Article Text
PDF
Abstract
Background Chimeric antigen receptor T (CAR-T) cell therapy has been shown remarkable efficacy in the treatment of hematological malignancies in recent years. However, a considerable proportion of patients would experience tumor recurrence and deterioration. Insufficient CAR-T cell persistence is the major reason for relapse. Multiple strategies to enhance the long-term antitumor effects of CAR-T cells have been explored and developed. In this study, we focused on tyrosine kinase inhibitors (TKIs), which have emerged immunomodulatory potential besides direct tumoricidal effects.
Methods Here, we screened 50 approved TKIs drugs and identified that afatinib (AFA) markedly enhanced the expressing of CD62L and inhibited reactive oxygen species level in T cells. And the underlying mechanisms of AFA medicating T cells were explored by detecting signal transduction, and metabolism pattern. Furthermore, we co-cultured AFA with CAR-T cells during the preparation stage and multianalyses of differentiation characteristics, metabolic profiling, and RNA sequencing revealed that AFA induce comprehensive metabolism remodeling and fate reprogramming. Based on it, we finally identified the antitumor efficacy of AFA-pretreatment CAR-T compared with negative-control CAR-T.
Results We identified that AFA blocked the T-cell receptor (TCR) and phosphoinositide 3-kinase-protein kinase B-mechanistic target of rapamycin signaling pathways, induced metabolic reprogramming and modulated T-cell differentiation. When combined with CAR-T cells, AFA inhibited the exhaustion and enhanced the persistence and cytotoxicity. Our results revealed that the pretreatment of AFA enables to boost CAR-T cells with strong antitumor cytotoxicity in leukemia mouse model.
Conclusions Our study systematically demonstrated that AFA pretreatment effectively enhanced CAR-T cells antitumor performance, which presents a novel optimization strategy for potent and durable CAR-T cell therapy.
- Chimeric antigen receptor - CAR
- Adoptive cell therapy - ACT
- Hematologic Malignancies
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Chimeric antigen receptor T (CAR-T) therapy has achieved remarkable success in the treatment of hematologic malignancies. However, limited persistence of CAR-T cells would impede the long-term antitumor efficacy.
WHAT THIS STUDY ADDS
We found that the tyrosine kinase inhibitor, Afatinib, inhibited T-cell receptor-phosphoinositide 3-kinase signaling, resulted in alteration of metabolic patterns, which prevented the terminal differentiation and exhaustion of CAR-T cells. These changes could enhance the persistence of CAR-T cells and efficacy of antitumor.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Our study has identified a novel and promising clinical compound, which could be combined with CAR-T cell therapy to augment its effectiveness, providing a potential strategy to enhance the CAR-T cells clinical efficacy.
Background
In recent years, chimeric antigen receptor T (CAR-T cells have shown remarkable efficacy in patients with refractory or relapsed B acute lymphoblastic leukemia, offering a promising therapeutic approach for patients who have not responded to traditional treatments.1 2 However, long-lasting recovery after CAR-T cell therapy is not guaranteed, with relapse observed in approximately 35–75% of patients.3 4 In relapsed patients, CAR-T cells frequently remain undetectable, indicating suboptimal persistence.5 CAR-T cells experience intrinsic dysfunction, such as exhaustion, terminal differentiation, and apoptosis, due to sustained tonic signaling or excessive activation during manufacturing or clinical therapy. These factors restrict the long-term antitumor performance of CAR-T cells.6 Consequently, enhancing the persistence of CAR-T cells is critical and urgent for improving the clinical efficacy of CAR-T cell therapy.
The enhancement of CAR-T cell persistence is fundamentally contingent on improvements in the intrinsic survival capabilities of T cells. The excessive activation of CAR-T cells during the preparation phase precipitates their rapid evolution into effector phenotypes on stimulation by CD3/CD28 antibodies and cytokines, especially interleukin-2 (IL-2).7 This leads to a preponderance of terminal effector T cells (Tef) within the CAR-T cell product, thereby diminishing the representation of memory T cell subsets. Compared with conventional CD19 CAR-T products, those derived from stem cell memory T (Tscm) cells or central memory T (Tcm) cells exhibit reduced terminal differentiation and exhaustion, superior persistence and sustained therapeutic responses in vivo.8 A multitude of interventions are being implemented to augment the memory subset composition of CAR-T cells, aiming to enhance their therapeutic efficacy. These interventions include the redesign of CAR constructs to attenuate toxicity signals, the strategic selection of distinct T-cell subsets for the generation of CAR-T products, the fine-tuning of T-cell metabolism to boost survival capacity, and the synergistic use of established pharmacological agents with CAR-T cells.9–11 These approaches have availably improved the persistence of CAR-T cells and have shown superior immunotherapeutic effects.
To improve the antitumor efficacy of CAR-T cell therapy, we focused on tyrosine kinase inhibitors (TKIs), which have garnered attention for their targeted antineoplastic properties by inhibiting aberrant kinase activities implicated in oncogenesis.12 In addition to their direct tumoricidal effects, emerging data highlight their immunomodulatory potential as they influence the tumor microenvironment and modulate immune checkpoint dynamics to potentiate the tumoricidal capacity of the immune system.13 This multifaceted role positions TKIs as integral components in cancer treatment, especially in combination therapies designed to maximize immune system engagement.14 In this study, we screened 50 approved TKIs drugs and explored their effects on phenotypic changes related to the exhaustion and differentiation state of T cells. We identified afatinib (AFA) as a potential medication for markedly enhancing the expressing of CD62L and inhibiting reactive oxygen species (ROS) level in T cells. We verified that AFA blocked T-cell receptor (TCR)-phosphoinositide 3-kinase (PI3K) signaling and modulated the metabolism pattern by decreasing glycolysis and increasing oxidative phosphorylation (OXPHOS), which further regulated the progression of differentiation, showed by inhibiting the exhaustion and improving survival capacity. During the CAR-T cells preparation stage, AFA pretreatment enhancing the antitumor efficacy. Multianalyses of differentiation characteristics, metabolic profiling, and RNA sequencing revealed that AFA induced comprehensive metabolism remodeling and fate reprogramming. Here, our study demonstrates that AFA intervention in CAR-T cells enables to improve antitumor efficacy with superior memory formation and resistance to exhaustion, which provide a new strategy for CAR-T therapy optimization.
Methods
Cell lines
Jurkat-T cells and Nalm-6-luc cells were cultivated in Roswell Park Memorial Insitute (RPMI) 1640 (Gibco) supplemented with 10% fetal bovine serum (FBS, Clark), 100 µg/mL penicillin and 100 µg/mL streptomycin at 37°C in a humidified atmosphere with 5% carbon dioxide.
T-cell isolation and CAR-T cell preparation
Following required ethical and safety procedures, fresh peripheral blood was collected from healthy donors (n=3 or more) aged between 18 and 45 years. The peripheral blood mononuclear cells were isolated by density gradient centrifugation. CD3+T cells were isolated and activated using Dynabeads Human T-Activator CD3/CD28 (Gibco) at a 1:1 bead-to-cell ratio in X-VIVO15 (LONZA) medium supplemented with 10% FBS and 200 IU/mL of recombinant human IL-2 (MedChemExpress (MCE)).
The CAR construct used in this study consisted of a single-chain antibody fragment (scFv, FMC63) specific for human CD19. This scFv was preceded by a CD8a leader peptide and followed by a CD8 hinge, 4-1BB costimulatory domain, and CD3ζ intracellular regions, all of which were linked to a P2A-eGFP sequence. Lentiviruses encoding CAR were produced by transient transfection of the 293 T-cell line, after which the supernatants were collected. These lentiviral stocks were then concentrated and stored at −80°C for future use.
Primary T cells activated for 2 days were transduced with the CD19 CAR lentiviral. After infection, the CAR-T cells were seeded (106 cells/mL) and expanded for 4 days. On the sixth day, the anti-CD3/CD28 beads were removed using a magnet, and the CAR-T cells were exposed to the indicated compounds for 3 days and then subjected to analysis or in vivo experiments. Unless otherwise indicated, the medium was changed every 2 days.
TKIs screens
There are 50 TKIs that mainly target EGFR/HER, VEGFR, ALK, JAK, ABL, RET, BTK, MET, FGFR, FLT3, TRK, SYK, CSF1R, KIT, SRC, etc, approved for marketing by the Food and Drug Administration (FDA) in 2021 or earlier. All the compounds were provided in vehicle at a 10 mM concentration from MCE. Briefly, T cells were incubated in 24-well plates containing the test compounds with at 1 µM for 3 days, after which the cell viability, ROS level, and CD62L level were assessed using flow cytometry (FCM).
Cell proliferation assays
The cell counting kit-8 (CCK-8) assay was used to determine the quantity of viable cells according to the manufacturer’s guidelines. CAR-T cells were distributed in triplicate in 96-well plates at a concentration of 50,000 cells/100 µL. Each well contained 90 µL of CAR-T cells and 10 µL of substances at final concentrations ranging from 0 to 5 µM. After 48 hours of incubation, 10 µL of CCK-8 solution were added, and the plates were incubated at 37°C for 4 hours. The absorbance at 450 nm was measured using a SpectraMax iD5 microplate reader.
Flow cytometry
Fluorochrome-conjugated antibodies were obtained from BioLegend and BD Biosciences. Cultured cells were washed with fluorescence-activated cell sorting (FACS) buffer (Dulbecco’s phosphate-buffered saline (DPBS) with 2% FBS) and then suspended in FACS buffer with antibodies at 4°C for 20 min. FCM data were processed using FlowJo V.10 software, which included typical filter settings to exclude doublets and dead cells. The T-cell surface phenotype was evaluated using the following specific antibodies: anti-CD69-APC (Clone FN50), anti-CD25-ACP-Cy7 (Clone M-A251), anti-CD45RO-PE (Clone UCHL1), anti-CD62L-APC (Clone DREG-56), anti-CD95-APC-Cy7 (Clone DX2), anti-CTLA-4-PE-Cy7 (Clone BNI3), anti-PD-1-PE (Clone NAT105), anti-TIM-3-APC (Clone F38-2E2), anti-INF-γ-PE (Clone 4S.B3), anti-TNF-α-APC (Clone Mab11), anti-CD107a-APC-Cy7 (Clone H4A3), anti-CD4-PE (Clone RPA-T4) and anti-CD8-PE-Cy7 (Clone SK1).
Western blot
Total protein was extracted using NP40-based lysis buffer (Beyotime) supplemented with a comprehensive suite of protease and phosphatase inhibitors (Beyotime) to ensure the preservation of protein integrity. The concentration of the extracted proteins was meticulously quantified with a bicinchoninic acid Protein Assay Kit (Thermo Scientific). Subsequently, a uniform quantity of the cell lysates was resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and electro-transferred to nitrocellulose membranes. The membranes were incubated with specific primary antibodies under stringent conditions at 4°C overnight. This was followed by subsequent incubation with horseradish peroxidase-linked secondary antibodies for 2 hours at ambient temperature. The resulting chemiluminescent signals were detected and documented with precision using a ChemiDoc MP Imaging System (Bio-Rad).
Messenger RNA library preparation and sequencing
CAR-T cells from healthy donors were prepared following a standard protocol. Total messenger RNA was extracted and purified from more than 1×106 CAR-T cells per sample using TRIzol reagent (Sigma, Darmstadt, Germany), adhering strictly to the instructions provided by the manufacturer. The RNA libraries were subsequently sequenced on an Illumina NovaSeq 6000 system (LC-BioTechnology, Hangzhou, China) via paired-end 150 sequencing. Differentially expressed genes (DEGs) demonstrating an absolute log2 fold change greater than 2 and a p value<0.05 were deemed significant.
In vitro co-culture cytotoxicity
In the in vitro cytotoxicity assay, 10,000 target Nalm-6-luc leukemia cells expressing luciferase were co-cultured with CAR-T cells in RPMI 1640 medium supplemented with 10% FBS but devoid of IL-2. The effector-to-target (E:T) ratios were established at 1:4, 1:8, and 1:16 in white-walled 96-well plates, with triplicates for each condition. After indicated time of incubation, cytotoxicity was assessed using a luciferase-based cytotoxic T lymphocyte assay. Concurrently, the luminescence intensity of the tumor cells was measured at 24-hour intervals. The specific lysis of target cells was quantified using D-luciferin salt (Vazyme) in accordance with the manufacturer’s protocol. Measurements were taken using a SpectraMax iD5 luminometer.
Cell survival assay
Following successful activation via CD3/CD28 antibodies, T cells were seeded at uniform densities in fresh X-VIVO15 medium lacking IL-2. All exogenous pharmacologic interventions were ceased, and the cells were subsequently maintained in culture for an uninterrupted period of 6 days. Cellular viability and counts were calculated by acridine orange and propidium iodide dual staining techniques.
ROS assay
The fluorescent probe H2DCFDA (MCE) was used to determine the levels of intracellular ROS. Cells were exposed to 5 µM H2DCFDA in fresh medium; the cells were shielded from light exposure for 30 min at a maintained temperature of 37°C. Subsequently, the cells were reconstituted in DPBS and subjected to immediate quantification through FCM.
Glucose uptake assay
Glucose Uptake Assay Kit (Abcam) were used by the fluorescent deoxyglucose analog 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)−2-deoxyglucose (2-NBDG) as a probe to detect glucose uptake in cultured cells. 5×105 cells were centrifuged to remove the supernatant and then resuspended in 100 µL of glucose-free medium, followed by incubation at 37°C for 2 hours. 10 min before the end of the treatment, 2-NBDG was added to the medium to a final concentration of 100 µg/mL. After 20 min of incubation at 37°C, the cells were centrifuged to remove the supernatant, resuspended in DPBS, and immediately analyzed by FCM.
ATP measurement
Cells were centrifuged to remove the supernatant and resuspended at a concentration of 1×106 cells in 200 µL of lysis buffer on ice for cell lysis. Then, the lysates were centrifuge at 4°C and 12,000 g for 5 min, and the supernatant was collected for subsequent assays. The ATP assay reagent (Beyotime) was diluted at a 1:4 ratio with the ATP assay reagent dilution solution according to the manufacturer’s protocol. Then, 100 µL of the ATP detection working solution was added to each well, and the plate was incubated at room temperature for 5 min. Then, 20 µL of the sample or standard solution was added to each well and quickly mixed with a micropipette, and after a 2 s interval, luminescence was measured using a luminometer.
Mitochondrial membrane potential measurement
Cells were centrifuged at 300 g for 5 min, after which the supernatant was discarded. The cell pellet was resuspended in Mito-Tracker Red CMXRos (Beyotime) staining solution (40 nM, prewarmed at 37°C) and incubated for 30 min. Following incubation, the cells were centrifuged, and the supernatant was discarded. The cells were then resuspended in DPBS and immediately analyzed using FCM.
Extracellular acidification rate and oxygen consumption rate assay
After treated with AFA for 2 days, CAR-T cells were resuspended using 1,640 medium and extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) detection working solution (Elabscience), maintained a cell concentration of 5×106 cells/mL. The cells were then added into a black-walled, clear-bottom culture plate. Following 30 min of incubation at 37°C in a microplate reader, measurements were taken by SpectraMax iD5 fluorescence.
In vivo models
Prior to in vivo experiments, the cell lines were regularly tested for Mycoplasma and determined to be negative. For the human leukemia model, female 6–8-week-old immunocompromised NOD.Cg-Prkdcscid mice (Vital River Laboratory Animal Technology) were used; all procedures were approved by the Institutional Animal Care and Use Committee of Army Medical University. The mice were intravenously injected with 2×106 Nalm-6-luc cells via the tail vein. On day 10, the mice were randomly divided into three distinct cohorts: Mock (four mice), negative-control (NC) CAR-T (six mice), and AFA-pretreatment CAR-T (six mice). Subsequently, each mouse in the latter two cohorts received an intravenous dose of 5×105 NC CAR-T or AFA-pretreatment CAR-T cells, respectively. Leukemia progression was measured every 10 days by bioluminescence imaging using a Bruker imaging system after the injection of 200 µL of D-luciferin and analyzed using Bruker MI SE V.7.2 (PerkinElmer). The body weight was recorded. In instances where either physiological or behavioral indices deteriorated beyond predefined thresholds or a reduction in body mass exceeded 15% of baseline measurements, humane euthanasia was promptly conducted.
Statistical analysis
All in vitro experiments were conducted with a minimum of three biological replicates per experimental group, and the results are presented as individual data points. In vivo studies, on the other hand, involved four to six mice per group. Statistical analyses were performed on data obtained from independent biological replicates. When comparing two groups, whether in vitro or in vivo, the unpaired Student’s t-test was used. In cases where biological replicates or independent experiments were being compared, the paired Student’s t-test was used. For comparisons involving more than two groups, one-way analysis of variance was used for multiple comparison. Flow cytometry measurements were quantified using FlowJo V.10 software. Statistical analyses to determine significant differences were conducted using Prism V.9 (GraphPad software). Survival curves were estimated using the Kaplan-Meier method, and the log-rank test was used for the statistical analysis of survival. The data are presented as the mean±SEM. A p value<0.05 was considered to indicate statistical significance (*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001).
Results
TKIs screening revealed that AFA regulated the differentiation and metabolism
To identify the impact of TKIs on T-cell differentiation and metabolism, we performed an in vitro screening using a TKIs library (figure 1A). CD3+T lymphocytes were isolated from healthy donors and co-cultured with the screened compounds for 1 day, followed by activation with CD3/CD28 antibodies for 2 days. FCM was then used for primary screening based on the levels of ROS and enrichment of CD62L+T cell subsets, which mainly included Tcm and naïve T (Tn) cells. 24 compounds increased the proportion of CD62L (fold change vs NC>1) (figure 1B), and 30 compounds reduced ROS levels (fold change vs NC<1) during T-cell expansion (figure 1C). In addition, we focused on the condition of the T cells and conducted a secondary analysis of their viability and proliferation rate. The results revealed that 12 compounds improved the viability of T cells (figure 1D), and 31 compounds increased the numbers of T cells, indirectly reflecting the impact of the drugs on the in vitro expansion rate of T cells (figure 1E).
TKIs screening revealed that AFA enhances the expression of CD62L and inhibits ROS. (A) Schematic of the T-cell screening process. After isolation, T cells were co-cultured with 50 compounds for 1 day, followed by supplementation with CD3/CD28 antibodies and subsequent CD62L and ROS analysis by FCM after 2 days. (B–E) The relative CD62L level (B), ROS content (C), viability (D), and cell counts (E) of T cells treated with 50 compounds (1 µM). The data are presented as means±SEMs of n=3 donors. (F) A gradient scatterplot of T-cell screening, displaying relative percentages of CD62L, ROS content, and viability for 50 compounds, as determined by FCM. (G) Histograms and plots of the naïve phenotype and ROS levels in T cells after treatment with AFA. T cells were exposed to 1 µM AFA for 3 days and then analyzed by FCM for CD62L expression and ROS content. The MFI of the CD62L+ and ROS populations was quantified. The data are presented as means±SEMs of n=3 donors. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001; ns, not significant by one-way analysis of variance with a multiple comparison test (B–E) or unpaired t tests (G). AFA, afatinib; FCM, flow cytometry; MFI, mean fluorescence intensity; NC, negative-control; PBMC, peripheral blood mononuclear cell; ROS, reactive oxygen species.
To select the most promising candidate compound, we performed an integrated analysis that considered the levels of CD62L+T (Tn and Tcm) cell subsets, along with the ROS content and viability. Among several compounds, AFA was the most effective (figure 1F). Compared with the NC, AFA significantly increased the expression levels of CD62L and decreased the concentration of ROS (figure 1G). These results suggested that among the 50 clinical TKIs, AFA pretreatment significantly influenced T-cell differentiation and decreased ROS levels.
AFA inhibited TCR-PI3K signaling and altered the metabolic pattern of T cells
To determine the potential mechanisms by which AFA modulates T-cell differentiation, we investigated the effect of AFA on TCR-mediated signal transduction in Jurkat-T cells, a prototypical model used to study TCR signaling.15 Jurkat-T cells maintained their viability when exposed to 1 µM AFA (online supplemental figure S1A). Next, we assessed the regulatory effects of AFA on cellular activation signals. In contrast to untreated Jurkat-T cells, those cultured with AFA for a brief period (48 hours) exhibited a relatively low level of activation (figure 2A). Reduced expression of CD69 and CD25, which serve as markers of early and middle activation states in T cells, respectively, is accompanied by alterations in TCR signaling pathways.16 Thus, we explored the impact of AFA on TCR signaling. CD3/CD28 stimulation, rapidly activated TCR signaling, which led to enhance PI3K pathway signaling, indicating that TCR activation signals are conveyed to PI3K downstream.17 AFA decreased the phosphorylation of lymphocyte-specific protein tyrosine kinase, zeta-chain-associated protein kinase 70, linker for activation of T cells, and phospholipase C while simultaneously decreasing the phosphorylation of PI3K, protein kinase B (AKT), mechanistic target of rapamycin (mTOR), and ribosomal protein S6 (RPS6) (figure 2B,C). The adjustment of the TCR and PI3K signaling pathways could reshape the metabolic state and alter differentiation in T cells.18 Therefore, we measured metabolic state after AFA pretreatment. The results showed that AFA significantly decreased the ROS content, mitochondrial membrane potential (Mito), and glucose uptake (figure 2D). Conversely, the ATP level increased (figure 2E). To further corroborate the findings of the present study, we explored the regulatory effect of AFA on primary T cells, and the results were consistent with those obtained with Jurkat-T cells (online supplemental figure S1B, figure 2F–J).
Supplemental material
AFA inhibited TCR-PI3K signaling and altered the metabolic program of T cells. (A) Density plots of Jurkat-T cells are shown. Jurkat-T cells were treated with 1 µM AFA for 72 hours, and the expression of CD25 and CD69 markers on T cells was determined by FCM. The data are reported as the means±SEMs (n=3). (B–C) Jurkat-T cells were treated with 1 uM AFA for 48 hours, followed by the addition of CD3/CD28 antibodies, and proteins were harvested at different time points. The expression of proteins involved in the TCR signaling pathway (B) and the PI3K-mTOR-AKT signaling pathway (C) was assessed by western blotting. (D–E) FCM was used to measure the MFI (D) of metabolic markers (ROS, Mito and 2-NBDG), and chemiluminescence was used to determine ATP concentration (E) in Jurkat-T cells. The data are reported as the means±SEMs (n=3). (F) The CD25 and CD69 expression levels on T cells were quantified using FCM. The data are reported as the means±SEMs (n=3). (G–H) Representative western blot of TCR (G) and PI3K-AKT-mTOR (H) in AFA-pretreatment T cells. (I–J) Levels of metabolism markers (ROS, Mito, 2-NBDG and ATP) in AFA-pretreatment T cells. The data are reported as the means±SEMs (n=3). (K) The protein expression of glycolytic enzymes in AFA-pretreatment T cells was analyzed through western blotting. (L) The ratio of CD4+ and CD8+ T cells after AFA treatment compared with that in the control group. The data are reported as the means±SEMs (n=3). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001; ns, not significant according to unpaired t-tests (A, D–F, I–J, and L). AFA, afatinib; AKT, protein kinase B; FCM, flow cytometry; Glut1, glucose transporter 1; HK II, hexokinase II; LAT, linker for activation of T cells; LCK, lymphocyte-specific protein tyrosine kinase; LDHA, lactate dehydrogenase A; MFI, mean fluorescence intensity; Mito, mitochondrial membrane potential; mTOR, mechanistic target of rapamycin; NC, negative-control; PFKP, platelet type; PI3K, phosphoinositide 3-kinase; PKM2, pyruvate kinase m2; PLC, phospholipase C; RPS6, ribosomal protein S6; ROS, reactive oxygen species; TCR, T-cell receptor; ZAP 70, zeta-chain-associated protein kinase 70; 2-NBDG, 2-(N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)−2-deoxyglucose.
Based on the inhibitory effect of AFA on AKT, a pivotal node within the PI3K signaling cascade, we investigated whether the re-activation of AKT could mitigate the effects of AFA. After co-cultured with AFA for 2 days, T cells were activated using CD3/CD28 antibodies, and the culture medium was supplemented with SC79, an activator of AKT. SC79 successfully reversed inhibitory effect of AFA by increasing the phosphorylation levels of AKT and RPS6 (online supplemental figure S1C). Additionally, pretreatment with SC79 could reversed the changes in the levels of Mito, concentrations of ROS, absorption rates of glucose, and content of ATP compared with those in the group without SC79 (online supplemental S1D). These results revealed that the regulatory effects of AFA on T-cell metabolism occur primarily via the inhibition of AKT and can be partially reversed through SC79. Furthermore, the reduction in glucose uptake suggested suppression of the glycolytic pathway.19 To further explore the variation in glycolytic pathways, we assessed the expression levels of enzymes associated with the glycolytic pathway in AFA-pretreatment T cells. The results showed a significant reduction in the expression levels of the enzymes glucose transporter, hexokinase II, phosphofructokinase, platelet type, pyruvate kinase m2, and lactate dehydrogenase A (figure 2K). These changes suggest that AFA inhibits the glycolytic pathway, potentially leading to alterations in the cellular metabolic patterns and energy acquisition mechanisms. Notably, alterations in the proportions of CD4+ and CD8+ T cells could also induce modifications in signaling pathways and metabolism.20 Thus, we further compared the abundance of CD4+T cells to that of CD8+T cells by FCM, and no significant differences were found (figure 2L). This shows that the regulation of CD4+ and CD8+ T cells by AFA is homogeneous. These data demonstrated that AFA inhibits the TCR and PI3K-AKT-mTOR signaling pathways, restricts glycolytic processes, and precipitates alterations in cellular metabolism.
AFA induced T-cell differentiation to improve survival capacity and reduce exhaustion
Distinct T-cell subsets exhibit diverse metabolic requirements, such as Tn cells being more dependent on OXPHOS and fatty acid oxidation in mitochondria,7 Tef cells being mainly dependent on aerobic glycolysis (Warburg effect)21 and Tscm cells being dependent on OXPHOS and low-level glycolysis.22 To investigate the potential impact of AFA-induced metabolic reprogramming on T-cell differentiation, we examined terminal differentiation phenotypes (marked by the expression of CD62L and CD45RO) to evaluate the degree of T-cell differentiation. In contrast to NC T cells, AFA-pretreatment T cells exhibited a change in T-cell subsets, including increased numbers of Tn cells, which possess high reproductive and survival abilities, and decreased numbers of Tef and effector memory T cells (Tem) cells (figure 3A). More significantly, Tscm cells could differentiate into other memory T-cell subsets on antigen re-exposure.23 To confirm whether AFA induces quantitative alterations in Tscm cells, the proportion of Tscm (CD62L+CD45RO–CD95+) cells was assessed via flow cytometric analysis. As expected, the administration of AFA additionally enhanced the expansion of Tscm cells (online supplemental figure S1E). In addition, we observed a similar trend of proliferation in Tn cells within both the CD4+ and CD8+ T-cell populations, accompanied by a reduction in the Tem and Tef subsets (online supplemental figure S1F).
AFA induced differentiation to improve T-cell survival capacity and reduce T-cell exhaustion. (A) FCM analysis of CD62L and CD45RO frequencies in AFA-pretreatment T cells. The subsets were defined as naïve T cells (CD62L+), central memory T cells (CD62L+CD45RO+), effector memory T cells (CD62L−CD45RO+), and effector T cells (CD62L−CD45RO−). The results are shown as changes relative control cells. The data are reported as the means±SEMs (n=3). (B) T cells were extensively washed and cultured in medium lacking exogenous IL-2 and AFA. The percentages and counts of living T cells as determined by acridine orange and propidium iodide staining at different time points after IL-2 withdrawal are shown. The data are reported as the means±SEMs (n=3). (C) FCM analysis of exhaustion marker frequency. The data are reported as the means±SEMs (n=3). (D–E) Western blot showing the modification of AFA-mediated expression levels of memory (D) and exhaustion (E) markers in T cells. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001; ns, not significant according to unpaired t-tests (A–C). AFA, afatinib; BCL-6, B-cell lymphoma 6; CCR7, C-C chemokine receptor 7; CTLA-4, cytotoxic T-lymphocyte-associated protein 4; FCM, flow cytometry; IL, interleukin; LAG3, lymphocyte-activation gene 3; LEF1, lymphoid enhancer-binding factor 1; MFI, mean fluorescence intensity; NC, negative-control; PD-1, programmed cell death protein 1; TCF7, T-cell factor 7; Tcm, central memory T; Tef, terminal effector T cells; Tem, effector memory T cells; Tim-3, T-cell immunoglobulin and mucin-domain containing-3; Tn, naïve T.
The long-term survival capacity of T cells is one of the prerequisites for the sustained antitumor efficacy of T cell-based immunotherapy.24 To assess whether AFA enhances the persistence of T cells, we activated human T cells and expanded them in the presence of IL-2 and AFA. Subsequently, the culture medium was replaced with the withdrawal of IL-2 and AFA, and T-cell viability and counts were assessed. The results demonstrated that compared with NC T cells, AFA-pretreatment T cells exhibited a slower decrease in both viability and number, with a transient increase in the number of T cells observed on day 3, which may be induced by the expanded population of Tn cells delaying the progression of terminal differentiation (figure 3B). On the other hand, terminal exhaustion compromises the in vivo efficacy of T cell-based immunotherapy. To determine the effect of AFA on exhaustion and memory phonotypes, we measured the expression of the inhibitory receptors programmed cell death protein 1 (PD-1), T-cell immunoglobulin and mucin-domain containing-3 (Tim-3), and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4). AFA-pretreatment T cells reduced levels of these inhibitory receptors (figure 3C). Furthermore, we detected the long-term effects of AFA on T-cell differentiation and exhaustion. The results demonstrated the upregulation memory-related phenotypes such as B-cell lymphoma 6BCL-6, C-C chemokine receptor 7 (CCR7), CD62L, T-cell factor 7 (TCF7) and lymphoid enhancer-binding factor 1 (LEF1), and the reduction in the expression of exhaustion-related proteins, like Tim-3, lymphocyte-activation gene 3 (LAG-3), CTLA-4 and PD-1 (figure 3D,E). Hence, the above evidence suggested that AFA pretreatment modulated the terminal differentiation of T cells, promoted the generation of Tn and Tscm subsets, exhibited improved resistance to exhaustion, and enhanced persistence, which implies an augmentation of antitumor activity.
AFA boosted CAR-T survival and enhanced antitumor capacity
To further investigate whether AFA adjusts the differentiation of CAR-T cells and improves persistence, we transfected T cells with a CD19 CAR to generate CAR-T cells and subsequently co-cultured them with AFA (figure 4A). Consistent with previous findings in T cells, CAR-T cells co-cultured with AFA exhibited significantly lower expression levels of CD69 and CD25 than that in NC CAR-T cells, and there was no appreciable difference in the ratio of CD4+ to CD8+ CAR T cells (online supplemental figure S2A, B). To confirm the effect of AFA on CAR transfection efficacy, the proportion of CAR+cells was measured by FCM. There was no difference in the percentage of CAR+cells between AFA-pretreatment CAR-T cells and NC CAR-T cells (figure 4B). We also evaluated the effects of AFA on the TCR-PI3K signaling pathways in CAR-T cells, and the results indicated that AFA also inhibited phosphorylation levels (figure 4C,D). Then, we monitored OCR and ECAR, which, respectively, indicate cellular oxygen consumption and glycolytic activity. AFA-pretreatment CAR-T cells showed an increase in OCR and a decrease in ECAR compared with NC CAR-T cells. This suggests that AFA preconditioning shifted cellular metabolism toward enhanced OXPHOS and reduced glycolysis (figure 4E). Additionally, the SC79 was still able to reverse the regulatory effects of AFA on CAR-T cells (online supplemental figure S2C). Then, we further examined the regulatory effect of AFA on CAR-T cell differentiation. Compared with NC CAR-T cells, AFA-pretreatment CAR-T cells exhibited a significant increase in the naïve and memory phenotype, accompanied by a reduction in the Tef and Tem phenotypes (figure 4F). We also assessed whether AFA induces alterations in the Tscm phenotype within CAR-T cells and results indicate that AFA enhance the Tscm ratio (online supplemental figure S2D). On repeated activation, CAR-T cells rapidly differentiated into Tef and diminished function. So, we further investigate the impact of AFA on the reactivation-induced CAR-T cells. The results indicated that AFA significantly reduced apoptosis in CAR-T cells on repeated activation and modulated the differentiation process, increased the generation of naïve and stem cell memory CAR-T cells (online supplemental figure S2E–G). These findings indicated that AFA pretreatment could similarly facilitate the generation of naïve and stem cell memory CAR-T cells, and that these alterations were not influenced by the CAR architecture.
AFA boosted the survival and enhanced the antitumor capacity of CAR-T cells. (A) Schematic of the experimental procedure. CAR-T cells were pretreated with AFA, after which CAR expression, differentiation, exhaustion and RNA sequencing were performed. The cytotoxicity of Nalm-6-luc cells and AFA-CAR-T cells was assessed in media without IL-2 and AFA. (B) The ratio of CAR expression in T and CAR-T cells after AFA pretreatment compared with that in the NC group. The data are reported as the means±SEMs (n=3). (C–D) Representative western blot of TCR (C) and PI3K-AKT-mTOR (D) in AFA-pretreatment CAR-T cells. (E) The relative fluorescence of OCR (L) and ECAR (M) in T cells treated with AFA. The data are reported as the means±SEMs (n=3). (F) Relative CD62L and CD45RO levels in AFA-pretreatment CAR-T cells. The data are reported as the means±SEMs (n=3). (G) The percentages and counts of living CAR-T cells. The data are reported as means±SEMs (n=3). (H) Cytotoxicity of AFA-pretreatment CAR-T cells in media containing E:T at ratio of 1:4, 1:8, or 1:16 for 48 hours, 72 hours, and 96 hours. The data are reported as the means±SEMs (n=3). (I) The expression levels of cytotoxicity-related factors of AFA-pretreatment CAR-T cells were co-cultured with Nalm-6-luc cells. The data are reported as the means±SEMs (n=3). (J) AFA-pretreatment CAR-T cells were categorized into Glu+ and Glu− groups based on the presence or absence of glucose in the culture medium and cultured for 2 days. Cytotoxicity of CAR-T cells in media containing E:T at ratio of 1:4, 1:8, or 1:16 for 72 hours. The data are reported as the means±SEMs (n=3). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001; ns, not significant according to unpaired t-tests (B, and E–J). AFA, afatinib; AKT, protein kinase B; CAR, chimeric antigen receptor; ECAR, extracellular acidification rate; E:T, effector-to-target; FCM, flow cytometry; Glut1, glucose transporter 1; INF, interferon; IL, interleukin; LAT, linker for activation of T cells; LCK, lymphocyte-specific protein tyrosine kinase; mTOR, mechanistic target of rapamycin; NC, negative-control; OCR, oxygen consumption rate; PI3K, phosphoinositide 3-kinase; PLC, phospholipase C; RPS6, ribosomal protein S6; Tcm, central memory T; Tef, terminal effector T cells; Tem, effector memory T cells; Tn, naïve T; TNF, tumor necrosis factor; ZAP 70, zeta-chain-associated protein kinase 70.
Distinct CAR-T cell subsets had specific metabolic requirements. Next, we further evaluated the regulatory effects of AFA on the proliferation and metabolic activity of CD62L+ (Tn and Tcm) and CD62L− (Tef and Tem) CAR-T cell subsets. The AFA pretreatment led the mean fluorescence intensity of carboxy fluorescein diacetate succinimidyl ester in CD62L+CAR T cells lower than CD62L− CAR-T cells, indicating that the proliferation of CD62L+CAR T cells was more rapid than CD62L− (online supplemental figure S2H). Additionally, AFA-induced CD62L+CAR T cells exhibited a decreased ROS content, Mito level, and glucose uptake level in CD62L+CAR T cells, which means AFA pretreatment could regulate the metabolism no matter the CD62L+or CD62L− subsets (online supplemental figure S2I). Low persistence, a major obstacle in CAR-T cell therapy, weakens the in vivo efficacy of CAR-T cells.25 To examine the impact of AFA on the persistence of CAR-T cells, we incubated CAR-T cells in a medium supplemented with AFA and IL-2 for three consecutive days and then replaced the medium (no AFA or IL-2) to assess the number and viability of the cells. AFA-pretreatment CAR-T cells exhibited a more gradual decrease in both viability and cell number than did NC CAR-T cells, indicating that AFA obviously increased the persistence of CAR-T cells (figure 4G). Moreover, CAR-T cells containing high fractions of Tn and Tscm cells are suggested to exhibit superior tumor killing ability.26 To investigate whether the AFA-pretreatment CAR-T cells could exert a more powerful antitumor effect, we co-cultured CAR-T cells treated with AFA for three consecutive days with Nalm-6-luc cells at different E:T ratios (1:4, 1:8, and 1:16). Although both groups of CAR-T cells were capable of killing tumor cells, the AFA-pretreatment CAR-T cells gradually outperformed the NC CAR-T cells as the co-culture period progressed, with the difference in efficacy becoming significantly greater. Compared with that of NC CAR-T cells, the killing efficiency of AFA-pretreatment CAR-T cells increased by approximately 15% (figure 4H). Thereafter, we co-cultured CAR-T cells treated with AFA with Nalm-6-luc cells at an E:T ratio of 1:1 for 4–6 hours and measured the expression ratios of interferon (IFN)-γ, tumor necrosis factor (TNF)-α and CD107a. The consequence presented that these cytokines intensity was increased, which implied the augmented antitumor ability in AFA-pretreatment CAR-T cells (figure 4I). In the tumor microenvironment, tumor cells excessively uptake nutrients. This metabolic competition for resources deprives T cells of essential metabolites such as glucose and amino acids, thereby limiting their ability to effectively immune response against tumors. Interestingly, in a glucose-deprived environment, we observed that AFA-pretreatment CAR-T cells maintained significantly enhanced antitumor efficacy even after co-incubation with Nalm-6-luc cells for 3 days (figure 4J). This finding suggests that AFA-pretreatment CAR-T cells are capable of sustaining their cytotoxic function under metabolic stress. Together, these data demonstrated that application of AFA in vitro reduced terminal differentiation in CAR-T cells, thereby augmenting CAR-T cell persistence and antitumor activity.
AFA regulated CAR-T cell transcriptome signature and inhibited exhaustion
To further study the influence of AFA on the exhaustion, memory, and activation of CAR-T cells, we performed RNA sequencing on CAR-T cells that were co-cultured with AFA. The results revealed distinct expression levels in the transcriptome profile and DEGs between AFA-pretreatment CAR-T cells and NC CAR-T cells (figure 5A,B). To determine the biological functions influenced by AFA, we performed Gene Ontology enrichment analysis. The DEGs were predominantly enriched in the regulation of leukocyte differentiation, alpha-beta T-cell differentiation, T-cell differentiation and cytokine binding, chemokine receptor activity, and G protein-coupled chemoattractant receptor activity, which are critical for biological processes and molecular functions (figure 5C). These findings indicate the modifications in the immune response and immunosurveillance functions, variation in the differentiation process, adjustments in the formation of immune memory, and the transitions in immune cell migration. Next, the Kyoto Encyclopedia of Genes and Genomes pathway analysis revealed 17 significant pathways (p<0.05). The top 10 pathways were cytokine–cytokine receptor interaction, inflammatory bowel disease, rheumatoid arthritis, viral protein interaction with cytokine and cytokine receptor, the IL-17 signaling pathway, hematopoietic cell lineage, antigen processing and presentation, leishmaniasis, malaria and JAK-STAT signaling pathway (figure 5D). Interestingly, we found that two IL-2 receptors, IL-2RA and IL-2RB were both inhibited by AFA treatment. Furthermore, the expression of memory transcripts (eg, BCL-6, TCF7, IL-7R, LEF1, and CCR7) significantly increased in the AFA-pretreatment CAR-T cells, while the expression of multiple effector transcripts, such as granzymes (GZMA and GZMB) and cytokines (IFN-γ), decreased (figure 5E). This was accompanied by a generalized suppression of genes involved in T-cell activation, along with a decrease in the levels of HAVCR2 (Tim-3), PDCD1 (PD-1), LAG-3, and CTLA-4, which are characteristic of the exhaustion phenotype. To corroborate the expression of the aforementioned genes, we validated the protein levels of memory and exhaustion markers, the expression of which was notably upregulated and downregulated, respectively, in CAR-T cells cultured with AFA (figure 5F,G). On the other hand, CAR-T cells exhibited decreased proportion of exhaustion markers compared with that in the NC CAR-T cells (figure 5H). Collectively, these results revealed that AFA altered the transcriptome signature of CAR-T cells and inhibited the exhaustion.
AFA regulated the transcriptome signature and inhibited the exhaustion of CAR-T cells. (A–B) The transcriptome profile and differentially expressed genes of AFA-pretreatment CAR-T cells are shown in the schematic. (C) The red and blue bars illustrate the pathways in which the upregulated or downregulated genes in the AFA group, respectively, are enriched. (D) Kyoto Encyclopedia of Genes and Genomes pathway analysis revealed trends in gene enrichment. (E) Expression of exhaustion-related, activation-related, cytotoxicity-related, and memory-related genes in AFA-pretreatment CAR-T cells, as determined by RNA sequencing. (F–H) Western blot showing the expression of memory (F) and exhaustion (G) markers in AFA-pretreatment CAR-T cells. FCM analysis of the exhaustion level of CAR-T cells treated with AFA (H). The data are reported as the means±SEMs (n=3). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001; ns, not significant according to unpaired t-tests (H). AFA, afatinib; BCL-6, B-cell lymphoma 6; CAR, chimeric antigen receptor; CCR7, C-C chemokine receptor 7; CTLA-4, cytotoxic T-lymphocyte-associated protein 4; IL, interleukin; LAG-3, lymphocyte-activation gene 3; LEF1, lymphoid enhancer-binding factor 1; MFI, mean fluorescence intensity; NC, negative-control; PD-1, programmed cell death protein 1; TCF7, T-cell factor 7; Tim-3, T-cell immunoglobulin and mucin-domain containing-3.
In vivo antitumor activity of CAR-T cells was augmented by AFA pretreatment
To better understanding of the in vivo persistence of AFA-pretreatment CAR-T cells, we monitored the CAR-T dynamics in mice with high-dose and low-dose CAR-T cells injected through tail vein. The results revealed that AFA-pretreatment CAR-T cells exhibited significantly prolonged persistence in vivo whether in the high-dose or low-dose groups (online supplemental figure S2J). We further confirm the in vivo antitumor effects of AFA on CAR-T cells, Nalm-6-luc leukemia mouse model were infused with NC CAR-T cells or AFA-pretreatment CAR-T cells (figure 6A). Despite the two groups of CAR-T cells displaying extraordinary antitumor activity post-Nalm-6-luc inoculation day 20, a notable relapse was observed in the NC CAR-T group on day 30, indicating that AFA-pretreatment CAR-T cells more strongly mitigated the tumor burden (figure 6B). On day 40, severe relapse and substantial mortality were evident in the mice treated with NC CAR-T cells. Although the AFA-pretreatment CAR-T group also exhibited signs of tumor recurrence, the extent of remission provided by this treatment was undeniable (figure 6C). Additionally, AFA-pretreatment CAR-T cells significantly prolonged the survival of recipients after infusion (figure 6D). Accordingly, these data indicate that AFA boosts the sustained antitumor efficacy of CAR-T cells in vivo.
The in vivo antitumor activity of CAR-T cells was augmented by AFA pretreatment. (A) Schematic of the experimental protocol. NOG mice were intravenously injected with 2×106 Nalm-6-luc cells for the leukemia model and intravenously administered Mock T cells, NC CAR-T cells or AFA-pretreatment CAR-T cells 10 days after tumor inoculation. Leukemia burden was measured using an IVIS imaging system. (B) Analysis of tumor clearance in a leukemia model by bioluminescence imaging. ‘‘X’’ represents the endpoint. (C) Quantification of leukemia burden by average luminescence (n=4 mice for the NC group and n=6 mice for the other groups). (D) Survival of the mice in (C). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001; ns, not significant by two-way analysis of variance with a multiple comparison test (C) or log-rank (Mantel-Cox) tests (D). AFA, afatinib; CAR, chimeric antigen receptor; NC, negative-control.
Discussion
Although CAR-T cell therapy has achieved breakthrough advancements in hematological malignancies, it still faces several vexing challenges. Relapse is regarded as the most significant obstacle, and it is induced by the rapid exhaustion and poor persistence of CAR-T cells after refusion.27 Abundant strategies have been explored to inhibit exhaustion and enhance persistence, including reconstructing CAR structures, selecting suitable T-cell resources, and remolding CAR-T cell metabolism, which have led to remarkable improvements in CAR-T cell persistence and antitumor efficacy.28 29 Moreover, several clinical drugs and natural products, such as decitabine,7 enasidenib,30 sulforaphane,31 ginsenoside Rg1,32 and metformin,33 have been shown to improve the antitumor capacity of CAR-T cells via intervening differentiation and strengthening persistence. These combination therapies have offered multifaceted approaches to overcome the current limitations of CAR-T cell therapy.34
In this study, we focused on TKIs, which have showed direct tumoricidal effects and recently emerged immunomodulatory potential.35 By a TKI library screening, we identified AFA as an ideal target to regulate T-cell metabolism and differentiation (figure 1). Mechanically, AFA blocked TCR and PI3K-AKT-mTOR signal, remolded the metabolism pattern via inhibiting glycolysis and enhancing OXPHOS (figure 2). During mitosis wherein progeny T cells exhibiting heightened PI3K-AKT-mTOR signaling undergo rapid differentiation into effector cells while those with diminished PI3K-AKT-mTOR signaling sustain a self-renewal capacity.36 The metabolism reprograming further modulated T-cell differentiation, showed by the increasing ratio of Tn and Tcm and decreasing ratio of Tef and Tem. Tn and Tcm cells are crucial for the initiation and maintenance of immune responses.37 Based on it, we further identified that AFA promoting the T-cell survival, which is a significant indicator for CAR-T cell persistence (figure 3). Therefore, we combined the AFA with CAR-T cells. During the preparation stage, the pretreatment AFA significant enhanced CAR-T cells survival and inhibited exhaustion. When co-cultured with tumor cells, AFA-pretreatment CAR-T cells showed more cytokines secretion, stronger cytotoxicity (figure 4). Further, multianalyses of differentiation characteristics, metabolic profiling, and RNA sequencing revealed that AFA induces comprehensive metabolism remodeling and fate reprogramming of CAR-T cells, which suggests a potential for improvement in CAR-T therapy persistence and long-term efficacy (figure 5). Finally, we verified that the pretreatment of AFA enhances the durability of CAR-T cells in mouse and the antitumor efficacy of CAR-T cell immunotherapy is significantly improved in leukemia mouse model (figure 6).
In summary, our study innovatively supported that AFA pretreatment enables to remold CAR-T cells metabolism and differentiation progression, which induced obviously sustained cell expansion and survival, reduced exhaustion, and enhanced antitumor cytotoxicity. Moreover, AFA has been approved by FDA, which applied the safety in clinical treatment. It is promising to combine AFA with CAR-T cell therapy in recent future for patients with hematological malignancies and even solid tumors.
Conclusion
Our study revealed that AFA could modulate CAR-T metabolism and differentiation, which further inhibit exhaustion and enhance persistence, and significantly improve the antitumor performance in vivo. These findings suggest that traditional medicine combined with CAR-T cells is a novel strategy for improving the efficacy of CAR-T cell immunotherapy.
In brief
AFA boosts CAR-T cell persistence and antitumor therapeutic efficacy by regulation of TCR-PI3K signaling, metabolism and fate reprogramming.
Supplemental material
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Medical Ethics Committee of The Second Affiliated Hospital of Army Medical University (2024-067-01). Participants gave informed consent to participate in the study before taking part. This study involves mouse experiments and was approved by Laboratory Animal Welfare and Ethics Committee of the Army Medical University (AMUWEC20234644).
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
YD and YL contributed equally.
Contributors SL and XZ conceptualized, designed and evaluated experiments. YD, YL, LA and FZ performed experiments. YD and YL wrote the manuscript. SL and XZ are guarantors.
Funding This work was supported by the National Natural Science Foundation of China (No. 82300256 and 82341201), the Natural Science Foundation of Chongqing, China (No. CSTB2022NSCQ-MSX1287), the Science and Technology Project of Chongqing Municipal Education Commission (KJQN202312804), and the Young Doctor Talent Incubation Program of Xinqiao Hospital (No. 2022YQB016), Discipline Talent Development Special Project of Second Affiliated Hospital, Army Medical University (2022XKRC001).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.