Moderate expression of CD39 in GPC3-CAR-T cells shows high efficacy against hepatocellular carcinoma

Abstract

CD39 serves as a crucial biomarker for neoantigen-specific CD8⁺ T cells and is associated with antitumor activity and exhaustion. However, the relationship between CD39 expression levels and the function of chimeric antigen receptor T (CAR-T) cells remains controversial. This study aimed to investigate the role of CD39 in the functional performance of CAR-T cells against hepatocellular carcinoma (HCC) and explore the therapeutic potential of CD39 modulators, such as mitochondrial division inhibitor-1 (mdivi-1), or knockdown CD39 through short hairpin RNA. Our findings demonstrated that glypican-3-CAR-T cells with moderate CD39 expression exhibited a strong antitumor activity, while high and low levels of CD39 led to an impaired cellular function. Methods modulating the proportion of CD39 intermediate (CD39^int)-phenotype CAR-T cells such as mdivi-1 and CD39 knockdown enhanced and impaired T cell function, respectively. The combination of mdivi-1 and CD39 knockdown in CAR-T cells yielded the highest proportion of infiltrated CD39^int CAR-T cells and demonstrated a robust antitumor activity in vivo. In conclusion, this study revealed the crucial role of CD39 in CAR-T cell function, demonstrated the potential therapeutic efficacy of combining mdivi-1 with CD39 knockdown in HCC, and provided a novel treatment strategy for HCC patients in the field of cellular immunotherapy.

Keywords

CD39 / CAR-T cells / mdivi-1 / hepatocellular carcinoma / antitumor activity

1 Introduction

Hepatocellular carcinoma (HCC) is the fourth leading cause of cancer and ranks sixth among cancer-related deaths globally [1]. Current immunotherapy options for HCC, particularly cellular therapy, have shown limited effectiveness [2]. Glypican-3 (GPC3), highly expressed on the surface of HCC cells but minimally present in normal tissues, makes chimeric antigen receptor (CAR)-T cells a promising therapeutic approach for HCC [3]. However, despite several ongoing clinical trials using GPC3 CAR-T cells and GPC3 peptide-specific cytotoxic T lymphocytes, the results have not met expectations [4–6]. The limitations in the effectiveness of CAR-T cell therapy in solid tumors such as HCC can be attributed to factors such as rapid T cell exhaustion and loss of cytotoxicity within the tumor microenvironment [7].

CD39, also known as ENTPD1, is expressed in various immune cells, including T cells, and it plays a dual role as a crucial marker of neoantigen-specific CD8⁺ T cells and as an indicator of T cell exhaustion [8]. Recent studies have revealed a correlation between CD39 expression and patient survival and prognosis, suggesting its significance in tumor neoantigen-specific CD8⁺ T cells [9,10]. Additionally, CD39 has been associated with the functional state of CD8⁺ T cells, representing the antitumor activity and T cell exhaustion in chronic infections and tumor-reactive CD8⁺ T cells [11,12].

In our previous studies, we found that the knockdown of CD39 weakened the proliferation and function of CAR-T cells in vitro during the early stages [13]. However, the role of CD39 in CAR-T cells in vivo remains unclear. Small chemical inhibitors of mitochondrial division dynamin, such as mitochondrial division inhibitor-1 (mdivi-1), CGS-21680, and 8-bromo-cAMP, play important roles in mitochondrial dynamics, ATP production, and Ca²⁺ homeostasis, and have been shown to influence CD39 function and mitochondrial dynamics, potentially impacting CD39-mediated immune responses [14–16].

In this study, we aimed to explore the impact of CD39 expression on the functional performance of CAR-T cells against HCC and examine the therapeutic potential of CD39 modulators, such as mdivi-1, or knockdown of endogenous CD39 using short hairpin RNA (shRNA). We hypothesized that the modulation of CD39 expression could affect CAR-T cell function and potentially enhance the antitumor activity against HCC. By investigating the effects of CD39 modulation on CAR-T cell function and evaluating the combination of mdivi-1 and CD39 knockdown, we aimed to elucidate the critical role of CD39 in CAR-T cell therapy and provide insights for improving treatment strategies for HCC in the field of cellular immunotherapy.

2 Materials and methods

2.1 Cell lines

Huh7 and HEK293T cell lines were obtained from ATCC and cultured in Dulbecco’s modified Eagle medium (DMEM) (Gibco, Thermo Fisher Scientific, Beijing, China) supplemented with 10% fetal bovine serum (FBS) (Gibco, Invitrogen, Carlsbad, CA). All cell culture media contained 100 U/mL of penicillin and 100 μg/mL of streptomycin. Cells were maintained in a humidified atmosphere containing 5% CO₂ at 37 °C. Cell lines were routinely tested for excluding mycoplasma contamination.

2.2 Isolation and culture of primary human T lymphocytes

Peripheral blood mononuclear cells (PBMCs) were derived from samples obtained from anonymous buffy coats of healthy donors (Guangzhou Blood Center, Guangzhou) by Ficoll-Hypaque gradient separation. Primary human CD8⁺ T cells were negatively purified with magnetic beads to a purity of > 98% from PBMCs with enrichment set DM (BD IMag^TM). T lymphocytes were activated by anti-CD3 antibody (Ab) (BioLegend, San Diego, CA, USA) and anti-CD28 Ab (BioLegend, San Diego, CA, USA) at 1 μg/mL and infected in RetroNectin-coated plates (Takara Bio Inc., Shiga, Japan). The transduced T cells were expanded in the conditioned medium containing 90% RPMI 1640 (Gibco, Invitrogen, Carlsbad, CA) supplemented with 10% FBS (Gibco, Invitrogen, Carlsbad, CA), 0.1 mmol/L nonessential amino acids (Gibco, Invitrogen, Carlsbad, CA), 2 mmol/L GlutaMAX (Gibco, Invitrogen, Carlsbad, CA), and 0.05 mmol/L 2-mercaptoethanol at an initial concentration of 1 × 10⁶ cells/mL. Cells were fed twice a week with recombinant interleukin-2 (IL-2) (10 ng/mL) (R&D Systems).

2.3 Construction of CAR-encoding lentiviral vector

Anti-GPC3 scFv was derived from the basic sequence of codrituzumab (US16649039, Chugai Seiyaku Kabushiki KaishaChugai Seiyaku Kabushiki KaishaF. Hoffmann-La Roche AG, Japan). As shown in Fig.1, the scFv region was fused with the transmembrane domain of CD8α (nucleotides 548–623; GenBank accession number: BC025715.1) and the endodomains of CD28 (nucleotides 753–882; GenBank accession number: NM_006139.3), CD137 (nucleotides 903–1026; NM_001561.5), and CD3ζ (nucleotides 307–637; NM_198053.2) in tandem with GGGGS sequences inserted between each domain. Her2-CAR, specifically targeting human Her2 antigen, was used as NC-CAR-T cells on the basis of a previous study [17]. A schematic of lentiviral vectors carrying a GPC3 CAR moiety and a cluster of shRNA targeting CD39 and the sequences of siRNA are shown in Fig.1 and Table S1.

2.4 Transduction of recombinant lentiviral particles

For the production of pseudotyped lentiviral supernatants, on the day before transduction, HEK293T cells were seeded in a dish with 8 × 10⁶ cells per 100 mmol/L. Twenty-four hours later, pseudoviruses were generated by co-transfecting HEK293T cells with plasmids encoding CAR with or without shCD39 moieties (13.5 μg), pMD.2G encoding VSV-G envelope (7.5 μg), and a packaging vector psPAX2 (16.5 μg) using a phosphate transfection system following the manufacturer’s instructions. Supernatants were harvested after 48 h and filtered through a 0.45 μm membrane to remove cell debris. Pseudoviruses were concentrated by ultracentrifugation (Optima XE-100, Beckmann) at 827 000× g for 2 h at 4 °C. Then, the activated CD8⁺ T lymphocytes were transduced with lentiviral supernatants using RetroNectin-coated plates, with polybrene (TR-1003-G, Sigma) at 8 μg/mL, followed by centrifugation for 90 min at 350 g and incubation at 37 °C. Twelve hours later, the recombinant viruses were removed, and T cells were expanded in the conditioned medium, as described above. The genetically modified T cells were maintained in a complete T cell medium in the presence of IL-2 (fed twice a week, 10 ng/mL) and used for functional assay 14 days after transduction.

2.5 Quantitative reverse transcription-polymerase chain reaction (qRT-PCR)

Total RNA was extracted from modified CAR-T cells with TRIzol reagent (Ambion, Carlsbad, CA, USA) and served as the template for preparing cDNA using a PrimeScript reverse transcription reagent kit (TaKaRa, Osaka, Japan). The primers for real-time qRT-PCR are listed in Table S1. The results of the RNA relative expression were reported using the CFX96 Real-Time System (Bio-Rad, Singapore) and normalized with β-actin housekeeping gene used as an endogenous control.

2.6 In vitro antigen stimulation experimental design

After being activated for 48 h, CD8⁺ T lymphocytes were transduced with pseudoviruses, supplemented with polybrene (Sigma-Aldrich, Saint Louis, MO, USA) at 8 μg/mL and centrifuged at 350× g and 37 °C for 90 min. After 12 h, CAR-T cells were washed twice, fed in an IL-2-free medium in the presence of recombinant human GPC3 (rhGPC3, 5 μg/mL), and used for flow cytometry experiment 24 h after rhGPC3 stimulation.

2.7 Cytotoxicity assay

Lactate dehydrogenase (LDH; Promega, Madison, WI, USA) release assay was used to evaluate the ability of modified CAR-T cells to kill antigen-specific tumor cells. Briefly, modified CAR-T cells were cocultured with tumor cell lines at different ratios (from 8:1 to 2:1) for 24 h in a 96-well V-bottom plate. The LDH released was measured using the CytoTox96 Non-Radioactive Cytotoxicity Assay (Promega), in accordance with the manufacturer’s instructions. The absorbance values of wells were measured at 490 nm and subtracted as the background from the values of cocultures. Cytotoxicity was calculated using the following formula: Cytotoxicity (%) = ((Experimental value − Effector spontaneous value − Target spontaneous value) / (Target maximum value − Target spontaneous value)) × 100%. In the spontaneous group, effector and target cells were cultured alone. Target cells alone were lysed with a lysis reagent at 37 °C for 30 min as a maximum control [17].

2.8 Patients and sample collection

Between 2018 and 2020, a cohort of patients with HCC at The First Affiliated Hospital of Sun Yat-sen University underwent needle biopsies and blood collection. The project was approved with the patients’ informed consent and ethical approval (permit number 2018 [43]). No treatment was given to patients when samples were collected, and patients had received surgery or no therapy previously. Three patients with puncture specimens showing GPC3 positivity were selected.

2.9 HCC tumor organoid culture

The procedures were developed by following previous procedures with minor modifications [18,19]. Briefly, tumor tissues derived from needle biopsies were mechanically dissociated into small pieces using needles and embedded in Matrigel (matrix basement membrane growth factor reduced, Corning #354230). After Matrigel was solidified for 20 min at 37 °C, cells were overlaid with a human HCC organoid medium. The organoid medium was composed of Advanced DMEM/F12 (GIBCO) supplemented with 2 mmol/L UltraGlutamine I (Lonza), 10 mmol/L HEPES (GIBCO), 100/100 U/mL of penicillin/streptomycin (GIBCO), 30% Wnt3a-conditioned medium, 1 × B27 supplement without vitamin A (GIBCO), 1 × N2 supplement (GIBCO), 2 mmol/L N-acetylcysteine (Sigma), 10 mmol/L nicotinamide (Sigma), 10 nmol/L recombinant human [Leu15]-Gastrin I (Sigma), 50 ng/mL of human recombinant EGF (PeproTech), 100 ng/mL of human recombinant FGF-10 (PeproTech), 50 ng/mL of human recombinant HGF (PeproTech), 50 ng/mL of human recombinant R-spondin1 (Novoprotein), 5 μmol/L A83-01 (Sigma), and 10 μmol/L MY-27632 (Sigma). The Wnt3a-conditioned medium was from L-Wnt3a cells. Organoids were passaged approximately every week by incubation in TrypLE Express (GIBCO) for 5–10 min at 37 °C to dissociate organoids to single cells and replate in fresh Matrigel. Organoids were cryopreserved in 10% FBS/DMSO, and organoids < passage 30 were used in experiments. All organoids were regularly checked for mycoplasma contamination using the MycoAlert Mycoplasma Detection Kit (Lonza).

2.10 Flow cytometry

For the evaluation of the cytotoxicity of T cells in vitro, cells were washed twice in FACS buffer and stained with the antibodies FITC anti-GGGGS (Cat No. GS-ARFT100, Hycells), APC anti-CD39 (Cat No. 328209, BioLegend), and Pacific Blue anti-PD-1 (Cat No. 329916, BioLegend) for 30 min at 4 °C. They were then washed twice in FACS buffer, fixed, and stained for intracellular PE anti-interferon-γ (IFNγ) (Cat No. 383304, BioLegend) using the Cytofix/Cytoperm kit (BD) in accordance with the manufacturer’s instructions.

For the evaluation of the tumor reactivity of T cells in a tumor organoid model, 10⁵ sorted CAR-T cells with or without mdivi-1 were cocultured with tumor organoids for 12 h. GolgiPlug (1:1000, BD) and GolgiStop (1:1500, BD) were added after 1 h. Cells were washed twice in FACS buffer and stained with the antibodies FITC anti-GGGGS (Cat No. GS-ARFT100, Hycells) and APC anti-CD39 (Cat No. 328209, BioLegend) for 30 min at 4 °C. Afterward, the cells were washed twice in FACS buffer, fixed, and stained for intracellular PE anti-IFNγ (Cat No. 383304, BioLegend) and PE anti-ki67 (Cat No. 350503, BioLegend) using the Cytofix/Cytoperm kit (BD) in accordance with the manufacturer’s instructions.

For the evaluation of the tumor reactivity of T cells in vivo, cells were washed twice in FACS buffer and stained with the antibodies FITC anti-GGGGS (Cat No. GS-ARFT100, Hycells) and APC anti-CD39 (Cat No. 328209, BioLegend) for 30 min at 4 °C. Subsequently, the cells were washed twice in FACS buffer, fixed, and stained for intracellular PE anti-IFNγ (Cat No. 383304, BioLegend) using the Cytofix/Cytoperm kit (BD) in accordance with the manufacturer’s instructions. Stained samples were tested by flow cytometric analysis and analyzed with FlowJo software.

2.11 Organoid killing assay

Organoids were isolated from Matrigel, and part of the organoids were dissociated into single cells and counted to infer the number of tumor cells per tumor organoid to allow coculture of organoids and T cells at 10:1 effector : target ratio. Tumor organoids were resuspended in X-VIVO^TM 15 Medium and seeded in triplicate of flat-bottom plate with 1 × 10⁵ autologous CAR-T cells. T cells were previously stained with 1 mmol/L CellTrace Far Red (Invitrogen) in PBS for 20 min at 37 °C, followed by blocking with human serum and washing in PBS, to facilitate visualization. At the start of coculture, a green fluorescent caspase 3/7 probe that binds DNA upon cleavage by caspase 3/7 (Invitrogen) was added at 1:2000 dilution to visualize cell apoptosis. After 24 h of coculture, microphotograph images were taken.

For the quantification of the results, the images of organoids were analyzed with Imaris software (Version 7.4) (Bitplane) using the Spot function to locate and enumerate T cells and apoptosis cells within the organoid threshold. Similarly, the absolute numbers of T cell and apoptotic cell spot per mm² of organoid areas were statistically analyzed with Imaris software (Version 7.4) (Bitplane) using the Spot function to locate and enumerate CAR-T cells, or apoptotic cells, on the basis of size and intensity threshold. The initial infiltration of T cells inside the organoids was considered zero. Finally, organoids were dissociated into single cells with TrypLE Express until the organoids were fully dissociated. Cells were washed in FACS buffer for FACS.

2.12 HCC xenograft mouse models

We used an NSG (NOD-Prkdc^scidIL2rg^tm1/Bcgen, Beijing Biocytogen Co., Ltd.) mouse model to evaluate the in vivo antitumor effects of transduced CAR-T cells. Mouse experiments were approved and carried out in accordance with the guidelines and regulations of the Laboratory Monitoring Committee of the First Affiliated Hospital of Sun Yet-sen University (SYSU-IACUC-2021-000675). For the observation of the antitumor activity of CAR-T cells in vivo, 6–8 week-old male NSG mice were inoculated subcutaneously (s.c.) into the right flank with 5 × 10⁶ Huh7 cells in 100 μL of 50:50 Matrigel (Corning) and PBS. On day 15, 1 × 10⁶ transduced CAR-T cells (in 200 μL of PBS) were adoptively transferred into mice via the tail vein, and mdivi-1 was administered to the mice via intraperitoneal injection at 20 mg/kg. The volume of tumors was measured by two-dimensional measurements (mm): Volume of tumor = length × width² / 2. Tissues were digested with collagenase type IV (2 mg/mL, Sigma) at 37 °C for 30 min, and tumor-infiltrating T cells were separated by centrifugation on a discontinuous Percoll gradient (Haoyang, China).

2.13 Immunofluorescence

Tumor tissue samples were fixed, processed, and stained in accordance with standard procedures. Briefly, the HCC tumor tissue was fixed with neutrally buffered 4% formaldehyde, and the following antibodies were used for automated staining on a BenchMark XT device: FITC anti-GGGGS (GS-ARFT100, Hycells), PE-conjugated goat anti-human Granzyme B (Cat No. 372208, BioLegend), and APC anti-CD39 (Cat No. 328209, BioLegend). DAPI (Thermo Fisher Scientific) was used for nucleus staining. Images of stained samples were obtained using a microscope (Leica, DM6000B). Fluorescent signals were detected using a laser scanning confocal microscope (ZEISS LSM 800).

2.14 Statistical analysis

Experiments were conducted independently at least three times. GraphPad Prism 7.0 (GraphPad Software, San Diego, CA, USA) was used to process and analyze the data, which are presented as mean ± standard error of the mean (SEM). Two-tailed Student’s t-test was adopted to compare two independent groups, while one-way ANOVA was used in Tukey’s multiple comparisons. Statistically significant differences were considered at P < 0.05.

3 Results

3.1 CD39^int CD8⁺ CAR-T cells exerted a potent antitumor activity in vivo

To investigate the impact of CD39 expression on CAR-T cells in vivo, we established a subcutaneous tumor model of HCC using GPC3⁺ Huh7 cells in NSG mice. Subsequently, we intravenously injected GPC3-CAR-T cells into the mice (Fig.1). The flow cytometry analysis revealed distinct subsets of CAR-T cells based on CD39 and PD-1 expression level, including CD39^hi PD-1^hi, CD39^int PD-1^int, CD39^int PD-1^-, CD39^- PD-1^-, and CD39^- PD-1^int populations. Here, hi, int and (-) indicated high, intermediate and negative subsets, respectively. Our results demonstrated that CD39^int CAR-T cells, specifically the CD39^int PD-1^int and CD39^int PD-1^hi subsets, exhibited higher levels of IFN-γ secretion (Fig.1 and 1C). These findings suggest that CD8⁺ CAR-T cells can maintain a potent antitumor function in vivo when CD39 is expressed at a moderate level.

3.2 CD39 knockdown compromised cytotoxicity and proliferation but reduced apoptosis in CAR-T cells in vitro

We hypothesized that CD39 knockdown in CAR-T cells could potentially decrease the terminal exhausted phenotype of CD39^hi subcluster and utilized shRNA to silence CD39 expression. The hsa-miR-106b cluster skeleton was used to place the shRNAs to knock down the expression of CD39, as previously described (Fig.2) [17,20]. The efficiency of CD39 knockdown in CAR-T cells (shCD39-CAR-T cells) was validated through qRT-PCR (Figure. S1A). Modified CAR-T cells and shCD39-CAR-T cells were separately cocultured with rhGPC3 in vitro (Fig.2). We observed a significant reduction in the frequency of CD39^hi CD8⁺ and CD39^int CD8⁺ T cells in the shCD39-CAR-T cells compared with that of the control (Fig.2). The expression of inhibitory receptors, such as PD-1, Tim-3, and Lag-3, was significantly decreased in the shCD39-CAR-T cells (Fig.2). Impaired cytokine secretion, proliferation, and cytotoxicity were observed in the cells (Fig.2, 2F, and S1B). Knocking down CD39 resulted in reduced apoptosis in CAR-T cells (Fig.2). These results suggest that CD39 downregulation leads to compromised cytotoxicity and proliferation in the early stages of CAR-T cell activity. Discovering and applying methods that effectively strengthen the function of shCD39-CAR-T cells are crucial.

3.3 Mdivi-1 enhanced the frequency of the CD39^intCD8⁺ subset and promoted cytokine secretion, proliferation, and cytotoxicity but induced exhaustion and apoptosis in CAR-T cells in vitro

Considering the potential impact of CD39 modulators on CAR-T cell function, we investigated the effects of three compounds: mdivi-1, CGS-21680, and 8-bromo-cAMP. These compounds are known to influence CD39 expression, mitochondrial dynamics, ATP production, and calcium homeostasis (Fig.3) [14–16]. We first assessed the toxicity of these compounds on GPC3-CAR-T and shCD39-GPC3-CAR-T cells in vitro and found that mdivi-1 displayed less toxicity than CGS-21680 and 8-bromo-cAMP (Fig.3 and S2A). Subsequently, we examined the correlation between drug concentration and the frequency of CD39^int shCD39-CAR-T cells. Mdivi-1 demonstrated a more potent ability to increase the frequency of CD39^int shCD39-CAR-T cells than the other compounds, with an optimal concentration of 50 µM (Fig.3 and S2B). These findings suggest that mdivi-1 exhibits superior efficacy and safety compared with the other compounds. Hence, we selected mdivi-1 for the following experiment and found that mdivi-1 reverted the CD39^- population into the CD39^int population and rather converted the CD39^hi subsets into the CD39^int population. The CD39^- and CD39^int CD8⁺ CAR-T cells exhibited significant increases in their CD39 fluorescence values after mdivi-1 treatment, with the mean fluorescence intensity respectively increasing from 2640 to 5559 and from 5469 to 8829. By contrast, no significant changes were observed in the CD39⁺ CAR-T cells (Fig.3). Further research revealed that mdivi-1 treatment significantly increased the frequency of CD39^int CAR-T cells (Fig.3 and 3F). Although the exhaustion levels of CAR-T cells were increased in the CAR-T cells treated with mdivi-1 (Fig.3), the INF-γ secretion and cell proliferation of the CAR-T cells were significantly enhanced (Fig.3). Mdivi-1 improved the cytotoxicity properties of CAR-T cells in the LDH experiment (Fig. S2C). However, it also increased the rate of apoptotic properties in the CAR-T cells (Fig.3). These results indicate a potential complementary relationship between mdivi-1 and CD39, offering possibilities for their combined use to enhance CAR-T cell function.

3.4 shCD39-CAR-T cells combined with mdivi-1 exhibited a synergistic effect and enhanced the antitumor activity in an HCC organoid model

To assess whether mdivi-1 could potentiate the antitumor activity of shCD39-CAR-T cells, we conducted an in vitro CAR-T cell killing assay using an HCC organoid model. HCC organoids were cocultured with CAR-T cells in the presence or absence of 50 µM mdivi-1 (Fig.4). After 24 h, a higher proportion of apoptotic organoids was observed in the group treated with shCD39-CAR-T cells combined with mdivi-1 than that in the group treated with CAR-T cells alone (Fig.4). The flow cytometry analysis of the cocultured CAR-T cells revealed a significantly higher frequency of the CD39^int subset with higher IFN-γ secretion in the shCD39-CAR-T cells combined with mdivi-1 than that in other groups (Fig.4). The CD39 knockdown markedly reduced the frequency of the CD39^hi subset, regardless of the presence of mdivi-1. The mdivi-1 treatment increased the frequency of the CD39^int subset in shCD39-CAR-T and CAR-T cell groups (Fig.4). The analysis of CAR-T cell cytotoxicity demonstrated that the addition of mdivi-1 significantly enhanced the antitumor activity of the CD39^int subset in CAR-T and shCD39-CAR-T cells (Fig.4). The evaluation of the proliferation capacity using Ki67 expression showed that mdivi-1 also augmented the proliferation of the CD39^int subset in CAR-T and shCD39-CAR-T cells (Fig.4). The combination of shCD39-CAR-T cells and mdivi-1 showed higher levels of IFN-γ secretion and proliferation than other groups.

3.5 shCD39-CAR-T cells combined with mdivi-1 exhibited a potent antitumor activity by increasing the proportion of infiltrated CD39^int CAR-T cells

To validate the role of mdivi-1 in enhancing the antitumor activity of shCD39-CAR-T cells in vivo, we constructed an HCC subcutaneous tumor model using the GPC3⁺ Huh7 cell line. The combination of shCD39-CAR-T cells and mdivi-1 demonstrated a potent antitumor activity compared with the other groups (Fig.5). Tumor growth was notably slower in terms of volume, and the tumors weighed less in the group treated with shCD39-CAR-T cells combined with mdivi-1 (Fig.5 and 5C). We observed a significant increase in the infiltration of CD39^int CAR-T cells into the tumors, whereas CD39^hi CAR-T cells were reduced in the shCD39-CAR-T cells combined with mdivi-1 (Fig.5). Moreover, the IFN-γ secretion by CD39^int CAR-T cells was increased in the shCD39-CAR-T cells combined with mdivi-1, whereas the IFN-γ production by CD39^hi CAR-T cells was not affected (Fig.5 and 5F).

Immunofluorescence staining assays confirmed these results, showing higher infiltration of CAR-T cells, particularly GZMB⁺ CAR-T cells, in tumors of shCD39-CAR-T combined with mdivi-1 (Fig.6 and 6B). Based on the enriched infiltration of CD39^int CAR-T cells with high IFN-γ production, the superior antitumor activity of the shCD39-CAR-T cells combined with mdivi-1 was due to the potent activity of CD39^int CAR-T cells. Collectively, our results demonstrate the critical role of CD39 expression in CAR-T cell function and highlight the potential therapeutic efficacy of combining mdivi-1 with CD39 knockdown in HCC. The combination approach yielded increased infiltration of CD39^int CAR-T cells and exhibited a robust antitumor activity. These findings provide insights into improving CAR-T cell therapy for HCC through the modulation of CD39 expression and offer a promising avenue for cellular immunotherapy in HCC treatment.

4 Discussion

The therapeutic potential of CAR-T cell therapy for HCC holds significant promise but has yet to achieve the desired outcomes. Overcoming the limitations of CAR-T cell therapy in solid tumors necessitates a thorough understanding of the factors influencing CAR-T cell function. In this study, we focused on investigating the role of CD39 expression in CAR-T cell function and explored the therapeutic potential of CD39 modulators, such as mdivi-1, or knockdown of endogenous CD39 using shRNA.

Our findings demonstrated that CAR-T cells with moderate expression of CD39 exhibited a strong antitumor activity against HCC, whereas high or low CD39 expression levels led to an impaired cellular function. These findings support previous studies that have underscored the complex dual role of CD39 as a marker of antitumor activity and T cell exhaustion [10,21,22]. The balance between CD39 expression and CAR-T cell function appears to be critical for achieving optimal therapeutic outcomes [23]. The moderate expression of CD39 in CAR-T cells may allow for proper regulation of ATP levels, optimizing T cell function and maintaining a balance between antitumor activity and exhaustion. Additionally, CD39^int CAR-T cells may exhibit optimized signaling and metabolic pathways, leading to enhanced functionality. Moreover, our results revealed that modulating the proportion of CD39^int CAR-T cells through the use of mdivi-1 or CD39 knockdown exerted significant effects on T cell function.

Mdivi-1, a small chemical inhibitor of mitochondrial division dynamin, has emerged as a potential therapeutic agent for neurodegeneration and the enhancement of CAR-T cell function [24,25]. Treatment with mdivi-1 increased the frequency of CD39^int CAR-T cells, demonstrating improved cytokine secretion, proliferation, and cytotoxicity. Although mdivi-1 also increased the rate of apoptosis in CAR-T cells, the overall enhancement of CAR-T cell function suggests a potential complementary relationship between mdivi-1 and CD39. This finding opens up avenues for leveraging the synergistic effects of mdivi-1 and CD39 knockdown to further enhance CAR-T cell activity.

By employing an HCC organoid model, we illustrated that the combination of shCD39-CAR-T cells and mdivi-1 led to amplified antitumor effects. The combination approach significantly increased the infiltration of CD39^int CAR-T cells into the organoids, which correlated with the increased production of IFN-γ and the improved tumor cell killing. Mdivi-1 bolstered the proliferation of CD39int CAR-T cells, further boosting their therapeutic prospects. These findings highlight the synergy between CD39 knockdown and mdivi-1 treatment in improving CAR-T cell function and support the potential application of this combination strategy for HCC treatment.

In vivo experiments using an HCC subcutaneous tumor model provided further evidence of the potent antitumor activity of shCD39-CAR-T cells combined with mdivi-1. The combination therapy resulted in slower tumor growth and reduced tumor weight compared with other treatment groups. Importantly, we observed increased infiltration of CD39^int CAR-T cells into the tumor microenvironment, accompanied with heightened IFN-γ secretion. Immunofluorescence staining confirmed the enhanced infiltration of CAR-T cells, particularly those expressing GZMB, in the tumors of the combination therapy group. These findings highlight the crucial role of CD39^int CAR-T cells in mediating the antitumor response in an in vivo setting.

Overall, our results underscore the critical role of CD39 expression in CAR-T cell function against HCC. We have demonstrated that modulating CD39 expression through mdivi-1 treatment and CD39 knockdown provides a means to enhance CAR-T cell activity and improve therapeutic outcomes. These findings offer valuable insights into the development of novel strategies for CAR-T cell immunotherapy in HCC. Further investigations are warranted to validate the safety and efficacy of shCD39-CAR-T cells and mdivi-1 prior to their clinical translation.

References

[1] Yang JD, Hainaut P, Gores GJ, Amadou A, Plymoth A, Roberts LR. A global view of hepatocellular carcinoma: trends, risk, prevention and management. Nat Rev Gastroenterol Hepatol 2019; 16(10): 589–604

[2] Llovet JM, Castet F, Heikenwalder M, Maini MK, Mazzaferro V, Pinato DJ, Pikarsky E, Zhu AX, Finn RS. Immunotherapies for hepatocellular carcinoma. Nat Rev Clin Oncol 2022; 19(3): 151–172

[3] Zhou F, Shang W, Yu X, Tian J. Glypican-3: a promising biomarker for hepatocellular carcinoma diagnosis and treatment. Med Res Rev 2018; 38(2): 741–767

[4] Taniguchi M, Mizuno S, Yoshikawa T, Fujinami N, Sugimoto M, Kobayashi S, Takahashi S, Konishi M, Gotohda N, Nakatsura T. Peptide vaccine as an adjuvant therapy for glypican-3-positive hepatocellular carcinoma induces peptide-specific CTLs and improves long prognosis. Cancer Sci 2020; 111(8): 2747–2759

[5] Sawada Y, Yoshikawa T, Nobuoka D, Shirakawa H, Kuronuma T, Motomura Y, Mizuno S, Ishii H, Nakachi K, Konishi M, Nakagohri T, Takahashi S, Gotohda N, Takayama T, Yamao K, Uesaka K, Furuse J, Kinoshita T, Nakatsura T. Phase I trial of a glypican-3-derived peptide vaccine for advanced hepatocellular carcinoma: immunologic evidence and potential for improving overall survival. Clin Cancer Res 2012; 18(13): 3686–3696

[6] Pang N, Shi J, Qin L, Chen A, Tang Y, Yang H, Huang Y, Wu Q, Li X, He B, Li T, Liang B, Zhang J, Cao B, Liu M, Feng Y, Ye X, Chen X, Wang L, Tian Y, Li H, Li J, Hu H, He J, Hu Y, Zhi C, Tang Z, Gong Y, Xu F, Xu L, Fan W, Zhao M, Chen D, Lian H, Yang L, Li P, Zhang Z. IL-7 and CCL19-secreting CAR-T cell therapy for tumors with positive glypican-3 or mesothelin. J Hematol Oncol 2021; 14(1): 118

[7] Mizukoshi E, Kaneko S. Immune cell therapy for hepatocellular carcinoma. J Hematol Oncol 2019; 12(1): 52

[8] Allard B, Longhi MS, Robson SC, Stagg J. The ectonucleotidases CD39 and CD73: novel checkpoint inhibitor targets. Immunol Rev 2017; 276(1): 121–144

[9] Krishna S, Lowery FJ, Copeland AR, Bahadiroglu E, Mukherjee R, Jia L, Anibal JT, Sachs A, Adebola SO, Gurusamy D, Yu Z, Hill V, Gartner JJ, Li YF, Parkhurst M, Paria B, Kvistborg P, Kelly MC, Goff SL, Altan-Bonnet G, Robbins PF, Rosenberg SA. Stem-like CD8 T cells mediate response of adoptive cell immunotherapy against human cancer. Science 2020; 370(6522): 1328–1334

[10] Liu T, Tan J, Wu M, Fan W, Wei J, Zhu B, Guo J, Wang S, Zhou P, Zhang H, Shi L, Li J. High-affinity neoantigens correlate with better prognosis and trigger potent antihepatocellular carcinoma (HCC) activity by activating CD39⁺CD8⁺ T cells. Gut 2021; 70(10): 1965–1977

[11] Gupta PK, Godec J, Wolski D, Adland E, Yates K, Pauken KE, Cosgrove C, Ledderose C, Junger WG, Robson SC, Wherry EJ, Alter G, Goulder PJ, Klenerman P, Sharpe AH, Lauer GM, Haining WN. CD39 expression identifies terminally exhausted CD8⁺ T cells. PLoS Pathog 2015; 11(10): e1005177

[12] Chow A, Uddin FZ, Liu M, Dobrin A, Nabet BY, Mangarin L, Lavin Y, Rizvi H, Tischfield SE, Quintanal-Villalonga A, Chan JM, Shah N, Allaj V, Manoj P, Mattar M, Meneses M, Landau R, Ward M, Kulick A, Kwong C, Wierzbicki M, Yavner J, Egger J, Chavan SS, Farillas A, Holland A, Sridhar H, Ciampricotti M, Hirschhorn D, Guan X, Richards AL, Heller G, Mansilla-Soto J, Sadelain M, Klebanoff CA, Hellmann MD, Sen T, de Stanchina E, Wolchok JD, Merghoub T, Rudin CM. The ectonucleotidase CD39 identifies tumor-reactive CD8⁺ T cells predictive of immune checkpoint blockade efficacy in human lung cancer. Immunity 2023; 56(1): 93–106.e6

[13] Zou F, Tan J, Liu T, Liu B, Tang Y, Zhang H, Li J. The CD39⁺ HBV surface protein-targeted CAR-T and personalized tumor-reactive CD8⁺ T cells exhibit potent anti-HCC activity. Mol Ther 2021; 29(5): 1794–1807

[14] Cui M, Ding H, Chen F, Zhao Y, Yang Q, Dong Q. Mdivi-1 protects against ischemic brain injury via elevating extracellular adenosine in a cAMP/CREB-CD39-dependent manner. Mol Neurobiol 2016; 53(1): 240–253

[15] Bao R, Shui X, Hou J, Li J, Deng X, Zhu X, Yang T. Adenosine and the adenosine A2A receptor agonist, CGS21680, upregulate CD39 and CD73 expression through E2F-1 and CREB in regulatory T cells isolated from septic mice. Int J Mol Med 2016; 38(3): 969–975

[16] Liao H, Hyman MC, Baek AE, Fukase K, Pinsky DJ. cAMP/CREB-mediated transcriptional regulation of ectonucleoside triphosphate diphosphohydrolase 1 (CD39) expression. J Biol Chem 2010; 285(19): 14791–14805

[17] Zou F, Lu L, Liu J, Xia B, Zhang W, Hu Q, Liu W, Zhang Y, Lin Y, Jing S, Huang M, Huang B, Liu B, Zhang H. Engineered triple inhibitory receptor resistance improves anti-tumor CAR-T cell performance via CD56. Nat Commun 2019; 10(1): 4109

[18] Broutier L, Mastrogiovanni G, Verstegen MM, Francies HE, Gavarró LM, Bradshaw CR, Allen GE, Arnes-Benito R, Sidorova O, Gaspersz MP, Georgakopoulos N, Koo BK, Dietmann S, Davies SE, Praseedom RK, Lieshout R, IJzermans JNM, Wigmore SJ, Saeb-Parsy K, Garnett MJ, van der Laan LJ, Huch M. Human primary liver cancer-derived organoid cultures for disease modeling and drug screening. Nat Med 2017; 23(12): 1424–1435

[19] Nuciforo S, Fofana I, Matter MS, Blumer T, Calabrese D, Boldanova T, Piscuoglio S, Wieland S, Ringnalda F, Schwank G, Terracciano LM, Ng CKY, Heim MH. Organoid models of human liver cancers derived from tumor needle biopsies. Cell Rep 2018; 24(5): 1363–1376

[20] Qiao Y, Chen J, Wang X, Yan S, Tan J, Xia B, Chen Y, Lin K, Zou F, Liu B, He X, Zhang Y, Zhang X, Zhang H, Wu X, Lu L. Enhancement of CAR-T cell activity against cholangiocarcinoma by simultaneous knockdown of six inhibitory membrane proteins. Cancer Commun (Lond) 2023; 43(7): 788–807

[21] Simoni Y, Becht E, Fehlings M, Loh CY, Koo SL, Teng KWW, Yeong JPS, Nahar R, Zhang T, Kared H, Duan K, Ang N, Poidinger M, Lee YY, Larbi A, Khng AJ, Tan E, Fu C, Mathew R, Teo M, Lim WT, Toh CK, Ong BH, Koh T, Hillmer AM, Takano A, Lim TKH, Tan EH, Zhai W, Tan DSW, Tan IB, Newell EW. Bystander CD8⁺ T cells are abundant and phenotypically distinct in human tumour infiltrates. Nature 2018; 557(7706): 575–579

[22] Qiao M, Zhou F, Liu X, Jiang T, Wang H, Jia Y, Li X, Zhao C, Cheng L, Chen X, Ren S, Liu H, Zhou C. Interleukin-10 induces expression of CD39 on CD8⁺ T cells to potentiate anti-PD1 efficacy in EGFR-mutated non-small cell lung cancer. J Immunother Cancer 2022; 10(12): e005436

[23] Oliveira G, Stromhaug K, Klaeger S, Kula T, Frederick DT, Le PM, Forman J, Huang T, Li S, Zhang W, Xu Q, Cieri N, Clauser KR, Shukla SA, Neuberg D, Justesen S, MacBeath G, Carr SA, Fritsch EF, Hacohen N, Sade-Feldman M, Livak KJ, Boland GM, Ott PA, Keskin DB, Wu CJ. Phenotype, specificity and avidity of antitumour CD8⁺ T cells in melanoma. Nature 2021; 596(7870): 119–125

[24] Xie N, Wang C, Lian Y, Wu C, Zhang H, Zhang Q. Inhibition of mitochondrial fission attenuates Aβ-induced microglia apoptosis. Neuroscience 2014; 256: 36–42

[25] Fan LF, He PY, Peng YC, Du QH, Ma YJ, Jin JX, Xu HZ, Li JR, Wang ZJ, Cao SL, Li T, Yan F, Gu C, Wang L, Chen G. Mdivi-1 ameliorates early brain injury after subarachnoid hemorrhage via the suppression of inflammation-related blood-brain barrier disruption and endoplasmic reticulum stress-based apoptosis. Free Radic Biol Med 2017; 112: 336–349

RIGHTS & PERMISSIONS

Higher Education Press