Metabolic dysregulation is a hallmark of cancer. Many tumors exhibit auxotrophy for various amino acids, such as arginine, because they are unable to meet the demand for these amino acids through endogenous production. This vulnerability can be exploited by employing therapeutic strategies that deplete systemic arginine in order to limit the growth and survival of arginine auxotrophic tumors. Pegzilarginase, a human arginase-1 enzyme engineered to have superior stability and enzymatic activity relative to the native human arginase-1 enzyme, depletes systemic arginine by converting it to ornithine and urea. Therapeutic administration of pegzilarginase in the setting of arginine auxotrophic tumors exerts direct antitumor activity by starving the tumor of exogenous arginine. We hypothesized that in addition to this direct effect, pegzilarginase treatment indirectly augments antitumor immunity through increased antigen presentation, thus making pegzilarginase a prime candidate for combination therapy with immuno-oncology (I-O) agents. Tumor-bearing mice (CT26, MC38, and MCA-205) receiving pegzilarginase in combination with anti–PD-L1 or agonist anti-OX40 experienced significantly increased survival relative to animals receiving I-O monotherapy. Combination pegzilarginase/immunotherapy induced robust antitumor immunity characterized by increased intratumoral effector CD8+ T cells and M1 polarization of tumor-associated macrophages. Our data suggest potential mechanisms of synergy between pegzilarginase and I-O agents that include increased intratumoral MHC expression on both antigen-presenting cells and tumor cells, and increased presence of M1-like antitumor macrophages. These data support the clinical evaluation of I-O agents in conjunction with pegzilarginase for the treatment of patients with cancer.

Tumor cells require an adequate supply of key nutrients, such as the amino acid arginine, which they can either endogenously synthesize or take up from the environment in order to accommodate their metabolic reprogramming, maintain abnormally elevated rates of proliferation, and generate biomass. However, amino acid auxotrophy, the state of dependence on exogenous amino acids, is a common metabolic abnormality found in tumors (1). Due to low or absent expression of key enzymes involved in arginine synthesis, most notably argininosuccinate synthetase 1 (ASS1; ref. 2), arginine auxotrophic tumors cannot endogenously synthesize sufficient arginine to meet their metabolic demands (3) and therefore depend on the uptake of extracellular arginine. Therapeutic strategies that exploit this vulnerability deplete systemic arginine, which limits arginine auxotrophic tumor growth and survival (Fig. 1A; ref. 3). Pegzilarginase, an engineered human arginase-1 enzyme that possesses superior serum stability and enzymatic activity relative to the native enzyme, depletes systemic arginine by converting it to ornithine and urea (4). For arginine auxotrophic tumors, therapeutic administration of an arginase, like pegzilarginase, deprives the tumor of exogenous arginine and results in direct antitumor activity. This mechanism of action underlies the single-agent antitumor activity of pegzilarginase in a variety of arginine auxotrophic murine and xenograft human tumor models (4). Clinically, a PEGylated recombinant human arginase, BCT-100, induced a sustained complete remission in a patient with immunotherapy–resistant melanoma (5).

Figure 1.

Pegzilarginase dosing depletes systemic and intratumoral arginine. A, Schematic depicting the ability of pegzilarginase to limit the intratumoral arginine pool by converting extracellular arginine to ornithine and urea. OTC, ornithine transcarbamylase. B, Pharmacodynamic time course of serum arginine following intraperitoneal (i.p.) administration of pegzilarginase and over the subsequent 6 days. Mean ± SEM; n = 5 animals/dose of pegzilarginase. C, Tumor interstitial fluid arginine concentration measurements 72 hours subsequent to vehicle or pegzilarginase treatment. n = 3 biological replicates/group. Mean ± SD. NN, nor-NOHA. D, Schematic depicting experimental protocol for samples in E and F. E, Measurement of early apoptosis (Annexin V) by flow cytometry in cells isolated from tumor 3 days after treatment. Median ± IQR; n = 5 biological replicates/group. IQR, interquartile range. F, Western blot of whole tumor lysate for the autophagy protein LC3 and densitometry of the Western blot. β-Actin expression was assessed as a loading control. Mean ± SD; n = 3 to 4 biological replicates/group. *, P < 0.05; **, P < 0.01; ****, P ≤ 0.0001.

Figure 1.

Pegzilarginase dosing depletes systemic and intratumoral arginine. A, Schematic depicting the ability of pegzilarginase to limit the intratumoral arginine pool by converting extracellular arginine to ornithine and urea. OTC, ornithine transcarbamylase. B, Pharmacodynamic time course of serum arginine following intraperitoneal (i.p.) administration of pegzilarginase and over the subsequent 6 days. Mean ± SEM; n = 5 animals/dose of pegzilarginase. C, Tumor interstitial fluid arginine concentration measurements 72 hours subsequent to vehicle or pegzilarginase treatment. n = 3 biological replicates/group. Mean ± SD. NN, nor-NOHA. D, Schematic depicting experimental protocol for samples in E and F. E, Measurement of early apoptosis (Annexin V) by flow cytometry in cells isolated from tumor 3 days after treatment. Median ± IQR; n = 5 biological replicates/group. IQR, interquartile range. F, Western blot of whole tumor lysate for the autophagy protein LC3 and densitometry of the Western blot. β-Actin expression was assessed as a loading control. Mean ± SD; n = 3 to 4 biological replicates/group. *, P < 0.05; **, P < 0.01; ****, P ≤ 0.0001.

Close modal

Immunotherapy has significantly improved outcomes for many patients by either boosting T-cell–mediated antitumor immunity by releasing the brakes on tumor-reactive T cells via checkpoint blockade (e.g., anti–CTLA-4, anti–PD-1, anti–PD-L1) or costimulating T cells via ligation of TNF receptor family members [e.g., OX40 (CD134); refs. 6–12]. The antitumor immune response generated by these treatments depends on neoantigen recognition and subsequent tumor cell killing; thus, neoantigen abundance profoundly influences immunotherapeutic efficacy. In addition to a sensitivity to depletion of extracellular arginine, urea cycle–disordered tumors, like those with reduced ASS1 expression, have more neoantigens, owing to a pyrimidine imbalance, and respond better to immune checkpoint blockade than do tumors in which enzymes involved in the urea cycle are expressed normally (13). Therefore, a therapeutic strategy that combines manipulation of extracellular arginine with exploitation of an enhanced immunogenicity may be an effective means to control arginine auxotrophic tumor growth. Here, we tested whether combining arginine depletion with immunotherapy enhances antitumor efficacy by exploiting the dual sensitivities of extracellular arginine dependence and increased antigenicity.

Arginase plays a complex role in immune modulation. T cells require sufficient nutrients, including amino acids like arginine, to carry out their essential functions (14, 15); thus, a lack of arginine results in downregulation of T-cell receptor zeta (16) and reduction of proliferation (17). In this vein, it is no surprise that some studies have shown that T-cell proliferation is inhibited by arginase secreted by myeloid-derived suppressor cells (MDSC) in the tumor microenvironment (TME; refs. 18, 19). However, other studies suggest that human arginase-1 possesses relatively poor catalytic activity and serum stability (20) and that MDSC-mediated suppression of T-cell activity does not require arginase (21). T cells compensate for low-arginine conditions by consuming citrulline and upregulating ASS1 (22, 23), an enzyme that converts citrulline and succinate to arginine (24), and producing arginine intracellularly. This allows T cells to proliferate in low-arginine environments and may explain several lines of evidence suggesting that arginine depletion is immune neutral or even immune promoting. For example, CD8+ T-cell cytotoxicity, perforin secretion, and IL2 secretion (25) are not impaired by the absence of arginine (26), and innate lymphoid 2 cells, which regulate proinflammatory functions in a number of tissues (27), are positively regulated by arginase-1. Thus, the effect of arginine depletion on immune cells may depend on the environment, the efficiency of upregulation of compensatory pathways, or the severity of depletion.

Herein, we demonstrated that pegzilarginase monotherapy increased intratumoral CD8+ T-cell number and activation relative to controls. Combining pegzilarginase with various immuno-oncology (I-O) agents, including anti–PD-1 and an agonist anti-OX40, uniformly resulted in increased therapeutic benefit compared with the respective I-O agents alone, including an increase in complete tumor regression and survival. Combining pegzilarginase with I-O agents evinced the greatest number of activated intratumoral CD8+ T cells and elevated systemic IFNγ among all treatment groups. Based in part on the data presented herein, we initiated a phase I to II trial examining the combination of pegzilarginase with pembrolizumab in patients with small-cell lung cancer (NCT03371979). Altogether, these data suggest that combining pegzilarginase with I-O agents may benefit patients harboring arginine auxotrophic tumors.

Mice

Wild-type female C57BL/6 and BALB/c mice (6–8 weeks of age) were purchased from Charles River or The Jackson Laboratory. All mice were maintained under specific pathogen-free conditions in the Providence Cancer Institute or Aeglea Research Labs animal facility. Experimental procedures were performed according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and in accordance with, and with the approval of, the Institutional Animal Care and Use Committee of either Aeglea BioTherapeutics or the Earle A. Chiles Research Institute (Animal Welfare Assurance No. A3913–01).

Chemicals and reagents

Pegzilarginase (AEB1102, Co-Arg1-PEG) was produced under good manufacturing practice conditions at KBI Biopharma as described previously (4). The drug was supplied in vehicle (10% glycerol in PBS, pH 7.4) at a concentration of 3 mg/mL and diluted as appropriate for each specific biological application. Pegzilarginase is available from Aeglea BioTherapeutics through a material transfer agreement.

Cell lines

The CT26 (BALB/c) murine colon carcinoma cell lines were purchased from and authenticated by the ATCC. MC38 (C57/Bl6) cells were obtained from and authenticated by the NIH. MCA-205 cells were kindly provided by Dr. Andrew Weinberg (Earle A. Chiles Research Institute). CT26 cells were maintained in complete RPMI 1640 (Thermo Fisher Scientific), MC38 cells in DMEM (Thermo Fisher Scientific), and MCA-205 cells in complete RPMI 1640 [10% FBS, 10 mmol/L HEPES, 1% nonessential amino acids, 1% sodium pyruvate (Lonza), and penicillin (100 IU/mL)–streptomycin (100 μg/mL)–glutamine (29.2 mg/mL; Invitrogen)]; the identity of this cell line was verified through monthly assessment of morphology and growth kinetics; in vitro MC38 cells were supplemented with 1% nonessential amino acids and 1% sodium pyruvate (Millipore Sigma). Hs5 cells were obtained from and authenticated by the ATCC, and were grown in SFEM2 media (Stem Cell Technologies) for use in generation of bone marrow–derived macrophages (BMDM). L-929 cells were obtained from and authenticated by the ATCC, and were grown in DMEM + 10% FBS. All cell lines were used at low passage (<10 passages in vitro after receipt from source). All cell lines were tested and screened negative for Mycoplasma using the MycoAlert test (Lonza).

In vivo tumor studies

Mice were inoculated with 2.5 × 105, 5 × 105, or 1 × 106 cells s.c. for MC38 (Crown Bioscience), CT26 (ATCC), or MCA-205 cells, respectively. Treatment was initiated when mean tumor volumes were 75 to 100 mm3. All dosing of drugs was done via intraperitoneal injection. Mice were randomized into groups based on average tumor volume using the following formula: L × W2/2, where L = Length, W = Width, and L > W. Mice were treated with PBS (1×/week), isotype control (10 mg/kg; 2×/week; catalog #BP0089; Bio-X-Cell), pegzilarginase (3 mg/kg; 1×/week), anti–CTLA-4 (10 mg/kg; 2×/week; clone 9D9, Bio-X-Cell), anti–PD-L1 (10 mg/kg; 2×/week; clone B7-H1; Bio-X-Cell), and/or anti-OX40 (10 mg/kg; 2×/week for 1 week; OX86). For CD8+ T-cell neutralization studies, anti-CD8 (10 mg/kg; 2×/week; clone 53-6.7; Bio-X-Cell) was administered. For CD4+ T-cell neutralization studies, anti-CD4 (10 mg/kg; 1×/week; clone GK1.5; Bio-X-Cell) was administered. For tumor rechallenge studies, animals with fully regressed tumors were inoculated s.c. opposite of the original site of tumor cell implantation with 1 × 106 CT26 cells, and observed for 30 days in order to assess the presence or absence of discernible tumorigenesis. For all experiments noted here, tumor volumes were measured 2×/week by calipers and calculated according to the formula noted above. Body weights were measured 2×/week. Animals were euthanized when either tumor volume reached 2,500 mm3 (2,000 mm3 in the MC38 model) or when tumors became ulcerated. Animals bearing ulcerated tumors were excluded from survival and mean tumor volume analyses.

Tissue processing and flow cytometry

Tumor-bearing animals were euthanized via CO2 asphyxiation, and the tumors were removed for analysis by flow cytometry. Tumors were minced on ice using razor blades and then mechanically and enzymatically dissociated using the mouse tumor dissociation Kit (Miltenyi Biotec) in conjunction with Miltenyi C tubes and the gentleMACS Octo Dissociator with heat unit (Miltenyi Biotec) as per the manufacturer's instructions (program “37C_m_TDK_1”) to yield single-cell suspensions. The cell suspension was sequentially filtered through 70- and 40-μm mesh filters (Corning), and then the cells were washed with cold PBS. Zombie Yellow fixable viability dye (BioLegend) was added as appropriate, and cells were blocked with anti-CD16/32 (BioLegend) for 10 minutes at 4°C and then stained with antibodies directed against cell surface antigens (see “Flow cytometry antibodies and reagents” below). For intracellular staining, cells were fixed and permeabilized using the FoxP3 Staining Buffer Kit (Thermo Fisher Scientific). Spleens were harvested and used to generate control cell suspensions by mechanical dissociation through a 70-μm filter. Cells were washed with cold PBS and stained as per tumor cells. Cells were either analyzed on the Guava 12-HT machine using Incyte v3.3 software (Millipore Sigma) or acquired on an LSR-II flow cytometer running FACSDiva software (BD Biosciences), and data were processed and analyzed with FlowJo (Treestar). All tissues were processed fresh with minimal storage time on ice.

Flow cytometry antibodies and reagents

Anti-mouse CD45 (clone 30-F11), CD3 (17A2), CD4 (GK1.5), CD8 (53-6.7), CD69 (H1.2F3), CD25 (3C7), KLRG1 (2F1/KLRG1), PD-1 (29F.1A12), CD11b (M1/70), CD11c (N418), F4/80 (BM8), CD206 (C06862), MHC II/I-A I-E (M5.114.15.2), Gr-1 (RB6-8C5), Ki-67 (16A8), F4/80 (BM8), Ly-6C (HK1.4), and GzmA (3G8.5) antibodies were from BioLegend. Anti-mouse CD274 (PD-L1, M1H5), CD24 (M1/69), and Ly-6G (1A8) were from BD Biosciences. Intracellular antibodies iNOS (CXNFT; Thermo Fisher Scientific) and Arg1 (Cat: IC5868A; R&D) were used. For viability of tumors, Fixable Viability Dye eFluor 780 (Thermo Fisher Scientific) was used. Anti-calreticulin (EPR3924) was purchased from Abcam. Antibodies against murine T-bet (4B10) and Eomes (Dan11mag) were obtained from Thermo Fisher Scientific. For bone marrow analysis, anti-human CD45 (clone HI30), CD3 (OKT3), and CD19 (HIB19) were from BioLegend. Anti-human CD66b (G10F5), CD11b (ICRF44), and CD14 (61D3) were obtained from Thermo Fisher Scientific. Annexin V reagent was used to assess apoptosis, and Zombie dyes were used to assess viability (BioLegend). gp70 pentamer/MHC I reagent (H-2Ld; SPSYVYHQF) was obtained from ProImmune.

In vivo pharmacodynamic studies

Female BALB/c mice (n = 5/group) were dosed i.p. with pegzilarginase formulated in 10% glycerol in PBS, pH 7.4, at a dose of 2 or 3 mg/kg. All animals were exsanguinated at their assigned time points by cardiac puncture and euthanized. Blood collected from all groups was split between EDTA tubes with or without 18 μg nor-NOHA (NN; Haematologic Technologies). Plasma was separated from whole blood by centrifugation. All plasma samples were neutralized with glacial acetic acid at 2% of the final concentration and frozen for subsequent arginine concentration assessment. Plasma arginine concentrations were assessed at Intertek Pharmaceutical Services using an LC/MS-MS assay (Intertek Pharmaceutical Services).

Immunohistochemistry

Immunohistochemistry (IHC) analysis of formalin-fixed and paraffin-embedded (FFPE) tumor sections from CT26 in vivo studies was performed with the Dako Autostainer platform. A monoclonal anti-CD3 (clone SP7; Abcam) was used to detect intratumoral T cells; a monoclonal anti-CD8 was used to detect CD8+ T cells (clone 4SM15; eBioscience). A chromogenic DAB ABC detection system (Abcam) was employed to produce a visual readout of antibody binding; positivity was defined as membrane-reactive chromogenic signal above background. Nontumor-bearing mouse spleens were used as a positive control for anti-CD3 staining. Stitched imaging and quantification were performed using a Keyence BZ-X700 microscope and analysis software (Keyence). CD8 IHC analysis of FFPE sections was performed by QualTek Molecular Laboratories.

Western blotting

For cell culture, media were removed, and cells were collected in cold PBS. For tissues and tumors, samples were homogenized using a FastPrep-24 homogenizer (MP Biomedicals) and lysed in RIPA buffer containing protease inhibitors (Thermo Fisher Scientific). Protein was quantified using the BCA assay (Thermo Fisher Scientific). Laemmli buffer (Bio-Rad) was added to samples, which were then boiled and run on a polyacrylamide gel and transferred to a nitrocellulose membrane. The membrane was blocked with 5% milk in TBST, and then primary antibodies against LC3 (Cell Signaling Technology) or β-actin (Cell Signaling Technology) were incubated with the membrane at 1:1,000 dilution overnight at 4°C. Horseradish peroxidase–conjugated secondary antibodies (Cell Signaling Technology) at 1:2,000 dilution were then incubated with the membrane for 1 to 2 hours at 25°C, and ECL or ECL-plus (Thermo Fisher Scientific) was used for detection on a G-Box (Syngene). Original full Western blots for H2Kd, LC3, and β-actin are available in Supplementary Fig. S1.

Cytokine analysis

Cytokines were analyzed using the MILLIPLEX MAP Mouse Cytokine/Chemokine Magnetic Bead Panel (Millipore Sigma) following the manufacturer's protocol. Mouse serum was diluted 1:2 with included assay buffer, and samples were incubated with beads overnight at 4°C and run on a Luminex 200 platform (Luminex) using assay standards provided in the MILLIPLEX Kit. Data were analyzed using MILLIPLEX Analyst software (Millipore).

Single-cell RNA sequencing and library preparation

CT26 tumors were harvested from mice 3 days after treatment start (n = 3/treatment group), as described above. The single-cell RNA sequencing (scRNA-seq) libraries were generated using the Chromium Single Cell 3′ Library & Gel Bead Kit v.2 (10x Genomics) according to the manufacturer's protocol. Briefly, 3  ×  104 live CD45+ live cells per tumor were sorted by flow cytometry and used to generate single-cell gel beads in emulsion (10x Genomics). After reverse transcription, gel beads in emulsion were disrupted using the kit's recovery agent following the manufacturer's instructions (10x Genomics). Barcoded complementary DNA was isolated and amplified by PCR (12 cycles). Following fragmentation, end repair, and A-tailing, sample indexes were added during index PCR (10x Genomics). Samples were resuspended at a concentration of 5 ng/μL and added in equal volumes to the Illumina sequencing chip. The raw data for this experiment have been deposited to Sequence Read Archive (accession number PRJNA691630).

scRNA-seq analysis

Single-cell sequencing was performed using Illumina NovaSeq 6000. We performed demultiplexing, alignment to GRCm38 reference sequence (GenBank accession number GCA_000001635.8), filtering, barcoding, and unique molecular identifier (UMI) counting using Cell Ranger v2.2.0 (10x Genomics). Data were further analyzed using SeqGeq software (SeqGeq v1.5.0, BD Life Science, Informatics). Briefly, for quality control, cells with the highest (top 0.2%) or lowest (bottom 0.2%) numbers of detected genes were excluded from the downstream analyses. Genes of interest, or highly variable genes to be used for downstream analysis, were selected based on a dispersion value greater than 3 (dispersion is a parameter calculated by SeqGeq). Dimensionality reduction by principal component analysis (PCA) was performed using variable genes. T-stochastic neighbor embedding (t-SNE) plots were generated based on PCA dimensions. Cell types were identified by gating on relevant gene expression.

Arginine auxotrophy screen

Two thousand CT26, MCA-205, or MC38 cells per well were seeded in a 96-well flat-bottom plate (Thermo Fisher Scientific). Cells were allowed to adhere overnight in 90 μL of media appropriate to each cell line (see “Cell lines”). Pegzilarginase was added to wells, yielding final concentrations of 3.0 μmol/L and 8 successive 10-fold serial dilutions. A final concentration 0.14 mmol/L citrulline (Millipore Sigma) was added to half of all wells. After 3 days of incubation, cell viability was assessed using the CellTiter-Glo Luminescent Cell Viability Assay (Promega). Biological triplicates were used for each condition. Luminescence was measured on a Promega GloMax plate reader (Promega). Percent viability was determined by comparing luminescence values of each well with the mean luminescent value of the control wells.

Tumor interstitial fluid collection

Tumor interstitial fluid (TIF) was isolated from tumors using a previously described centrifugal method (29–32). Briefly, tumor-bearing animals were anesthetized via isoflurane inhalation (2%–3% IsoFlo #5260-04-05, Zoetis, mixed with O2), bled via cardiac puncture, and euthanized by cervical dislocation, and tumors were rapidly dissected from the animals. Dissections took <1 minute to complete. Following collection, blood was placed immediately in EDTA tubes (Sarstedt) and centrifuged to separate plasma. Plasma was frozen in liquid nitrogen and stored at −80°C until further analysis. Tumors were then weighed and briefly rinsed in cold saline (150 mmol/L NaCl) and blotted on filter paper (VWR, 28298-020). The entire process of preparing the tumor prior to isolation of TIF took <2 minutes. The tumors were then put onto 20-μm nylon filters (Spectrum Labs, 148134), affixed atop 50 mL conical tubes, and centrifuged for 10 minutes at 4°C at 106 × g. TIF was then collected from the conical tube, frozen in liquid nitrogen, and stored at −80°C until further analysis. To neutralize residual pegzilarginase activity, NN was added to appropriate collection tubes (18 μg NN, #10006861, Cayman Chemical; 2.6 mg of K2-EDTA; 0.1% Mannitol, all dissolved in DI H2O and lyophilized).

Arginine measurements in biofluids

Quantification of arginine in plasma was performed as previously described (4). Quantification of metabolites in TIF was performed as previously described (28). Briefly, a custom library of 152 chemical standards arranged in 7 separate pools (termed external standard pools) was assembled for major polar metabolites found in mammalian plasma. Serial dilutions, prepared in high-performance liquid chromatography (HPLC)–grade water, were made of these external standard pools. External standard pools were used to confirm m/z and retention time for each analyte and to quantitate stable isotope–labeled internal standards for use in downstream analysis (see below).

To extract polar metabolites from plasma, TIF, or the external standard pools, 5 μL of TIF, plasma, or external sample pools was mixed with 45 μL of acetonitrile:methanol:formic acid (75:25:0.1) extraction buffer (all solvents used were HPLC grade), including the following isotopically labeled internal standards: 13C-labeled yeast extract (Cambridge Isotope Laboratory, ISO1), 13C3 lactate (Sigma Aldrich, 485926), 13C3 glycerol (Cambridge Isotope Laboratory, CLM-1510), 13C615N2 cystine (Cambridge Isotope Laboratory, CNLM-4244), 2H9 choline (Cambridge Isotope Laboratory, DLM-549), 13C4 3-hydroxybutyrate (Cambridge Isotope Laboratory, CLM-3853), 13C6 glucose (Cambridge Isotope Laboratory, CLM-1396), 13C215N taurine (Cambridge Isotope Laboratory, CNLM-10253), 2H3 creatinine (Cambridge Isotope Laboratory, DLM-3653), 8-13C adenine (Cambridge Isotope Laboratory, CLM-1654), 13C5 hypoxanthine (Cambridge Isotope Laboratory, CLM-8042), 8-13C guanine (Cambridge Isotope Laboratory, CLM-1019), 13C3 serine (Cambridge Isotope Laboratory, CLM-1574), and 13C2 glycine (Cambridge Isotope Laboratory, CLM-1017). After addition of extraction buffer to the samples, they were then vortexed for 10 minutes at 4°C, and insoluble material was sedimented by centrifugation at 15 × g for 10 minutes at 4°C. Twenty microliter of the polar metabolite extract was added to sample vials for LC/MS analysis.

LC/MS analysis was performed using a Q Exactive Orbitrap Mass Spectrometer using an Ion Max source and heated electrospray ionization (HESI) probe coupled to a Dionex UltiMate 3000 UPLC system (Thermo Fisher Scientific). External mass calibration was performed every 7 days. Two microliter of each sample was injected onto a ZIC-pHILIC 2.1 × 150 mm analytic column equipped with a 2.1 × 20 mm guard column (both 5 μmol/L particle size, EMD Millipore). The autosampler and column oven were held at 4°C and 25°C, respectively. Buffer A was 20 mmol/L ammonium carbonate and 0.1% ammonium hydroxide; buffer B was acetonitrile. The chromatographic gradient was run at a flow rate of 0.150 mL/min as follows: 0 to 20 minutes: linear gradient from 80% to 20% B; 20 to 20.5 minutes: linear gradient from 20% to 80% B; 20.5 to 28 minutes: hold at 80% B. The mass spectrometer was operated in full scan, polarity-switching mode with the spray voltage set to 3.0 kV, the heated capillary held at 275°C, and the HESI probe held at 350°C. The sheath gas flow rate was set to 40 units, the auxiliary gas flow was set to 15 units, and the sweep gas flow was set to 1 unit. The MS data acquisition was performed in a range of 70 to 1,000 m/z, with the resolution set to 70,000, the automatic gain control (AGC) target at 1 × 106, and the maximum injection time at 20 msec.

Metabolite identification and quantification were performed with Xcalibur 2.2 software (Thermo Fisher Scientific) using a 5 ppm mass accuracy and a 0.5-minute retention time window. For metabolite identification, external standard pools were used for assignment of metabolites to peaks at given m/z and retention time (see Supplementary Table S1 for the m/z and retention time for each metabolite analyzed). For metabolite quantification, comparison of the peak areas of the stable isotope–labeled internal standards with the external standard pools at known concentrations allowed for determination of the concentration of labeled internal standards in the extraction buffer. For quantification of metabolite concentration in biological samples, we subsequently compared the peak area of a given metabolite in the TIF and plasma samples with the peak area of the now-quantified internal standard to determine the concentration of that metabolite in the TIF or plasma sample.

In vitro effects of pegzilarginase

CT26, MC38, or MCA-205 cells were seeded in a 96-well flat-bottom plate (3595, Corning) at a density of 1 × 105 cells/well. After 24 hours to adhere, cells were treated with 3 μmol/L pegzilarginase for 48 hours, at which point the cells were harvested and stained for viability and antibodies for calreticulin, MHC I, and MHC II (see “Flow cytometry antibodies and reagents”). Supernatant from the same wells was collected to determine ATP concentration (ATP Determination Kit, Thermo Fisher Scientific), following the manufacturer's instructions.

Ex vivo human bone marrow culture and analysis

Primary human bone marrow (AllCells) was obtained from healthy patients and cultured as follows. Briefly, Hs5 cells were cultured for 72 hours in SFEM2 media (Stem Cell Technologies) to obtain Hs5-conditioned media. Red blood cells were lysed from the human bone marrow cell population using a hypotonic lysis buffer (BioLegend). The remaining cells were plated in media (75% SFEM2 and 25% Hs5-conditioned media) and cultured at 37°C for 72 hours. Flow cytometry was performed to assess the effects of treatment on leukocyte populations of interest.

BMDMs and coculture with CT26

Femurs were collected from 6- to 8-week-old BALB/c mice. Bones were cut open at each end and flushed with PBS through a 27G needle into a sterile conical vial. Bone marrow was centrifuged at 400 × g for 5 minutes at 4°C. Cells were resuspended in BMDM growth medium [RPMI-1640 supplemented with 1 mol/L HEPES buffer, 1% penicillin–streptomycin, 10% FBS, and 15% 0.2-μm filtered L-929 cell line culture supernatant (taken at 24 hours of cell culture)] and cultured in a T175 culture flask. After incubating bone marrow cultures at 37°C for 3 days, growth media were changed to fresh growth-conditioned media without antibiotics. After 7 days in culture, cells were scraped with a cell scraper, 2 × 105 cells per well were plated in 6-well culture dishes in triplicate per culture condition, and cells were dosed with either PBS, pegzilarginase at 0.02 mg/mL, or 0.12 mg/mL. After 72-hour incubation at 37°C, cells were scraped, collected, and processed for flow cytometry.

For coculture experiments, BMDMs were cultured for 72 hours in the presence of IL4 (10 ng/mL) and IL13 (10 ng/mL; R&D Systems) to stimulate macrophage polarization toward the M2 (immunosuppressive) phenotype. M2 BMDMs (1 × 104) were cocultured with 1 × 105 CT26 cells in triplicate in the presence of PBS or pegzilarginase (0.03 or 0.12 mg/mL) for 72 hours.

Statistical analysis

All group sample sizes and biological replicate numbers are noted in the figure legends as appropriate. For comparison of more than two groups, statistical significance was determined by one-way ANOVA comparing each group against the combination treatment (for comparisons between groups). For comparison of two groups, statistical significance was determined by a Student t test. The Kaplan–Meier survival estimate was used to determine significance for tumor survival studies. Data were plotted and analyzed, and statistical tests were performed using GraphPad Prism software (GraphPad) or RStudio. A P value of <0.05 was considered statistically significant.

Pegzilarginase administration limits systemic and intratumoral arginine availability

To test the extent of extracellular arginine depletion after pegzilarginase treatment, we measured plasma arginine every 24 hours for 1 week following one dose of pegzilarginase, given at either 2 or 3 mg/kg. Across both concentrations, arginine was undetectable for 2 days after treatment and then recovered, albeit incompletely, by 144 hours (6 days; Fig. 1B). Thus, animals dosed with pegzilarginase experienced cyclical periods of arginine withdrawal and relative replenishment. At 3 mg/kg, pegzilarginase had single-agent antitumor efficacy (4) and lowered serum arginine to a nadir of approximately 1 μmol/L for 72 hours after treatment; therefore, we chose 3 mg/kg pegzilarginase for subsequent studies.

To test whether TME arginine concentrations were similarly modulated by pegzilarginase, we performed LC/MS on TIF (28) derived from CT26 tumor–bearing mice. We collected TIF in the presence of NN, an arginase inhibitor, because enzyme (wild-type arginase and/or pegzilarginase) activity during fluid collection and processing would continue to degrade arginine and result in an overestimation of the pegzilarginase and/or endogenous arginase–mediated decrease in arginine. We found that pegzilarginase treatment significantly decreased TIF arginine concentration, whereas arginine in TIF from control (PBS+NN) tumors was comparable with blood concentrations (Fig. 1C).

For arginine auxotrophic tumors, which cannot be rescued by citrulline through the ASS1-mediated compensatory pathway, the reduction of extracellular arginine due to pegzilarginase may result in increased autophagy and apoptosis (33, 34). In vitro, we observed a reduction in the viability of CT26, MCA-205, and MC38 tumor cells cultured in the presence of pegzilarginase; this compromised viability was not rescued by citrulline for CT26 and MC38, but a partial rescue for MCA-205 was observed. However, it is clear that MCA-205 cells cannot synthesize sufficient arginine to meet their metabolic requirements (Supplementary Fig. S2). To determine whether the pegzilarginase induced immunogenic cell death (ICD), we assessed two hallmarks of ICD, ATP secretion and calreticulin exposure, in CT26, MCA-205, and MC38 cells and found, indeed, that pegzilarginase significantly increased calreticulin and ATP secretion in all lines (Supplementary Fig. S3). We subsequently tested whether pegzilarginase had a similar direct tumor cell killing effect in vivo, by examining tumor cell and lymphocyte viability from CT26 tumors 3 days after treatment (Fig. 1D), when plasma arginine was at a nadir (Fig. 1B), and found no difference in the viable cell fraction (Supplementary Fig. S4). We also detected a significantly larger fraction of early apoptotic cells within the total viable cell fraction of the pegzilarginase treatment group, a finding consistent with the ability of arginine withdrawal to induce apoptosis in arginine auxotrophic tumor cells (Fig. 1E). We then asked whether the tumor cell fraction (CD45) and CD8+ T cells were similarly susceptible to increased apoptosis due to arginine depletion. We saw a significant increase in apoptosis within the CD45 compartment following pegzilarginase treatment, indicating that tumors were susceptible to cell death. In contrast, the CD8+ T cells from tumors experienced only a small increase in apoptosis due to pegzilarginase treatment (Fig. 1E). However, splenic CD8+ T cells experienced a reduction in apoptosis (Supplementary Fig. S5A and S5B), indicating that acute arginine deprivation was detrimental to intratumoral, but not systemic, CD8+ T cells. The T-cell arginine compensatory mechanism maintains T-cell proliferative capacity through upregulation and activity of ASS1 (22). However, we did not observe differences in ASS1 due to pegzilarginase treatment in intratumoral CD8+ T cells at this time point (Supplementary Fig. S5C).

Autophagy is another hallmark of ICD; thus, we examined the extent of pegzilarginase-induced autophagy in CT26 tumors in vivo by assessing LC3 protein expression (Fig. 1F) and by sequencing bulk RNA from CD45- cells to examine expression of autophagy-related genes (Supplementary Fig. S6). Combined, these data revealed pegzilarginase-induced increased autophagy within 3 days after treatment. Taken together, these data demonstrate that pegzilarginase has direct effects on arginine auxotrophic CT26 tumors, which undergo early apoptosis and autophagy during acute windows of systemic arginine deprivation.

Pegzilarginase antitumor activity requires CD8+ T cells

Because autophagy facilitates antigen processing and loading onto MHC I and II molecules, we hypothesized that pegzilarginase indirectly enhances antitumor immunity in part by increasing antigen presentation. To test this in vitro, we treated MCA-205, CT26, and MC38 tumors with 3 μmol/L pegzilarginase and monitored MHC I and MHC II expression. We found pegzilarginase-induced increases in both MHC I and MHC II across all cell lines (Supplementary Fig. S7). To test this in vivo, we harvested CT26 tumors 3 days after treatment (Fig. 2A) and determined changes in H2Kd (MHC I) expression. Indeed, we detected increased H2Kd expression following pegzilarginase treatment (Fig. 2B). Although pegzilarginase did not significantly increase macrophage frequency (Fig. 2C), it did significantly increase macrophage MHC II expression (Fig. 2C). Thus, providing evidence that pegzilarginase increases MHC I and MHC II expression.

Figure 2.

Pegzilarginase's antitumor activity requires CD8+ T cells. A, Schematic depicting experimental schema for experiments in B–F. B, MHC class I (H2Kd allele) expression measured by Western blot 3 days after treatment start. β-Actin measurement was used to ensure equal loading of samples. C,, Flow cytometry of cells isolated from tumors 3 days after treatment. n = 3 biological replicates/group. Median ± IQR. Mac, macrophage; MFI, mean fluorescence intensity. D, Flow cytometry of cells isolated from tumors 3, 7, 10, and 17 days after treatment. n = 5 biological replicates/group. Median ± IQR. TIL, tumor-infiltrating lymphocyte. E, Flow cytometry of tumor-specific cells isolated from tumors 3 days after treatment. n = 8 to 10 from two independent experiments. Median ± IQR. F, Tumor growth comparing pegzilarginase treatment against pegzilarginase treatment administered with a CD8-depleting mAb. Tumor measurement for all mice ended when the first mouse from that group reached maximum tumor volume. n = 8 biological replicates; mean ± SD. G, Human bone marrow was cultured in the presence of pegzilarginase for 3 days, and then leukocyte populations were measured by flow cytometry. n = 3 biological replicates/group. For all panels, Student t test: *, P < 0.05; **, P < 0.01; ****, P < 0.0001. IQR, interquartile range.

Figure 2.

Pegzilarginase's antitumor activity requires CD8+ T cells. A, Schematic depicting experimental schema for experiments in B–F. B, MHC class I (H2Kd allele) expression measured by Western blot 3 days after treatment start. β-Actin measurement was used to ensure equal loading of samples. C,, Flow cytometry of cells isolated from tumors 3 days after treatment. n = 3 biological replicates/group. Median ± IQR. Mac, macrophage; MFI, mean fluorescence intensity. D, Flow cytometry of cells isolated from tumors 3, 7, 10, and 17 days after treatment. n = 5 biological replicates/group. Median ± IQR. TIL, tumor-infiltrating lymphocyte. E, Flow cytometry of tumor-specific cells isolated from tumors 3 days after treatment. n = 8 to 10 from two independent experiments. Median ± IQR. F, Tumor growth comparing pegzilarginase treatment against pegzilarginase treatment administered with a CD8-depleting mAb. Tumor measurement for all mice ended when the first mouse from that group reached maximum tumor volume. n = 8 biological replicates; mean ± SD. G, Human bone marrow was cultured in the presence of pegzilarginase for 3 days, and then leukocyte populations were measured by flow cytometry. n = 3 biological replicates/group. For all panels, Student t test: *, P < 0.05; **, P < 0.01; ****, P < 0.0001. IQR, interquartile range.

Close modal

Given the effects on MHC I and II expression and the known increased antigenicity of arginine auxotrophic tumors, we investigated the impact of pegzilarginase on T cells. CT26 tumor–bearing mice were treated with pegzilarginase, and then tumors were harvested 3, 7, 10, and 17 days after treatment for immune profiling (Fig. 2A). CD4+ and CD8+ T cells decreased 3 days after treatment with pegzilarginase, which was significant for CD8+ T cells and corresponded with increased Annexin V staining (Fig. 1E) and known T-cell sensitivity to arginine depletion (Fig. 2D; refs. 14, 15, 26). However, by day 7 after treatment, both the CD4+ and CD8+ T-cell populations had recovered, and by days 10 and 17 after treatment, intratumoral CD8+ T cells in the pegzilarginase-treated group increased compared with controls (Fig. 2D). A slightly increased proportion of CD8+ T cells were tumor specific (Fig. 2E), perhaps resulting from increased tumor antigen presentation. We saw a reduction in tumor size beginning approximately 7 days after treatment (4, 21), which occurred in a CD8+ T-cell–dependent manner, as evidenced by a loss of tumor growth control following CD8+ T-cell depletion (Fig. 2F).

These results suggested that the population of intratumor CD8+ T cells increased over time despite arginine deprivation, which was surprising considering reports demonstrating T-cell dependence on arginine for proliferation (17). We reasoned that T cells conditioned in low-arginine environments may respond in a manner distinct from T cells that circulate in arginine-replete conditions. To test this hypothesis, we cultured human bone marrow, which expresses abundant arginase (35), ex vivo in the presence of increasing concentrations of pegzilarginase. We found that bone marrow–derived T-cell numbers increased slightly in response to treatment with 100 nmol/L pegzilarginase for 72 hours, whereas B cells and monocytes decreased in number (Fig. 2G). All together, these data indicated that pegzilarginase increases tumor-infiltrating CD8+ T cells over time, suggesting that pegzilarginase may support the activity of I-O agents.

Combining pegzilarginase with I-O agents improves antitumor efficacy relative to monotherapies

To test whether pegzilarginase enhances I-O agent efficacy, we examined the combination of pegzilarginase plus two different immunotherapy approaches: (i) Checkpoint blockade (anti–PD-L1 and anti–CTLA-4) and (ii) T-cell costimulation (agonist anti-OX40). To test the former, CT26 or MC38 tumor–bearing mice (75–100 mm3) received anti–PD-L1, pegzilarginase, or combined therapy (Fig. 3A). In both models, combined pegzilarginase/anti–PD-L1 was significantly more effective than I-O therapy alone (Fig. 3B). The pegzilarginase/anti–PD-L1 combination resulted in a pronounced increase in complete responses (defined as a tumor that exhibited positive growth trajectory prior to complete regression; 0/8 in monotherapy groups; 3/8 in combination group). Similar experiments were carried out with anti–CTLA-4, which trended toward increased efficacy for combination therapy, but statistical significance was not reached for the pegzilarginase/anti–CTLA-4 combination therapy over monotherapies (Supplementary Fig. S8).

Figure 3.

Combining pegzilarginase with I-O agents improves antitumor efficacy relative to monotherapies. A, Schematic depicting the tumor implantation and dosing regimen for in vivo studies involving pegzilarginase and anti–PD-L1. Pegzilarginase was given 1×/week (3 mg/kg; orange arrows), and anti–PD-L1 was given 2×/week (10 mg/kg; green arrows). B, Tumor growth (mean ± SD) and survival of CT26 (top) or MC38 (bottom) tumor–bearing mice. n = 10/group. C, Schematic depicting the tumor implantation and dosing regimen for in vivo studies involving pegzilarginase and anti-OX40. Pegzilarginase was given 1×/week (3 mg/kg; orange arrows) and anti-OX40 was given twice (250 μg/dose; green arrows). D, Tumor growth (mean ± SD) and survival of CT26 (top) or MCA-205 (bottom) tumor–bearing mice. n = 8/group (anti–PD-L1) or n = ≥13/group from two independent experiments (anti-OX40). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Figure 3.

Combining pegzilarginase with I-O agents improves antitumor efficacy relative to monotherapies. A, Schematic depicting the tumor implantation and dosing regimen for in vivo studies involving pegzilarginase and anti–PD-L1. Pegzilarginase was given 1×/week (3 mg/kg; orange arrows), and anti–PD-L1 was given 2×/week (10 mg/kg; green arrows). B, Tumor growth (mean ± SD) and survival of CT26 (top) or MC38 (bottom) tumor–bearing mice. n = 10/group. C, Schematic depicting the tumor implantation and dosing regimen for in vivo studies involving pegzilarginase and anti-OX40. Pegzilarginase was given 1×/week (3 mg/kg; orange arrows) and anti-OX40 was given twice (250 μg/dose; green arrows). D, Tumor growth (mean ± SD) and survival of CT26 (top) or MCA-205 (bottom) tumor–bearing mice. n = 8/group (anti–PD-L1) or n = ≥13/group from two independent experiments (anti-OX40). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Close modal

Based on our findings that pegzilarginase treatment led to increased tumor cell death, antigen presentation, and CD8+ T-cell infiltration, we hypothesized that administration of pegzilarginase in combination with an I-O agent that directly activates T cells may be beneficial. We tested this hypothesis by dosing pegzilarginase in combination with anti-OX40 in CT26 or MCA-205 (sarcoma) tumor–bearing mice (Fig. 3C). We noted significantly enhanced antitumor activity in the combination therapy group relative to anti-OX40 monotherapy in both models (Fig. 3D). Overall, these data suggest that arginine depletion through pegzilarginase results in an increased number of intratumoral CD8+ T cells within the TME, which help drive increased therapeutic efficacy of combined pegzilarginase/immunotherapy treatment (Fig. 3E; Table 1). We performed rechallenge experiments in mice that cleared tumors, and 100% were protected against new tumor growth, indicating formation of immunologic memory after both combination therapies.

Table 1.

Number of complete responses.

ModelI-O agentIsotope controlPegzilarginase single therapyI-O single therapyCombination therapy
CT26 Anti–PD-L1 0/8 0/8 0/8 3/8 
MC38 Anti–PD-L1 0/8 0/8 0/7 0/6 
CT26 Anti-OX40 0/14 0/13 1/13 7/13 
MCA-205 Anti-OX40 0/14 0/14 5/14 10/14 
ModelI-O agentIsotope controlPegzilarginase single therapyI-O single therapyCombination therapy
CT26 Anti–PD-L1 0/8 0/8 0/8 3/8 
MC38 Anti–PD-L1 0/8 0/8 0/7 0/6 
CT26 Anti-OX40 0/14 0/13 1/13 7/13 
MCA-205 Anti-OX40 0/14 0/14 5/14 10/14 

Note: Bold text highlights the increased number of complete responses in the combination therapy groups.

Therapeutic efficacy of pegzilarginase/I-O therapy depends on CD8+ T-cell activity

Because we observed the greatest antitumor activity and survival benefit when combining pegzilarginase with anti–PD-L1 or anti-OX40 in the CT26 model, we characterized the resulting immune activity in response to combination therapy. CT26 tumor–bearing mice were treated with pegzilarginase and anti–PD-L1, and then tumors were harvested 7, 10, and 17 days after treatment for immune profiling (Fig. 3A and C). These data revealed that pegzilarginase/anti–PD-L1 therapy increased CD4+ T cells on days 7 and 17 after treatment, which was driven by pegzilarginase treatment (Fig. 4A). In contrast, the CD8+ T-cell population increased only in response to the combination therapy 7 days after treatment and remained high throughout the time course (Fig. 4A). CD8-specific IHC on tumors collected 22 days after treatment confirmed our flow cytometry results (Fig. 4B). The gating strategy for these results and for subsequent markers is presented in Supplementary Fig. S9.

Figure 4.

Pegzilarginase/I-O agent combination therapy efficacy depends on CD8+ T-cell activity. A, Flow cytometric analysis of the intratumoral lymphocytes from CT26 tumors on days 7, 10, and 17 after treatment with pegzilarginase/anti–PD-L1 therapy. Number of cells per 70,000 total cells shown. n ≥ 5 biological replicates/experiment; median ± IQR. B, Representative images of intratumoral CD8 IHC 22 days after initiation of treatment. Clockwise from top left: PBS control, pegzilarginase, pegzilarginase + anti–PD-L1, and anti–PD-L1 groups. Images taken under 20× objective. C and D, Flow cytometric analysis of the intratumoral lymphocytes from CT26 tumors on day 7 after treatment with pegzilarginase/anti-OX40 therapy. Number of cells per 70,000 total cells shown. n ≥ 7 biological replicates from two independent experiments; median ± IQR. Statistics for A–D: ANOVA. E, Tumor growth over time comparing pegzilarginase/anti–PD-L1 combination treatment against pegzilarginase/anti–PD-L1 combination treatment and a CD8-depleting mAb. Tumor measurement for all mice ended when the first mouse from that group reached maximum tumor volume. n = 8/group; mean ± SD. F, Tumor growth over time comparing pegzilarginase/anti-OX40 against pegzilarginase/anti-OX40 with a CD8- or CD4-depleting mAb. Tumor measurement for all mice ended when the first mouse from that group reached maximum tumor volume. n = 10; mean ± SD. Statistics for E and F: ANOVA of groups at each time point.*, P < 0.05; **, P < 0.01; ***, P < 0.001. IQR, interquartile range.

Figure 4.

Pegzilarginase/I-O agent combination therapy efficacy depends on CD8+ T-cell activity. A, Flow cytometric analysis of the intratumoral lymphocytes from CT26 tumors on days 7, 10, and 17 after treatment with pegzilarginase/anti–PD-L1 therapy. Number of cells per 70,000 total cells shown. n ≥ 5 biological replicates/experiment; median ± IQR. B, Representative images of intratumoral CD8 IHC 22 days after initiation of treatment. Clockwise from top left: PBS control, pegzilarginase, pegzilarginase + anti–PD-L1, and anti–PD-L1 groups. Images taken under 20× objective. C and D, Flow cytometric analysis of the intratumoral lymphocytes from CT26 tumors on day 7 after treatment with pegzilarginase/anti-OX40 therapy. Number of cells per 70,000 total cells shown. n ≥ 7 biological replicates from two independent experiments; median ± IQR. Statistics for A–D: ANOVA. E, Tumor growth over time comparing pegzilarginase/anti–PD-L1 combination treatment against pegzilarginase/anti–PD-L1 combination treatment and a CD8-depleting mAb. Tumor measurement for all mice ended when the first mouse from that group reached maximum tumor volume. n = 8/group; mean ± SD. F, Tumor growth over time comparing pegzilarginase/anti-OX40 against pegzilarginase/anti-OX40 with a CD8- or CD4-depleting mAb. Tumor measurement for all mice ended when the first mouse from that group reached maximum tumor volume. n = 10; mean ± SD. Statistics for E and F: ANOVA of groups at each time point.*, P < 0.05; **, P < 0.01; ***, P < 0.001. IQR, interquartile range.

Close modal

For pegzilarginase/anti-OX40 combination therapy, we examined immune infiltration 7 days after treatment because of the rapid separation in tumor growth curves between the monotherapy and combination treatment cohorts. The high rate of complete responses following combination therapy (Fig. 3D and E) precluded examination of later time points. Pegzilarginase/anti-OX40 treatment elicited a pegzilarginase-driven increase in CD4+ T cells similar to that observed for the pegzilarginase/anti–PD-L1 combination (Fig. 4C). Within the CD4+ T-cell population, anti-OX40 increased the frequency of T-effector cells (FoxP3, Teff), which was not present in the combination therapy (Fig. 4C). Intratumoral CD8+ T cells only increased slightly in the anti-OX40 combination arm relative to controls 7 days after treatment, and this treatment demonstrated similar trends, albeit less significant, to the CD8+ T-cell increase in response to pegzilarginase/anti–PD-L1 therapy (Fig. 4D). Despite this small increase in CD8+ T cells in response to both combination therapies, treatment with a CD8-depleting mAb abrogated the antitumor effects of both (Fig. 4E and F). Due to the effects of anti-OX40 on effector and regulatory CD4+ T-cell subsets, we asked whether CD4+ T cells were necessary for pegzilarginase/anti-OX40 combination treatment efficacy by treating with a CD4-depleting mAb. CD4+ T cells appeared to be dispensible for the therapeutic efficacy of pegzilarginase/anti-OX40 combination therapy, despite an early increase in tumor growth in the absence of CD4+ T cells, by 20 days after treatment tumor growth, as it was unaffected by the absence of CD4+ T cells (Fig. 4F). This variable curve might be the result of the differing effects of depleting both Th1 effector CD4+ T cells and regulatory CD4+ T cells.

Because CD8+ T cells are required for the efficacy of pegzilarginase/I-O combination therapy, we asked whether this effect was due to increased numbers of CD8+ T cells (Fig. 4) or whether the combination therapies also influenced CD8+ T-cell activation and/or function. CT26 tumor–bearing mice were treated with pegzilarginase and/or anti–PD-L1, and subsequently tumors were harvested 7, 10, and 17 days after treatment for immune profiling (Fig. 3A and C). Differences in CD8+ T-cell memory and exhaustion were measured based upon KLRG1+PD-1hi subpopulations, and differences in short-term memory cells were measured by T-bet+Eomes subpopulations (36); however, by 17 days after treatment, these subsets were unchanged across treatment groups (Supplementary Fig. S10A–10C). We observed a pegzilarginase-driven upregulation of CD69 and CD25 by day 7 after treatment, which was maintained until day 17 after treatment (Fig. 5A). In the pegzilarginase/anti-OX40–treated cohort, we saw similar increases in CD8+CD25+ T cells (Supplementary Fig. S10D) and significant increases in the proportion of Ki-67+CD8+ T cells 7 days after treatment (Fig. 5B). We further examined effector responses by measuring serum IFNγ by ELISA following pegzilarginase/anti–PD-L1 treatment or granzyme A+CD8+ T cells following pegzilarginase/anti-OX40 treatment. In both cases, we detected significant increases in effector function in the combination therapy treatment groups compared with either agent alone (Fig. 5C and D). We were able to confirm these phenotypes by scRNA-seq of CD4+ and CD8+ T cells isolated 3 days after therapy (Supplementary Fig. S11). This analysis revealed increased markers of exhaustion, activation, and reactive oxygen species metabolism after pegzilarginase/anti-OX40 combination therapy. This suggests combination therapy made lymphocytes uniquely adapted to survive and function in a more inflamed environment. Taken together, these data indicate that pegzilarginase/I-O combination therapy induces more robust antitumor CD8+ T-cell responses, in terms of cell numbers and function, than either monotherapy.

Figure 5.

Enhanced activity of combined pegzilarginase/I-O agent therapy is marked by increased intratumoral CD8+ T-cell activation and function. A, Flow cytometric analysis of CD69+ and CD25+ CD8+ T cells from CT26 tumors on days 7, 10, and 17 after treatment with pegzilarginase/anti–PD-L1 combination therapy. Number of cells per 70,000 total cells shown. n = 10/group from two independent experiments; median ± IQR. B, Flow cytometric analysis of intratumoral Ki-67+ CD8+ T cells from CT26 tumors on day 7 after treatment with pegzilarginase/anti-OX40 combination therapy. n = 7/group from two independent experiments; median ± IQR. C, Serum IFNγ in tumor-bearing mice measured by ELISA 17 days after initiation of pegzilarginase/anti–PD-L1 combination treatment. n = 5/group; median ± IQR. D, Flow cytometric analysis of intratumoral granzyme A (GzmA)+CD8+ T cells from CT26 tumors on day 7 after treatment with pegzilarginase/anti-OX40 combination therapy. n = 11/group from two independent experiments; median ± IQR. Statistics for all panels, ANOVA: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. IQR, interquartile range.

Figure 5.

Enhanced activity of combined pegzilarginase/I-O agent therapy is marked by increased intratumoral CD8+ T-cell activation and function. A, Flow cytometric analysis of CD69+ and CD25+ CD8+ T cells from CT26 tumors on days 7, 10, and 17 after treatment with pegzilarginase/anti–PD-L1 combination therapy. Number of cells per 70,000 total cells shown. n = 10/group from two independent experiments; median ± IQR. B, Flow cytometric analysis of intratumoral Ki-67+ CD8+ T cells from CT26 tumors on day 7 after treatment with pegzilarginase/anti-OX40 combination therapy. n = 7/group from two independent experiments; median ± IQR. C, Serum IFNγ in tumor-bearing mice measured by ELISA 17 days after initiation of pegzilarginase/anti–PD-L1 combination treatment. n = 5/group; median ± IQR. D, Flow cytometric analysis of intratumoral granzyme A (GzmA)+CD8+ T cells from CT26 tumors on day 7 after treatment with pegzilarginase/anti-OX40 combination therapy. n = 11/group from two independent experiments; median ± IQR. Statistics for all panels, ANOVA: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. IQR, interquartile range.

Close modal

Pegzilarginase/anti-OX40 combination therapy alters the TME

Based upon the pegzilarginase-induced changes in tumor infiltration of activated and functional CD8+ T cells (Figs. 4 and 5), we asked whether there were changes to the TME that would promote CD8+ T-cell activity. To address this, we performed scRNA-seq on CD45+ cells isolated from CT26 tumors harvested 3 days after treatment with pegzilarginase and/or anti-OX40. The distribution of cells was visualized using a two-dimensional t-SNE projection (37). T-SNE projections from the different samples within each treatment group were similar (Supplementary Fig. S12A); thus for subsequent analyses, all cells from mice within each treatment group were pooled. The t-SNE projection revealed different cellular distributions for the combination therapy group (Fig. 6A). Clustering all cells from all samples revealed six major clusters: (i) macrophages, (ii) inflammatory monocytes, (iii) natural killer cells, (iv) dendritic cells (DC), (v) CD8+ T cells, and (vi) CD4+ T cells (Fig. 6B; Supplementary Fig. S12B). The combination therapy group was enriched in inflammatory monocytes, which appeared to be driven by anti-OX40 treatment (Fig. 6C and D), and uniquely enriched in macrophages (Fig. 6C and D). We confirmed these changes by flow cytometry (Fig. 6E). Flow cytometry further indicated that combination therapy with pegzilarginase/anti-OX40 did not enhance the absolute number of macrophages (Supplementary Fig. S13A and S13B); we saw a similar trend following pegzilarginase/anti–PD-L1 therapy (Fig. 6F).

Figure 6.

Pegzilarginase/anti-OX40 combination therapy alters the TME. A, CD45+ cells were sorted from CT26 tumors 3 days after treatment and used for scRNA-seq. Two thousand, one hundred cells downsampled from the cells sequenced for each mouse/treatment group (group numbers indicated). Mice within treatment groups were combined and visualized using a t-SNE projection. B, A contour plot displayed over the t-SNE projection indicating clusters of indicated cell types. iMono, inflammatory monocytes; macs, macrophages; NK, natural killer cells; TAMs, tumor-associated macrophages. C, Each treatment group visualized on the t-SNE plot, with cell type subset colored according to the contour plot in B. D, Percentage of each major cell type, of their parent population, across treatment groups. E, Flow cytometric analysis from the same tumors used for scRNA-seq; n = 3 biological replicates/group; median ± IQR. F, Flow cytometric analysis of macrophages from CT26 tumors on day 3 after treatment with pegzilarginase/anti–PD-L1 combination therapy. Number of cells per 70,000 total cells shown. n = 5 biological replicates/group; median ± IQR. G, Each treatment group visualized on the t-SNE plot, with a heatmap for expression of the sum of three genes, Ass1, Asl, and Slc7a2, overlaid on the t-SNE plot (left). Histograms of the sum of Ass1, Asl, and Scl7a2 expression from TAMs and inflammatory monocytes across all treatment groups (right). MFI, mean fluorescence intensity. H, From scRNA-seq data, histograms depicting expression of M1 and M2 genes in TAMs across all treatment groups. IQR, interquartile range.

Figure 6.

Pegzilarginase/anti-OX40 combination therapy alters the TME. A, CD45+ cells were sorted from CT26 tumors 3 days after treatment and used for scRNA-seq. Two thousand, one hundred cells downsampled from the cells sequenced for each mouse/treatment group (group numbers indicated). Mice within treatment groups were combined and visualized using a t-SNE projection. B, A contour plot displayed over the t-SNE projection indicating clusters of indicated cell types. iMono, inflammatory monocytes; macs, macrophages; NK, natural killer cells; TAMs, tumor-associated macrophages. C, Each treatment group visualized on the t-SNE plot, with cell type subset colored according to the contour plot in B. D, Percentage of each major cell type, of their parent population, across treatment groups. E, Flow cytometric analysis from the same tumors used for scRNA-seq; n = 3 biological replicates/group; median ± IQR. F, Flow cytometric analysis of macrophages from CT26 tumors on day 3 after treatment with pegzilarginase/anti–PD-L1 combination therapy. Number of cells per 70,000 total cells shown. n = 5 biological replicates/group; median ± IQR. G, Each treatment group visualized on the t-SNE plot, with a heatmap for expression of the sum of three genes, Ass1, Asl, and Slc7a2, overlaid on the t-SNE plot (left). Histograms of the sum of Ass1, Asl, and Scl7a2 expression from TAMs and inflammatory monocytes across all treatment groups (right). MFI, mean fluorescence intensity. H, From scRNA-seq data, histograms depicting expression of M1 and M2 genes in TAMs across all treatment groups. IQR, interquartile range.

Close modal

Given that pegzilarginase depleted arginine in the TME, we asked whether we could detect differences in gene expression within the arginine synthesis compensatory pathway (Ass1, Asl, and Slc7a2). Pegzilarginase did not result in significant changes in expression of these genes; however, anti-OX40 monotherapy did increase the expression and frequency of cells expressing these genes (Fig. 6G). Combination pegzilarginase/anti-OX40 therapy further expanded the expression and percentage of cells expressing genes that compensate for the lack of extracellular arginine (Fig. 6G). Cells having the highest expression of Ass1/Asl/Slc7a2 appeared to be macrophages and inflammatory monocytes (Fig. 6G). Of these, the percentage of tumor-associated macrophages expressing Ass1/Asl/Slc7a2 increased to 71% in the combination therapy cohort and 53% in the anti-OX40 monotherapy cohort. We also observed that combination therapy drove increased expression of M1-associated genes (STAT1, CXCL9, and NOS2) in macrophages (Fig. 6H). We confirmed these data by flow cytometry, which demonstrated decreased PD-L1 and increased iNOS on DCs and macrophages after pegzilarginase/anti-OX40 therapy (Supplementary Fig. S13C). Flow cytometry also revealed that pegzilarginase/anti-OX40 combination therapy resulted in fewer absolute numbers of suppressive monocytic MDSCs (Supplementary Fig. S13B). Combined with the single-cell data, these data confirmed that after pegzilarginase/anti-OX40, the TME was less suppressive and more supportive of a proinflammatory, antitumor immune response. We observed a similar finding ex vivo wherein unpolarized BMDMs cocultured with CT26 cells in the presence or absence of pegzilarginase for 72 hours displayed an increase in the M1 macrophage fraction in the presence of pegzilarginase (Supplementary Fig. S13D). These data suggest that the anti-OX40–mediated induction of genes regulating the arginine compensatory pathway was primarily within M1-polarized macrophages, which may enable their continued proliferation and/or survival under scarce extracellular arginine, thus contributing to the enhanced antitumor activity observed between these two treatment modalities.

Herein, we demonstrated that weekly dosing of pegzilarginase resulted in cyclical periods of arginine withdrawal and replenishment that increases CD8+ T-cell infiltration leading to enhanced antitumor immunity. This insight, in addition to the reported enhanced effect of immune checkpoint blockade on urea cycle–disordered tumors (13), prompted us to investigate the combination of pegzilarginase with I-O agents. Administering a combination of pegzilarginase with either anti–PD-L1 or anti-OX40 to mice bearing established syngeneic, arginine auxotrophic tumors uniformly resulted in increased therapeutic benefit compared with the respective I-O agents alone, including an increase in complete tumor regression and survival. The therapeutic efficacy of combination therapy relied on tumor-reactive CD8+ T cells exhibiting increased effector function. Altogether, these data provide a more nuanced exploration of arginine depletion in the setting of solid tumors. We found that the cyclical periods of arginine depletion and restoration present in our therapeutic model resulted in an acute window of diminished CD8+ T-cell proliferation, when available circulating arginine was at a nadir; this acute window was followed by CD8+ T-cell expansion, perhaps aided by recovering arginine levels.

Pegzilarginase activity depletes circulating arginine, leading to a corresponding reduction of arginine in the TME. The pegzilarginase-induced reduction of TME arginine levels is in contrast to endogenous MDSC-secreted arginase present in CT26 tumors, which possess a robust MDSC population (38) that does not create a low-arginine TME. This finding is significant as we believe that this is the first reported measurement of arginine in an MDSC-rich TME, and although this observation may be unexpected, it is not surprising because endogenous arginase-1 possesses poor catalytic activity and serum stability (20). This result, therefore, challenges the notion that MDSC-secreted arginase restrains T-cell activity by depleting intratumoral arginine and supports studies reporting that MDSC-driven T-cell suppression may be arginase-independent and instead requires cell-to-cell contact between MDSCs and T cells (21, 39).

The cyclical nature of arginine depletion in our in vivo studies is experimentally different from literature investigating the effects of continuous arginine depletion on immune function. Many of these studies examined chronic arginine depletion using in vitro systems in which arginine was completely absent or extremely limited (16, 23, 40–42). Clinically, however, it is not feasible to completely suppress patients' circulating arginine levels for extended intervals without incurring significant adverse events; thus, the dosing paradigm that we used in these studies and the consequent cycling of arginine levels between troughs where arginine is scarcely detectable and peaks that are below baseline but in excess of starvation conditions represent a clinically relevant scheme that produces only mild side effects in human patients (43, 44). Although chronic arginine depletion may be expected to result in tumor cell death and immunosuppression due to a lack of lymphocyte proliferation and activity, the result of a cyclical fluctuation of systemic and intratumoral arginine has not been explored. Despite this experimental difference, our results build upon established literature concerning arginine's role in immune system function in several respects. First, when plasma arginine levels were at the initial nadir (day 3 after treatment start), more intratumoral CD8+ T cells expressed apoptosis markers, and the CD8+ T-cell population was diminished in number, suggesting that T-cell proliferation required arginine (17, 40, 42). In contrast to intratumoral CD8+ T-cell populations, splenic CD8+ T cells from the same tumor-bearing animals were not reduced in number when systemic arginine was diminished. Similarly, prior research demonstrates that homeostatic naïve T-cell numbers are not decreased by systemic arginine depletion (45). This may be due to the stark difference in metabolic requirements between quiescent and activated T cells, including a differential requirement for L-arginine (46). Although arginine depletion may affect CD8+ T-cell proliferation, CD8+ T-cell activation is not impaired upon arginine depletion (26, 45); likewise, we found that intratumoral CD8+ T cells were activated during periods of cyclic arginine depletion. Thus, our data pertaining to the early posttreatment window, when arginine levels are depleted, agree with much of the literature detailing the effects of arginine depletion on T cells; however, these effects are fundamentally different at later time points, after the system experiences cycles of arginine depletion and recovery.

Given the impact of pegzilarginase on increasing CD8+ T-cell infiltration and antigen presentation machinery, we tested whether pegzilarginase augmented the antitumor efficacy of the I-O agents anti–PD-L1 and anti-OX40. Paradoxically, both systemic arginase inhibition (19) and administration of an engineered arginase enzyme (47) have been reported to elicit antitumor activity in combination with anti–PD-L1. This may be resolved by considering that an arginase inhibitor and an arginase enzyme may affect similar biological outcomes through different mechanisms. For example, an arginase inhibitor likely abrogates arginase produced by MDSCs and M2-polarized macrophages (19) that contribute to an immunosuppressive TME (42, 48), thereby reducing suppression of antitumor T cells. Administering an engineered arginase enzyme to arginine auxotrophic tumors likely induces tumor cell death, releasing tumor antigens and altering the TME to support antitumor adaptive immunity. Thus, arginase inhibition in either circumstance may result in a more polarized, proinflammatory TME that supports antitumor adaptive immunity.

Pegzilarginase-induced arginine deprivation may promote an immune-stimulatory TME through two distinct and independent mechanisms. First, pegzilarginase, through arginine depletion, appeared to act directly on the tumor by increasing ICD and releasing tumor-associated antigens, thereby enhancing immune function. Second, pegzilarginase indirectly affects T cells by (i) enhancing antigen presentation machinery on antigen-presenting cells and (ii) potentially increasing the repertoire of tumor-associated antigens available for processing and presentation. This model is consistent with the concept of altered nucleotide pools in arginine auxotrophic cells affecting neoantigen presentation (13) and supports exploiting arginine depletion in ASS1-low tumors for both direct anticancer and indirect immune-modulatory reasons. Considering these effects of pegzilarginase, our work revealed two potential mechanisms of synergy between pegzilarginase and I-O agents. The described outcomes of pegzilarginase treatment—increased tumor cell death and increased MHC expression—may both result in a greater number of tumor-specific T cells and/or increased T-cell activation and differentiation. Subsequently, those cells can be supported by administration of I-O agents to elicit a sustained antitumor response. Anti-OX40–induced expression of genes capable of compensating for the lack of extracellular arginine, along with repolarizing M2 to M1 macrophages, would allow beneficial antitumor M1 macrophages to persist in the low-arginine environment after pegzilarginase treatment. A similar mechanism may be at play in CD8+ T cells, as T-cell costimulation induces the expression of L-type amino acid transporter 1, a protein responsible for citrulline import (17), thus activating the compensatory arginine synthesis pathway. This costimulation-specific promotion of the compensatory pathway may explain the increase in antiumor activity in combination with pegzilarginase compared with the combination of pegzilarginase with anti–PD-L1. We suspect that each of these mechanisms contributes to the beneficial outcomes we observed following pegzilarginase/I-O combination therapy.

The antitumor activity of arginase as a single agent is shown to correlate with the ornithine transcarbamylase (OTC) and/or ASS1 status of the tumor (4); however, one limitation of our study is that it is unclear whether the enhanced antitumor activity we observed when combining cyclical arginine depletion with I-O therapy depends upon tumor cell sensitivity to arginine withdrawal. In several in vivo studies, including our combination with anti-OX40 in the MCA-205 model, we noted enhanced combination activity, despite minimal single-agent pegzilarginase antitumor activity; in vitro, MCA-205 cell death upon arginine withdrawal was not fully rescued by citrulline supplementation. This suggests that the antitumor mechanism of therapeutic arginine depletion in combination with I-O agents may consist of elements that are dependent and independent of increased tumor cell death. Future studies, such as those that employ arginine prototrophic tumor models in lieu of arginine auxotrophic models, should address the question of whether single-agent antitumor activity or tumor cell killing by arginine depletion is necessary for full combinatorial antitumor activity. Similarly, the syngeneic models used in these studies induce a fairly robust adaptive immune response and therefore are considered immune “warm” or “hot”; it remains unclear whether immune “cold” tumors, which provoke only a minimal response from the adaptive immune system, would be amenable to these combination therapeutic regimens.

Clinically, targeting tumor cell metabolism is not novel, as drugs such as methotrexate and various asparaginases are established standard-of-care therapies. However, the success of I-O agents in the clinic has intensified the need to understand how immune cell metabolism is modulated by anticancer agents, particularly those that act by targeting the aberrant metabolic profile of the tumor cells themselves. This work, in total, suggests that a dosing regimen that depletes arginine periodically can combine effectively with checkpoint inhibitors and T-cell activators alike. Biomarker-aided identification of appropriate tumor histologies that are both sensitive to arginine depletion and immune activation will be key to unlocking the potential of this combination therapy in the clinic.

M.D. Badeaux reports grants from Cancer Prevention and Research Institute of Texas (CPRIT) during the conduct of the study, as well as a patent for US10729752B2 issued. A.S. Rolig reports grants and nonfinancial support from Aeglea BioTherapeutics during the conduct of the study, as well as grants and nonfinancial support from Nektar Therapeutics outside the submitted work. G. Agnello reports grants from CPRIT during the conduct of the study, as well as a patent for US10729752B2 issued. D. Enzler reports grants from CPRIT during the conduct of the study, as well as a patent for US10729752B2 issued. M.J. Kasiewicz reports grants from Aeglea BioTherapeutics during the conduct of the study. L. Priddy reports grants from CPRIT during the conduct of the study, as well as a patent for US10729752B2 issued. J.F. Wiggins reports grants from CPRIT during the conduct of the study, as well as a patent for US10729752B2 issued. C. Daige reports grants from CPRIT during the conduct of the study, as well as a patent for US10729752B2 issued. M.G. Vander Heiden reports personal fees from Aeglea BioTherapeutics, Agios Pharmaceuticals, iTeos Biotherapeutics, Faeth Therapeutics, and Auron Therapeutics outside the submitted work. J.E. Wooldridge reports receiving grants from CPRIT (to institution) during the conduct of the study and was an employee of and owns stock in Aeglea BioTherapeutics. W.L. Redmond reports grants from Aeglea BioTherapeutics and Providence Portland Medical Foundation during the conduct of the study; grants from Galectin Therapeutics, OncoSec, Calibr, Mina Therapeutics, Nektar Therapeutics, Merck, Bristol-Myers Squibb, GlaxoSmithKline, Shimadzu, Veana Therapeutics, and Inhibrx outside the submitted work; and is on the advisory board of Vesselon and receives licensing fees from Galectin Therapeutics. S.W. Rowlinson reports grants from CPRIT during the conduct of the study; a patent for US10729752B2 issued; and owns stock in Aeglea BioTherapeutics. No disclosures were reported by the other authors.

M.D. Badeaux: Conceptualization, formal analysis, supervision, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing. A.S. Rolig: Conceptualization, data curation, validation, investigation, visualization, writing–original draft, writing–review and editing. G. Agnello: Conceptualization, writing–review and editing. D. Enzler: Investigation, methodology, writing–review and editing. M.J. Kasiewicz: Validation, investigation. L. Priddy: Investigation, writing–review and editing. J.F. Wiggins: Investigation, project administration, writing–review and editing. A. Muir: Investigation, methodology, writing–review and editing. M.R. Sullivan: Investigation. J. Van Cleef: Investigation, writing–review and editing. C. Daige: Conceptualization, formal analysis, supervision, methodology, project administration, writing–review and editing. M.G. Vander Heiden: Conceptualization, writing–review and editing. V. Rajamanickam: Data curation, software. J.E. Wooldridge: Conceptualization, writing–review and editing. W.L. Redmond: Conceptualization, resources, supervision, funding acquisition, project administration, writing–review and editing. S.W. Rowlinson: Conceptualization, supervision, writing–review and editing.

This research was funded in part by a grant (DP140031) awarded to Aeglea BioTherapeutics from CPRIT.

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.
Icard
P
,
Lincet
H
. 
A global view of the biochemical pathways involved in the regulation of the metabolism of cancer cells
.
Biochim Biophys Acta
2012
;
1826
:
423
33
.
2.
Rabinovich
S
,
Adler
L
,
Yizhak
K
,
Sarver
A
,
Silberman
A
,
Agron
S
, et al
Diversion of aspartate in ASS1-deficient tumours fosters de novo pyrimidine synthesis
.
Nature
2015
;
527
:
379
83
.
3.
Patil
MD
,
Bhaumik
J
,
Babykutty
S
,
Banerjee
UC
,
Fukumura
D
. 
Arginine dependence of tumor cells: targeting a chink in cancer's armor
.
Oncogene
2016
;
35
:
4957
72
.
4.
Agnello
G
,
Alters
S
,
Rowlinson
SW
. 
Preclinical safety and antitumor activity of the arginine-degrading therapeutic enzyme pegzilarginase
.
Transl Res
2020
;
217
:
11
22
.
5.
De Santo
C
,
Cheng
P
,
Beggs
A
,
Egan
S
,
Bessudo
A
,
Mussai
F
. 
Metabolic therapy with PEG-arginase induces a sustained complete remission in immunotherapy-resistant melanoma
.
J Hematol Oncol
2018
;
11
:
68
.
6.
Wilky
BA
. 
Immune checkpoint inhibitors: the linchpins of modern immunotherapy
.
Immunol Rev
2019
;
290
:
6
23
.
7.
Heinhuis
KM
,
Ros
W
,
Kok
M
,
Steeghs
N
,
Beijnen
JH
,
Schellens
JHM
. 
Enhancing antitumor response by combining immune checkpoint inhibitors with chemotherapy in solid tumors
.
Ann Oncol
2019
;
30
:
219
35
.
8.
Abu Eid
R
,
Razavi
GS
,
Mkrtichyan
M
,
Janik
J
,
Khleif
SN
. 
Old-school chemotherapy in immunotherapeutic combination in cancer, a low-cost drug repurposed
.
Cancer Immunol Res
2016
;
4
:
377
82
.
9.
Tang
C
,
Wang
X
,
Soh
H
,
Seyedin
S
,
Cortez
MA
,
Krishnan
S
, et al
Combining radiation and immunotherapy: a new systemic therapy for solid tumors?
Cancer Immunol Res
2014
;
2
:
831
8
.
10.
Mumm
JB
,
Emmerich
J
,
Zhang
X
,
Chan
I
,
Wu
L
,
Mauze
S
, et al
IL-10 elicits IFNgamma-dependent tumor immune surveillance
.
Cancer Cell
2011
;
20
:
781
96
.
11.
Charych
DH
,
Hoch
U
,
Langowski
JL
,
Lee
SR
,
Addepalli
MK
,
Kirk
PB
, et al
NKTR-214, an engineered cytokine with biased IL2 receptor binding, increased tumor exposure, and marked efficacy in mouse tumor models
.
Clin Cancer Res
2016
;
22
:
680
90
.
12.
Sullivan
RJ
. 
Back to the future: rethinking and retooling IL2 in the immune checkpoint inhibitor era
.
Cancer Discov
2019
;
9
:
694
5
.
13.
Lee
JS
,
Adler
L
,
Karathia
H
,
Carmel
N
,
Rabinovich
S
,
Auslander
N
, et al
Urea cycle dysregulation generates clinically relevant genomic and biochemical signatures
.
Cell
2018
;
174
:
1559
70
.
14.
MacIver
NJ
,
Michalek
RD
,
Rathmell
JC
. 
Metabolic regulation of T lymphocytes
.
Annu Rev Immunol
2013
;
31
:
259
83
.
15.
Buck
MD
,
O'Sullivan
D
,
Pearce
EL
. 
T cell metabolism drives immunity
.
J Exp Med
2015
;
212
:
1345
60
.
16.
Rodriguez
PC
,
Zea
AH
,
Culotta
KS
,
Zabaleta
J
,
Ochoa
JB
,
Ochoa
AC
. 
Regulation of T cell receptor CD3zeta chain expression by L-arginine
.
J Biol Chem
2002
;
277
:
21123
9
.
17.
Werner
A
,
Amann
E
,
Schnitzius
V
,
Habermeier
A
,
Luckner-Minden
C
,
Leuchtner
N
, et al
Induced arginine transport via cationic amino acid transporter-1 is necessary for human T-cell proliferation
.
Eur J Immunol
2016
;
46
:
92
103
.
18.
Bronte
V
,
Brandau
S
,
Chen
SH
,
Colombo
MP
,
Frey
AB
,
Greten
TF
, et al
Recommendations for myeloid-derived suppressor cell nomenclature and characterization standards
.
Nat Commun
2016
;
7
:
12150
.
19.
Steggerda
SM
,
Bennett
MK
,
Chen
J
,
Emberley
E
,
Huang
T
,
Janes
JR
, et al
Inhibition of arginase by CB-1158 blocks myeloid cell-mediated immune suppression in the tumor microenvironment
.
J Immunother Cancer
2017
;
5
:
101
.
20.
Stone
EM
,
Glazer
ES
,
Chantranupong
L
,
Cherukuri
P
,
Breece
RM
,
Tierney
DL
, et al
Replacing Mn(2+) with Co(2+) in human arginase i enhances cytotoxicity toward l-arginine auxotrophic cancer cell lines
.
ACS Chem Biol
2010
;
5
:
333
42
.
21.
Bian
Z
,
Abdelaal
AM
,
Shi
L
,
Liang
H
,
Xiong
L
,
Kidder
K
, et al
Arginase-1 is neither constitutively expressed in nor required for myeloid-derived suppressor cell-mediated inhibition of T-cell proliferation
.
Eur J Immunol
2018
;
48
:
1046
58
.
22.
Werner
A
,
Koschke
M
,
Leuchtner
N
,
Luckner-Minden
C
,
Habermeier
A
,
Rupp
J
, et al
Reconstitution of T cell proliferation under arginine limitation: activated human T cells take up citrulline via L-type amino acid transporter 1 and use it to regenerate arginine after induction of argininosuccinate synthase expression
.
Front Immunol
2017
;
8
:
864
.
23.
Bansal
V
,
Rodriguez
P
,
Wu
G
,
Eichler
DC
,
Zabaleta
J
,
Taheri
F
, et al
Citrulline can preserve proliferation and prevent the loss of CD3 zeta chain under conditions of low arginine
.
JPEN J Parenter Enteral Nutr
2004
;
28
:
423
30
.
24.
Husson
A
,
Brasse-Lagnel
C
,
Fairand
A
,
Renouf
S
,
Lavoinne
A
. 
Argininosuccinate synthetase from the urea cycle to the citrulline-NO cycle
.
Eur J Biochem
2003
;
270
:
1887
99
.
25.
Van de Velde
LA
,
Subramanian
C
,
Smith
AM
,
Barron
L
,
Qualls
JE
,
Neale
G
, et al
T cells encountering myeloid cells programmed for amino acid-dependent immunosuppression use Rictor/mTORC2 protein for proliferative checkpoint decisions
.
J Biol Chem
2017
;
292
:
15
30
.
26.
Munder
M
,
Engelhardt
M
,
Knies
D
,
Medenhoff
S
,
Wabnitz
G
,
Luckner-Minden
C
, et al
Cytotoxicity of tumor antigen specific human T cells is unimpaired by arginine depletion
.
PLoS One
2013
;
8
:
e63521
.
27.
Monticelli
LA
,
Buck
MD
,
Flamar
AL
,
Saenz
SA
,
Tait Wojno
ED
,
Yudanin
NA
, et al
Arginase 1 is an innate lymphoid-cell-intrinsic metabolic checkpoint controlling type 2 inflammation
.
Nat Immunol
2016
;
17
:
656
65
.
28.
Sullivan
MR
,
Danai
LV
,
Lewis
CA
,
Chan
SH
,
Gui
DY
,
Kunchok
T
, et al
Quantification of microenvironmental metabolites in murine cancers reveals determinants of tumor nutrient availability
.
Elife
2019
;
8
:
e44235
.
29.
Eil
R
,
Vodnala
SK
,
Clever
D
,
Klebanoff
CA
,
Sukumar
M
,
Pan
JH
, et al
Ionic immune suppression within the tumour microenvironment limits T cell effector function
.
Nature
2016
;
537
:
539
43
.
30.
Ho
PC
,
Bihuniak
JD
,
Macintyre
AN
,
Staron
M
,
Liu
XJ
,
Amezquita
R
, et al
Phosphoenolpyruvate is a metabolic checkpoint of anti-tumor T cell responses
.
Cell
2015
;
162
:
1217
28
.
31.
Haslene-Hox
H
,
Oveland
E
,
Berg
KC
,
Kolmannskog
O
,
Woie
K
,
Salvesen
HB
, et al
A new method for isolation of interstitial fluid from human solid tumors applied to proteomic analysis of ovarian carcinoma tissue
.
PLoS One
2011
;
6
:
e19217
.
32.
Wiig
H
,
Aukland
K
,
Tenstad
O
. 
Isolation of interstitial fluid from rat mammary tumors by a centrifugation method
.
Am J Physiol Heart Circ Physiol
2003
;
284
:
H416
24
.
33.
Kim
RH
,
Bold
RJ
,
Kung
HJ
. 
ADI, autophagy and apoptosis: metabolic stress as a therapeutic option for prostate cancer
.
Autophagy
2009
;
5
:
567
8
.
34.
Gong
H
,
Zolzer
F
,
von Recklinghausen
G
,
Havers
W
,
Schweigerer
L
. 
Arginine deiminase inhibits proliferation of human leukemia cells more potently than asparaginase by inducing cell cycle arrest and apoptosis
.
Leukemia
2000
;
14
:
826
9
.
35.
Bhatta
A
,
Sangani
R
,
Kolhe
R
,
Toque
HA
,
Cain
M
,
Wong
A
, et al
Deregulation of arginase induces bone complications in high-fat/high-sucrose diet diabetic mouse model
.
Mol Cell Endocrinol
2016
;
422
:
211
20
.
36.
Lazarevic
V
,
Glimcher
LH
,
Lord
GM
. 
T-bet: a bridge between innate and adaptive immunity
.
Nat Rev Immunol
2013
;
13
:
777
89
.
37.
Van Der Maaten
L
,
Hinton
G
. 
Visualizing data using t-SNE
.
J Mach Learn Res
2008
;
9
:
2579
605
.
38.
Yu
JW
,
Bhattacharya
S
,
Yanamandra
N
,
Kilian
D
,
Shi
H
,
Yadavilli
S
, et al
Tumor-immune profiling of murine syngeneic tumor models as a framework to guide mechanistic studies and predict therapy response in distinct tumor microenvironments
.
PLoS One
2018
;
13
:
e0206223
.
39.
Crook
KR
,
Jin
M
,
Weeks
MF
,
Rampersad
RR
,
Baldi
RM
,
Glekas
AS
, et al
Myeloid-derived suppressor cells regulate T cell and B cell responses during autoimmune disease
.
J Leukoc Biol
2015
;
97
:
573
82
.
40.
Rodriguez
PC
,
Quiceno
DG
,
Ochoa
AC
. 
L-arginine availability regulates T-lymphocyte cell-cycle progression
.
Blood
2007
;
109
:
1568
73
.
41.
Feldmeyer
N
,
Wabnitz
G
,
Leicht
S
,
Luckner-Minden
C
,
Schiller
M
,
Franz
T
, et al
Arginine deficiency leads to impaired cofilin dephosphorylation in activated human T lymphocytes
.
Int Immunol
2012
;
24
:
303
13
.
42.
Munder
M
,
Schneider
H
,
Luckner
C
,
Giese
T
,
Langhans
CD
,
Fuentes
JM
, et al
Suppression of T-cell functions by human granulocyte arginase
.
Blood
2006
;
108
:
1627
34
.
43.
Uy
GL
,
Savona
MR
,
Tomlinson
B
,
Carraway
H
. 
Phase 1 trial of pegzilarginase in patients with relapsed/refractory AML or MDS refractory to hypomethylating agents
. J Clin Oncol 36:15s, 2018 (suppl; abstr 7031).
44.
Phillips
MM
,
Sheaff
MT
,
Szlosarek
PW
. 
Targeting arginine-dependent cancers with arginine-degrading enzymes: opportunities and challenges
.
Cancer Res Treat
2013
;
45
:
251
62
.
45.
Fletcher
M
,
Ramirez
ME
,
Sierra
RA
,
Raber
P
,
Thevenot
P
,
Al-Khami
AA
, et al
l-Arginine depletion blunts antitumor T-cell responses by inducing myeloid-derived suppressor cells
.
Cancer Res
2015
;
75
:
275
83
.
46.
Geiger
R
,
Rieckmann
JC
,
Wolf
T
,
Basso
C
,
Feng
Y
,
Fuhrer
T
, et al
L-arginine modulates T cell metabolism and enhances survival and anti-tumor activity
.
Cell
2016
;
167
:
829
42
.
47.
Brin
E
,
Wu
K
,
Lu
HT
,
He
Y
,
Dai
Z
,
He
W
. 
PEGylated arginine deiminase can modulate tumor immune microenvironment by affecting immune checkpoint expression, decreasing regulatory T cell accumulation and inducing tumor T cell infiltration
.
Oncotarget
2017
;
8
:
58948
63
.
48.
Roszer
T
. 
Understanding the mysterious M2 macrophage through activation markers and effector mechanisms
.
Mediators Inflamm
2015
;
2015
:
816460
.