Immunotherapy has revolutionized cancer treatment. Unfortunately, most tumor types do not respond to immunotherapy due to a lack of immune infiltration or “cold” tumor microenvironment (TME), a contributing factor in treatment failure. Activation of the p53 pathway can increase apoptosis of cancer cells, leading to enhanced antigen presentation, and can stimulate natural killer (NK) cells through expression of stress ligands. Therefore, modulation of the p53 pathway in cancer cells with wild-type TP53 has the potential to enhance tumor immunogenicity to NK cells, produce an inflammatory TME, and ultimately lead to tumor regression. In this study, we report simultaneous targeting of the AKT/WEE1 pathways is a novel and tolerable approach to synergistically induce p53 activation to inhibit tumor development. This approach reduced the growth of melanoma cells and induced plasma membrane surface localization of the ER-resident protein calreticulin, an indicator of immunogenic cell death (ICD). Increase in ICD led to enhanced expression of stress ligands recognized by the activating NK-cell receptor NKG2D, promoting tumor lysis. WEE1/AKT inhibition resulted in recruitment and activation of immune cells, including NK cells, in the TME, triggering an inflammatory cascade that transformed the “cold” TME of B16F10 melanoma into a “hot” TME that responded to anti–programmed cell death protein 1 (anti–PD-1), resulting in complete regression of established tumors. These results suggest that AKT/WEE1 pathway inhibition is a potential approach to broaden the utility of class-leading anti–PD-1 therapies by enhancing p53-mediated, NK cell–dependent tumor inflammation and supports the translation of this novel approach to further improve response rates for metastatic melanoma.

Immunotherapies targeting CD8+ T cells are successful in a minor proportion of patients with advanced-stage melanoma. The majority of patients, however, remain difficult to treat due to a combination of factors, including low mutational burden, insufficient immune infiltration, suppressive tumor microenvironment (TME), frequent mutations in β2-microglobulin (1), and reduced expression of HLA class I molecules (2, 3). Effectively engaging NK cell–mediated immunity is predicted to address some of these issues because NK cells actively drive tumor inflammation and can effectively eliminate HLA class I–negative tumors (3, 4). However, established solid tumors usually have poor NK-cell infiltration likely due to limited homing to tumors (5) and an immunosuppressive microenvironment (6, 7).

Despite their low abundance compared with CD8+ T cells within most tumor types, high intratumoral NK-cell frequency correlates with improved overall survival in melanoma, lung, head and neck, gastric, and colorectal carcinomas, underscoring their role in immunosurveillance (8, 9). Furthermore, NK-cell and conventional type I dendritic cell (cDC1) infiltration in melanoma can predict response to class-leading immunotherapy with anti–PD-1 in both preclinical and clinical settings. This highlights an emerging role for NK cells in priming subsequent melanoma neoantigen–specific T-cell immunity that could have the potential for improving the 5-year survival rates of patients with metastatic melanoma (10, 11).

The role of NK-cell cytotoxicity in tumor immunity has been extensively studied; however, it is becoming clear that NK cells contribute several distinct immune effector functions during the cancer–immunity cycle (8, 9). NK cells must first recognize tumor cells with the activating receptor NKG2D, the best characterized NK-cell receptor in detecting stress ligands on tumor cells, leading to potent effector function and tumor lysis (12, 13). NK-cell cytotoxicity occurs through the release of proinflammatory cytokines, such as IFNγ and TNFα, expression of death receptor ligands, and delivery of granzyme serine-proteases (14–16).

Emerging evidence suggests that restoration of p53 signaling within tumor cells might promote NK-cell recruitment, activation, and function in the tumor microenvironment (TME; refs. 13, 17, 18). For instance, p53-mediated tumor cell senescence leads to chemokine production by tumor cells, such as CCL2, which can recruit NK cells (13). Also, increased p53 activity can cause tumor release of cytokines, such as IL12, IL15, and IL18, all of which activate NK cells (13). Furthermore, the p53 pathway is important for upregulation of NKG2D ligands, such as ULBP1 and ULBP2, leading to increased tumor recognition and lysis (17–20). Finally, p53 activity within NK cells themselves has been shown to be critical for functional maturation (21).

A prominent mechanism to functionally inactivate the p53 pathway in melanoma occurs through Murine Double Minute (MDM) proteins, which are overexpressed in 80% of patients with melanoma. The consequence is reduced responsiveness to DNA damage, decreased apoptosis, and impaired antitumor immunity (22–26). Targeting MDM protein signaling could potentially restore these p53-mediated activities. Unfortunately, therapies directly targeting MDM proteins tend to be toxic or have to be combined with other approaches for improved efficacy, which further increase toxicity (27). MDM antagonists have been reported to induce major hematopoietic defects including bone marrow suppression (27).

Targeting AKT and WEE1 signaling is a potentially nontoxic and novel approach to increase p53 activity in cancers with wild-type TP53 such as melanoma. AKT is a member of the serine/threonine protein kinase family and is activated in approximately 70% of sporadic melanomas (28), and its activation promotes cancer cell survival by deregulating apoptotic signaling (28). AKT as a monotherapy is not effective for melanoma treatments (28, 29), but in vitro data suggest that simultaneous inhibition of AKT and WEE1 could synergistically kill melanoma cells (30, 31). WEE1 kinase is located downstream of V600E-BRAF in the MAPK pathway, which is a frequent genetic alteration in sporadic melanomas (∼50% cases; refs. 32, 33). It normally functions to phosphorylate and inactivate cyclin-dependent kinase 1 in the presence of cellular DNA damage, halting proliferation at the G2–M checkpoint until the damage is repaired (34, 35). Inhibition leads to premature progression into the cell cycle, with the accrual of DNA damage resulting in mitotic catastrophe (36). Simultaneous inhibition of both AKT and WEE1 synergistically reduces cellular proliferation and increases apoptosis in melanoma cells in vitro, which is mediated by activation of the p53 pathway in melanomas and is typically silenced by MDM protein overexpression (30, 31).

In this study, we assessed the efficacy of modulating the p53 pathway in cancer cells containing wild-type (WT) TP53 (where the pathway is inactivated by MDM protein overexpression) to induce NK cell–driven immunotherapy responses. Pharmacologic inhibition of AKT and WEE1 using AZD5363 (capivasertib) and MK1775 (adavosertib), respectively, led to significant synergistic inhibition of melanoma cell growth and increased calreticulin signals indicative of immunogenic cell death (ICD). Increase in ICD further increased NK-cell ligand expression on melanoma and enhanced recognition, leading to NK-cell activation and cytotoxicity. Combining p53 pathway activation with anti–PD-1 immunotherapy increased NK-cell, cDC1, and CD8+ T-cell recruitment into the TME, without toxicity, and led to regression of syngeneic melanomas, which are typically anti–PD-1–insensitive. Depletion of NK cells, CD8+ T cells, or inactivation of p53 in melanoma cells ablated the therapeutic efficacy, thus, validating the importance of the p53 pathway in NK cell–based potentiation of anti–PD-1 immunotherapy responses.

Cell lines, culture conditions, and chemicals

The human melanoma cell lines 1205 Lu, WM164, SK-MEL-28, 451 Lu, and murine melanoma line B16F10 were provided by Dr. Meenhard Herlyn (Wistar Institute, Philadelphia, PA). The human melanoma cell line UACC 903 was provided by Dr. Mark Nelson (University of Arizona, Tucson, AZ). The NK-92 cell line was procured from ATCC. Mouse melanoma cell lines M1 (Mel114433), M3 (HCmel1274), and M4 (B2905) were provided by Dr. Glenn Merlino, National Cancer Institute (NCI, Rockville, Maryland). M1 tumors were induced by UV radiation in BrafCA/+; Ptenfox/+; Cdkn2afox/+; Tyr-CreERT2-tg mice whereas M4 tumors were induced by UV radiation in Hgftg mice (37). Cell lines were maintained in a 37°C humidified 5% CO2 incubator and periodically monitored for phenotypic and genotypic characteristics, as well as tumorigenic potential, to validate and confirm cell line identity. Cell lines were authenticated and tested for Mycoplasma contamination periodically and were used within 15 passages after authentication. Clinical-grade pharmacologic inhibitors of AKT (AZD5363/capivasertib) and WEE1 (MK1775/adavosertib) were provided by AstraZeneca. AZD5363 was discovered by AstraZeneca subsequent to a collaboration with Astex Therapeutics (and its collaboration with the Institute of Cancer Research and Cancer Research Technology Limited).

Lentiviral packaging and infection

Lentiviral constructs were transfected into 1 million 293T (ATCC) cells using Lipofectamine 2000 (Thermo Fisher). Lentivirus was packaged using the ViraPower Kit (Invitrogen) following the manufacturer's instructions. B16F10 and 1205 Lu cells were seeded in 6-well plates at the density of 500,000 cells/well and infected with 400 μL lentiviral media containing pLKO.1 empty vector or pLKO.1-shp53 (Pennsylvania State University, College of Medicine, shRNA genomic core, Hershey, PA; human shp53: CCGGGTTCCTGCATCTTGACCAATACTCGAGTATTGGTCAAGATGCAGGAACTTTTT, CCGGCGGCGCACAGAGGAAGAGAATCTCGAGATTCTCTTCCTCTGTGCGCCGTTTTT, mouse shp53: CCGGCCAGTCTACTTCCCGCCATAACTCGAGTTATGGCGGGAAGTAGACTGGTTTTT). B16F10 and 1205 Lu cells were cultured in 6-well plates and selected with puromycin (Millipore Sigma; 3 μg/mL) for 7 days.

Cell viability assays

Cell viability assays of melanoma cells treated with p53 modulators, AZD5363 and MK1775, were performed as described previously (38–40). Briefly, 5,000 B16F10 and 1205 Lu cells (parental and p53 KD) per well were plated in a 96-well plate and incubated overnight at 37°C in a 5% CO2 atmosphere. Cells were treated with AZD5363 and MK1775 at concentrations indicated in Fig. 1B along with 0.05% DMSO which was the highest amount of DMSO in the treatment and used as controls, and incubated for 72 hours. Twenty microliters of MTS reagent (2 μg/μL MTS, Promega, and 0.05 μg/μL PMS, Millipore Sigma) was then added into each well, and formation of tetrazolium was measured by absorbance after 1 hour at 492 nm using the spectrometer, SPECTRA max M2 (Molecular Devices). IC50 values or % cells for each experimental group were measured in three independent experiments (41, 42).

Figure 1.

Modulating p53 pathway signaling in tumors to induce immunogenic cell death. B16F10 p53 knockdown (KD) and 1205 Lu p53 KD cells were created using shRNA. A, 1205 Lu cells (control) and 1205 p53 KD cells were treated with doxorubicin and AKT/WEE1 inhibitors for 24 hours. Cell lysates were probed for Western blot analysis with p53, and ERK2 served as a loading control. Representative of two independent experiments is shown. B, Melanoma cells with wildtype p53 (1205 Lu and B16F10) and those lacking p53 (1205 Lu p53 KD and B16F10 p53 KD) were treated with the indicated concentrations (μM) of AKT (AZD5363) and WEE1 (MK1775) inhibitors for 72 hours, followed by estimation of cell viability using an MTS assay. Representative of two independent experiments is shown. C, Nude mice bearing 1205 Lu (control) and 1205 Lu p53 KD subcutaneous tumors were treated with AZD5363 and MK1775 at 150 mg/kg and 50 mg/kg, respectively, daily through oral gavage. Tumor growth was measured over time. Mean + SEM is shown for 6 mice per group. Experiment was repeated three times. A representative is shown. D, B16F10 (control) and B16F10 p53 KD were treated with AKT (6 μmol/L AZD5363) and WEE1 (3 μmol/L MK1775) inhibitors alone and in combination for 24 hours. Calreticulin was measured by flow cytometry. DMSO-treated cells served as controls. Representative of two independent experiments is shown. E, 1205 Lu cells (control) and 1205 p53 KD cells were treated with doxorubicin and AKT/WEE1 inhibitors for 24 hours. Cell lysates were probed with p53, HMGB1, ERp57, and ERK2 by Western blot. Representative of two independent experiments is shown. ***, P < 0.001 (Statistical test: ANOVA + Dunnett post hoc). Error bars represent mean + SEM.

Figure 1.

Modulating p53 pathway signaling in tumors to induce immunogenic cell death. B16F10 p53 knockdown (KD) and 1205 Lu p53 KD cells were created using shRNA. A, 1205 Lu cells (control) and 1205 p53 KD cells were treated with doxorubicin and AKT/WEE1 inhibitors for 24 hours. Cell lysates were probed for Western blot analysis with p53, and ERK2 served as a loading control. Representative of two independent experiments is shown. B, Melanoma cells with wildtype p53 (1205 Lu and B16F10) and those lacking p53 (1205 Lu p53 KD and B16F10 p53 KD) were treated with the indicated concentrations (μM) of AKT (AZD5363) and WEE1 (MK1775) inhibitors for 72 hours, followed by estimation of cell viability using an MTS assay. Representative of two independent experiments is shown. C, Nude mice bearing 1205 Lu (control) and 1205 Lu p53 KD subcutaneous tumors were treated with AZD5363 and MK1775 at 150 mg/kg and 50 mg/kg, respectively, daily through oral gavage. Tumor growth was measured over time. Mean + SEM is shown for 6 mice per group. Experiment was repeated three times. A representative is shown. D, B16F10 (control) and B16F10 p53 KD were treated with AKT (6 μmol/L AZD5363) and WEE1 (3 μmol/L MK1775) inhibitors alone and in combination for 24 hours. Calreticulin was measured by flow cytometry. DMSO-treated cells served as controls. Representative of two independent experiments is shown. E, 1205 Lu cells (control) and 1205 p53 KD cells were treated with doxorubicin and AKT/WEE1 inhibitors for 24 hours. Cell lysates were probed with p53, HMGB1, ERp57, and ERK2 by Western blot. Representative of two independent experiments is shown. ***, P < 0.001 (Statistical test: ANOVA + Dunnett post hoc). Error bars represent mean + SEM.

Close modal

Apoptosis assay

The Annexin-V-PE/7-AAD Kit (Millipore Sigma) was used to distinguish live cells from apoptotic cells as described previously (41–44). Briefly, 1205 Lu parental and p53 KD cells were incubated with indicated concentrations of the p53 modulator, AZD5363 (3 or 6 μmol/L) and MK1775 (1 or 2 μmol/L) or 0.05% DMSO for 24 hours. Cells were washed with PBS and stained with Annexin-V-PE and 7-AAD solution per the manufacturer's instructions. Cells were acquired by BD Fortessa flow cytometer and gated for four distinct regions, namely, live cells (Annexin V-7AAD), early apoptotic (Annexin V-7+AAD), late apoptotic (Annexin V-7+AAD+), and necrotic (Annexin V-7AAD+) cells using the analyzing software, FLOWJO, v10.7.

NK-cell cytotoxicity assay

100,000 1205 Lu parental and p53 KD cells were treated with indicated concentrations of p53 modulator, AZD5363 (3 or 6 μmol/L) and MK1775 (1 or 2 μmol/L) or 0.05% DMSO for 24 hours. NK-92 cells were then added at the ratio of 8:1 or 4:1 of NK cells:melanoma cells and incubated for 4 hours. The number of healthy melanoma cells was then evaluated using the apoptosis assay described above.

NK-cell ligands

B16F10 and 1205 Lu (parental and p53 KD) cells were treated with p53 modulators, 6 μmol/L AZD5363, 2 μmol/L MK1775, 20 μmol/L nutlin (Tocris), 40 μmol/L SJ172550 (AdooQ Bioscience), and 0.05% DMSO and incubated for 24 hours. Cells were trypsinized, washed with PBS (Hyclone), and labeled with 7-aminoactinomycin D (7AAD; 0.25 μg, BD Pharmingen 51-68981E) and incubated for 15 minutes in the dark to exclude nonviable cells. Cells were washed twice with FACS buffer [2% FBS (R&D Systems) and 0.1% NaN3 (Millipore Sigma) in PBS] and stained with different NK ligand antibodies, anti-MICA-PE (FAB1300P, R&D Systems), anti-MICB-AF488 (FAB1599G, R&D Systems), anti-ULBP-1-PerCP (FAB1380C, R&D Systems), anti-ULBP2/5/6-AF647 (FAB1298R, R&D Systems), anti-ULBP3-AF450 (FAB1517S, R&D Systems), anti-Nec2-PE-Cy7 (clone: TX31, 337414, BioLegend) at 1:100 dilution and incubated for 30 minutes at room temperature in the dark. Cells were washed twice with FACS buffer and all samples were immediately run on a BD LSR Fortessa flow cytometer, and the data were analyzed using FlowJo software (v10.7).

Western blot analysis

1205 Lu parental and p53 KD cells were harvested by the addition of RIPA lysis buffer (Thermo Fisher) and protease and phosphatase inhibitors (Thermo Fisher), and samples were processed as previously described (30, 45). Briefly, 1–2 × 106 cells were incubated overnight at 37°C in a 5% CO2 atmosphere. For experiments with p53 modulators, 0.5 μg/mL doxorubicin (LC Laboratories), AZD5363, MK1775 (concentrations indicated in Fig. 1A and E) were added, and protein lysates collected via RIPA lysis buffer following 24 hours of treatment. Thirty micrograms of proteins/sample were loaded into wells of a NuPage 4% to 12% Bis-Tris gel (Invitrogen) and separated using the MES SDS running buffer (Invitrogen) in the Invitrogen gel running system (Invitrogen). Proteins were transferred from gels to polyvinylidene difluoride membranes (Millipore Sigma) and blocked with 5% nonfat milk in TBS (25 mmol/L Tris, 150 mmol/L NaCL, 2 mmol/L KCl, pH 7.4). Blots were probed with primary antibodies, anti-p53 (Santa Cruz Biotechnology), anti-HMGB1 (Santa Cruz Biotechnology), anti-peIF2α (Santa Cruz Biotechnology), and anti-ERK2 (Santa Cruz Biotechnology) at 1:1,000 dilution at 4°C overnight. After washing with TBST [TBS + 0.5% Tween-20 (Fisher Scientific)] three times, blots were probed with secondary antibodies (Genesee Scientific) at 1:10,000 ratio for 2 hours at room temperature followed by three times TBST wash. Blots were developed using the Enhanced Chemiluminescence Detection System (Thermo Fisher Scientific).

Calreticulin expression

In vitro calreticulin expression assay was performed as described previously (46). Briefly, mouse melanoma cells, B16F10 parental and p53 KD, were seeded at 1.5–2.0 × 105 cells per well in 24-well plates in RPMI1640 medium (Gibco) and 10% FBS and incubated overnight at 37°C, 5% CO2. 3 μmol/L AZD5363, 1 μmol/L MK1775, the combination, and 0.05% DMSO were added to cells and incubated for a further 24 hours at 37°C, 5% CO2. Cells were trypsinized and washed with FACS buffer and seeded at 1–2 × 105 per well in 96-well round-bottom plates. Cells were stained with polyclonal rabbit anti-calreticulin (1:100, Abcam, ab2907) in FACS buffer for 30 minutes at 4°C. Following two washes with FACS buffer, cells were labeled with goat anti-rabbit Alexa Fluor 488 (1:500, Thermo Fisher A11070) in FACS buffer for 30 minutes at 4°C. Cells were washed twice with FACS buffer and labeled with 7-AAD (0.25 μg, BD Pharmingen 51-68981E) to exclude nonviable cells. All samples were immediately run on a BD LSR Fortessa flow cytometer, and the data analyzed using FlowJo software (v10.7).

Tumor efficacy and toxicity assessment

All animal studies were conducted in accordance with, and with the approval of, Penn State Institutional Animal Care and Use Committee (IACUC). Efficacy and toxicity studies were performed in nude mice (Envigo) or C57BL/6J mice (Jackson Laboratory) as described previously (30, 47–49). Briefly, 1 million human melanoma 1205 Lu cells were injected in both flanks of 4- to 6-week-old female nude Balb/c mice. Similarly, 100,000 B16F10 or 1 million M1/M4 melanoma cells were injected subcutaneously into both female and male C57BL/6J mice. Power analysis was performed to identify the number of animals per group. NK-cell or CD8+ T-cell depletion was achieved in C57BL/6J mice by treatment on days -7, -3, -1 and one day after tumor cell injection with 100 μg/mouse anti-NK1.1 (clone PK136; BioXCell) or anti-CD8α (clone 2.43; BioXCell) intraperitoneally. Depletion was maintained throughout the experiment by treating the animals once a week with the antibodies. After a week, when the tumors were vascularized, animals were randomized on the basis of the tumor volumes and treated with different drug combinations as indicated: AZD5363 at 150 mg/kg, oral, daily or MK1775 at 50 mg/kg, oral, daily (control: DMSO solvent). Mice also received anti–PD-1 (clone RMP1-14; BioXCell) or control rat IgG (Sigma) at 200 μg/day intraperitoneally twice per week. Tumor volumes measured by a caliper, animal weight, and behavior were monitored continuously every other day. Animals were sacrificed after tumor volumes exceeded 2,000 mm3, and tumors and spleens were removed for further analysis.

Immune cell analysis

The spleens and tumors from B16F10-treated mice were dissected at from day 18 onwards (dictated by ethical tumor size limit) and placed into RPMI supplemented with 10% FBS and maintained on ice. Tumors were dissociated by digesting with collagenase-IV (Gibco). Briefly, spleens were minced with razor blades and passed through a wire mesh to form single-cell suspensions. Tumors were minced, resuspended in 3 mL of collagenase mixture [1 mg/mL collagenase-IV + 50 U/mL DNAse I (Roche) in RPMI media + 2% FBS] and gently rocking for 30 to 45 minutes at 37°C before passing samples through a wire mesh to form single-cell suspension. Samples were centrifuged at 1,200 rpm for 7 minutes, resuspended in 2 mL Tris-NH4Cl2 and incubated at 37°C for 5 minutes. 4 mL RPMI+2% FBS were added to stop the reaction and samples were centrifuged at 1,200 rpm for 7 minutes before resuspended in 5.5 mL media. Aliquots of 2 × 106 cells were stained with commercially available antibodies prior to analysis on an LSR Fortessa flow cytometer (BD Biosciences) in the Penn State Flow Cytometry Shared Resource. Samples were stained with fixable viability stain (BD Biosciences) prior to analysis to eliminate dead cells. Data were analyzed using FlowJo software (v10.4, FlowJo, LLC). Antibodies used included: anti-CD45.2-BV480 (clone 104), anti-CD4-BB700 (clone RM4-5), anti-CD8-BV786 (clone 53–6.7), anti-NK1.1-FITC (clone PK136), anti-CD3-PE (clone 145-2C11), anti-CD11c-APC-Cy7 (clone HL3), anti-CD11b-APC (clone M1/70), anti-F4/80-PerCP-Cy5.5 (clone T45-2342), anti-Ly6C-BV785 (clone HK1.4), anti-Ly6G-VF450 (clone1A8), anti-NKG2D-BV711 (clone CX5), anti-CD25-PE-texas Red (PC61), anti-PD1-PE/Dazzle (RMP1-30), and anti-FoxP3-APC (MF23). Antibodies were purchased from BD Biosciences: anti-CD45, anti-CD4, anti-CD8, anti-CD11c, anti-CD11b, and anti-F4/80; BioLegend: anti-NK1.1, anti-CD3, anti-Ly6C, anti-NKG2D, anti-CD25, and anti–PD-1; and Tonbo Biosciences: anti-Ly6G, anti-FoxP3. Live cells were gated on the CD45.2+/FVS population. Gating strategy for tumor–immune infiltrating cells: singlets (using FSC-W vs. SSC-A plots), live cells (Fixable viability stain-FVS700), CD45+ cells (CD45.2-BV480), CD3+ (CD3e-PE) cells were identified from single-cell suspension of tumors followed by CD4 and CD8 (CD4-BB700 and CD8-BV786). NK cells were identified from CD45+CD3 cells using NK1.1 (NK1.1-FITC). Subsets such as Ki67+ (BV421), GranzymeB+ (PE-Cy7), NKG2D+ (BV711), NKp46+ (PerCPCy5.5), IFNγ+ (PE-Cy5) were evaluated within NK cells. Similarly, Ki67+ (BV421), GranzymeB+ (PE-Cy7), IFNγ+ (PE-Cy5) were identified within CD8 T-cell subsets. Regulatory T cells (Treg) were identified from CD4+ T cells using CD25 (PerCPCy5.5) and FoxP3 (APC). In addition, CD11b cells (singlet/live/CD45+/CD11b+) cells were further characterized as macrophages (F4/80+ APC), DCs (CD11c+ APC-Cy7), PMN-MDSCs (Ly6g+ VFluor 450) and monocytic-MDSCs (LY6c+ BV786). To validate MHC-I–related gene expression, cells were stained with H2-Kb-APC antibody (clone 28-8-6, BioLegend) and analyzed by flow cytometry.

RNA extraction

Tumors were removed at the end of the experiment, followed by the RNA extraction. A bead mill homogenizer (Bullet Blender, Next Advance) was used to homogenize the tissue. Tumors from multiple mice per treatment group were pooled because very little tumor mass was present in individual mice following combination and triple therapy. Approximately 30 to 60 mg of tissue sample was transferred to a safe-lock microcentrifuge tube (Eppendorf). A mass of stainless-steel beads (Next Advance, catalog no. SSB14B) equal to the mass of the tissue was added to the tube. Two volumes of TRI Reagent (Zymo Research) were added to the tube, and samples were immediately mixed in the Bullet Blender (Next Advance, Troy, New York) for 1 minute at a speed of ten. Samples were visually inspected to confirm desired homogenization and then incubated at 37°C for 5 minutes. The TRI Reagent was added up to 0.6 mL, and samples were mixed in the Bullet Blender for 1 minute. Total RNA was extracted using Direct-zol RNA Miniprep Kit (Zymo Research). Optical density values of extracted RNA were measured using a NanoDrop (Thermo Scientific) to confirm an A260:A280 ratio above 1.9. RNA integration number (RIN) was measured using TapeStation (Agilent Technologies) RNA Kit. Measured RIN scores were 3.8 ± 0.2 (mean ± SD). We utilized 3′mRNA-Seq Library Prep Kit FWD for Illumina (Lexogen), which is designed to generate Illumina compatible libraries of sequences close to the 3′ end of polyadenylated RNA and proven to be suitable for moderately degraded RNA.

Library preparation and sequencing for mRNA

The cDNA libraries were prepared using the QuantSeq 3′mRNA-Seq Library Prep Kit FWD for Illumina (Lexogen) as per the manufacturer's instructions. Briefly, 64 to 1,360 ng of total RNA was reverse transcribed using oligo (dT) primers. The second cDNA strand was synthesized by random priming, in a manner that DNA polymerase is efficiently stopped when reaching the next hybridized random primer, so only the fragment close to the 3′ end gets captured for later indexed adapter ligation and PCR amplification. The processed libraries were assessed for size distribution and concentration using the BioAnalyzer High Sensitivity DNA Kit (Agilent Technologies). Pooled libraries were diluted to 2 nmol/L in EB buffer (Qiagen) and then denatured using the Illumina protocol. The denatured libraries were loaded onto an MiSeq v3 flow cell on an Illumina MiSeq and run for 151 cycles using a single-read recipe according to the manufacturer's instructions. Demultiplexed sequencing reads were generated using Illumina bcl2fastq (released version 2.20.0.422, Illumina), allowing no mismatches in the index read. After the quality and polyA trimming by BBDuk and alignment by HISAT2 (version 2.1.0), read counts were calculated using HTSeq by supplementing Ensembl gene annotation (GRCm38.78). Raw fastq and counts data generated during this study are available at GEO (GSE199096).

RNA-sequencing analysis

After the quality and polyA trimming by BBDuk and alignment by HISAT2 (version 2.1.0), read counts were calculated using HTSeq by supplementing an Ensembl gene annotation (GRCm38.78). The GENCODE M24 annotation was downloaded and used to retrieve effective gene lengths based on exons. The edgeR (version 3.32.1) Bioconductor package (50) was used to process the data in R (versions > 3.6). We retained only protein-coding genes that had a count-per-million (CPM) > 2 in at least one sample; genes in the molecular signatures used in this study (indicated below) were maintained. Trimmed mean of M-value (TMM) normalization was performed, and logRPKM values were calculated. This data where then used for single-sample scoring and data visualizations.

We used the singscore method through its Bioconductor package (version 1.10.0; 51) and several molecular signatures to quantify the concordance of each individual sample with each signature. Immune cell signatures were obtained from Jerby-Arnon (52), general IMMUNE and STROMA signatures were obtained from Aran and colleagues (53), and the rest of the signatures were available from MSigDB (54) and imsig (55) R packages. ComplexHeatmap R package (version 2.6.2; ref. 56) was used to visualize scores and gene expression values. Data visualizations in this study were performed using tidyverse (version 3.1.3) packages (https://tidyverse.tidyverse.org/articles/paper.html), specifically ggplot2 (version 3.3.3).

Statistical analysis

Statistical analysis was undertaken using the one-way/two-way ANOVA in GraphPad PRISM version 7.04 software. Dunnett post hoc analysis was performed when there was a significant difference. Results were considered significant at a P < 0.05.

Targeting p53 activity in melanoma to induce immunogenic cell death for NK cell–mediated immunity

Targeting the p53 pathway to induce NK cell–mediated immunity for melanoma treatment has largely been unexplored. Approximately 80% of melanomas have wild-type p53, with high MDM2/4 proteins leading to the functional inactivation of the p53 pathway (22–26). The MDM proteins reduce the transcriptional transactivation of proapoptotic proteins in tumor cells and promote an immune-suppressive TME. Therapies directly targeting MDM proteins tend to be toxic or must be combined with other approaches for improved efficacy, further increasing toxicity. One alternative is to indirectly increase p53-mediated transcriptional activity in tumor cells and reverse the immunosuppressive TME by targeting AKT/WEE1 pathways (30, 31). This approach synergistically increased p53 (Fig. 1A).

Pharmacologic inhibitors of AKT (AZD5363/capivasertib) and WEE1 (MK1775/adavosertib) can synergistically reduce the kinetics of melanoma tumor development (30, 31). To demonstrate the efficacy of activating the p53 pathway by targeting AKT and WEE1, melanoma survival was evaluated in the presence of MK1775 (1 μmol/L) and AZD5363 (0.63, 2.5, 10 μmol/L). Combination of AZD5363 and MK1775 was synergistic at killing 1205 Lu melanoma cells (WT TP53; Fig. 1B). To evaluate the role of TP53 in this process, melanoma lines expressing shRNA targeting TP53/p53 were generated. In contrast to WT TP53-expressing cells, the combination of AZD5363 and MK1775 had no effect on Lu1205 shp53–expressing cells [p53 knockdown (KD); Fig. 1B]. Similar results were also observed with the mouse melanoma cell line B16F10, where the combination of AZD5363 and MMK1775 were synergistic in inducing melanoma cell death only in p53-expressing cells (Fig. 1B). Furthermore, the combination of AZD5363 and MMK1775 was not effective in human melanoma cells with a mutant TP53 gene, such as WM164, 451Lu, and SK-MEL-28 cells, as the addition of AZD5363 did not enhance the potency of MMK1775 treatment alone (Supplementary Fig. S1A).

To validate the effect of p53 on the AKT/WEE1-targeting combination in tumors, 1205 Lu and 1205 Lu p53 KD cells were injected into nude mice. After the tumor volumes reached 50 mm3, the mice were randomized to different groups and treated with vehicle or combination AZD5363 (150 mg/kg) and MK1775 (50 mg/kg), daily via oral gavage. Tumor growth was monitored every other day (Fig. 1C). The combination was not toxic and did not have an effect on animal weights (Supplementary Fig. S1B), whereas reduced tumor volumes (by ∼80%) were observed compared with controls in 1205 Lu cell xenografts. The drug combination had a nonsignificant effect on 1205 Lu p53 KD cells (Fig. 1C), demonstrating the importance of p53 pathway on the efficacy of AKT/WEE1 treatment. To further elucidate the effects of p53 modulation on ICD, calreticulin expression was also evaluated in B16F10 p53 KD cells. p53 KD in B16F10 cells reduced calreticulin relocation induced by the AKT/WEE1 combination (Fig. 1D). We failed to detect a noticeable increase in ERp57, which is required for calreticulin surface expression following AKT/WEE1-inhibitor treatment of 1205 Lu melanoma cells; however, another protein characteristic of ICD, HMGB1, was clearly induced in 1205 Lu melanoma cells but not in 1205 Lu p53 KD cells (Fig. 1E). p53 KD was confirmed by Western blot analysis, where p53 expression was compared between p53 WT and KD cells induced using doxorubicin or the AKT/WEE1 inhibitor drug combination (Fig. 1A and E).

WEE1/AKT inhibition leads to p53-dependent expression of ligands to activating NK-cell receptors

Reports indicate that the p53 pathway is important for upregulation of NKG2D ligands, such as ULBP1 and ULBP2, leading to increased tumor recognition and destruction (17–20). Furthermore, p53 activity within NK cells has been shown to be critical for functional maturation (21). To evaluate the effect of increasing p53 activity on NK ligands in melanoma cells, expression of different NK ligands was evaluated with or without AZD5363 and MK1775 combination treatment (Fig. 2A). Treatment with AKT/WEE1 inhibitors significantly increased the expression of ULBP1, ULBP2, and ULBP3, along with MICB and Nec2 in 1205 Lu cells, whereas no significant difference was seen in the surface expression of MICA, although this could be masked by shedding of soluble MICA. There was a 3.1-fold, 2.7-fold, 1.6-fold, 1.3-fold, and 2.6-fold increase in ULBP1, ULBP2, ULBP3, MICB, and Nec2, respectively, with AZD5363+MK1775 combination at 3 μmol/L:1 μmol/L ratio. In contrast, treatment with MDM2 or MDM4 inhibitors, nutlin or SJ172550, significantly upregulated only ULBP1, ULPB2, and Nec2. Nutlin increased ULBP1, ULBP2, and Nec2 by 2.1-fold, 1.4-fold, and 2.5-fold, respectively, whereas SJ172550 increased ULBP1, ULKBP2, and Nec2 by 2-fold, 1.3-fold, and 2-fold, respectively. The increased expression of ligands to NKG2D and DNAM1 was abrogated in p53 KD cells, demonstrating the importance of p53 signaling in tagging stressed or malignant cells for recognition and destruction by NK cells. Similar results were also observed in B16F10 cell lines (Fig. 2B). Given the enhanced ICD and upregulation of NKG2D ligands following inhibition of AKT/WEE1 pathways in melanoma cells, we next investigated whether these events significantly increased their sensitivity to NK-cell cytotoxicity by coincubating tumor cells treated with p53 modulators and NK cells in vitro. NK cells effectively killed 1205 Lu cancer cells at an effector to target ratio of 8:1. The AKT+WEE1 inhibitor combination significantly increased NK-cell cytotoxicity to target cells, and knockdown of p53 in cancer cells abrogated the efficacy of AKT/WEE1 inhibitors in increasing melanoma sensitivity to NK-cell killing (Fig. 2C). Taken together, these data demonstrate the advantage of targeting AKT/WEE1 pathways for directly increasing NKG2D/DNAM1 ligand expression and potential recognition and killing by NK cells.

Figure 2.

Increasing p53 pathway signaling in tumor cells induces NK-cell ligands and enhances immunogenicity. 1205 Lu, 1205 Lu p53 KD, B16F10, and B16F10 p53 KD cells were treated with p53 modulators (6 μmol/L AZD5363+2 μmol/L MK1775, 20 μmol/L nutlin, or 40 μmol/L SJ172250) and incubated for 24 hours. Mean fluorescence intensities (MFI) of the indicated NK-cell ligands in 1205 Lu and 1205 Lu p53 KD (A) or B16F10 and B16F10 p53 KD cells (B) following treatment were measured by flow cytometry. C, 1205 Lu and 1205 Lu p53 KD cells were cultured with the indicated doses of AZD5363+MK1775 for 24 hours, washed, and cocultured with NK-92 cells for 4 hours at a NK-cell:tumor cell ratio of 8:1. Tumor cell viability was monitored by Annexin V by flow cytometry. Representative of two independent experiments is shown for all panels. Statistical test: ANOVA + Dunnett post hoc. Error bars represent mean + SEM.

Figure 2.

Increasing p53 pathway signaling in tumor cells induces NK-cell ligands and enhances immunogenicity. 1205 Lu, 1205 Lu p53 KD, B16F10, and B16F10 p53 KD cells were treated with p53 modulators (6 μmol/L AZD5363+2 μmol/L MK1775, 20 μmol/L nutlin, or 40 μmol/L SJ172250) and incubated for 24 hours. Mean fluorescence intensities (MFI) of the indicated NK-cell ligands in 1205 Lu and 1205 Lu p53 KD (A) or B16F10 and B16F10 p53 KD cells (B) following treatment were measured by flow cytometry. C, 1205 Lu and 1205 Lu p53 KD cells were cultured with the indicated doses of AZD5363+MK1775 for 24 hours, washed, and cocultured with NK-92 cells for 4 hours at a NK-cell:tumor cell ratio of 8:1. Tumor cell viability was monitored by Annexin V by flow cytometry. Representative of two independent experiments is shown for all panels. Statistical test: ANOVA + Dunnett post hoc. Error bars represent mean + SEM.

Close modal

Modulating p53 enhances immune infiltration and NK-cell activity to potentiate anti–PD-1 responsiveness

There is emerging evidence that response to immune checkpoint blockade is positively influenced by the magnitude of NK-cell infiltration in the tumor at the time of therapy (11, 57). Targeting the PD-1/PD-L1 axis can lead to regression of immune-infiltrated tumors, and the role of CD8+ T cells in regression is established in most settings. Several preclinical tumor models demonstrate that both CD8+ T cells and NK cells are essential for optimal tumor regression following anti–PD-1 therapy (11). Furthermore, anti–PD-1 nonresponders can eventually respond to anti–PD-1 therapy following treatment of systemic NK cell–stimulating agents (11). At present, there are no established targeted therapeutic approaches to increase NK-cell infiltration into solid tumors. To evaluate the effect of NK-cell activation following p53-targeting with AKT/WEE1 inhibition on anti–PD-1–mediated tumor regression, we chose a preclinical melanoma model that does not respond to anti–PD-1 alone, B16F10. After tumors reached 50 mm3, mice were randomized into groups and treated with AKT/WEE1 (150/50 mg/kg, oral, daily), anti–PD-1 (200 μg/mouse, twice a week, i.p.), or a combination of both. Animal weights, tumor volumes, and behavioral changes of mice were monitored regularly. There was no significant difference in tumor volumes following treatment with anti–PD-1 alone, as was previously reported (ref. 58; Fig. 3A). AKT/WEE1 inhibitor treatment reduced tumor volumes by 65%, and the triple combination of AKT/WEE1/PD-1 inhibitors reduced tumor volumes by >95%, along with significantly increasing the survival of animals for >40 days compared with the individual treatments (Fig. 3A and B). None of the treatment groups had effects on animal weights, indicating the safety of these regimens (Supplementary Figs. S1B and S2A). Immune phenotype analysis of the TME by flow cytometry revealed a significant increase in immune cells, including NK cells, cDC1s, and CD8+ T cells with triple combination treatment compared with individual treatments (Fig. 3C). Anti–PD-1 treatment alone did not have an effect on tumor-infiltrating immune cells. The AZD5363+MK1775 combination increased tumor-resident NK cells by 2-fold, whereas the combination with anti–PD-1 increased NK-cell infiltration by 2.5-fold. A significant increase in the expression of granzyme B–expressing NK cells was observed in tumors, suggesting that the modulation of p53 along with checkpoint inhibition has a profound impact on the infiltration, activity, and function of NK cells (Fig. 3C). In addition, triple combination therapy significantly increased the infiltration of CD8+ T cells, and DCs (Fig. 3C) in the TME, whereas there was no significant difference in CD4+ T cells, Tregs, or macrophages (Supplementary Fig. S2B).

Figure 3.

AKT/WEE1 targeting enhances the effectiveness of anti–PD-1 treatment via NK-cell recruitment and activity in tumors. B16F10 cells were injected subcutaneously into male (n = 10) or female (n = 20) C57BL6 mice. Experiments were repeated two and four times, respectively, with 5 mice per group. After tumors reached 50 mm3, mice were treated with IgG (control), anti–PD-1 (200 μg twice a week), AZD5363 (150 mg/kg) + MK1775 (50 mg/kg), or triple combination daily. A, Tumor growth was measured over time until tumor size in any group reached ethical limits. B, Animal survival was also monitored every three days in a separate cohort until tumors reached ethic size limits in each group or stopped at day 40 in the AZD+MK group. Log-rank test was performed. Experiment was repeated twice with 5 mice per group. At day 21, tumors from female mice were excised and phenotyped by FACS for total NK-cell frequency (C), granzyme B+ NK-cell frequency, CD8+ T-cell and DC frequencies amongst mononuclear cells. D, Single-cell suspensions from tumors (1 tumor per group) were then subjected to bulk RNA-seq. Single-sample scores were calculated using the singscore method and available molecular signatures; higher scores indicate higher concordance of samples with signatures. E, Expression (logRPKM) of selected genes across the four samples. *, P < 0.05; **, P < 0.01, and ***, P < 0.001 (statistical test: ANOVA + Dunnett post hoc). Error bars represent mean + SEM.

Figure 3.

AKT/WEE1 targeting enhances the effectiveness of anti–PD-1 treatment via NK-cell recruitment and activity in tumors. B16F10 cells were injected subcutaneously into male (n = 10) or female (n = 20) C57BL6 mice. Experiments were repeated two and four times, respectively, with 5 mice per group. After tumors reached 50 mm3, mice were treated with IgG (control), anti–PD-1 (200 μg twice a week), AZD5363 (150 mg/kg) + MK1775 (50 mg/kg), or triple combination daily. A, Tumor growth was measured over time until tumor size in any group reached ethical limits. B, Animal survival was also monitored every three days in a separate cohort until tumors reached ethic size limits in each group or stopped at day 40 in the AZD+MK group. Log-rank test was performed. Experiment was repeated twice with 5 mice per group. At day 21, tumors from female mice were excised and phenotyped by FACS for total NK-cell frequency (C), granzyme B+ NK-cell frequency, CD8+ T-cell and DC frequencies amongst mononuclear cells. D, Single-cell suspensions from tumors (1 tumor per group) were then subjected to bulk RNA-seq. Single-sample scores were calculated using the singscore method and available molecular signatures; higher scores indicate higher concordance of samples with signatures. E, Expression (logRPKM) of selected genes across the four samples. *, P < 0.05; **, P < 0.01, and ***, P < 0.001 (statistical test: ANOVA + Dunnett post hoc). Error bars represent mean + SEM.

Close modal

To gain further mechanistic insight into the almost complete response of B16F10 to AZD5363+MK1775+anti–PD-1 therapy, we next performed global RNA-seq on B16F10 tumors following treatment. Tumors from multiple mice per treatment group were pooled because very little tumor mass was present in individual mice following combination and triple therapy. Immune cell score and signature analysis confirmed the flow cytometry data and showed that therapy resulted in substantial accumulation of DCs, CD8+ T cells, and NK cells within tumors, as well as enhanced type I and II IFN responses (Fig. 3D). NK-cell signatures (activation, residency, exhaustion) were highest in AZD+MK-treated tumors, whereas CD8+ T-cell signatures were highest with AZD+MK+anti–PD-1 treatment, highlighting a switch from dominant NK-cell effector responses to CD8+ T-cell responses with the addition of anti–PD-1 (Fig. 3D). We then selected genes involved in immune infiltration. Triple combination therapy resulted in clear increases in Cxcl10, Cxcl9, Ccl5, and Ccl2, which have all been implicated in the recruitment of CD8+ T cells and NK cells (Fig. 3E). Several NKG2D and DNAM1 ligands were also upregulated, with AZD5363 + MK1775 + anti–PD-1 treatment inducing the most upregulation (Fig. 3E). Evidence of CD8+ T-cell activation was only evident following triple combination treatment, with increased expression of Pdcd1 (PD-1), Hvcra2 (Tim-3), Gzmb (granzyme B), Prf1 (perforin-1), Tnf, Il2, an IFNγ (i.e., IFNγ_Response) signature, Interferon_imsig, and subsequent IFNγ-induced Cd274 (PD-L1; Fig. 3D and E). Taken together, these data demonstrate that cold, anti–PD-1–nonresponsive tumors can be successfully turned into hot, anti–PD-1–responsive tumors by targeting p53, which leads to ICD induction and promotion of immune cell infiltration and NK-cell activity. These responses then ultimately prime neoantigen-specific CD8+ T-cell responses.

CD8+ T cells, NK cells, and p53 are required for anti–PD-1 responses in preclinical models of cold tumors

Our in vitro data convincing demonstrated a role for p53 signaling in the expression of NKG2D ligands on melanoma and their sensitivity to NK-cell lysis following AKT/WEE1 inhibition. We next investigated how critical p53-induced NK-cell activity was for the beneficial effects of triple therapy on B16F10 tumor regression in vivo. We also evaluated the effect of p53 modulation in the efficacy of AKT/WEE1 + anti–PD-1 therapy using shRNA p53 KD B16F10 melanoma cells (anti–PD-1–resistant) and compared with parental B16F10 in vivo in response to the combination therapy (Fig. 4A). B16F10 tumor volumes were reduced by 95% with the triple combination in parental B16F10 cells compared with treatment-naïve tumors (Fig. 4A). In contrast, B16F10 p53 KD tumor volumes were not significantly different to treatment-naive mice following combination therapy (Fig. 4A). P53 expression by melanoma cells was essential for the enhanced tumor immune infiltration following combination therapy. Again, AKT/WEE1 + anti–PD-1 therapy resulted in a robust increase in the frequency of tumor-infiltrating NK cells and CD8+ T cells, with the fraction of granzyme B–expressing NK cells and CD8 T cells, and IFNγ+ or Ki67+ CD8+ T cells, being significantly increased compared with treatment-naïve mice (Fig. 4B; Supplementary Fig. S3). In contrast, AKT/WEE1 + anti–PD-1 therapy failed to increase immune infiltration when melanoma cells lacked p53 expression (Fig. 4B; Supplementary Fig. S3).

Figure 4.

AKT+WEE1+anti–PD-1 therapeutic response requires P53, NK cells, and CD8+ T cells. B16F10 or B16F10 p53 KD tumors were engrafted subcutaneously into C57BL/6 mice. NK cells and CD8+ T cells were depleted in mice by treating mice with anti-NK1.1 and anti-CD8α, respectively, on days −7, −3, −1, 1, 3, 7, 14, and 21, with tumors being inoculated on day 0. After tumors reached 50 mm3 (day 9), mice were treated with anti–PD-1 twice a week, or AZD5363 (150 mg/kg) + MK1775 (50 mg/kg), or triple combination daily. A, Tumor size was monitored once every three days. P53KD experiment: n = 15. Experiments were repeated 3 times with n = 5 for each group. NK-cell depletion experiment and control groups: n = 10. Experiments were repeated twice with n = 5 for each group. T cell depletion experiment: n = 8. Experiments were repeated twice with n = 4 for each group. B, At the end of the experiment, tumors were excised, and cells were phenotyped by FACS for frequency and activity of NK cells and CD8+ T cells. C, MHC-I (H-2Kb) expression on various mouse melanoma cell lines was determined by FACS. MFI: mean fluorescent intensity. Representative of two independent experiments is shown for all panels. D, M1 and M4 cells were injected subcutaneously into C57BL/6 mice on day 0. After tumors reached 50 mm3 (day 9), mice were treated with anti–PD-1 twice a week, AZD5363 (150 mg/kg)+MK1775 (50 mg/kg), or triple combination daily. Tumor kinetics were monitored once every three days in female mice (n = 10). *, P < 0.05; **, P < 0.01; and ***, P < 0.001 (statistical test: ANOVA + Dunnett post hoc). Error bars represent mean + SEM.

Figure 4.

AKT+WEE1+anti–PD-1 therapeutic response requires P53, NK cells, and CD8+ T cells. B16F10 or B16F10 p53 KD tumors were engrafted subcutaneously into C57BL/6 mice. NK cells and CD8+ T cells were depleted in mice by treating mice with anti-NK1.1 and anti-CD8α, respectively, on days −7, −3, −1, 1, 3, 7, 14, and 21, with tumors being inoculated on day 0. After tumors reached 50 mm3 (day 9), mice were treated with anti–PD-1 twice a week, or AZD5363 (150 mg/kg) + MK1775 (50 mg/kg), or triple combination daily. A, Tumor size was monitored once every three days. P53KD experiment: n = 15. Experiments were repeated 3 times with n = 5 for each group. NK-cell depletion experiment and control groups: n = 10. Experiments were repeated twice with n = 5 for each group. T cell depletion experiment: n = 8. Experiments were repeated twice with n = 4 for each group. B, At the end of the experiment, tumors were excised, and cells were phenotyped by FACS for frequency and activity of NK cells and CD8+ T cells. C, MHC-I (H-2Kb) expression on various mouse melanoma cell lines was determined by FACS. MFI: mean fluorescent intensity. Representative of two independent experiments is shown for all panels. D, M1 and M4 cells were injected subcutaneously into C57BL/6 mice on day 0. After tumors reached 50 mm3 (day 9), mice were treated with anti–PD-1 twice a week, AZD5363 (150 mg/kg)+MK1775 (50 mg/kg), or triple combination daily. Tumor kinetics were monitored once every three days in female mice (n = 10). *, P < 0.05; **, P < 0.01; and ***, P < 0.001 (statistical test: ANOVA + Dunnett post hoc). Error bars represent mean + SEM.

Close modal

To evaluate the contribution of NK cells in the efficacy of AKT/WEE1 + anti–PD-1 therapy, C57BL/6J mice were treated with anti-NK1.1 thrice before, and once every week after, tumor implantation and depletion was validated by flow cytometry (Supplementary Fig. S4). Depletion of NK cells significantly blunted the antitumor efficacy of AKT/WEE1 + anti–PD-1 therapy, with tumor volumes reduced by 45% in the absence of NK cells compared with 95% with triple combination in presence of NK cells (Fig. 4A). Tumor-resident NK cells were largely absent following anti-NK1.1 treatment, and CD8+ T-cell frequency was reduced compared with NK cell–sufficient mice (Fig. 4B).

Lastly, we confirmed the role of CD8+ T cells in AKT/WEE1+anti-PD-1 therapy responses induced in B16F10 tumors by treating C57BL/6J mice with anti-CD8a twice before, and once every week after, tumor implantation. CD8+ T cells were essential for tumor control mediated by AKT/WEE1+anti-PD-1 therapy. Tumor volumes were reduced by only 25% in the absence of CD8+ T cells compared to 95% with the triple combination therapy in presence of CD8+ T cells (Fig. 4A). The frequency and activity of tumor-infiltrating NK cells following AKT/WEE1 + anti–PD-1 therapy were not altered by anti-CD8α treatment; however, CD8+ T cells were largely absent from treated tumors (Fig. 4B). In addition, depletion of either NK cells, CD8+ T cells, or knockdown of p53 in tumor cells had no effect no effect on the frequency of tumor-infiltrating CD4+ T cells, Tregs, and macrophages (Supplementary Figs. S3–S5).

Combination of p53 modulation and anti–PD-1 is synergistic in preclinical melanoma models with variable MHC-I expression

B16F10 mouse melanoma cells are sensitive to NK-cell lysis in experimental metastasis models (intravenous transplant) and in vitro, with their basal low MHC-I expression a likely contributing factor, and thus are not representative of melanoma patients with high or moderate MHC-I expression. To validate the efficacy the drug combination at inducing robust endogenous antitumor immunity against high MHC-I–expressing orthotopic melanoma, M1 and M4 melanoma lines were utilized (see Materials and Methods; ref. 37). M1 and M4 cell lines had high MHC expression, as evaluated by the surface expression of H2-K1 (Fig. 4C). 1 × 106 cells were injected into C57BL/6J mice, and when tumors reached 50 mm3, mice were randomized into groups and treated with AKT/WEE1 (150/50 mg/kg, oral, daily), anti–PD-1 (200 μg/mouse, twice a week, i.p.), or a combination of both. Animal weights, tumor volumes, and behavioral changes of mice were monitored regularly. There was no significant difference in tumor volumes in M1 tumors when treated with anti–PD-1 alone, as was previously reported (37), whereas treatment of M1 tumors with AZD5363 + MK1775 reduced tumor volumes by 62% (Fig. 4D). Modulation of p53 in combination with anti–PD-1 treatment reduced the tumor volumes by 88% at day 21 postinoculation (Fig. 4D). None of the treatments had effects on animal weights, indicating the safety of the dosage regimen (Supplementary Fig. S6A). Immune phenotyping of tumors revealed the greatest increase in NK cells in the triple combination group compared with individual treatments (Supplementary Fig. S6B). Anti–PD-1 treatment increased tumor-infiltrating NK cells by 1.5-fold. The AZD5363 + MK1775 combination increased NK cells by 1.8-fold in the TME, whereas the combination with anti–PD-1 increased NK-cell infiltration by 2.1-fold. Consistently, there was a significant increase in the expression of granzyme B+ NK cells infiltrating into the TME (Supplementary Fig. S6B). In addition, the triple combination significantly increased tumor-infiltrating CD8+ T cells and DCs, whereas there was no significant difference in CD4+ T cells, Tregs, or macrophages (Supplementary Fig. S7). Furthermore, the triple combination was also evaluated in M4 xenografts. In this tumor model, anti–PD-1 treatment alone significantly inhibited tumor volumes, as reported previously (37). AZD5363 + MK1775 + anti–PD-1 further improved tumor control compared with either drug regimen alone, again validating the efficacy of the combination in multiple xenograft models with different genetic backgrounds (Fig. 4D).

The therapeutic mAbs, termed immune checkpoint inhibitors (ICI), have dramatically changed the treatment landscape for several types of tumors, with greatest efficacy shown in metastatic melanoma. However, response rate with these agents still remains relatively low (59). A contributing factor to ICI treatment failure is a relative lack of immune cell infiltration within the tumors, often referred to as “cold tumors.” The mechanisms controlling immune trafficking to the TME are not fully understood, but likely involve tumor mutational burden, cell death pathways, pathogen-associated and danger-associated molecular patterns (PAMP/DAMP, respectively), antigen presentation efficiency, and innate immune cells. Indeed, melanoma tends to have β2-microglobulin mutations (1) and reduced HLA class I expression (2, 3), which is a major inhibitory ligand for NK cells and could compromise their protection from endogenous NK-cell immunity (9). Data suggest NK cells are key inflammatory mediators that drive recruitment of additional antitumor effector cells and thus play a key role in priming the TME to respond to ICI (60). As such, targeted approaches to recruit and/or activate tumor-resident NK cells are predicted to result in higher ICI response rates and more durable antitumor immunity (59).

NK cells contribute different effector functions during cancer evolution and was recently summarized in The cancer–natural killer cell immunity cycle (9). Besides direct recognition and lysis of early transformed cells and continual immune surveillance of metastases, activated NK cells are a unique source of cDC1- and CD8+ T cell–attracting chemokines, namely XCL1/2, Flt3L, and CCL5 (9). Given these two cell types are the targets of ICIs and key for tumor regression, it is understandable that the degree of NK-cell infiltration in solid tumors is prognostic of patient outcomes (9, 61). By this logic, therapies that result in greater NK-cell infiltration and/or activation to drive tumor inflammation are needed. Emerging evidence suggests restoration of p53 signaling within tumor cells can lead to NK-cell recruitment, activation, and function in the TME (13, 17, 18). Furthermore, increased p53 activity causes tumor cell release of cytokines, such as IL12, IL15, and IL18, all of which can be potent inducers of NK-cell function whether alone and in combination (13). Numerous therapies exist that upregulate p53, such as ionizing radiation and DNA-damaging chemotherapeutics (62, 63). However, p53 activity is often suppressed by factors including the MDM2 and MDM4 proteins, leading to resistance (64). Thus, targeting MDM protein signaling has been pursued in an effort to restore p53 activity and has shown efficacy in preclinical models (27). Unfortunately, therapies directly targeting MDM proteins, such as nutlin-3 (MDM2) and SJ172550 (MDM4), are toxic or have limited efficacy necessitating combinatorial therapy, which can further increase toxicity (27). A promising dual MDM2/MDM4 inhibitor (ALRN-6924), which is being evaluated for efficacy in various cancers, has been reported to have a minimal toxicity profile (65).

A novel approach to target p53 pathway and avoid serious MDM2/4-mediated adverse effects include the dual targeting of AKT/WEE1 pathways. The importance of these pathways and pharmacologic inhibition synergistically inhibit melanoma tumor progression through p53 activation has been reported (30, 31, 49) and was confirmed in our studies. A key step in tumor immunity, and likely in the efficacy of our combination therapy, is the induction of ICD (66, 67). A study by Guo and colleagues shows that local activation of p53 within EL4 and B16F10 cells using nutlin-3a, an MDM2 antagonist, increased all biochemical hallmarks of ICD and limited tumor progression (68). Treatment of melanoma cells with AKT/WEE1 inhibitors had a significant increase in markers associated with ICD, including calreticulin and HMGB1, resulting in significantly reduced tumor cell survival, effects that required the expression of p53. AKT/WEE1 inhibitor treatment was efficient at reducing melanoma growth in vivo in both immune-competent and -deficient mice, highlighting the tumor-intrinsic role of p53 on cell death. Indeed, p53 knockdown or TP53-null mutation–bearing melanoma where insensitive to AKT/WEE1 inhibitor treatment.

Several studies have shown that the p53 pathway is important for the upregulation of NKG2D ligands on tumor cells, particularly ULBP1 and ULBP2, leading to increased tumor recognition and destruction (17–20, 69). Activation of p53 through AKT/WEE1 inhibition significantly increased the expression of ULBP1, ULBP2, and ULBP3, along with inducing MICB and Nec2, increasing their sensitivity to NK-cell detection and lysis. Again, this process required p53 expression. Induction of some of these ligands on melanoma was not observed following treatment with either MDM2 or MDM4 inhibitors, thus demonstrating an advantage of AKT/WEE1 inhibition increasing tumor immunogenicity to NK cells. The induction of stress ligands on melanoma cell lines following p53 upregulation appeared adequate to overcome NK-cell inhibitory signaling via high MHC-I expression, as the AKT/WEE1 inhibitor therapy was effective in melanoma with high basal MHC-I expression. It would be of interest to determine if targeting additional NK cell–suppressive mechanisms, such as TGFβ or cytokine-induced SH2 containing protein (Cish; CIS), in combination with AKT/WEE1 inhibition could augment NK-cell activity and tumor inflammation to a degree that efficiently induces control of MHC-I–mutant tumors, such as those that adaptively resistant to anti–PD-1 therapy.

We observed greater expression of NKG2D and DNAM1 ligands following triple therapy compared with AKT/WEE1 inhibition alone. A possible explanation for this is the large reduction in tumor RNA relative to immune RNA due to melanoma cell death and dominant NK cell–mediated immuno-editing, which would lead to preferential loss of tumors cells expressing high DNAM1/NKG2D-liagnds in vivo. However, NKG2D and DNAM1 ligand expression was clearly increased following AZD+MK+anti–PD-1 treatment (which had the highest immune:tumor RNA ratio) and likely reflected a switch in tumor immune-editing, along with CD8+ T-cell killing of neoantigen+/MHCI+ targets being more dominant than NK-cell immune-editing on NKG2D/DNAM1 ligands. In vivo, the tumor-intrinsic ICD mediated by AKT/WEE1 inhibitors in a p53-dependent manner clearly drove immune cell accumulation, including NK cells and cDC1s, most likely due to recruitment rather than tumor-resident proliferation. This has not been described for TME-resident NK cells or cDC1s. Temporally speaking, NK-cell and cDC1 activity and recruitment would be a prerequisite for the potent anti–PD-1 responses, and thus we hypothesize that anti–PD-1 treatment prior to AKT/WEE1 therapy would be ineffective. Similarly, we would predict that NK-cell depletion at later stages of therapy would not have the same nullifying effect as prior to therapy, as NK cells are hypothesized to prime the tumor for anti–PD-1 responsiveness but are not a target of anti–PD-1 therapy itself.

In summary, p53-mediated NK-cell activation following AKT/WEE1 inhibition was sufficient to switch immune “cold,” anti–PD-1–insensitive melanomas into immune “hot,” anti–PD-1–responsive tumors. This switch was best characterized by increased NK-cell infiltration, which was evident following AKT/WEE1 inhibition alone and further amplified following anti–PD-1 treatment. This effect was accompanied by increased cDC1 and CD8+ T-cell infiltration and activation. Both NK cells and T cells were essential for tumor regression following combination therapy, as was p53 expression by melanoma cells. Supporting the emerging evidence of the role for NK cells in driving CD8+ T-cell recruitment into tumors, NK-cell depletion impaired CD8+ T-cell infiltration following combination therapy, whereas CD8+ T-cell depletion had little impact on NK-cell infiltration. Thus, this result would benefit approximately 80% of patients with melanoma whose tumors have WT p53, high AKT and WEE1, and high MDM protein expression (22, 31). This approach could also significantly reduce melanoma mortality rates, improve treatments, and could be translated to other cancers having WT p53, such as ovarian, lung, and colorectal cancers or subsets of other cancer types with WT p53 (27).

S. Dinavahi reports grants from Melanoma Research Foundation during the conduct of the study. M. Foroutan reports other support from Prostate Cancer Foundation (PCF) during the conduct of the study. N.D. Huntington reports personal fees from ONKO-INNATE during the conduct of the study; personal fees from oNKo-Innate outside the submitted work; and founder and CSO of ONKO-INNATE PTY LTD. G.P. Robertson reports a patent for US2018/0325902A1 pending and issued. No disclosures were reported by the other authors.

S.S. Dinavahi: Data curation, formal analysis, investigation, methodology. Y.-Y. Chen: Data curation, formal analysis, investigation, visualization. K. Punnath: Data curation, formal analysis, investigation. A. Berg: Data curation, formal analysis, investigation. M. Herlyn: Resources, data curation, investigation, methodology. M. Foroutan: Data curation, formal analysis, investigation, methodology, writing–original draft. N.D. Huntington: Conceptualization, resources, supervision, funding acquisition, writing–original draft, project administration, writing–review and editing. G.P. Robertson: Conceptualization, resources, formal analysis, supervision, funding acquisition, validation, writing–original draft, project administration, writing–review and editing.

Our coauthor, Dr. Rajasekharan Somasundaram from the Wistar Institute, died in February 2021. We are grateful for his contributions to this manuscript and dedicate this publication to him. G.P. Robertson is supported by funding from The Foreman Foundation for Melanoma Research, The Geltrude Foundation, The Chocolate Tour Cancer Research Fund, Melanoma Research Alliance, and NCI (R01CA241148) of the NIH. S.S. Dinavahi received research grants from the Melanoma Research Foundation to support this project. N.D. Huntington is supported by project grants from the National Health and Medical Research Council (NHMRC) of Australia (GNT1124784, GNT1066770, GNT1057852, GNT1124907, GNT1057812, GNT1049407, GNT1027472, GNT1184615, to N.D. Huntington) and an NHMRC Investigator Fellowship (GNT1195296). N.D. Huntington is a recipient of a Melanoma Research Grant from the Harry J Lloyd Charitable Trust, Melanoma Research Alliance Young Investigator Award, National Foundation for Medical Research and Innovation (NFMRI) John Dixon Hughes Medal, and a CLIP grant from Cancer Research Institute.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Del Campo
AB
,
Kyte
JA
,
Carretero
J
,
Zinchencko
S
,
Méndez
R
,
González-Aseguinolaza
G
, et al
.
Immune escape of cancer cells with beta2-microglobulin loss over the course of metastatic melanoma
.
Int J Cancer
2014
;
134
:
102
13
.
2.
Ferrone
S
,
Marincola
FM
.
Loss of HLA class I antigens by melanoma cells: molecular mechanisms, functional significance and clinical relevance
.
Immunol Today
1995
;
16
:
487
94
.
3.
Schrörs
B
,
Lübcke
S
,
Lennerz
V
,
Fatho
M
,
Bicker
A
,
Wölfel
C
, et al
.
HLA class I loss in metachronous metastases prevents continuous T cell recognition of mutated neoantigens in a human melanoma model
.
Oncotarget
2017
;
8
:
28312
27
.
4.
Sucker
A
,
Paschen
A
.
Deciphering the genetic evolution of T-cell resistance in melanoma
.
Oncoimmunology
2015
;
4
:
e1005510
.
5.
Melero
I
,
Rouzaut
A
,
Motz
GT
,
Coukos
G
.
T-cell and NK-cell infiltration into solid tumors: a key limiting factor for efficacious cancer immunotherapy
.
Cancer Discov
2014
;
4
:
522
6
.
6.
Cantoni
C
,
Huergo-Zapico
L
,
Parodi
M
,
Pedrazzi
M
,
Mingari
MC
,
Moretta
A
, et al
.
NK cells, tumor cell transition, and tumor progression in solid malignancies: new hints for NK-based immunotherapy?
J Immunol Res
2016
;
2016
:
4684268
.
7.
Gonzalez-Gugel
E
,
Saxena
M
,
Bhardwaj
N
.
Modulation of innate immunity in the tumor microenvironment
.
Cancer Immunol Immunother
2016
;
65
:
1261
8
.
8.
Pahl
J
,
Cerwenka
A
.
Tricking the balance: NK cells in anti-cancer immunity
.
Immunobiology
2017
;
222
:
11
20
.
9.
Huntington
ND
,
Cursons
J
,
Rautela
J
.
The cancer–natural killer cell immunity cycle
.
Nat Rev Cancer
2020
;
20
:
437
54
.
10.
Barry
KC
,
Hsu
J
,
Broz
ML
,
Cueto
FJ
,
Binnewies
M
,
Combes
AJ
, et al
.
A natural killer-dendritic cell axis defines checkpoint therapy-responsive tumor microenvironments
.
Nat Med
2018
;
24
:
1178
91
.
11.
Zemek
RM
,
De Jong
E
,
Chin
WL
,
Schuster
IS
,
Fear
VS
,
Casey
TH
, et al
.
Sensitization to immune checkpoint blockade through activation of a STAT1/NK axis in the tumor microenvironment
.
Sci Transl Med
2019
;
11
:
eaav7816
.
12.
Guerra
N
,
Tan
YX
,
Joncker
NT
,
Choy
A
,
Gallardo
F
,
Xiong
N
, et al
.
NKG2D-deficient mice are defective in tumor surveillance in models of spontaneous malignancy
.
Immunity
2008
;
28
:
571
80
.
13.
Iannello
A
,
Thompson
TW
,
Ardolino
M
,
Lowe
SW
,
Raulet
DH
.
p53-dependent chemokine production by senescent tumor cells supports NKG2D-dependent tumor elimination by natural killer cells
.
J Exp Med
2013
;
210
:
2057
69
.
14.
Artis
D
,
Spits
H
.
The biology of innate lymphoid cells
.
Nature
2015
;
517
:
293
301
.
15.
Krzewski
K
,
Strominger
JL
.
The killer's kiss: the many functions of NK cell immunological synapses
.
Curr Opin Cell Biol
2008
;
20
:
597
605
.
16.
Martín-Fontecha
A
,
Thomsen
LL
,
Brett
S
,
Gerard
C
,
Lipp
M
,
Lanzavecchia
A
, et al
.
Induced recruitment of NK cells to lymph nodes provides IFN-gamma for T(H)1 priming
.
Nat Immunol
2004
;
5
:
1260
5
.
17.
Li
H
,
Lakshmikanth
T
,
Garofalo
C
,
Enge
M
,
Spinnler
C
,
Anichini
A
, et al
.
Pharmacological activation of p53 triggers anticancer innate immune response through induction of ULBP2
.
Cell Cycle
2011
;
10
:
3346
58
.
18.
Textor
S
,
Fiegler
N
,
Arnold
A
,
Porgador
A
,
Hofmann
TG
,
Cerwenka
A
.
Human NK cells are alerted to induction of p53 in cancer cells by upregulation of the NKG2D ligands ULBP1 and ULBP2
.
Cancer Res
2011
;
71
:
5998
6009
.
19.
Ciaglia
E
,
Laezza
C
,
Abate
M
,
Pisanti
S
,
Ranieri
R
,
D'alessandro
A
, et al
.
Recognition by natural killer cells of N6-isopentenyladenosine-treated human glioma cell lines
.
Int J Cancer
2018
;
142
:
176
90
.
20.
Soriani
A
,
Borrelli
C
,
Ricci
B
,
Molfetta
R
,
Zingoni
A
,
Fionda
C
, et al
.
p38 MAPK differentially controls NK activating ligands at transcriptional and post-transcriptional level on multiple myeloma cells
.
Oncoimmunology
2017
;
6
:
e1264564
.
21.
Collin
R
,
St-Pierre
C
,
Guilbault
L
,
Mullins-Dansereau
V
,
Policheni
A
,
Guimont-Desrochers
F
, et al
.
An unbiased linkage approach reveals that the p53 pathway is coupled to NK cell maturation
.
J Immunol
2017
;
199
:
1490
504
.
22.
Gembarska
A
,
Luciani
F
,
Fedele
C
,
Russell
EA
,
Dewaele
M
,
Villar
S
, et al
.
MDM4 is a key therapeutic target in cutaneous melanoma
.
Nat Med
2012
;
18
:
1239
47
.
23.
Jochemsen
AG
.
Reactivation of p53 as therapeutic intervention for malignant melanoma
.
Curr Opin Oncol
2014
;
26
:
114
9
.
24.
Kawashima
H
,
Takatori
H
,
Suzuki
K
,
Iwata
A
,
Yokota
M
,
Suto
A
, et al
.
Tumor suppressor p53 inhibits systemic autoimmune diseases by inducing regulatory T cells
.
J Immunol
2013
;
191
:
3614
23
.
25.
Lu
M
,
Miller
P
,
Lu
X
.
Restoring the tumour suppressive function of p53 as a parallel strategy in melanoma therapy
.
FEBS Lett
2014
;
588
:
2616
21
.
26.
Park
J-S
,
Lim
M-A
,
Cho
M-L
,
Ryu
J-G
,
Moon
Y-M
,
Jhun
J-Y
, et al
.
p53 controls autoimmune arthritis via STAT-mediated regulation of the Th17 cell/Treg cell balance in mice
.
Arthritis Rheum
2013
;
65
:
949
59
.
27.
Tisato
V
,
Voltan
R
,
Gonelli
A
,
Secchiero
P
,
Zauli
G
.
MDM2/X inhibitors under clinical evaluation: perspectives for the management of hematological malignancies and pediatric cancer
.
J Hematol Oncol
2017
;
10
:
133
.
28.
Cheung
M
,
Sharma
A
,
Madhunapantula
SV
,
Robertson
GP
.
Akt3 and mutant V600E B-Raf cooperate to promote early melanoma development
.
Cancer Res
2008
;
68
:
3429
39
.
29.
Pal
SK
,
Reckamp
K
,
Yu
H
,
Figlin
RA
.
Akt inhibitors in clinical development for the treatment of cancer
.
Expert Opin Investig Drugs
2010
;
19
:
1355
66
.
30.
Dinavahi
SS
,
Noory
MA
,
Gowda
R
,
Drabick
JJ
,
Berg
A
,
Neves
RI
, et al
.
Moving synergistically acting drug combinations to the clinic by comparing sequential versus simultaneous drug administrations
.
Mol Pharmacol
2018
;
93
:
190
6
.
31.
Kuzu
OF
,
Gowda
R
,
Sharma
A
,
Noory
MA
,
Kardos
G
,
Madhunapantula
SV
, et al
.
Identification of WEE1 as a target to make AKT inhibition more effective in melanoma
.
Cancer Biol Ther
2018
;
19
:
53
62
.
32.
Sharma
A
,
Madhunapantula
SV
,
Gowda
R
,
Berg
A
,
Neves
RI
,
Robertson
GP
.
Identification of aurora kinase B and Wee1-like protein kinase as downstream targets of (V600E)B-RAF in melanoma
.
Am J Pathol
2013
;
182
:
1151
62
.
33.
Sullivan
RJ
,
Flaherty
KT
.
BRAF in melanoma: pathogenesis, diagnosis, inhibition, and resistance
.
J Skin Cancer
2011
;
2011
:
423239
.
34.
Madhunapantula
SV
,
Sharma
A
,
Gowda
R
,
Robertson
GP
.
Identification of glycogen synthase kinase 3alpha as a therapeutic target in melanoma
.
Pigment Cell Melanoma Res
2013
;
26
:
886
99
.
35.
Watanabe
N
,
Broome
M
,
Hunter
T
.
Regulation of the human WEE1Hu CDK tyrosine 15-kinase during the cell cycle
.
EMBO J
1995
;
14
:
1878
91
.
36.
Mir
SE
,
De Witt Hamer
PC
,
Krawczyk
PM
,
Balaj
L
,
Claes
A
,
Niers
JM
, et al
.
In silico analysis of kinase expression identifies WEE1 as a gatekeeper against mitotic catastrophe in glioblastoma
.
Cancer Cell
2010
;
18
:
244
57
.
37.
Pérez-Guijarro
E
,
Yang
HH
,
Araya
RE
,
El Meskini
R
,
Michael
HT
,
Vodnala
SK
, et al
.
Multimodel preclinical platform predicts clinical response of melanoma to immunotherapy
.
Nat Med
2020
;
26
:
781
91
.
38.
Dinavahi
SS
,
Nyayapathy
S
,
Perumal
Y
,
Dharmarajan
S
,
Viswanadha
S
.
Combined inhibition of PDE4 and PI3Kdelta modulates the inflammatory component involved in the progression of chronic obstructive pulmonary disease
.
Drug Res (Stuttg)
2014
;
64
:
214
9
.
39.
Rao
PC
,
Begum
S
,
Jahromi
MAF
,
Jahromi
ZH
,
Sriram
S
,
Sahai
M
.
Cytotoxicity of withasteroids: withametelin induces cell cycle arrest at G2/M phase and mitochondria-mediated apoptosis in non-small cell lung cancer A549 cells
.
Tumour Biol
2016
;
37
:
12579
87
.
40.
Rao
PC
,
Begum
S
,
Sahai
M
,
Sriram
DS
.
Coptisine-induced cell cycle arrest at G2/M phase and reactive oxygen species-dependent mitochondria-mediated apoptosis in non-small-cell lung cancer A549 cells
.
Tumour Biol
2017
;
39
:
1010428317694565
.
41.
Pulla
VK
,
Alvala
M
,
Sriram
DS
,
Viswanadha
S
,
Sriram
D
,
Yogeeswari
P
.
Structure-based drug design of small molecule SIRT1 modulators to treat cancer and metabolic disorders
.
J Mol Graph Model
2014
;
52
:
46
56
.
42.
Pulla
VK
,
Sriram
DS
,
Soni
V
,
Viswanadha
S
,
Sriram
D
,
Yogeeswari
P
.
Targeting NAMPT for therapeutic intervention in cancer and inflammation: structure-based drug design and biological screening
.
Chem Biol Drug Des
2015
;
86
:
881
94
.
43.
Dinavahi
SS
,
Gowda
R
,
Bazewicz
CG
,
Battu
MB
,
Lin
JM
,
Chitren
RJ
, et al
.
Design, synthesis characterization and biological evaluation of novel multi-isoform ALDH inhibitors as potential anticancer agents
.
Eur J Med Chem
2020
;
187
:
111962
.
44.
Dinavahi
SS
,
Gowda
R
,
Gowda
K
,
Bazewicz
CG
,
Chirasani
VR
,
Battu
MB
, et al
.
Development of a novel multi-isoform ALDH inhibitor effective as an antimelanoma agent
.
Mol Cancer Ther
2020
;
19
:
447
59
.
45.
Dinavahi
SS
,
Prasanna
R
,
Dharmarajan
S
,
Perumal
Y
,
Viswanadha
S
.
A novel, potent, small molecule AKT inhibitor exhibits efficacy against lung cancer cells in vitro
.
Cancer Res Treat
2015
;
47
:
913
20
.
46.
Kokolus
KM
,
Haley
JS
,
Koubek
EJ
,
Gowda
R
,
Dinavahi
SS
,
Sharma
A
, et al
.
Schweinfurthin natural products induce regression of murine melanoma and pair with anti-PD-1 therapy to facilitate durable tumor immunity
.
Oncoimmunology
2019
;
8
:
e1539614
.
47.
Gowda
R
,
Dinavahi
SS
,
Iyer
S
,
Banerjee
S
,
Neves
RI
,
Pameijer
CR
, et al
.
Nanoliposomal delivery of cytosolic phospholipase A2 inhibitor arachidonyl trimethyl ketone for melanoma treatment
.
Nanomedicine
2018
;
14
:
863
73
.
48.
Gowda
R
,
Inamdar
GS
,
Kuzu
O
,
Dinavahi
SS
,
Krzeminski
J
,
Battu
MB
, et al
.
Identifying the structure-activity relationship of leelamine necessary for inhibiting intracellular cholesterol transport
.
Oncotarget
2017
;
8
:
28260
77
.
49.
Kuzu
OF
,
Gowda
R
,
Sharma
A
,
Noory
MA
,
Dinavahi
SS
,
Kardos
G
, et al
.
Improving pharmacological targeting of AKT in melanoma
.
Cancer Lett
2017
;
404
:
29
36
.
50.
Robinson
MD
,
Mccarthy
DJ
,
Smyth
GK
.
edgeR: a Bioconductor package for differential expression analysis of digital gene expression data
.
Bioinformatics
2010
;
26
:
139
40
.
51.
Foroutan
M
,
Bhuva
DD
,
Lyu
R
,
Horan
K
,
Cursons
J
,
Davis
MJ
.
Single sample scoring of molecular phenotypes
.
BMC Bioinf
2018
;
19
:
404
.
52.
Jerby-Arnon
L
,
Shah
P
,
Cuoco
MS
,
Rodman
C
,
Su
M-J
,
Melms
JC
, et al
.
A cancer cell program promotes T cell exclusion and resistance to checkpoint blockade
.
Cell
2018
;
175
:
984
97
.
53.
Aran
D
,
Sirota
M
,
Butte
AJ
.
Systematic pan-cancer analysis of tumour purity
.
Nat Commun
2015
;
6
:
8971
.
54.
Liberzon
A
,
Birger
C
,
Thorvaldsdóttir
H
,
Ghandi
M
,
Mesirov
JP
,
Tamayo
P
.
The Molecular Signatures Database (MSigDB) hallmark gene set collection
.
Cell Syst
2015
;
1
:
417
25
.
55.
Nirmal
AJ
,
Regan
T
,
Shih
BB
,
Hume
DA
,
Sims
AH
,
Freeman
TC
.
Immune cell gene signatures for profiling the microenvironment of solid tumors
.
Cancer Immunol Res
2018
;
6
:
1388
400
.
56.
Gu
Z
,
Eils
R
,
Schlesner
M
.
Complex heatmaps reveal patterns and correlations in multidimensional genomic data
.
Bioinformatics
2016
;
32
:
2847
9
.
57.
Bonavita
E
,
Bromley
CP
,
Jonsson
G
,
Pelly
VS
,
Sahoo
S
,
Walwyn-Brown
K
, et al
.
Antagonistic inflammatory phenotypes dictate tumor fate and response to immune checkpoint blockade
.
Immunity
2020
;
53
:
1215
29
.
58.
Hsu
J
,
Hodgins
JJ
,
Marathe
M
,
Nicolai
CJ
,
Bourgeois-Daigneault
M-C
,
Trevino
TN
, et al
.
Contribution of NK cells to immunotherapy mediated by PD-1/PD-L1 blockade
.
J Clin Invest
2018
;
128
:
4654
68
.
59.
Bonaventura
P
,
Shekarian
T
,
Alcazer
V
,
Valladeau-Guilemond
J
,
Valsesia-Wittmann
S
,
Amigorena
S
, et al
.
Cold tumors: a therapeutic challenge for immunotherapy
.
Front Immunol
2019
;
10
:
168
.
60.
Zemek
RM
,
Chin
WL
,
Nowak
AK
,
Millward
MJ
,
Lake
RA
,
Lesterhuis
WJ
.
Sensitizing the tumor microenvironment to immune checkpoint therapy
.
Front Immunol
2020
;
11
:
223
.
61.
Cursons
J
,
Souza-Fonseca-Guimaraes
F
,
Foroutan
M
,
Anderson
A
,
Hollande
F
,
Hediyeh-Zadeh
S
, et al
.
A gene signature predicting natural killer cell infiltration and improved survival in melanoma patients
.
Cancer Immunol Res
2019
;
7
:
1162
74
.
62.
Cui
Y
,
Guo
G
.
Immunomodulatory function of the tumor suppressor p53 in host immune response and the tumor microenvironment
.
Int J Mol Sci
2016
;
17
:
1942
.
63.
Suek
N
,
Campesato
LF
,
Merghoub
T
,
Khalil
DN
.
Targeted APC activation in cancer immunotherapy to enhance the Abscopal effect
.
Front Immunol
2019
;
10
:
604
.
64.
Duffy
MJ
,
Synnott
NC
,
Crown
J
.
Mutant p53 as a target for cancer treatment
.
Eur J Cancer
2017
;
83
:
258
65
.
65.
Carvajal
LA
,
Neriah
DB
,
Senecal
A
,
Benard
L
,
Thiruthuvanathan
V
,
Yatsenko
T
, et al
.
Dual inhibition of MDMX and MDM2 as a therapeutic strategy in leukemia
.
Sci Transl Med
2018
;
10
:
eaao3003
.
66.
Radogna
F
,
Diederich
M
.
Stress-induced cellular responses in immunogenic cell death: Implications for cancer immunotherapy
.
Biochem Pharmacol
2018
;
153
:
12
23
.
67.
Wang
Y-J
,
Fletcher
R
,
Yu
J
,
Zhang
L
.
Immunogenic effects of chemotherapy-induced tumor cell death
.
Genes Dis
2018
;
5
:
194
203
.
68.
Guo
G
,
Yu
M
,
Xiao
W
,
Celis
E
,
Cui
Y
.
Local activation of p53 in the tumor microenvironment overcomes immune suppression and enhances antitumor immunity
.
Cancer Res
2017
;
77
:
2292
305
.
69.
Heinemann
A
,
Zhao
F
,
Pechlivanis
S
,
Eberle
J
,
Steinle
A
,
Diederichs
S
, et al
.
Tumor suppressive microRNAs miR-34a/c control cancer cell expression of ULBP2, a stress-induced ligand of the natural killer cell receptor NKG2D
.
Cancer Res
2012
;
72
:
460
71
.

Supplementary data