI-1306 treated MOC2 tumor bearing mice stained with DAPI (blue), CD8 (red), and Gzmb (green). Composite image is also shown to indicate CD8 cells expressing Gzmb. Bar graph depicting number of Gzmb producing CD8+ cells in tumors of vehicle control or CPSI-1306 treated mice. Quantification of graphical images were taken from four fields with N = 5 animals per group. Data are presented as mean ± SE. ^**P value < 0.01, * P value < 0.05.

3.5. – Effect of CPSI-1306 on MIF mediated signaling pathways on HNSCC cells

MIF interactions with CD74 leads to activation of the PI3K/AKT and ERK1/2 MAPK pathways, which mediate expression of cancer associated apoptotic and proliferative factors and impact the tumor micro-environment. Given that MIF mediates cancer growth and progression via the PI3K/AKT and ERK1/2 MAPK cascades, we determined the expression of pAKT, ERK, and pERK after treatment with CPSI-1306. We first confirmed expression of MIF in normal (TE1177), and HNSCC (SCC83, CA83, CAL27 and MOC-2) cell lines (Figure 5A-E). Our analysis for the non-cancer human epithelial cell line TE1177 showed a reduction in pAKT and pERK levels (Figure 5F). Interestingly, in MOC-2, CA83, SCC83, and CAL27 cell lines, we observed that pAKT and pERK were not significantly reduced with treatment of CPSI-1306 at 1 or 10 μM (Figure 5G-J).

Figure 5:

(A-E): Histogram plots of A) human oral epithelial cell line TE1177, B) human cancer cell lines SCC83, C) CA83, D) CAL27, and E) mouse oral cancer cell line MOC2, stained with MIF antibody or IgG isotype control antibody, and analyzed by flow cytometry. (F-J): Western blot images and graphical quantification of pERK, pAKT, and total ERK protein expression in (F) TE1177 normal oral epithelial cell line, (G) SCC83, (H) CA83, (I) CAL 27 and (J) MOC2 HNSCC cell lines treated with vehicle or CPSI-1306 for 24 hours at 1 and 10 μM. Western blot images were quantified with Image J, with proteins were normalized against GAPDH. * p value < 0.05; **p value < 0.01; ***p value < 0.001

The data suggests that MIF inhibition via CPSI-1306 may not significantly affect these pathways in vitro in these oral cancer cell lines. Given that other receptor-ligand interactions may exert influence on the PI3K/AKT and ERK1/2 MAPK axis, such as EGFR, our results demonstrate that MIF inhibition with CPSI-1306, may have minimal direct effects on cell signaling pathways associated with MAPK and AKT signaling in HNSCC cells. Further, CPSI-1306 did not inhibit the proliferation of HNSCC cells in vitro (Supplementary Figure 2). Taken together, our data suggests that MIF inhibition using CPSI-1306 indirectly inhibits cancer progression, by enhancing T cell mediated cytotoxicity of HNSCC cells in the tumor microenvironment.

4. Discussion

MIF is a pluripotent cytokine with proinflammatory characteristics which promote HNSCC carcinogenesis. We demonstrated high expression of MIF in HNSCC cell lines, as well as in MOC-2 HNSCC in vivo model. Noticeably, the increased expression of MIF in the more aggressive CA83 and MOC-2 cancer cell lines suggest a positive correlation between MIF overexpression and increased HNSCC aggressiveness. This agrees with previous studies demonstrating the potential for MIF as a biomarker for HNSCC and target for HNSCC treatment [12, 29, 34, 35].

Although MIF inhibitors in general have shown positive results in cell-based analysis, our focus on CPSI-1306, a special isooxaline class of inhibitors, was based on its non-toxic and well tolerated pharmacokinetic profile. We did not observe significant anti-proliferative effects of CPSI-1306 on mouse and human HNSCC cells in vitro, which suggested that the anti-tumoral effects of CPSI-1306 in HNSCC is largely due to enhancement of T cell immune response pathways. Interestingly, while CPSI-1306 did not reduce MOC-2 cell proliferation in vitro, we observed that CPSI-1306 significantly reduced MOC-2 tumor volumes in vivo. Along with MOC-2, we also evaluated the effect of CPSI-1306 on another HNSCC cell line, an indolent cell line MOC-1, where MOC-1 tumor volumes were also significantly reduced with CPSI-1306 treatment. Additionally, tumors were found to be less inflamed and visibly less vascular for both tumor models, suggesting an anti-tumoral effect on the HNSCC tumor microenvironment.

The absence of an inhibitory effect on pERK and pAKT phosphorylation in human and mouse oral cancer cell lines by CPSI-1306 mediated MIF inhibition, suggests active MIF independent activation of these pathways in HNSCC cancer cells. Interestingly, DDT (MIF 2), another member of the MIF family, is not inhibited by CPSI-1306, and has been shown to promote tumor proliferation and development via similar mechanisms as MIF [36, 37]. As such, dual inhibition of MIF and DDT, may prove to be more beneficial than MIF inhibition alone in HNSCC treatment and warrants further exploration. Our results may also suggest a potentially viable strategy via combinatorial approaches targeting both MIF and DDT as well as combining MIF inhibition with EGFR inhibitors currently used in HNSCC treatment.

A key finding of this study was a complete abrogation of exhaustion markers CTLA-4, TIGIT, and TIM3 in CD8⁺ T-cells in the tumor microenvironment. In addition to CTLA4, TIGIT and TIM3 inhibitors are growingly becoming used in clinical studies to target cancer [38], and our data demonstrates the potential of CPSI-1306 mediated MIF inhibition to target checkpoint markers expressed on T cells. Our results indicate that CPSI-1306 contributes to reducing immunosuppressive markers that downregulate cytotoxic T-cell function, thereby restoring T cell anti-tumoral properties in HNSCC. Checkpoint inhibition has been a large focus of immunotherapy, and current FDA approved inhibitors target CTLA-4, PD-1, and PD-L1, with PD-1 inhibitors pembrolizumab and nivolumab being approved for HNSCC. These immune checkpoint inhibitors (ICIs) often are minimally effective in patients in “cold” tumors, with only a fraction of HNSCC patients responding to ICIs [39]. However, combination therapies of ICIs with MIF inhibitors in melanoma have demonstrated enhanced reprogramming of innate immunity and increased potentiation of CD8⁺ T cells, leading to reduced tumor progression [40]. Additionally, in non-small cell lung cancer, esophageal squamous carcinoma, and melanoma patients undergoing PD-1 blockade, high MIF levels correlate with poor progression-free survival, suggesting that MIF inhibition may be combined with PD-1 blockade for a better response [40–42]. Our results depicted that MIF inhibition reprogrammed CD8⁺ T cells by repressing the exhaustion markers except for PD-1 within the tumor microenvironment of both MOC-2 and MOC-1 tumor-bearing mice. Hence, we plan to explore a combinatorial approach involving CPSI-1306 and PD-1 blockade in our future studies. Lower T cell infiltration and expression of other immunosuppressive exhaustion markers contribute to the resistance towards PD-1 blockade response [43]. Since MIF inhibition through CPSI-1306 was able to enhance CD8⁺ T cell infiltration and repress the other exhaustion markers, this combined regimen strategy with PD-1 blockade could offer a promising therapeutic avenue for HNSCC by making PD-1 blockade more responsive.

As evidenced by increased levels of chemokines CCL5 and CXCL9, linked to Th1 and T effector cell responses, MIF inhibition via CPSI-1306 may contribute to a reprograming of the tumor microenvironment characterized by increased infiltration of antitumoral T cells to facilitate an improved tumor response. CXCL9, is associated with Th1 responses and correlates with enhanced anti-tumoral immune cell infiltration [44]. CCL5, is also known to recruit anti-tumoral T cells to the tumor microenvironment, thereby enhancing the immunotherapy response in various cancers [45]. CD8⁺ T cells producing pro-inflammatory and anti-tumoral cytokines, TNF-α and IFN-γ, were significantly elevated in the lymph nodes of MOC-2 tumor-bearing mice treated with CPSI-1306 as compared to vehicle control, whereas the levels of these cytokines remained unchanged in the lymph nodes of MOC-1 tumor-bearing mice. This difference can be attributed to the distinct metastatic behaviors of the two cell lines where MOC-2, a metastatic cell line, readily spreads to the lymph nodes, while MOC-1, an indolent cell line, does not metastasize [46]. Metastasis of MOC2 to the lymph node activates the immune response which when inhibited by MIF inhibition through CPSI-1306 enhances the anti-tumoral response, notably by increasing the population of CD8⁺ T cells producing higher levels of TNF-α and IFN-γ. Conversely, as MOC-1 does not metastasize to the lymph nodes, CD8⁺ T cells are not significantly induced, and MIF inhibition by CPSI-1306 has no effect on their cytokine production in the lymph node of these mice. Moreover, IFNγ is linked to immune activation when secreted by T cells, and promotes production of Gzmb by CTLs leading to apoptosis in cancer cells [47]. Therefore, our observed increase in CD8⁺ Gzmb producing cells within the tumor of CPSI-1306 treated mice in our study suggests a more robust and efficient anti-tumoral immunity against HNSCC, in response to CPSI-1306 treatment.

In conclusion, our findings suggest for the first time that CPSI-1306 can reverse immunosuppressive T cell phenotypes in the tumor microenvironment in an in vivo murine model of HNSCC. CPSI-1306 mediated MIF inhibition led to significant antitumor response in HNSCC, due to heightened immune cell filtration coupled with enhanced T cell antitumoral activity and is associated with reduced tumor growth in vivo. Therefore, targeting MIF via the small-molecule inhibitor CPSI-1306 is a potent strategy against HNSCC – which currently lacks formidable and effective targeted therapy options – and could especially be a promising therapeutic in combination with existing therapies.

Highlights.

• Small molecule MIF inhibitor CPSI-1306 inhibits HNSCC growth in vivo

• CPSI-1306 potently inhibits T cell immunosuppression in the tumor microenvironment

• CPSI-1306 completely abrogates CTLA4, TIGIT and TIM3 in cytotoxic CD8+ T cells

Data availability:

Data will be made available on request.