WHAT IS ALREADY KNOWN ON THIS TOPIC

Several studies have shown that lactate can enhance the immunosuppressive function of Treg cells. Additionally, high expression of TNFR2 is closely related to the immunosuppressive function of Treg cells. However, the relationship between lactate and TNFR2 expression remains unclear.

WHAT THIS STUDY ADDS

Lactate upregulated TNFR2 expression on Treg cells and enhanced their immunosuppressive function in Malignant pleural effusion. Lactate modulated the gene transcription of transcription factor nuclear factor-κB p65 (NF-κB p65) through histone H3K18 lactylation (H3K18la). Lactate metabolism blockade combined with immune checkpoint blockade (ICB) therapy effectively enhanced the efficacy of ICB therapy.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICYBackground

Treg cells are crucial for maintaining immune homeostasis, promoting self-tolerance and immunosuppression, making them a promising target for cancer immunotherapy. However, the prevalence of immunosuppressive tumor-infiltrating Treg cells poses a challenge to immune checkpoint blockade (ICB) therapy, contributing to drug resistance.1 2 Therefore, the exploration of strategies to target Treg cells alone or in combination with ICB therapy becomes attractive for cancer immunology. Numerous literature reviews therapeutic strategies that target immunosuppressive molecules by Treg cells (such as anti-CD25, anti-cytotoxic lymphocyte-associated antigen 4 (CTLA-4), GITR agonists, and anti-OX40) and block chemokine/chemokine receptors (such as anti-TGFβ, anti-CCR4, and anti-CCR8).3–6 Nevertheless, there are still significant obstacles to the clinical application of Treg cell targeted therapy such as severe autoimmunity resulting from clearance of Treg cells.7 Thus, it is imperative to identify potential targets on Treg cells. TNFR2, as a member of the tumor necrosis factor receptor superfamily (TNFRSF), modulates the activation and functionality of immune cells through interacting with the ligand tumor necrosis factor (TNF), promoting cell survival, proliferation, differentiation, and immune regulation.8 The abnormally elevated TNFR2+Treg subsets in various tumors contribute to an immunosuppressive tumor microenvironment (TME), weaken the antitumor immune response, and facilitate immune evasion.9 Particularly, our previous research has indicated that the aberrant accumulation of TNFR2+ Treg cells in Malignant pleural effusion (MPE) is closely associated with rapid tumor growth and continued progression of MPE.10 Blocking TNFR2 selectively eliminates Treg cells in the TME, offering a highly promising avenue for tumor immunotherapy.11

MPE is typically attributed to the infiltration of the pleura by malignant pleural mesothelioma, lung cancer, or other malignancies.12 It commonly occurs in advanced-stage non-small cell lung cancer (NSCLC), with limited effective treatments.13 14 As a typical “cold tumor”, MPE is characterized by an accumulation of immunosuppressive cell populations, particularly Treg cells.15 Therefore, targeting immunosuppressive cells in MPE holds great promise for significantly enhancing the antitumor immune response and improving the prognosis of patients with MPE. Meanwhile, MPE represents a persistently hypoxic environment,16 in which tumor cells exhibit enhanced glycolysis, leading to the generation of significant amounts of glycolytic byproduct lactate based on the Warburg effect.16 17 Lactate, an important metabolite in MPE, has been identified as a key regulator of tumorigenesis, maintenance and progression. Numerous studies have reported a strong association between lactate and the immunosuppressive function of Treg cells.18–21 Furthermore, Treg cells can uptake lactate from the TME through a lactate transporter monocarboxylate transporter 1 (MCT1)20 as an energy source, thereby promoting their immunosuppressive function.

The metabolic reprogramming of Treg cells within the TME has gained significant attention in recent years.22–25 Under the action of acetyltransferase p300, lactate facilitates histone lactylation modification (Kla) at the amino terminus of lysine residues on histones, thereby regulating gene transcription.26–31 The previous studies have documented that histone lactylation exerts a regulatory function in the transcription of genes associated with tumorigenesis, such as YTHDF2,32 thereby facilitating tumor progression, metastasis, and invasion. Additionally, tumor-derived lactate has been shown to induce METTL3 (RNA methyltransferase 3) expression in tumor-infiltrating myeloid cells (TIMs) through histone Kla, promoting the immunosuppressor function of TIMs and mediating tumor immune escape.33 Consequently, it is of paramount importance to investigate the role and regulatory mechanism of lactate modification in cancer development for a more comprehensive understanding of the immunosuppressive TME and its clinical implications in immunotherapy.

In this study, we investigated the impact of lactate on the expression of TNFR2 on Treg cells and their immunosuppressive function in MPE. Our results indicated that Treg cells with elevated TNFR2 expression displayed heightened levels of immunosuppressive molecules and demonstrated enhanced immunosuppressive functionality. Additionally, elevated levels of TNFR2 expression in MPE patients were associated with an unfavorable prognosis. Within the context of MPE, Treg cells actively absorbed lactate from the TME. Simultaneously, they produced a modest amount of lactate through glycolysis. The presence of lactate in turn modulated the gene transcription of NF-κB p65 through histone H3K18la. This process upregulated the expression of TNFR2 and immunosuppressive molecules on Treg cells, thereby enhancing their immunosuppressive function on CD8+T cells and promoting the progression of MPE. What’s more, AZD3965 or Oxamate effectively decreased histone lactylation levels and suppressed TNFR2 expression on Treg cells. Combining AZD3965 or Oxamate with an anti-PD-1 antibody revealed a superior antitumor effect compared with using an anti-PD-1 antibody alone in the MPE mouse model of NSCLC. This discovery highlights the molecular mechanism underlying lactylation-mediated TNFR2 expression and suggests the promising potential of metabolic immunotherapy in the clinical treatment of MPE.

Materials and methodsCell lines

Lewis lung carcinoma (LLC) cells (ATCC, CRL-1642), LLC-LUC (ATCC, CRL-1642-LUC2) were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (FBS). A549 cells (ATCC, CRL-CCL-185). All cells were incubated at 37℃ in 5% CO2 and were regularly examined for mycoplasma contamination using the GMyc-PCR Mycoplasma Test Kit (40 601ES10, Yeasen).

Patients and samples

In this study, pleural fluid samples and peripheral blood (PB) were collected from patients diagnosed lung adenocarcinoma patients with MPE at the Department of Respiratory and Critical Care Medicine of the Union Hospital of Tong Medical College (Wuhan, China) between January 2021 and January 2022. The diagnosis of MPE was confirmed with the demonstration of malignant cells in pleural fluid (cytopathological test), closed pleural biopsy, or employing a combination of both diagnostic methods. None of the patients had undergone any anticancer therapy, corticosteroids, or other non-steroidal anti-inflammatory drugs at the time of sample collection. The obtained MPE and PB specimens were promptly placed on ice and then subjected to centrifugation at 1500 r for 6 min. The serum cell-free supernatants were promptly frozen at −80℃ following centrifugation for subsequent lactate concentration determination. The cell pellets were reconstituted in 1×phosphate-buffered saline (PBS), pleural effusion mononuclear cells (PEMCs) and peripheral blood mononuclear cells (PBMCs) were isolated through Ficoll-Hypaque gradient centrifugation (Stem Cell Technologies, Vancouver, British Columbia, Canada). The clinicopathological data is presented in online supplemental table 1.

Cell isolation and culture

The CD4+CD25+FOXP3+ Treg cells were purified from PEMCs with a human CD4+CD25+ Regulatory T Cell Isolation Kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s instructions.TNFR2high Treg cells (the ratio of TNFR2 on CD4+CD25+FOXP3+Treg cells ≥60%) and TNFR2low Treg cells (TNFR2<60%) were isolated using fluorescence-activated cell sorting (FACS) by BD FACSAria II Cell Sorter. CD8+ T cells were purified using human CD8+ T-cell isolation kit (STEMCELL). They were cultured in RPMI medium supplemented with 10% FBS. A comprehensive list of all antibodies and reagents can be found in online supplemental table S1.

Co-culture of Treg cells and tumor cells

Treg cells (1×107) were isolated and co-cultured with A549 (5×106) cells at 37℃, 5% CO2 condition.

Flow cytometry

Cells were resuspended in a staining buffer and subjected to staining with surface antibodies and a fixable viability dye (Thermo Fisher Scientific) at room temperature (RT) for 30 min in the dark. After washing, the cells were fixed and permeabilized using Fixation/Permeabilization Concentrate and Fixation/Perm Diluent (eBioscience) for 30 min at RT in the dark according to the instructions provided by the manufacturer. Subsequently, they were washed twice with diluted 1×Permeabilization Buffer. Then, intracellular staining was performed with intracellular antibodies at RT for 30 min. Finally, Fixation Buffer was employed to fix the cells, and they were subsequently analyzed using a flow cytometer (BD Biosciences). Data were analyzed using FlowJo software (BD Biosciences). The antibodies used in the FCM analyses were summarized in key resources (online supplemental table S1).

T cell cytokine production experiments

To detect cytokines secreted by CD8+ T cells, a cell activation cocktail (BioLegend) was applied to stimulate the cells for 4 hours. Subsequently, the staining procedure described above was performed.

The levels of interleukin-10 (IL-10) and transforming growth factor β (TGF-β) of Treg cells were assessed by flow cytometry.

ELISA

The concentrations of IL-10 and TGF-β in cell culture fluid were measured with an ELISA kit (R&D Systems, USA), according to the manufacturer’s protocol. All samples were repeated three times in independent replicates.

Real-time quantitative PCR experiments/real-time quantitative PCR

Total RNA of Treg cells was extracted using Trizol reagent (Invitrogen, Cat# 15596018) according to the manufacturer’s protocol. Extracted RNA was transcribed into cDNA using the HiScript quantitative real-time PCR (qRT-PCR) SuperMix (Vazyme, Nanjing, China), according to the manufacturer’s protocol. The AceQ qPCR SYBR Green Master Mix (Vazyme) and gene-specific primers were used for PCR amplification and detection on ABI 7500 real-time PCR system. The expression level of GAPDH was employed for normalization. The 2−ΔΔCt method of analysis was used to determine the relative gene expression levels. The RT-qPCR primers are listed in online supplemental table S3.

Suppressive assay with human T cells

CD8+ T cells were sorted from PEMCs using a CD8+ T cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany). TNFR2high and TNFR2low Treg cells were sorted from PEMCs using FACS. Afterward, CD8+ T cells were labeled with carboxyfluorescein succinimidyl ester (CFSE) (Invitrogen, Carlsbad, California, USA) and stimulated with 2 ug/mL anti-CD3 (eBioscience) and 5 ug/mL anti-CD28 (eBioscience) antibodies. Then, CD8+ T cells were co-cultured with TNFR2high Treg cells, TNFR2low Treg cells, or LA-treated Treg cells at 37℃, 5% CO2. After 4 days, proliferation was assessed by dilution of CFSE-labeled cells with FCM. Cells were stained with anti-CD8 PE-Cy7 and tested through flow cytometry.

Measurement of lactate concentration

The supernatant was obtained by centrifugation from clinical specimens at 1500 r for 6 min. The concentrations of lactate were assessed with a Lactate assay kit (Lactate Colorimetric/Fluorometric Assay Kit Abcam Cat# ab65330). The enzyme working solution and chromogenic agent were added to the incubation supernatant or cell culture supernatant. After adding the Terminator, the absorbance was measured at the wavelength of 530 nm on the enzyme labeling instrument. The concentration of lactate in the supernatant was calculated according to the absorbance.

Western blotting analysis

The CD4+CD25+FOXP3+ Treg cells were lysed in a radioimmunoprecipitation assay buffer with protease inhibitors and phosphatase inhibitor cocktails on ice for 30 min. The supernatant (containing Treg cells protein) was collected after centrifugation. The protein concentration in the extracts was analyzed by BCA Protein Assay Kit (Beyotime Biotechnology, Shanghai, China). Equal amounts of proteins were separated using Sodium dodecyl sulfate-Polyacrylamide Gel Electrophoresis and then transferred to a polyvinylidene difluoride membrane (Millipore) on ice. The membranes were blocked with 5% non-fat milk in Tris-buffered saline containing 0.05% Tween-20 (TBST) for 1 hour at RT and were incubated with indicated antibodies overnight at 4°C. After being washed three times with TBST at RT for 5 min each time, they were incubated with secondary antibody at RT for 1 hour. Finally, chemiluminescent substrate solution (Vazyme, Nanjing, China) was added and the membranes were imaged on UVP chemiluminescence imaging system (Upland, California, USA). The antibodies used for western blotting were listed in key resources (online supplemental table S1).

Treg differentiation

The naive CD4+ T cells (CD45RA+ CD45RO–) were isolated from PBMCs of MPE patients by Naive CD4+ T Cell Isolation Kit II, human (Miltenyi Biotec, Bergisch Gladbach, Germany) following the manufacturer’s instructions. The purity of naive CD4+ T cells was more than 90%, as measured by flow cytometry.

Purified naive CD4+ T cells (5×105) were cultured in 1 mL complete medium containing human IL-2 (2 ng/mL) in 48-well plates and stimulated with plate-bound anti-CD3 (5 ug/mL) and soluble anti-CD28 mAbs (2 ug/mL) for 7 days. The exogenous cytokines used were TGF-β (2 ng/mL). Recombinant human IL-2 and TGF-β were purchased from R&D Systems.

Mouse MPE models

We established a mouse model of lung cancer with MPE by inoculating 2×105 LLC-LUC in 100 mL of PBS into the thoracic cavity of C57BL/6J mice. The experimental mice were all sex-matched (4–5 weeks old, 16–18 g). Bioluminescence imaging was performed to ensure that MPE models were established successfully and uniformly. Subsequently, on days 6, 9, 12, and 15, we administered intrapleural injections of the following treatments: PBS, AZD3965 (100 mg/kg), anti-PD-1 monoclonal antibody (10 mg/kg), and a combination of AZD3965 and anti-PD-1 monoclonal antibody for treatment. Similarly, mice were randomly divided into four groups (n=6 mice/group) and treated with PBS, Oxamate (750 mg/kg), anti-PD-1mAb (10 mg/kg), or Oxamate and anti-PD-1mAb groups, respectively. Oxamate was intraperitoneally injected at 750 mg/kg on days 6, 9, 12, and 15. The MPE was collected for analysis on day 14 and survival status of the mice was observed daily until a fatal outcome.

Bioluminescence imaging with the Bruker system

For the bioluminescence experiment, MPE mice were first anesthetized with 0.5% pentobarbital sodium. Whole-body bioluminescence imaging was performed using the Bruker In Vivo FX PRO Imager 15 min after intraperitoneally injected with D-luciferin (150 mg/kg in PBS). Bioluminescent images were obtained with an exposure time of 3 min, and reflectance images were obtained with an exposure time of 0.175 s. Total photon flux was analyzed using Bruker MI SE software.

Chromatin immunoprecipitation assay

Chromatin immunoprecipitation (ChIP) assay was performed on CD4+CD25+FOXP3+Treg cells by SeqHealth (Wuhan, China). The cells were fixed in 1% formaldehyde for 10 min at RT, after which 0.125 M glycine was added and the mixture was sat for 5 min to terminate the crosslinking reaction. The tissue was then collected and frozen in liquid nitrogen. The grinded cells were treated with cell lysis buffer and nucleus were collected by centrifuging at 2000 g for 5 min. Then, nucleus was treated with nucleus lysis buffer and sonicated to fragment chromatin DNA. The 10% lysis sonicated chromatin was stored and named “input”, and 80% was used in immunoprecipitation reactions with anti-H3K18la antibody (PTM Bio) and named “IP”, and 10% was incubated with rabbit IgG (Cell Signaling Technology) as a negative control and named “IgG”, respectively. Fold enrichment was quantified using qRT-PCR and calculated as a percentage of input chromatin (percentage of input). The sites of NF-kBp65 binds to promoters of TNFR2 were predicted by JASPAR (http://jaspar.genereg.net). The primers used for ChIP PCR are shown in online supplemental table S2.

Chip-seq analysis

Raw sequencing data was first processed by Trimmomatic (V.0.36) to filter out low-quality reads. Subsequently, the clean reads were used for protein binding site analysis. They were mapped to the reference genome of human from NCBI using STAR software (V.2.5.3a) with default parameters. Following the alignment, the RSeQC (V.2.6）was used for reads distribution analysis. The peak calling was performed using the MACS2 software (V.2.1.1). The identified peaks were then labeled and subjected to peak distribution analysis using bedtools (V.2.25.0).

Cell viability assay

Treg cells were cultured with different concentrations of lactate. After 24 hours treatment, cell viability was evaluated by CCK8 assay (Biosharp Cat# BS350A) as described above. After incubation with CCK8 (1:10) for 2 hour, cell counts were determined by measuring the absorbance at 450 nm with a Microplate reader (SpectraMaxiD5, Molecular Devices, USA). At least three duplicate wells were used for each sample, and the experiments were conducted independently in triplicate.

Single cell RNA-seq analysis

Single-cell RNA sequencing (scRNA-seq) analysis of regulatory T (Treg) cells was conducted using our previously published MPE scRNA-seq data (PRJNA970083) along with publicly available datasets from the GEO database (GSE1311907, GSE99254, GSE146771). Low-quality cells were excluded based on criteria established in a prior study. Primary analyses were performed using the Seurat R package (V.4.2.0). Data normalization was achieved using a scaling factor of 10,000, and the top 2000 variable genes were identified using the FindVariableFeatures function for subsequent principal component analysis. The first 30 principal components were employed in the FindNeighbors algorithm, and cell clusters were determined using the FindClusters function. Visualization of identified clusters was accomplished using the t-distributed stochastic neighbor embedding method. Differentially expressed genes within clusters were identified using the FindAllMarkers function. Enrichment analysis was conducted using the clusterProfiler package (V.4.2.0), and gene set variation analysis (GSVA) package (V.1.42.0), using gene sets from the MSigDB database.

Statistical analysis

GraphPad Prism V.8 software was used for statistical analyses. The relations between groups were compared using unpaired Student’s t-test, one-way analysis of variance; test or Pearson correlation coefficient. Survival was analyzed with the Kaplan-Meier method and was compared with the log-rank test. P values <0.05 were considered statistically significant. (ns: not significant, *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001).

ResultsTNFR2high Treg cells exhibit stronger immunosuppressive functions in MPE

Our previous study has suggested that TNFR2-expressing Tregs abnormally accumulate in MPE, sustaining the immunosuppressive activity of MPE.10 The high frequency of TNFR2+ Tregs in MPE is closely associated with a greater number of tumor cells, larger MPE volume, and unfavorable prognosis of patients.10 Therefore, we confirmed significant variations in TNFR2 levels on Treg cells among different MPE cohorts. Based on this observation, we have categorized the cohort into two distinct groups: TNFR2high and TNFR2low Treg cells for further investigation (figure 1A,B). Besides, patients with TNFR2high Treg cells exhibited poorer survival outcomes (figure 1C). In order to investigate the characteristics of Treg cells expressing high levels of TNFR2, we conducted an analysis of the previous MPE scRNA-seq results to explore the gene expression atlas of TNFR2+ and TNFR2− Treg cells within the same patient. Based on gene signatures, a subset with the high expression of CD4, CD25 (IL2RA), and FOXP3 was identified as Treg cells (figure 1D). Gene set enrichment analysis revealed significant enrichment of gene sets associated with immune inhibitory molecules and Treg activity markers in TNFR2+ Treg cells compared with TNFR2− Treg (figure 1E,F). In addition, we identified four immune checkpoint molecules which were significantly differentially expressed between TNFR2+ Treg cells and TNFR2− Treg cells by integrating RNA-seq data analysis from MPE and public databases of NSCLC (GSE1311907, GSE99254, GSE146771), including glucocorticoid-induced tumor necrosis factor receptor family-related gene (GITR, also known as TNFRSF18), Inducible T-cell Costimulatory (ICOS), cytotoxic lymphocyte-associated antigen 4 (CTLA-4), and T-cell immunoreceptor with Ig and immunoreceptor tyrosine-based inhibition motif domains (TIGIT) (figure 1G,H, online supplemental figure 1 A–C). The upregulation of these immune suppressive molecules is closely associated with the immunosuppressive function of Treg cells, thereby promoting tumor progression, invasion, and metastasis.34 35 The high expression of GITR has been believed to inhibit the suppressive function of Treg cells and is associated with ICB resistance.34 36–38 Additionally, the expression of ICOS is intricately linked to the activation of Treg cells and plays a crucial role in maintaining the viability and functional effectiveness of effector Treg cells.34 36 CTLA4 also plays a role in Treg-mediated suppression.34 39 Similar to other coinhibitory molecules, TIGIT is highly expressed on Treg cells and competitively inhibits the binding of CD226 to CD112 and CD155, resulting in reduced activation and proliferation of effector T cells.35 Therefore, we speculated that TNFR2+ Treg cells exhibited strong immunosuppressive functionality.

Figure 1

TNFR2high Treg cells exhibit stronger immunosuppressive functions in MPE (A, B) Treg cells from different MPE patients with lung cancer were stratified into two subgroups: TNFR2high Treg cells (TNFR2 positivity for ≥60% of Treg cells, n=19) or TNFR2low Treg cells (TNFR2 positivity for <60% of Treg cells, n=20). The expression level of TNFR2 on Tregs in MPE from patients with lung cancer was measured using flow cytometry Representative plots (A) and a representative histogram (B) are shown. (C) Correlation between TNFR2high Treg cells and survival time in MPE was analyzed using Kaplan-Meier analysis (n=18–22). The median survival time of TNFR2high group was 9 months, while the median survival time of TNFR2low group was 17 months. (D) MPE specimens of untreated MPE patients with lung cancer were collected. CD45+ immune cells were isolated using magnetic activated cell sorting for scRNA-seq. A t-distributed stochastic neighbor embedding (t-SNE) plot based on scRNA-seq clustered the CD4+CD25+FOXP3+ Treg cells. (E, F) Treg cells from the same MPE patients with lung cancer were stratified into two subgroups: TNFR2+ Treg cells or TNFR2- Treg cells. Differentially expressed genes were identified between TNFR2+ and TNFR2−Treg cells by scRNA-seq data of MPE (E). including inhibitory markers and Treg activity markers were shown (F). (G, H) Four immune checkpoint molecules were identified to be significantly different in MPE and NSCLC, including GITR, ICOS, CTLA4 and TIGIT. A Venn diagram was shown by integrating RNA-seq data from MPE and public databases of NSCLC (GSE1311907, GSE99254, GSE146771) (G). Violin plot analyzed the differential genes enriched between TNFR2+ and TNFR2−Treg cells in MPE (H). (I) The relative mRNA expression levels of immune checkpoint molecules between TNFR2high and TNFR2low Treg cells were measured by q-PCR assay. (J, K) The CD4+CD25+FOXP3+ Treg cells were isolated from human MPE. The immune checkpoint molecules of TNFR2high and TNFR2low Treg cells were measured by flow cytometry. Representative plots (J) and representative histograms (K) are shown. (L, M) Suppression assays with human CD8+ T cells were performed (L). TNFR2high and TNFR2low Treg cells were isolated from corresponding human MPE by magnetic-activated cell sorting (MACS), CD8+ T cells were isolated from peripheral blood of healthy donors by MACS. CFSE assay was performed to detect the proliferation of CD8+ T cells after co-cultured with Treg cells. The tumor-killing activities (INF-γ, TNF-α, granzyme B and perforin) of CD8+ T cells were analyzed by flow cytometry after co-culturing with Treg cells for 48 hours (M). Statistical analysis was performed using unpaired two-tailed Student’s t-test (B, I, J, M). Data represent mean±SD of independent experiments (n=3 biological replicates). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. CFSE, carboxyfluorescein succinimidyl ester; MPE, malignant pleural effusion; ns, not significant; NSCLC, non-small cell lung cancer; TNFR2, tumor necrosis factor receptor type II.

To validate the relationship between the expression levels of immune suppressive molecules and TNFR2, we performed qPCR and flow cytometry analyses on Treg cells among different patients. Compared with TNFR2low Treg cells, TNFR2high Treg cells highly expressed GITR, ICOS, CTLA4, and TIGIT in MPE patients (figure 1I–K). To validate whether Treg cells with elevated suppressive molecule expression possess stronger immunosuppressive functions, human CD8+ T cells were isolated from healthy donors’ PB, labeled with CFSE, and then co-cultured with TNFR2high or TNFR2low Treg cells from different patients. Subsequently, the proliferation of CD8+ T cells were assessed. The results demonstrated that co-culturing with TNFR2high Treg cells suppressed proliferation of CD8+ T cells (figure 1L) and diminished the expression of tumor necrosis factor-α (TNF-α), interferon (INF-γ), granzyme B, and perforin of CD8+ T cells (figure 1M, online supplemental figure 1D), compared with co-culture with TNFR2low Treg cells. Simultaneously, there was an increase in levels of exhaustion molecules PD-1, T-cell immunoglobulin domain and mucin domain 3 (TIM-3) and TIGIT on CD8+T cells (online supplemental figure 1E). The CD4+CD25+Foxp3+Treg cells exert inhibitory effects on the proliferation and immune reactivity of CD8+ T cells through the production of immunosuppressive factors, such as IL-10 and TGF-β.40–43 Therefore, we employed ELISA and flow cytometry to assess the levels of IL-10 and TGF-β in TNFR2high and TNFR2low Treg cells after coculture with CD8+ T cells for 48 hours, revealing that cytokine secretion by TNFR2high Treg cells was significantly higher compared with TNFR2low Treg cells. This further substantiates the enhanced immunosuppressive function exhibited by TNFR2high Treg cells (online supplemental figure 1F). In summary, TNFR2high Treg cells exhibit elevated expression levels of multiple inhibitory molecules and possess a greater capacity to suppress the proliferation of T cells and induce functional exhaustion of CD8+T cells, thereby exerting more potent immunosuppressive functions. The above findings indicate that elevated TNFR2 expression enhances the immunosuppressive capacity of Treg cells.

Lactate increases TNFR2 expression and enhances the immunosuppressive function of Treg cells

To uncover the factors which influence TNFR2 expression on Treg cells in MPE, we conducted a GSVA using our scRNA-seq data of MPE and public database of NSCLC to explore the metabolic alterations occurring in Treg cells. The result of GSVA unveiled an upregulation of the glycolytic pathway in Treg cells, accompanied by an elevation in the gene expression of SLC16A1, encoding MCT1 (figure 2A). Notably, MCT1 facilitates lactate uptake and subsequently converts it into pyruvate through LDH enzymatic activity, contributing to energy production. Lactate, the primary product of glycolysis, plays a crucial role in sustaining the survival and proliferation of malignant tumor cells. Additionally, it alters the atlas of TME and modulates the function of immune-suppressive cells. Gu et al have reported that lactate increased the expression of FOXP3 and enhanced the function of Treg cells.21 Meanwhile, investigations have highlighted the significant role of TNF/TNFR2 signaling in maintaining FOXP3 expression and the stability of Treg cells.44 Consequently, the regulation of TNFR2 expression on Treg cells may be intricately linked to lactate metabolism, establishing it as a pivotal factor in function of Treg cells.

Figure 2

Lactate increases TNFR2 expression and enhances the immunosuppressive function of Treg cells. (A) Gene-set enrichment analysis was performed on gene sets of glycolysis signaling pathway. Positive NES indicates higher expression in TNFR2+ Treg cells. (B) Comparison of the concentration of lactate in MPE and PB from untreated patients with lung cancer (n=45–46). Lactate concentration was measured by a Lactate Colorimetric/Fluorometric Assay Kit. (C) Correlation between the proportions of TNFR2+cells present in CD4+CD25+FOXP3+ Treg cells and lactate levels in MPE. (D) Correlation between lactate level and survival in MPE was analyzed using Kaplan-Meier analysis (n=18–24). The median survival time of patients with high lactate level (≥4 mM) was 10 months. While the median survival time of patients with low lactate level (<4 mM) was 20 months. (E) The expression of MCT-1 and LDHB in TNFR2high and TNFR2low Treg cells were analyzed using flow cytometry. (F, G) Treg cells from human MPE individuals were stimulated with the indicated concentration of LA for 24 hours. The relative expression levels of TNFR2 genes were measured by qRCR in sorted Treg cells isolated from human MPE (F). The expression level of TNFR2 was assessed by flow cytometry (n=3–7) (G). (H) The ratio of FOXP3 was assessed by flow cytometry after being treated with the indicated concentration of LA for 24 hours (n=5). (I, J) Immune inhibitory molecules (GITR, ICOS, CTLA4, TIGIT) in Treg cells of human MPE treated with 10 mM LA or α-TNFR2 for 24 hours were detected by flow cytometry. Representative plots (I) and representative histograms (J) are shown (n=3–5). (K) Suppression assays with human CD8+ T cells were performed. (K) Treg cells were isolated from corresponding human MPE by MACS, CD8+ T cells were isolated from peripheral blood of healthy donors by MACS. CFSE assay was performed to detect the proliferation of CD8+ T cells after co-cultured with Treg or Treg (LA) cells. Data shown in (A−K) are representative of at least three independent experiments (mean±SD). Statistical analysis was performed using unpaired two-tailed Student’s t-test (B, D, E) or one-way ANOVA (F, G, H, J, K). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. ANOVA, analysis of variance; CFSE, carboxyfluorescein succinimidyl ester; LA, lactate; LDHB, lactate dehydrogenase B; NES, Normalized Enrichment Score; NS, not significant.

Subsequently, we explore the role of lactate metabolism in Treg cells. Research indicated that the lactate concentration in tumor tissue was 5–20 times higher than that in normal tissue (1.8–2.0 mM).45 Therefore, we collected MPE samples from multiple untreated lung cancer patients, along with PB samples from the corresponding patients, to assess lactate concentration. Our analysis revealed a significant enrichment of lactate in MPE compared with the corresponding patients’ PB (figure 2B), and the proportion of TNFR2 on Treg cells increases in direct correlation with the lactate concentration in MPE (figure 2C). Additionally, an increased concentration of lactate in MPE is associated with poorer patient survival outcomes in our study (figure 2D), which aligns with the findings of previous studies.21 32 46 Besides, lactate dehydrogenase B (LDHB) has been identified as a critical enzyme involved in the glycolytic pathway. Aberrant expression of LDHB leads to elevated lactate production, consequently promoting tumor cell proliferation and metastasis.47 48 Given the pivotal roles of MCT1 and LDHB in exogenous transport and intracellular synthesis respectively, we examined the expression of MCT1 and LDHB protein in Treg cells. Flow cytometry analysis showed significantly higher expression levels of MCT1 and LDHB in TNFR2high Treg cells compared with TNFR2low Treg cells (figure 2E). Studies have shown that MCT4 (SLC16A3), the key lactate transporter, also plays a crucial role in tumor metabolism.49–51 Hence, we have also taken into consideration the disparity in MCT4 expression between TNFR2high and TNFR2low Treg cells. However, both scRNA-seq analysis (online supplemental figure 2A) and Western blot (online supplemental figure 2B) revealed no significant difference in MCT4 expression levels between TNFR2high and TNFR2low Treg cells. Consequently, we mainly investigated the regulatory role of lactate in Treg cells through MCT1. Taken together, we hypothesize that lactate may be involved in regulating the expression of TNRF2 on Treg cells through exogenous transport and intracellular synthesis.

The direct correlation between lactate and TNFR2 expression in Treg cells was subsequently investigated in vitro. The CD4+CD25+FOXP3+Treg cells were isolated by magnetic-activated cell sorting from PEMC and then treated with different concentrations of lactate in vitro. The addition of exogenous lactate resulted in a significant increase in the expression of TNFR2 on Treg cells with increasing lactate concentrations byqPCR and flow cytometry (figure 2F,G, online supplemental figure 2C). Moreover, the presence of lactate slightly increases the proportion of FOXP3 (figure 2H), suggesting that lactate may impact the proportion of Treg cells in MPE. In the same way, the introduction of exogenous sodium lactate produces comparable results (online supplemental figure 2D,E). To investigate the variations in lactate concentration within the TME, we measured lactate concentration in the supernatant at different time points. Our findings revealed a progressive rise in lactate concentration as the culture time increased (online supplemental figure 2F). Subsequently, a notable upregulation of TNFR2 on Treg cells was observed when Treg cells were cultured with conditioned media (CM) derived from A549 cells. The upregulation trend was then hindered by AZD3965 (online supplemental figure 2G). Likewise, on co-culturing the Treg cells with A549 cells, the same phenomenon was observed (online supplemental figure H,I), providing evidence for the association between lactate and TNFR2 expression on Treg cells. To investigate the impact of lactate-induced upregulation of TNFR2 on the immunosuppressive function of Treg cells, we assessed the expression levels of immunosuppressive molecules on Treg cells following the addition of exogenous lactate. Our results indicated a substantial upregulation in the expression of GITR, ICOS, CTLA4, and TIGIT. The upregulation was partially reversed after blocking TNFR2 (figure 2I–J, online supplemental figure 2J). In addition, we observed a significant reduction in the proliferative capacity and secretion levels of TNF-α, IFN-γ, granzyme B, and perforin of CD8+ T cells following co-cultured with lactate-treated Treg cells, compared with untreated Treg cells. In line with previous results, TNFR2 blocking antibodies rescued this trend (figure 2K, online supplemental figure 2K). The immunosuppressive factors IL-10 and TGF-β, produced by Treg cells themselves during co-culture, were also closely monitored. It was observed that the secretion levels of IL-10 and TGF-β in lactate-treated Treg cells significantly increased compared with untreated Treg cells. However, this trend could be reversed with the administration of TNFR2 blocking antibodies (online supplemental figure 3A). Collectively, these results suggest that lactate may enhance the function of Treg cells by activating the TNFR2 pathway, thereby upregulating the expression of immune molecules to facilitate Treg-mediated immunosuppression.

Blocking lactate metabolism pathway downregulates TNFR2 expression on Treg cells

Given that extracellular lactate can stimulate TNFR2 upregulation on Treg cells, we hypothesized that regulating intracellular lactate production might impact the levels of TNFR2. AZD3965, a potent MCT1 inhibitor, effectively hinders the transport of lactate, leading to a notable reduction in intracellular lactate levels. As a result, it significantly diminishes the expression of TNFR2 on FOXP3+ Treg cells when cultured in 5 mM medium (figure 3A–D). The production of lactate is determined by the balance between glycolysis and mitochondrial metabolism.26 Consequently, we investigated whether the activity of enzymes associated with these pathways could govern lactate levels and subsequently modulate TNFR2 expression on FOXP3+ Treg cells. The non-metabolizable glucose analog 2-deoxy-d-glucose (2-DG) and Oxamate were used to inhibit lactate production by modulating the activities of Hexokinase and pyruvate dehydrogenase, respectively (figure 3A). As anticipated, both lactate production and expression of TNFR2 on FOXP3+ Treg cells were decreased by these two compounds (figure 3E–H). Additionally, rotenone, an inhibitor of the mitochondrial respiratory chain complex I that drives cells towards glycolysis, increased levels of both intracellular lactate and TNFR2 on FOXP3+ Treg cells (figure 3I,J). It is noteworthy that the proportion of Treg cells within the CD4+ T cell population (as indicated by FOXP3 expression level) was also influenced by level of lactate, in line with the observed fluctuation of TNFR2 (online supplemental figure 4A–D). Nevertheless, the indicated glycolysis modulators used in vitro exert a certain degree of influence on Treg survival (online supplemental figure 4E–H). These results suggest that lactate can regulate the expression of TNFR2 on Treg cells.

Figure 3

Blocking lactate metabolism pathway downregulates TNFR2 expression on Treg cells. (A) Regulation of glycolysis and lactate production by diverse metabolic modulators. (B–D) CD4+CD25+FOXP3+Treg was separated from human MPE by Miltenyi Biotec Kit. Intracellular lactate levels were measured in Treg cells treated with different concentrations of AZD3965 for 24 hours (n=3) (B). The expression level of TNFR2 of Treg cells was measured by flow cytometry after being treated with different concentrations of AZD3965 for 24 hours (n=3) (C). The expression level of TNFR2 of Treg cells was measured by flow cytometry after being treated with 10 mM LA combined with 10 mM AZD3965 for 24 hours (n=3) (D). (E, F) Intracellular lactate levels were measured in Treg cells treated with different concentrations of Oxamate (E) or 2-DG (F) for 24 hours (n=3). (G, H) The expression level of TNFR2 of Treg cells was measured by flow cytometry after being treated with different concentrations of Oxamate (G) or 2-DG (H) for 24 hours (n=4–6). Representative plots (left) and representative histograms (right) are shown. (I) Intracellular lactate levels were measured in Treg cells treated with different concentrations of Rotenone for 24 hours (n=3). (J) The expression level of TNFR2 of Treg cells was measured by flow cytometry after being treated with different concentrations of Rotenone for 24 hours (n=4). Data shown in (B–J) are representative of at least three independent experiments (mean±SD). Statistical analysis was performed using one-way ANOVA (B–J). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. ANOVA, analysis of variance; 2-DG, 2-deoxy-d-glucose; MPE, Malignant pleural effusion; ns, not significant.

Elevated histone lactylation regulates TNFR2 expression on Treg cells

We next explored the underlying mechanism by which lactate regulates the high expression of TNFR2 on Treg cells. Research has demonstrated that post-translational lactate modification is a crucial mechanism for the functional role of lactate, and lactate-derived lactylation of histone lysine residues was described as an epigenetic modification to promote gene transcription.26 Since the MPE microenvironment showed active glycolysis, in which Treg cells generate large amounts of lactate as substrates for histone lactylation, we examined the protein lactylation levels in Treg cells using the pan anti-Kla antibody. The results confirmed the presence of histone Kla and showed significantly high levels of histone Kla in Treg cells from MPE patients than in Treg cells from PB; the predominant band appeared near 15 kD