RPMI-1640 medium, fetal bovine serum, penicillin/streptomycin/Amphotericin B, and red blood cell (RBC) lysis buffer were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Quantikine Human CXCL10/IP-10 ELISA Kit was purchased from R&D Systems (Minneapolis, MN, USA). CB-839 (telaglenastat) was purchased from Selleck Chemicals (Houston, TX, USA).

CL1–0 and 786–0 MTAP-intact (WT) and MTAP-knockout (KO) cells as well as CL1–5 and ACHN control (Mock) and MTAP-overexpressing (MTAP) cells were established and maintained at 37°C in a humidified atmosphere of 5% CO₂ as previously described (7–9). To establish MTAP-knockout Lewis lung carcinoma (LLC) cells, we used lentiviruses generated by co-transfecting HEK293T cells with an MTAP sgRNA-containing lentiviral vector and a packaging DNA mix using Lipofectamine 2000. The lentiviral vector was constructed by synthesizing, annealing, and cloning the following oligonucleotides into the LentiCRISPRv2 expression vector: oligo 1 (5’- CACCGTCTCACCTTCACCGCCGTGC-3’) and oligo 2 (5’- AAACGCACGGCGGTGAAGGTGAGAC-3’). Cells were infected at three different Multiplicities of Infection (MOIs) in polybrene (8 µg/mL)-containing medium. Twenty-four hours after infection, the cells were treated with puromycin (final concentration 2 µg/mL) and puromycin-resistant clones were selected and sequenced to confirm the gene-editing results.

All mouse experiments and procedures were approved and periodically reviewed by the Institutional Animal Care and Use Committee (IACUC) at UC Davis. Eight-week-old C57BL/6J mice were purchased from the Jackson Laboratory and housed four mice per cage and fed autoclaved food ad libitum. For tumorigenicity assay, 1×10⁶ cells were suspended in 100 μl PBS and implanted subcutaneously into the dorsal region of mice. Tumor growth was examined twice or thrice a week, and tumor volume was estimated by the formula LW ²/2, where L is the length and W is the width of the tumor. After 21 days, the mice were euthanized by CO₂ inhalation at a displacement rate of 20% of the chamber volume per minute, and the tumor xenografts were removed, weighted, and photographed.

Formalin-fixed, paraffin-embedded (FFPE) tissue sections (4 µm) were deparaffinized, rehydrated, and stained using the Hematoxylin and Eosin Stain Kit (Vector Laboratories, Burlingame, CA, USA) according to the manufacturer’s protocol. FFPE sections were also used for immunohistochemical analysis of MTAP and CD45 expression. For immunohistochemistry, the protocol was adapted from the manufacturer’s paraffin immunohistochemistry guidelines (Cell Signaling Technology, Danvers, MA, USA). Tissue sections were deparaffinized in a xylene substitute, rehydrated through graded alcohol solutions, and subjected to antigen retrieval using 10 mM sodium citrate (pH 6.0) at a sub-boiling temperature. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide, followed by serum blocking. Sections were incubated overnight at 4°C with anti-MTAP antibody (Novus Biologicals, Littleton, CO, USA) or anti-CD45 antibody (Cell Signaling Technology, Danvers, MA, USA). Immunostaining was detected using the VECTASTAIN ABC system (Vector Laboratories, Burlingame, CA, USA) following the manufacturer’s instructions. Stained sections were mounted and examined under a light microscope.

PBMCs were isolated from whole blood (STEMCELL Technologies, Cambridge, MA, USA) using a density gradient centrifugation method. For co-culture assays, cancer cells were mixed with isolated PBMCs at a ratio of 1:10 and co-cultured for 24 hours. After co-culture, cancer cells and PBMCs were separately harvested, washed with PBS, and processed for further assays.

Total RNA was extracted from cancer cells using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA). cDNAs were reversely transcribed using SuperScript III Reverse Transcriptase (Invitrogen). The primers used were as follows: CXCL9 forward primer 5’-CCAGTAGTGAGAAAGGGTCGC-3’ and reverse primer 5’-AGGGCTTGGGGCAAATTGTT-3’; CXCL10 forward primer 5’-CCAATTTTGTCCACGTGTTGAG-3’ and reverse primer 5’-GCTCCCCTCTGGTTTTAAGGA-3’; CXCL11 forward primer 5’-TTAAACAAACATGAGTGTGAAGGG-3’ and reverse primer 5’-CGTTGTCCTTTATTTTCTTTCAGG-3’; SLC38A2 forward primer 5’-AGCCAACAGCTCTTGTACCTGC-3’ and reverse primer 5’-GGAAGAACAGCAGGATGACAGAC-3’; SLC3A2 forward primer 5’-CCAGAAGGATGATGTCGCTCAG-3’ and reverse primer 5’-GAGTAAGGTCCAGAATGACACGG-3’; SLC7A5 forward primer 5’-GCCACAGAAAGCCTGAGCTTGA-3’ and reverse primer 5’-ATGGTGAAGCCGATGCCACACT-3’; SLC7A11 forward primer 5’-TCCTGCTTTGGCTCCATGAACG-3’ and reverse primer 5’-AGAGGAGTGTGCTTGCGGACAT-3’; TBP forward primer 5’-CACGAACCACGGCACTGATT-3’ and reverse primer 5’-TTTTCTTGCTGCCAGTCTGGAC-3’. The housekeeping gene TBP was utilized as the reference gene in quantitative real-time RT-PCR assay. Quantitative real-time RT-PCR was conducted using the SYBR Green system and performed according to the manufacturer’s instructions of the ViiA 7 Real-Time PCR System (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA, USA). The relative expression level of the target gene compared with that of the TBP was defined as –ΔCT = – [CT_{target gene} – CT_TBP]. The target gene/TBP mRNA ratio was calculated as 2^-ΔCT x K, where K is a constant.

The constructs containing 1000 bp (1K) or 2000 bp (2K) upstream of the CXCL10 transcription start site were cloned into the pGL3-Basic luciferase reporter vector (Promega Corporation, Madison, WI, USA). The resulting constructs were verified by DNA sequencing. CL1–0 cells were co-transfected with the CXCL10 promoter-luciferase constructs and the phRL-TK vector, which served as an internal control to normalize for transfection efficiency. The pGL3-Basic empty vector was used as a negative control. Luciferase activity was measured using the Dual-Luciferase Reporter Assay Kit (Promega), following the manufacturer’s instructions. Firefly luciferase activity was normalized to Renilla luciferase activity to calculate relative promoter activity for each condition.

The cells were collected and prepared as whole-cell lysates with lysis buffer (50 mM Tris-HCl [pH 7.4], 1% Triton X-100, 10% glycerol, 150 mM NaCl, 1 mM EDTA, 20 μg/mL leupeptin, 1 mM PMSF, and 20 μg/mL aprotinin). Total proteins were separated via SDS-PAGE and transferred to polyvinylidene fluoride (PVDF) membranes, followed by immunoblotting with anti-phospho-STAT1, anti-STAT1, anti-phospho-p65, anti-p65, anti-β-actin (Cell Signaling Technology, Danvers, MA, USA), anti-MTAP (Novus Biologicals, Littleton, CO, USA) antibodies and chemiluminescence detection. The protein expression levels were quantified by ImageJ software (National Institutes of Health, Bethesda, MD, USA).

The LOPAC 1280 library (Sigma-Aldrich, Cat# LO1280) was used to identify selective inhibitors for MTAP-deficient cancer cells. CL1–0 MTAP-WT and MTAP-KO cells were seeded in 96-well plates and treated the following day with either vehicle (DMSO) or individual compounds at 2 or 20 μM for 72 hours. Cell viability was then assessed using MTT assays. Compounds exhibiting dose-dependent cytotoxicity and a viability difference greater than 40% between MTAP-WT and MTAP-KO cells were selected for further analysis.

Data from at least three independent experiments are presented as the mean ± standard deviation (SD). Quantitative variables were analyzed using an unpaired two-tailed Student’s t-test. All analyses were performed using GraphPad Prism software (version 10.4.1, Boston, MA, USA). Statistical tests were two-sided, and p-values < 0.05 were considered statistically significant.

To assess the role of MTAP in regulating antitumor immunity, we employed an immunocompetent murine model using Lewis lung carcinoma (LLC) cells with or without MTAP expression. MTAP-intact (wild-type, WT) and MTAP-knockout (KO) cells were subcutaneously injected into the contralateral flanks of syngeneic C57BL/6J mice. By day 21, tumors derived from MTAP-KO cells exhibited significantly accelerated growth and greater tumor volume compared to their WT counterparts ( Figure 1A ). Consistently, excised MTAP-KO tumors were markedly larger and heavier than MTAP-WT tumors across all animals ( Figures 1B,C ). Histological evaluation revealed notable differences in tumor architecture. Hematoxylin and eosin (H&E) staining indicated increased vascularization and stromal remodeling in MTAP-KO tumors relative to WT ( Figure 1D , top), top. Immunohistochemical analysis confirmed strong MTAP expression in WT tumors and its absence in KO tumors ( Figure 1D, middle ). Notably, CD45⁺ immune cell infiltration was significantly reduced in MTAP-deficient tumors, suggesting impaired recruitment of leukocytes ( Figure 1D, bottom ). These findings suggest that MTAP loss enhances tumor progression and is associated with reduced immune cell infiltration, highlighting the importance of MTAP in antitumor immune responses.

To elucidate the molecular mechanisms by which MTAP regulates antitumor immune responses, we performed bulk RNA sequencing to analyze transcriptomic alterations in MTAP-WT (WT) and MTAP-KO (KO) cancer cells co-cultured with peripheral blood mononuclear cells (PBMCs) (WT+PBMC and KO+PBMC). We focused on genes with a fragments per kilobase of transcript per million mapped reads (FPKM) value ≥1 and identified those with at least a two-fold change in expression when comparing WT vs. WT+PBMC, KO vs. KO+PBMC, and WT+PBMC vs. KO+PBMC. This analysis yielded 61 differentially expressed genes, ranked by fold change between WT+PBMC and KO+PBMC ( Figure 2A ). Of these, 43 genes were upregulated in the WT+PBMC condition, while only 18 genes were upregulated in KO+PBMC cells, supporting our in vivo findings that MTAP deficiency suppresses immune activation. Notably, CXCL10 (IP-10, interferon gamma-induced protein 10), a chemokine critical for T cell recruitment and a positive prognostic marker for response to immune checkpoint blockade (23), was significantly downregulated in KO+PBMC cells. Additionally, several interferon-inducible (IFI) genes were also suppressed in MTAP-deficient cells, suggesting impaired interferon signaling in the absence of MTAP.

Since CXCL9, CXCL10, and CXCL11 are ligands for chemokine receptor CXCR3, and play a crucial role in leukocyte migration and activation (24), we further assessed their expression using RT-qPCR analysis. Baseline expression levels of CXCL9, CXCL10, and CXCL11 were relatively low and comparable between MTAP-WT and MTAP-KO CL1–0 and 786–0 cancer cells, as well as between CL1–5 Mock and MTAP-expressing cells ( Figures 2B–D ). However, upon co-culture with PBMCs, expression of these chemokines was robustly induced in MTAP-expressing cells but significantly blunted in MTAP-KO and Mock cells. Among the three, CXCL10 exhibited the most pronounced differential induction, further underscoring its central role in MTAP-mediated immune responses. To determine the source of CXCL10 production, we performed ELISA assays on culture supernatants and confirmed that CXCL10 secretion was significantly reduced in MTAP-KO cells, indicating that cancer cells, not PBMCs, were the primary contributors to CXCL10 levels ( Figure 2E ). Further analysis using the TCGA pan-cancer dataset (25) consolidated the positive correlation between MTAP and CXCL10 expression ( Figure 2F ). To explore upstream regulatory mechanisms, we conducted luciferase reporter assays using cloned CXCL10 promoter constructs. MTAP-KO cancer cells displayed significantly lower luciferase activity, and deletion mapping localized the responsible regulatory elements to the proximal 1-kb region of the CXCL10 promoter ( Figure 2G ). We further investigated potential upstream regulators of CXCL10 by examining STAT1 and NF-κB p65, two key transcription factors known to drive CXCL10 expression. Phosphorylation of both factors following PBMC co-culture was minimally affected by MTAP status, suggesting that CXCL10 suppression is not primarily mediated through impaired STAT1 or NF-κB signaling ( Figure 2H ). Collectively, these findings demonstrate that MTAP deficiency suppresses CXCL10 transcription and secretion in response to immune cell contact, thereby limiting chemokine-mediated immune cell recruitment. This suggests that MTAP plays a critical role in shaping tumor immunogenicity through regulation of interferon-driven chemokine expression.

Having uncovered the mechanistic role of MTAP deficiency in promoting immune evasion, we next sought to identify therapeutic strategies that selectively target MTAP-deficient cancer cells while potentially enhancing antitumor immune responses. To this end, we utilized the LOPAC1280, a well-characterized high-throughput screening library composed of pharmacologically active small molecules ( Figure 3A ). CL1–0 MTAP-WT and MTAP-KO cells were screened across multiple dosages to evaluate compound specificity and differential sensitivity. Initial screening using MTT assays identified 53 candidate compounds with selective cytotoxicity against MTAP-KO cancer cells. Remarkably, four of these compounds were involved in glutamate-related signaling pathways ( Figure 3B ): spermidine trihydrochloride and spermine tetrahydrochloride are modulators of NMDA-type glutamate receptors ( Figures 3C,D ); AIDA acts as an antagonist of metabotropic glutamate receptors ( Figure 3E ); and BPTES inhibits glutaminase, a key enzyme in glutamine-to-glutamate conversion ( Figure 3F ). These findings suggest that MTAP-deficient cancer cells are particularly susceptible to disruption of glutamate signaling. To further explore this metabolic vulnerability, we tested CB-839, a clinically advanced analog of BPTES with improved solubility and potent antiproliferative activity (26). CB-839 was evaluated across four matched pairs of MTAP-expressing and MTAP-deficient cancer cell lines. Strikingly, CB-839 exhibited strong selective efficacy in MTAP-deficient cells at concentrations as low as 0.3 μM ( Figure 3G ). Together, these findings demonstrate that MTAP-deficient cancer cells display a critical dependence on glutamate metabolism. Targeting glutamate signaling may therefore represent a promising therapeutic strategy in MTAP-deleted malignancies.

To further understand the regulatory mechanisms underlying glutamate dependency in MTAP-deficient cancer cells, we examined the expression of glutaminase 1 (GLS1), the target of CB-839, and key glutamine/glutamate transporters. RT-qPCR analysis revealed a slight decrease in GLS1 expression in MTAP-deficient cells, despite their increased sensitivity to CB-839 treatment. Interestingly, MTAP-deficient cells showed a modest upregulation of the glutamine transporter SLC38A2, while the glutamine antiporter complex SLC3A2-SLC7A5 and the glutamate-cystine antiporter SLC7A11 were moderately downregulated ( Figure 4A ). These changes suggest a metabolic shift favoring increased glutamine uptake and reduced glutamate export, potentially reinforcing intracellular glutamate dependence in MTAP-deficient cells. Given the selective vulnerability of MTAP-deficient cancer cells to CB-839 treatment and the immunoregulatory role of CXCL10, we next evaluated whether glutaminase inhibition could enhance CXCL10 expression. As shown in Figure 4B , basal level of CXCL10 was minimal in cancer cells cultured alone (green bar). Co-culture with PBMCs induced CXCL10 expression in both MTAP-expressing and MTAP-deficient cells, but the magnitude of induction was significantly higher in MTAP-expressing cells (purple bar). While CB-839 treatment had little effect on CXCL10 expression (blue bar), pre-treatment with CB-839 followed by PBMC co-culture led to a marked increase in CXCL10, particularly in MTAP-deficient cells (red bar). These findings suggest that CB-839 effectively enhances CXCL10 expression in an immune-competent context, with preferential effects in MTAP-deficient tumors. This finding supports the potential of glutaminase inhibition not only to impair tumor metabolism but also to augment chemokine-driven immune responses, offering a dual therapeutic strategy for MTAP-deficient malignancies.