Skip to main content
Download PDF

The miR-23a/27a/24 − 2 cluster drives immune evasion and resistance to PD-1/PD-L1 blockade in non-small cell lung cancer

Molecular Cancer volume 23, Article number: 285 (2024) Cite this article

Abstract

Programmed cell death protein ligand-1 (PD-L1) and major histocompatibility complex I (MHC-I) are key molecules related to tumor immune evasion and resistance to programmed cell death protein 1 (PD-1)/PD-L1 blockade. Here, we demonstrated that the upregulation of all miRNAs in the miR-23a/27a/24 − 2 cluster was correlated with poor survival, immune evasion and PD-1/PD-L1 blockade resistance in patients with non-small cell lung cancer (NSCLC). The overexpression of all miRNAs in the miR-23a/27a/24 − 2 cluster upregulated PD-L1 expression by targeting Cbl proto-oncogene B (CBLB) and downregulated MHC-I expression by increasing the level of eukaryotic initiation factor 3B (eIF3B) via the targeting of microphthalmia-associated transcription factor (MITF). In addition, we demonstrated that the expression of the miR-23a/27a/24 − 2 cluster of miRNAs is maintained in NSCLC through increased Wnt/β-catenin signaling-regulated interaction of transcription factor 4 (TCF4) and the miR-23a/27a/24 − 2 cluster promoter. Notably, pharmacologic targeting of the eIF3B pathway dramatically increased sensitivity to PD-1/PD-L1 blockade in patients with high expression of the miR-23a/27a/24 − 2 cluster in NSCLC. This effect was achieved by increasing MHC-I expression while maintaining high expression of PD-L1 induced by the miR-23a/27a/24 − 2 cluster. In summary, we elucidate the mechanism by which the miR-23a/27a/24 − 2 cluster miRNAs maintain their own expression and the molecular mechanism by which the miR-23a/27a/24 − 2 cluster miRNAs promote tumor immune evasion and PD-1/PD-L1 blockade resistance. In addition, we provide a novel strategy for the treatment of NSCLC expressing high levels of the miR-23a/27a/24 − 2 cluster.

Introduction

Lung cancer is the most common malignancy and the leading cause of cancer death worldwide [1]. Non-small cell lung cancer (NSCLC) is the most common type of lung cancer, accounting for 85% of all lung cancer cases [2]. Despite the improvement in treatment in recent years, the mortality of patients with NSCLC is still high, with less than 20% of patients surviving after 5 years [2, 3]. Increasing evidence shows that immune evasion is a critical factor leading to cancer progression and a major obstacle in effective anticancer therapy [4]. Thus, immunotherapy, such as immune checkpoint blockade (ICB), has become an important therapeutic intervention for a range of cancer types, including NSCLC, and promising results have been reported [3, 5]. However, with either ICB alone or in combination with chemotherapy, only a subset of cancer patients achieve long-term remission [3, 6]. Therefore, a clearer understanding of the mechanisms of tumor immune evasion and resistance to ICB is critical for the development of future therapies.

Many factors, including microRNAs (miRNAs), are involved in the regulation of tumor immune evasion. Notably, miRNA clusters synergistically affect tumor immune evasion, as different members of the cluster simultaneously target either the same gene or different genes within the same signaling pathway [7, 8]. The miR-23a/27a/24 − 2 cluster includes miR-23a, miR-27a, and miR-24-2; our previous study revealed that miRNAs in the miR-23a/27a/24 − 2 cluster are upregulated in early-stage NSCLC and promote postoperative recurrence through the activation of Wnt/β-catenin signaling [9]. Importantly, the activation of Wnt/β-catenin signaling is associated with immune escape and resistance to ICB treatment in NSCLC [10]. In addition, studies have shown that the miR-23a/27a/24 − 2 cluster can suppress interferon-γ (IFN-γ) expression and antigen-specific cytotoxicity in CD8+ T cells [11]. Based these findings, we hypothesize that upregulated miRNAs in the miR-23a/27a/24 − 2 cluster may be involved in immune evasion and ICB resistance in NSCLC.

In this study, we demonstrated that upregulated expression of the miR-23a/27a/24 − 2 cluster miRNAs was closely associated with immune evasion and PD-1/PD-L1 blockade resistance in NSCLC. Our data revealed that miRNAs in the miR-23a/27a/24 − 2 cluster significantly inhibited IFN-γ secretion by T cells, suppressed CD8+ T cell infiltration into tumor tissues, downregulated major histocompatibility complex class I (MHC-I) expression and upregulated programmed cell death ligand 1 (PD-L1) expression in NSCLC. Mechanistically, miRNAs in the miR-23a/27a/24 − 2 cluster upregulate PD-L1 expression by targeting the PD-L1 negative regulator Cbl proto-oncogene B (CBLB) and downregulate MHC-I expression by increasing eukaryotic initiation factor 3B (eIF3B) levels through the suppression of its negative regulator, microphthalmia transcription factor (MITF), in NSCLC. Furthermore, we revealed that the miR-23a/27a/24 − 2 cluster miRNAs maintain their expression by increasing the interaction of transcription factor 4 (TCF4) and the miR-23a/27a/24 − 2 cluster promoter through activation of Wnt/β-catenin signaling. Finally, we demonstrated that pharmacologic targeting of the eIF3B pathway or the β-catenin/TCF4 axis can improve the efficacy of PD-1/PD-L1 blockade in NSCLC with high expression of miRNAs in the miR-23a/27a/24 − 2 cluster. Notably, targeting eIF3B more effectively promoted the therapeutic efficacy of PD-1/PD-L1 blockade.

Methods

Human samples and cell culture

Blood was collected from 20 healthy human donors for isolation of peripheral blood mononuclear cells (PBMCs), and NSCLC specimens were collected from 82 patients with NSCLC during surgery or by biopsy (Table S1) at Daping Hospital under a protocol approved by the ethical review committees. All cell lines used in this study was purchased from Shanghai Cell Bank of Chinese Academy of Sciences (Shanghai, China). 293T and Lewis lung cancer (LLC) cells were cultured in Dulbecco`s modified Eagle`s medium supplemented with 10% fetal bovine serum (FBS) (HyClone, Logan, UT). H1299 and H1650 cells were cultured in RPMI1640 medium with 10% of FBS.

Constructs

MITF (EX-B5124-M35), CBLB (EX-Z7996-M35), eIF3B shRNA expressing constructs (HSH066232), and their empty or scramble vectors were purchased from iGene Biotechnology Co., Ltd. (Guangzhou, China). β-catenin expressing construct (HG11279-CF) was obtained from Sino Biological Inc. (Beijing, China). Constructs that expressing miR-23a/27a/24 − 2 cluster or shRNA and scramble were obtained from Genechem Co. Ltd. (Shanghai, China). For miRNA luciferase reporter assay, the 3`-untranslational region (3`-UTR) of CBLB or MITF were amplified from cDNA obtained from human normal lung tissue RNA, and inserted into the MluI and HindIII sites of the pMIR-REPORT™ miRNA expression reporter vector (Thermo Fisher Scientific, Waltham, MA, USA). For promoter luciferase reporter assay, miR-23a/27a/24 − 2 cluster promoter was amplified from human genomic DNA and inserted into the NheI and BglII sites of pGL4.10 reporter vector (Promega, Madison, WI, USA).

Coculture of T cells and tumor cells

PBMCs were isolated from fresh blood of healthy human donors with lymphocyte separation medium according to manufacturer`s protocol (Tian Jin Hao Yang Biological Manufacture Co., Ltd, Tianjin, China), and immune cells were activated as described previously [12]. Then, the CD8+ T cells were sorted using a CD8+ T cells isolation kit (Miltenyi Biotec) according to manufacturer`s instruction.

T cell-induced cytotoxicity was measured as described previously [12]. Briefly, NSCLC cells transfected with vector or miR-23a/27a/24 − 2 cluster expressing constructs. After 48 h of transfection, NSCLC cells were co-cultured with activated T cells at a ratio of 1:20 for 6 h, then cell death was measured by flow cytometry.

CD8+ T cell migration assay was performed as described by Shang et al. [13]. Briefly, NSCLC cells were transfected with vector or miR-23a/27a/24 − 2 cluster overexpressing construct, and cell culture medium was changed after 48 h of transfection. After 48 h of cell culture with fresh medium, the supernatant was transferred into the lower chamber of transwell and 5 × 105 activated CD8+ T cells in 100ul medium that have been incubated with anti-human CD8α APC-Cy7 antibody (Invitrogen, Cat#: A15448) added into the upper of transwell. After 2 h co-incubation the migrated T cells were enumerated by flow cytometry.

To investigate the effects of miR-23a/27a/24 − 2 cluster overexpressing NSCLC cells on the secretion of interferon-gamma (IFN-γ) by T cells, NSCLC cells were transfected with indicated constructs. After 48 h of transfection, NSCLC cells were co-cultured with activated T cells at a ratio of 1:10 for 12 h, then the culture medium was collected for detection of IFN-γ concentration. IFN-γ concentration was measured using human IFN-γ ELISA kit (Invitrogen) according to the manufacturer`s instructions.

Luciferase report assay

For miRNA reporter assay, miR-23a/27a/24 − 2 cluster expressing constructs or vectors were transfected into 293T cells that were transfected with a firefly luciferase reporter constructs containing the 3`-UTR of CBLB or MITF. For miR-23a/27a/24 − 2 cluster promoter luciferase assay, TCF4 expressing constructs or β-catenin expressing constructs or vectors were transfected into 293T cells that were transfected with a firefly luciferase reporter constructs containing miR-23a/27a/24 − 2 cluster promoter. The Renilla luciferase plasmid was cotransfected as a transfection control. After 72 h of transfection, cells were harvested and subjected to measurement of luciferase activity. Luciferase activity was measured with a Dual-luciferase assay kit (Promega).

Western blot (WB), immunohistochemistry (IHC), immunofluorescence (IF), and coimmunoprecipitation (Co-IP) assay

Western blot, IHC, IF and Co-IP were performed as described previously [14]. The WB band intensities were quantified by image J. The assessment of the staining was scored independently by two pathologists without knowledge of the clinicopathological findings. Primary antibodies against to eIF3B (Cat No: bs-14542R, Lot No: BC12254757; for IHC), MITF (Cat No: bsm-51339 M, Lot No: BD06051693), MHC-I (Cat No: bs-18070R, Lot No: BD07151765, for WB; bs-2355R, Lot No: BD07151765, for IHC and IF), and PD-L1 (Cat No: bs-22022R, Lot No: BD7153735, for flow cytometry) were purchased from Beijing Biosynthesis Biotechnology Co., Ltd (Beijing, China). β-catenin (Cat No:8480 S, Lot No:5), Histone H3 (Cat No:9715 S, Lot No:20) and Actin (Cat No:4970 S, Lot No:20) antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA). CD8 (Cat No: ab217344, Lot No:00121531; Cat No: ab17147, Lot No:3258634-5) and PD-L1 (Cat No: ab279292, Lot Not: 1007824-2, for WB) antibodies were purchased from Abcam (Cambridge, MA, USA). CBLB (Cat No: 66353-1, Lot No: 10003654), TCF4 (Cat No: 22337-1-AP, Lot No: 00121531), eIF3B (Cat No: 68202-1, Lot No: 10028876, for WB) and PD-L1 (Cat No: 66248-1, Lot No: 10022147, for IHC) antibodies were from Proteintech (Wuhan, Hubei, China), respectively. Rabbit IgG (Cat No:30000-0-AP, Lot No: 00140749) and mouse IgG (Cat No: BS-0296P. Lot No: BC06261797) was obtained from Proteintech and Beijing Biosynthesis Biotechnology Co., Ltd, respectively.

Real-time quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) and miRNA in situ hybridization (ISH)

Total RNAs were isolated using TRIzol reagent (Beyotime, Shanghai, China), and mRNA expression level of genes was measured by qRT-PCR using SYBR Green One-Step qRT-PCR kit (Beyotime Biotechnology) according to manufacturer`s instruction. Primers of U6, has-miR-23a, has-miR-24-2 and has-miR-27a were purchased from GeneCopoeia (Rockville, MD, USA), and detected their expression using All-in-One™ miRNA qRT-PCR detection kit 2.0 (GeneCopoeia) according to manufacturer`s instruction. The relative expression of miRNA was normalized to U6 expression.

miRNA ISH and score were performed as described by Nuovo [15] and Guo et al. [16], respectively. Briefly, paraffin embedded tissues deparaffinized and antigen retrieval was performed by boiling in citric acid buffer for 15 min. Tissues were further treated with proteinase K for 15 min, and then, treated with pre-hybridization buffer for 1 h at 37℃. Next, tissues were treated with hybridization buffer containing miR-23a or miR-24-2 or miR-27a probes or without probes (negative control) for overnight at 37℃. miRNA probes for ISH were synthesized at Zoonbio Biotechnology Co., Ltd (Nanjing, China). Sequences of primmer and probes used in this study are summarized in Table S2.

RNA sequencing and proteomics analysis

mRNA sequencing and analysis was conducted by Shanghai Genechem Co., Ltd as described by Li et al. [17]. Briefly, using NEBNext Ultra RNA library prep kit constructed the Illumina RNA libraries, and then, sequenced by Illumina NovaSeq platform. Raw data of FASTQ format were processed via In-house Perl scripts, and the Hisat2 was selected as a mapping tool to construct the database of splice junctions following the gene model annotation file.

Proteomics analysis was performed as described previously [18]. Briefly, H1299 cells were transfected with miR-23a/27a/24 − 2 cluster expressing construct or vector. After 72 h of transfection, proteins were extracted from cells and denatured proteins. Proteins were further digested and labeled with iTRAQ reagent. Then, the labeled peptides were fractionated using an Agilent 1260 infinity II HPLC system. Mass spectrometry data were collected on a Q-Exactive plus mass spectrometer (Thermo Fisher Scientific) coupled with an Easy-nLC system (Thermo Fisher Scientific). The data were analyzed using Mascot 2.6 and Proteome Discoverer 2.1.

Electrophoretic mobility shift assay (EMSA)

TCF4 binding site in miR-23a/27a/24 − 2 cluster was predicted using hTFtarget [19]. EMSA was performed using the LightShift chemiluminescence kit (Thermo Fisher Scientific) according to manufacturer`s instruction as described previously [14]. Sequences of probes are shown in Table S2.

Chromatin immunoprecipitation (ChIP)-qPCR assay

For ChIP-qPCR analysis, 293T cells were transfected with β-catenin expressing construct or vector. After 48 h of transfection, cells were cross-linked incubating with 1% of formaldehyde solution at room temperature followed by glycine addition to stop crosslinking. Cells were lysed in ChIP lysis buffer, sonicated, precleared, and incubated overnight with 2ug of TCF4 antibodies or IgG, and precipitated with protein G/A beads. The beads were washed with low and high salt buffer, and immunoprecipitations were eluted with elution buffer at 65℃ for 15 min. Collect the supernatant by centrifugation, add NaCl solution (final concentration is 0.2 M), and incubated at 65℃ for overnight. Then, purified DNA and subjected to qPCR.

Animal study

6-weeks old C57BL/6J female mice and SCID male mice (Beijing HFK Bioscience Co., Ltd, Beijing, China) were used for animal experiments. The effect of miR-23a/27a/24 − 2 cluster on lung cancer immune evasion was examined using C57BL/6J xenograft and SCID xenograft models. For generate C57BL/6J xenograft model, 1 × 106 LLC cells that transfected with empty vector or miR-23a/27a/24 − 2 cluster plasmid or scramble or shRNA of miR-23a/27a/24 − 2 cluster in 100ul of phosphate-buffered saline (PBS) were transplanted subcutaneously on the backs of mice. For generate SCID xenograft model, 2 × 106 H1299 cells that transfected with scramble plasmid or plasmid that expressing shRNA of miR-23a/27a/24 − 2 cluster in 100ul PBS were transplanted subcutaneously on the backs of mice. After 3days of tumor cell transplantation, mice were treated with or without 3 × 106 PBMC through the lateral tail vein injection. The tumor volumes were measured once a week.

The effects of miR-23a/27a/24 − 2 cluster on PD-1/PD-L1 blockade resistance was investigated using C57BL/6J xenograft models. 1 × 106 LLC cells that transfected with empty vector or miR-23a/27a/24 − 2 cluster plasmid in 100ul of PBS were transplanted subcutaneously on the backs of mice. When tumor volumes are reached about 100mm3, mice were treated with PD-L1 mAb (1 mg/Kg body weight) or rat IgG (Bio X cell) every three days by intravenous (i.v.) injection (Fig. S1A).

To investigate the therapeutic effect of LF3 and PD-1/PD-L1 blockade combination on lung cancer with high expressing miR-23a/27a/24 − 2 cluster, C57BL/6J xenograft models were generated using miR-23a/27a/24 − 2 cluster high expressing LLC cells as described above. When tumor volumes are reached about 100mm3, mice were randomly divided into four groups and started to drug treatment. Control group mice were treated with IgG, and other groups were treated with PD-L1 mAb alone (1 mg/Kg body weight, every three days) or LF3 alone (50 mg/Kg body weight, every two days) or combination of PD-L1 mAb and LF3 by i.v. injection (Fig. S1B).

To compare the effects of LF3 and 4EGI-1 treatment on PD-1/PD-L1 blockade therapeutic efficacy, C57BL/6J xenograft models were generated using miR-23a/27a/24 − 2 cluster high expressing LLC cells as described above. When tumor volumes are reached about 100mm3, mice were randomly divided into three groups and started to drug treatment. Control group mice were treated with PD-L1 mAb alone (1 mg/Kg body weight, every three days), other groups were treated with PD-L1 mAb and LF3 or PD-L1 mAb and 4EGI-1 (75 mg/Kg body weight, i.p. 5days a week) (Fig. S1C). All animal experiment complied with the Army Medical University Policy on the Care and Use of Laboratory Animals.

Statistical analysis

Data were presented as the mean ± standard deviation of least three independent experiments. Statistical differences were analyzed using SPSS software. Differences were considered statistically significant at a p value of less than 0.05.

Results

High expression of miRNAs in the miR-23a/27a/24 − 2 cluster is strongly associated with disease progression and immune evasion in NSCLC

To investigate the correlation between the expression levels of miRNAs in the miR-23a/27a/24 − 2 cluster and NSCLC progression, patients with higher expression of all miRNAs in the miR-23a/27a/24 − 2 cluster in tumors compared with adjacent tissues were divided into the high-expression group. In contrast, patients with lower or equal expression levels of all miRNAs in the miR-23a/27a/24 − 2 cluster in tumors compared with adjacent tissues were divided into the low-expression group. Clinical data revealed that the miRNA expression levels of the miR-23a/27a/24 − 2 cluster were negatively correlated with the overall survival rate (Fig. 1A) and the disease-free survival rate (Fig. 1B) in NSCLC patients. In addition, we found that the high-expression group, with high expression in all the miRNAs in the miR-23a/27a/24 − 2 cluster, had a lower number of infiltrating CD8+ T cells in the tumor tissues (Fig. 1C). Furthermore, Gene Ontology (GO) analysis of mRNA sequencing data from the miR-23a/27a/24 − 2 cluster-silenced H1299 cells and control cells revealed that silencing of the miR-23a/27a/24 − 2 cluster affects T-cell activation and differentiation (Fig. 1D). Importantly, gene set enrichment analysis (GSEA) revealed that the miRNA expression levels of the miR-23a/27a/24 − 2 cluster were negatively correlated with T-cell-mediated immunity in NSCLC cells (Fig. 1E). Consistent results were observed from the analysis of the GEO dataset. As shown in Fig. 1F, GEO dataset analysis revealed that the miR-23a/27a/24 − 2 cluster expression level was associated with T-cell activation and differentiation in NSCLC. In addition, target-based miRNA functional analysis [20] revealed that miRNAs in the miR-23a/27a/24 − 2 cluster were associated with immune system processes (Fig. 1G). Taken together, these findings suggest that upregulated miRNAs in the miR-23a/27a/24 − 2 cluster contribute to NSCLC progression by inhibiting T-cell-mediated immunity.

Fig. 1
figure 1

High expression of the miR-23a/27a/24 − 2 cluster closely correlated with poor progression and tumor immune evasion in NSCLC. (A) Kaplan-Maier analysis showing that NSCLC patients with high expression of miR-23a/27a/24 − 2 cluster had shorter overall survival and (B) shorter disease-free survival. Analysis was performed in a cohort of 82 NSCLC patients and the log-rank test used for comparisons. (C) Representative immunohistochemical images of CD8+ T cells and the number of CD8+ T cell infiltration in human NSCLC tissues (100 × 100 μm square area). p value was calculated by t test. (D) Gene Ontology (GO) analysis was performed using mRNA sequencing data that conducted with miR-23a/27a/24 − 2 cluster silenced H1299 cells and their control cells. (E) Gene Set Enrichment Analysis (GSEA) showing that miR-23a/27a/24 − 2 cluster expression level was negatively correlated with T cell medicated immunity in NSCLC cells. GSEA was performed using mRNA sequencing results of miR-23a/27a/24 − 2 cluster silenced H1299 cells and their control cells. (F) GO analysis was performed using GEO dataset GSE151103. (G) miRNA target-based Gene Ontology analysis (http://www.microrna.gr/miRPathv4). miR-cluster, miR-23a/27a/24 − 2 cluster

Full size image

Overexpression of the miR-23a/27a/24 − 2 cluster in NSCLC inhibits the T-cell-mediated immune response

To investigate whether the miR-23a/27a/24 − 2 cluster-overexpressing NSCLC cells directly regulate the T-cell-mediated immune response, T cells were cocultured with miR-23a/27a/24 − 2 cluster-overexpressing NSCLC cells (Fig. S2A), and T-cell function was subsequently assessed. Our results revealed that NSCLC cells with high expression of the miR-23a/27a/24 − 2 cluster significantly reduced T cell released IFN-γ (Fig. 2A and Fig. S3A), T-cell-induced NSCLC cell death (Fig. 2B and Fig. S3B), and T-cell migration (Fig. 2C and Fig. S3C). The effect of the miR-23a/27a/24 − 2 cluster on tumor immune evasion was further investigated in animal models generated from Lewis lung cancer (LLC) cells. Because LLC cells can form tumors in C57BL/6J mice without immune deficiency, this model is used for research related to tumor immunity [21]. In the C57BL/6J xenograft model, the overexpression of all the miRNAs in the miR-23a/27a/24 − 2 cluster (Fig. S2B) in LLC cells significantly promoted tumor growth (Fig. 2D and Fig. S4A), inhibited the infiltration of CD8+ T cells into tumor tissue (Fig. 2E), and suppressed the IFN-γ secretion of T cells (Fig. 2F). In contrast, the silencing of the miR-23a/27a/24 − 2 cluster in LLC cells (Fig. S2C) inhibited tumor growth (Fig. 2G and Fig. S4B) and increased CD8+ T-cell infiltration (Fig. 2H) and IFN-γ secretion (Fig. 2I). These results were further confirmed in other xenograft models. The results revealed that the silencing of the miR-23a/27a/24 − 2 cluster in H1299 cells (Fig. S2D) significantly inhibited tumor growth when the cells were inoculated into immunodeficient SCID mice treated with (Fig. 2J and Fig. S4C) or without peripheral blood mononuclear cells (PBMCs) (Fig. 2L and Fig. S4D). T-cell infiltration was also increased by inhibition of the miR-23a/27a/24 − 2 cluster in SCID model mice treated with PBMCs (Fig. 2K). Notably, PBMC treatment more effectively increased the tumor inhibition rate of miR-23a/27a/24 − 2 cluster silencing in SCID model mice (Fig. 2M). Together, these data indicate a potential role for the miR-23a/27a/24 − 2 cluster of miRNAs in the inhibition of T-cell-mediated antitumor immune responses.

Fig. 2
figure 2

The miR-23a/27a/24 − 2 cluster inhibits T cell mediated immune response in NSCLC. (A) The concentration of IFN-γ in the culture medium after co-culturing CD8+ T cells and NSCLC cells in a 10:1 ratio for 12 h. (B) CD8+ T cell-induced NSCLC cells death was measured by flow cytometry. Indicated NSCLC cells were transfected with vector or miR-23a/27a/24 − 2 cluster, and then co-cultured with activated CD8+ T cells at 1:20 ratio for 6 h. (C) Relative migration of human CD8+ T cells incubated with culture medium supernatant from H1299 or H1650 cells that transfected with vectors or miR-23a/27a/24 − 2 cluster expressing constructs. (D) The growth and weight of subcutaneous tumors in C57BL/6J mouse xenograft model constructed using LLC cells transfected with vectors or miR-23a/27a/24 − 2 cluster expressing constructs. (E) Representative immunofluorescence (IF) image of CD8+ T cells, and the number of infiltrating CD8+ T cells in C57BL/6J xenograft tumors (100 × 100 μm square area). Tumor tissues from (D) (scale bar: 50 μm) and green is CD8+ T cells. (F) IFN-γ concentration in tumor tissues of C57BL/6J xenograft model. Tumor tissues from (D). (G) The growth and weight of subcutaneous tumors in C57BL/6J mouse xenograft model that constructed using LLC cells transfected with scramble or miR-23a/27a/24 − 2 cluster shRNA expressing constructs. (H) Representative IF images of CD8+ T cells and the number of infiltrating CD8+ T cells (100 × 100 μm square area) in C57BL/6J xenograft tumor tissues. Tumor tissues from (G) and green is CD8+ T cells (scale bar: 50 μm). (I) IFN-γ concentration in C57BL/6J xenograft tumor tissues. Tumor tissues from (G). (J) The growth and weight of subcutaneous tumors in SCID mouse xenograft models that treated with PBMCs. Xenograft models were constructed in SCID mouse using H1299 cells transfected with scramble or miR-23a/27a/24 − 2 cluster shRNA expressing constructs. (K) Representative IF images of CD8+ T cells and the number of infiltrating CD8+ T cells (100 × 100 μm square area) in SCID xenograft tumor tissues. Tumor tissues from (J) (scale bar: 50 μm) and red is CD8+ T cells. (L) The growth and weight of subcutaneous tumors in SCID mouse xenograft models without PBMCs treatment. Xenograft models were constructed in SCID mouse using H1299 cells transfected with scramble or miR-23a/27a/24 − 2 cluster shRNA expressing constructs. (M) Tumor inhibition rate of miR-23a/27a/24 − 2 cluster silencing on SCID xenograft models with or without PBMCs treatment. Tumor growth data from (J) and (L). In animal experiments, each group consists of 10 mice. In vitro experiments were conducted in three independent experiments. Data shown are means ± SD. p value was calculated by t test. *, p < 0.05 compared to vector or scramble control; **, p < 0.01 compared to vector or scramble control; ***, p < 0.001 compared to vector or scramble control. LLC, Lewis lung cancer; IFN-γ, interferon-γ; miR-cluster, miR-23a/27a/24 − 2 cluster; sh miR-cluster, shRNA of miR-23a/27a/24 − 2 cluster; PBMCs, peripheral blood mononuclear cells

Full size image

miRNAs in the miR-23a/27a/24 − 2 cluster upregulated PD-L1 expression and downregulated MHC-I expression in NSCLC

To investigate which proteins are involved in the inhibitory effects of miRNAs in the miR-23a/27a/24 − 2 cluster on the T-cell immune response, we performed a proteomics analysis using NSCLC cells with high expression of all miRNAs in the miR-23a/27a/24 − 2 cluster and their corresponding control cells. As shown in Fig. 3A, overexpression of the miR-23a/27a/24 − 2 cluster affects the expression of many proteins, including PD-L1 and MHC-I, which play key roles in immune evasion and immune checkpoint blockade (ICB) resistance. In addition, GO analysis revealed that miR-23a/27a/24 − 2 cluster overexpression was correlated with MHC-I-mediated antigen processing and presentation pathways (Fig. 3A). Notably, Western blot analysis revealed that the miR-23a/27a/24 − 2 cluster upregulated PD-L1 expression and downregulated MHC-I expression in both the presence and absence of IFN-γ (Fig. 3B and Fig. S5A) compared with the control group. These findings suggest that the miR-23a/27a/24 − 2 cluster directly regulates PD-L1 and MHC-I expression in NSCLC cells, beyond simply inhibiting IFN-γ secretion by T cells. We also demonstrated that simultaneously overexpressing all miRNAs in the miR-23a/27a/24 − 2 cluster had a more significant effect on the expression of PD-L1 and MHC-I than single miRNA overexpression did in NSCLC cells (Fig. S7), suggesting that the miRNAs in the miR-23a/27a/24 − 2 cluster have a synergistic effect on the regulation of PD-L1 and MHC-1. Furthermore, the effects of the miR-23a/27a/24 − 2 cluster on the expression of PD-L1 and MHC-I were confirmed by flow cytometry (Fig. 3C) and immunofluorescence analysis (Fig. 3D). Consistent results were observed in the proteomic analysis. In addition, immunohistochemistry (IHC) analysis of tumor tissues from a lung cancer model constructed using LLC cells from C57BL/6J mice revealed high expression of PD-L1 and low expression of MHC-I in the miR-23a/27a/24 − 2 cluster-overexpressing group (Fig. 3E). These results were further confirmed in human NSCLC samples via IHC analysis (Fig. 3F). Collectively, these data suggest that the miR-23a/27a/24 − 2 cluster may inhibit the T-cell-mediated tumor immune response by upregulating PD-L1 and downregulating MHC-I in NSCLC.

Fig. 3
figure 3

The miR-23a/27a/24 − 2 cluster regulated PD-L1 and MHC-I expression in NSCLC. (A) Heatmap and Gene Ontology analysis were performed using proteomics data from miR-23a/27a/24 − 2 cluster overexpressing H1299 cells and their control cells. (B) Western blot analysis showing the miR-23a/27a/24 − 2 cluster increased PD-L1 and downregulated MHC-I expression in NSCLC cells at both absence or presence of 20 ng/ml IFN-γ (24 h). (C) Flow cytometry analysis, and (D) Immunofluorescence analysis (scale bar: 20 μm) show the miR-23a/27a/24 − 2 cluster downregulates MHC-I, while, upregulates PD-L1 expression in NSCLC cells. Indicated NSCLC cells were transfected with vector or miR-23a/27a/24 − 2 cluster expressing constructs. After 72 h of transfection, cells were subjected to analysis (B-D). Three independent experiments were conducted. (E) Representative immunohistochemical images and IHC scores of PD-L1 and MHC-I in C57BL/6J xenograft tumors. Tumor tissues from Fig. 2D (scale bar: 50 μm). (F) Representative immunohistochemical images and IHC scores of PD-L1 and MHC-I in tumor tissues that from NSCLC patients with high level miR-23a/27a/24 − 2 cluster (n = 67) or low level miR-23a/27a/24 − 2 cluster (n = 15) (scale bar: 50 μm). p value was calculated by Wilcoxon signed rank test (E) and Wilcoxon rank sum test (F). miR-cluster, miR-23a/27a/24 − 2 cluster; miR-OE, miR-23a/27a/24 − 2 cluster overexpression; CTRL, control; miR high, high expression of miRNAs in miR-23a/27a/24 − 2 cluster; miR low, low expression of miRNAs in miR-23a/27a/24 − 2 cluster

Full size image

miRNAs in the miR-23a/27a/24 − 2 cluster upregulated the expression of PD-L1 in NSCLC by targeting CBLB

Next, we investigated the regulatory mechanism of the miR-23a/27a/24 − 2 cluster of miRNAs on PD-L1 expression in NSCLC. Proteomic analysis revealed that CBLB was downregulated in the miR-23a/27a/24 − 2 cluster overexpression group (Fig. 3A), which is a negative regulator of PD-L1 in NSCLC [22]. The inhibitory effects of miRNAs in the miR-23a/27a/24 − 2 cluster on CBLB expression in NSCLC cells at the protein (Fig. 4A, Fig. S5B and Fig. S6) and mRNA levels (Fig. 4B) were further confirmed by Western blot and qRT‒PCR, respectively. Notably, our data revealed that CBLB overexpression suppressed the miRNA expression of the miR-23a/27a/24 − 2 cluster-induced upregulation of PD-L1 in NSCLC cells (Fig. 4C and Fig. S5C), suggesting that the miRNAs in the miR-23a/27a/24 − 2 cluster upregulate PD-L1 expression by inhibiting CBLB. In fact, we used computational algorithm analysis (targetscan.org) to predict that the 3`-untranslated region (3`-UTR) of CBLB contains sequences that can bind with miR-23a and miR-27a (Fig. 4D). Importantly, a luciferase reporter assay revealed that the overexpression of the miR-23a/27a/24 − 2 cluster significantly suppressed the luciferase level regulated by the wild-type 3`-UTR of CBLB but did not affect the luciferase expression level regulated by the mutant 3`-UTR of CBLB (Fig. 4E). The negative and positive correlations of the miR-23a/27a/24 − 2 cluster with CBLB and PD-L1 expression, respectively, were further confirmed by IHC analysis of tumor tissues from C57BL/6J lung cancer models constructed from miR-23a/27a/24 − 2 cluster-overexpressing LLC cells (Fig. 4F) and patients with NSCLC (Fig. 4G). Taken together, these findings indicate that miRNAs in the miR-23a/27a/24 − 2 cluster upregulate PD-L1 expression in NSCLC through the inhibition of CBLB by directly targeting the 3`-UTR of CBLB.

Fig. 4
figure 4

The miR-23a/27a/24 − 2 cluster upregulate PD-L1 expression by targeting CBLB in NSCLC. (A) Western blot and, (B) qRT-PCR analysis show that overexpression of miR-23a/27a/24 − 2 cluster downregulate CBLB expression at both protein and mRNA levels. (C) Overexpression of CBLB inhibited miR-23a/27a/24 − 2 cluster-induced upregulation of PD-L1 in H1299 cells. Indicated NSCLC cells were transfected with vector or miR-23a/27a/24 − 2 cluster expressing constructs or/and CBLB expressing constructs. After 72 h of transfection, cells were subjected to analysis. Three independent experiments were conducted (A-C). (D) Sequence alignment of miRNAs in the miR-23a/27a/24 − 2 cluster with the 3′ UTRs of the CBLB. (E) miRNA luciferase reporter assay showing that overexpression of miR-23a/27a/24 − 2 cluster suppressed wild type 3′ UTR of CBLB-regulated luciferase expression in NSCLC cells, but did not affected mutated 3′ UTRs of CBLB-regulated luciferase expression. (F) Representative immunohistochemical (IHC) images and IHC scores of CBLB and PD-L1 in tumors that from C57BL/6J xenograft. Tumor tissues from Fig. 2D (scale bar: 50 μm). (G) Representative IHC images and scores of CBLB and PD-L1 in tumor tissues that from NSCLC patients with high level miR-23a/27a/24 − 2 cluster (n = 67) or low level miR-23a/27a/24 − 2 cluster (n = 15) (scale bar: 50 μm). Data shown are means ± SD. p value was calculated by t test (B and E), Wilcoxon signed rank test (F) and Wilcoxon rank sum test (G-left) and one way ANOVA test (G-right). miR-cluster, miR-23a/27a/24 − 2 cluster; miR high, high expression of miR-23a/27a/24 − 2 cluster miRNAs; miR low, low expression of miR-23a/27a/24 − 2 cluster miRNAs

Full size image

miRNAs in the miR-23a/27a/24 − 2 cluster inhibit the expression of MHC-I in NSCLC by targeting MITF

We also investigated the regulatory mechanism of the miR-23a/27a/24 − 2 cluster on MHC-I expression. Proteomic results revealed that eIF3B was upregulated in the miR-23a/27a/24 − 2 cluster overexpression group (Fig. 3A). In addition, computational algorithm analysis revealed that the 3`-UTR of MITF contains sequences that can bind with all miRNAs in the miR-23a/27a/24 − 2 cluster (Fig. 5A). According to Santasusagna et al., MITF inhibits the expression of eIF3B, whereas eIF3B inhibits the expression of MHC-I [23], suggesting that miRNAs in the miR-23a/27a/24 − 2 cluster in NSCLC may downregulate MHC-I expression through the upregulation of eIF3B by targeting MITF. As expected, the in vitro experimental results revealed that the overexpression of the miR-23a/27a/24 − 2 cluster dramatically downregulated the expression of MITF and MHC-I in NSCLC cells but upregulated eIF3B expression (Fig. 5B, Fig. S5D and Fig. S6). Additionally, downregulated expression of MITF by the miR-23/27a/24 − 2 cluster at the mRNA level was demonstrated in NSCLC cells (Fig. 5C). Importantly, overexpression of MITF or silencing of eIF3B restored MHC-I expression inhibited by the miR-23a/27a/24 − 2 cluster (Fig. 5D and Fig. S5E). The miR-23a/27a/24 − 2 cluster-induced upregulation of eIF3B was also inhibited by the overexpression of MITF (Fig. 5D). Furthermore, we used a luciferase reporter assay to investigate whether miRNAs in the miR-23a/27a/24 − 2 cluster inhibit MITF expression by targeting the 3`-UTR of MITF. As shown in Fig. 5E, overexpression of the miR-23a/27a/24 − 2 cluster significantly suppressed the luciferase level regulated by the wild-type 3`-UTR of MITF but did not affect the luciferase expression level regulated by the mutant 3`-UTR of MITF. Finally, the correlations among the expression levels of the miR-23a/27a/24 − 2 cluster, MITF, eIF3B and MHC-I were further confirmed via IHC analysis of tumors from C57BL/6J xenograft models generated from miR-23a/27a/24 − 2 cluster-overexpressing LLC cells (Fig. 5F) and patients with NSCLC (Fig. 5G). Collectively, these findings suggest that miRNAs in the miR-23a/27a/24 − 2 cluster inhibit MHC-I expression in NSCLC through the upregulation of eIF3B by directly targeting MITF.

Fig. 5
figure 5

The miR-23a/27a/24 − 2 cluster suppressed MHC-I expression through MITF/eIF3B axis in NSCLC. (A) Sequence alignment of miRNAs in the miR-23a/27a/24 − 2 cluster with the 3′-UTRs of MITF. (B) Western blot analysis showed that overexpression of miR-23a/27a/24 − 2 cluster suppressed MITF and MHC-I expression and upregulated eIF3B expression in NSCLC cells. (C) qRT-PCR analysis showed that overexpression of miR-23a/27a/24 − 2 cluster in NSCLC cells inhibited MITF expression at mRNA levels. (D) Western blot analysis showed that miR-23a/27a/24 − 2 cluster-induced downregulation of MHC-I was restored by MITF overexpression or silencing of eIF3B in H1299 cells. Indicated NSCLC cells were transfected with indicated constructs. After 72 h of transfection, cells were subjected to analysis. Three independent in vitro experiments were conducted (B-D). (E) miRNA luciferase reporter assay showing that overexpression of miR-23a/27a/24 − 2 cluster suppressed 3′ UTRs of MITF-regulated luciferase expression in NSCLC cells, but did not affected mutated 3′ UTRs of MITF-regulated luciferase expression. (F) Representative immunohistochemical (IHC) images and IHC scores of MITF, eIF3B and MHC-I in C57BL/6J xenograft tumors. Tumor tissues from Fig. 2D (scale bar: 50 μm). (G) Representative IHC images and scores of MITF, eIF3B and MHC-I in tumor tissues that from NSCLC patients with high (n = 67) or low level miR-23a/27a/24 − 2 cluster (n = 15) (scale bar: 50 μm). Data are shown means ± SD. p value was calculated by t test (C and D), Wilcoxon signed rank test (F) and Wilcoxon rank sum test (G). miR-cluster, miR-23a/27a/24 − 2 cluster; miR high, high expression of miR-23a/27a/24 − 2 cluster miRNAs; miR low, low expression of miR-23a/27a/24 − 2 cluster miRNAs

Full size image

miRNAs in the miR-23a/27a/24 − 2 cluster maintain their expression through the β-catenin/TCF4 axis in NSCLC

To investigate the regulatory mechanism of the miR-23a/27a/24 − 2 cluster of miRNAs in NSCLC, we screened candidate transcription factors that may be involved in the regulation of miR-23a/27a/24 − 2 cluster expression and found that TCF4 can bind to the promoter of the miR-23a/27a/24 − 2 cluster (Fig. 6A). TCF4 is an effector of Wnt/β-catenin signaling and is involved in miRNA expression by binding to the promoter region of miRNAs [24]. In addition, our previous study revealed that miRNAs in the miR-23a/27a/24 − 2 cluster stimulate TCF/LEF complex-regulated gene expression by activating Wnt/β-catenin signaling in NSCLC [9]. Thus, we propose that the miR-23a/27a/24 − 2 cluster maintained its expression through the β-catenin/TCF4 axis. To test this hypothesis, we first confirmed that TCF4 regulates miR-23a/27a/24 − 2 cluster expression. Our results revealed that overexpression of TCF4 significantly upregulated the expression of miR-23a, miR-27a and miR-24-2 (Fig. 6B) and promoted luciferase transcription via the miR-23a/27a/24 − 2 cluster promoter (Fig. 6C). In addition, electrophoretic mobility shift assay (EMSA) results revealed that TCF4 directly interacts with the miR-23a/27a/24 − 2 cluster promoter site, which contains TCF4 binding sequences, whereas mutation of the binding sequence significantly reduced the interaction between TCF4 and the miR-23a/27a/24 − 2 cluster promoter (Fig. 6D and Fig. S5F). Collectively, these results indicate that TCF4 promotes miRNA expression in the miR-23a/27a/24 − 2 cluster by interacting with the miR-23a/27a/24 − 2 cluster promoter. Next, we investigated whether β-catenin promotes TCF4-mediated expression of miR-23a/27a/24 − 2 cluster miRNAs. Our results revealed that overexpression of β-catenin increased the interaction between TCF4 and the miR-23a/27a/24 − 2 cluster promoter (Fig. 6E, F and Fig. S5G), regulated luciferase transcription from the miR-23a/27a/24 − 2 cluster promoter (Fig. 6G), and enhanced miRNA expression of the miR-23a/27a/24 − 2 cluster (Fig. 6H). Notably, inhibition of the interaction between β-catenin and TCF4 by the small molecule LF3 dramatically inhibited the β-catenin overexpression-induced upregulation of miR-23a/27a/24 − 2 cluster miRNAs (Fig. 6I). Together, these findings indicate that the miR-23a/27a/24 − 2 cluster maintains its expression through the β-catenin/TCF4 axis in NSCLC.

Fig. 6
figure 6

The miR-23a/27a/24 − 2 cluster maintains its expression through β-catenin/TCF4 axis in NSCLC. (A) The sequences in the promoter region of miR-23a/27a/24 − 2 cluster that may bind to TCF4. (B) qRT-PCR analysis showed that overexpression of TCF4 increases the expression of all miRNAs at the miR-23a/27a/24 − 2 cluster in 293T cells. (C) Luciferase reporter assay showed that overexpression of TCF4 increased luciferase expression in 293T cells that regulated by wild type miR-23a/27a/24 − 2 cluster promoter. (D) EMSA indicate that interaction between TCF4 and miR-23a/27a/24 − 2 cluster promoter in 293T cells. (E) ChIP-qPCR results showing that overexpression of β-catenin increased the interaction of TCF4 and miR-23a/27a/24 − 2 cluster promoter in 293T cells. **, p < 0.01 compared to Vector control (IgG); ##, p < 0.01 compared to Vector control (TCF4 Ab). (F) EMSA showing that overexpression of β-catenin increased the interaction of TCF4 and miR-23a/27a/24 − 2 cluster promoter in 293T cells. (G) Luciferase report assay showing that overexpression of β-catenin increased wild type promoter of miR-23a/27a/24 − 2 cluster regulated luciferase expression in 293T cells, but did not affect mutated promoter of miR-23a/27a/24 − 2 cluster regulated luciferase expression. (H) qRT-PCR analysis showed that β-catenin overexpression increased all miRNAs expression in miR-23a/27a/24 − 2 cluster. (I) qRT-PCR analysis showed that inhibition of β-catenin and TCF4 interaction by 30 μm LF3 (6 h) suppressed β-catenin-induced upregulation of miRNAs at miR-23a/27a/24 − 2 cluster in 293T cells. 293T cells were transfected with indicated constructs for 72 h, and then subjected to analysis (B-H). Data shown are means ± SD. p value was calculated by t test (B, C, G and H) or one-way ANOVA and Duncan`s multiple range test (E and I). miR-cluster, miR-23a/27a/24 − 2 cluster; ns, no significance; WT, wild type; MT, mutant; WCL, whole cell lysate

Full size image

Targeting the eIF3B pathway dramatically enhances the therapeutic efficacy of PD-1/PD-L1 blockade in lung cancers with high expression of miRNAs in the miR-23a/27a/24 − 2 cluster

Given that miRNAs in the miR-23a/27a/24 − 2 cluster modulate the expression of key molecules associated with PD-1/PD-L1 blockade therapy efficacy, we investigated whether high expression of miRNAs in the miR-23a/27a/24 − 2 cluster leads to resistance to PD-1/PD-L1 blockade. As expected, the overexpression of miR-23/27a/24 − 2 cluster miRNAs reduced the sensitivity of tumors to PD-L1 mAb treatment (Fig. 7A) and suppressed the infiltration of CD8+ T cells into tumors in C57BL/6J model mice generated from LLC cells (Fig. 7B).

Fig. 7
figure 7

Overexpression of miR-23a/27a/24 − 2 cluster induced PD-1/PD-L1 blockade resistance in C57BL/6J xenograft models that constructed by LLC cells, and targeting eIF3B recovered miR-23a/27a/24 − 2 cluster-induced ICB resistance. (A) Tumor weight of vector or miR-23a/27a/24 − 2 cluster overexpressing LLC tumors in C57BL/6J mice that treated with IgG or PD-L1 mAb. (B) Representative immunofluorescence (IF) images of CD8+ T cells, and the number of infiltrating CD8+ T cells (100 × 100 μm square area; scale bar: 50 μm) in C57BL/6J xenograft tumor tissues. Tumor tissues from (A). (C) Co-IP analysis show that LF3 significantly reduced the interaction of β-catenin and TCF4 in C57BL/6J xenograft tumors. Tumor tissues from (D). (D) tumor volumes, (E) tumor weight and, (F) miR-23a/27a/24 − 2 cluster expression in C57BL/6J xenograft tumors that treated with indicated drugs. C57BL/6J xenograft models were generated with miR-23a/24a/27 − 2 cluster overexpressing LLC cells. When mean tumor volumes were reached about 100mm3, mice were treated with IgG or PD-L1 mAb or LF3 or combination of PD-L1 mAb and LF3. (G) Representative immunohistochemical (IHC) images and IHC scores of PD-L1 and MHC-I in C57BL/6J xenograft tumors from indicated groups. Tumor tissues from (D) (scale bar: 50 μm). (H) Representative IF images of CD8+ T cells and the number of infiltrating CD8+ T cells (100 × 100 μm square area) in C57BL/6J xenograft tumor tissues that from indicated groups. Tumor tissues from (D) (scale bar: 50 μm). (I) Tumor weight, (J) representative IF mages of CD8+ T cells (scale bar: 50 μm) and the number of infiltrating CD8+ T cells (100 × 100 μm square area), and (K) representative IHC images (scale bar: 50 μm) and IHC scores of PD-L1 and MHC-I in C57BL/6J xenograft tumors from indicated groups. miR-23a/27a/24 − 2 cluster high expressing C57BL/6J xenograft models constructed by miR-23a27a/24 − 2 cluster overexpressing LLC cells and mice were treated with indicated drugs when tumor volume reached about 100mm3. After 18 days from first day of drug treatment tumors were collected and subjected to analysis (C-K). Each group of animal experiments consists of 10 mice. Data shown are means ± SD. p value was calculated by one-way ANOVA and Duncan`s multiple range test (A, B, D, E, F, H, I and J), and Kruskal-Wallis test and Dunn`s test (G and K). miR-cluster, miR-23a/27a/24 − 2 cluster; miR-cluster OE, overexpression of miR-23a/27a/24 − 2 cluster; PD-L1 mAb, PD-L1 monoclonal antibodies; ***, p < 0.001 compared to IgG group; ###, p < 0.001, compared to single drug treatment group

Full size image

Previous studies have shown that activation of the β-catenin pathway [25] and the eIF3B/MHC-I pathway [23] in cancer are key factors contributing to resistance to PD-1/PD-L1 blockade therapy. Furthermore, given that high expression of the miR-23a/27a/24 − 2 cluster contributes to the resistance of NSCLC to PD-1/PD-L1 blockade and that the miR-23a/27a/24 − 2 cluster maintains its expression through the β-catenin/TCF4 axis and downregulates MHC-I expression through the eIF3B pathway, we investigated whether these axes could be pharmacologically targeted to increase the therapeutic efficacy of PD-1/PD-L1 blockade in lung cancer patients with high expression of the miR-23a/27a/24 − 2 cluster. The results of animal experiments revealed that blockade of the interaction between β-catenin and TCF4 by LF3 (Fig. 7C and Fig. S5H) significantly suppressed the growth of tumors with high expression of miR-23a/27a/24 − 2 cluster miRNAs compared with that of the control group (Fig. 7D-E and Fig. S4E). Notably, compared with single drug treatment, the combination of LF3 and the PD-L1 mAb more significantly inhibited tumor growth in lung cancers with high expression of the miR-23a/27a/24 − 2 cluster (Fig. 7D-E). In addition, LF3 treatment suppressed miRNA expression in the miR-23a/27a/24 − 2 cluster (Fig. 7F) and PD-L1 expression and upregulated MHC-I expression (Fig. 7G). Notably, the combination of the PD-L1 mAb and LF3 more significantly increased the infiltration of CD8+ T cells into tumor tissues (Fig. 7H). Next, we compared the effects of targeting the β-catenin/TCF4 axis and the eIF3B pathway in PD-1/PD-L1 blockade therapy. Because the small molecule 4EGI-1 impedes eIF3B binding to translation initiation assembly and inhibits the MITF/eIF3B pathway [23, 26], we used 4EGI-1 as an inhibitor of the eIF3B pathway. The results of the animal experiments revealed that targeting the eIF3B pathway with 4EGI-1 more significantly enhanced PD-L1 mAb therapeutic efficacy in lung cancers with high expression of the miR-23a/27a/24 − 2 cluster, compared with inhibiting the β-catenin and TCF4 interaction with LF3 treatment (Fig. 7I). In addition, we demonstrated that the combination of 4EGI-1 and PD-L1 mAb significantly increased CD8+ T-cell infiltration into tumor tissues (Fig. 7J). We also demonstrated that 4EGI-1 treatment upregulated MHC-I expression but did not suppress the miR-23a/27a/24 − 2 cluster-induced increase in PD-L1 expression (Fig. 7K). Taken together, these data indicate that targeting eIF3B can significantly increase the therapeutic efficacy of PD-1/PD-L1 blockade in miR-23a/27a/24 − 2 cluster-overexpressing lung cancer.

Discussion

In this study, we investigated the molecular mechanism by which miRNAs in the miR-23a/27a/24 − 2 cluster lead to immune evasion and PD-1/PD-L1 blockade resistance in NSCLC. Increasing evidence has shown that PD-L1 [27] and MHC-I [28] are key molecules associated with the tumor immune response. PD-L1 is an immune checkpoint protein that is expressed on the surface of tumor cells and inactivates T cells by binding to PD-1 on the T-cell surface. Thus, abnormally high expression of PD-L1 is an important mechanism for cancer immune evasion. In addition, downregulation of MHC-I is an important tumor immune evasion mechanism. MHC-I loads peptides and presents them on the surface of cells, which provides the immune system with the signal to detect and eliminate abnormal cells [29]. When peptides loaded with MHC-I are tumor-associated antigens, CD8+ T cells are activated following antigen recognition and subsequently induce immune killing of cancer cells [29]. Unfortunately, low expression or loss of MHC-I is a frequent event in various cancers, including lung cancer [30, 31], which leads to T cells’ inability to detect cancer cells, allowing cancer cells to survive [31]. In this study, we demonstrated that overexpression of miRNAs in the miR-23a/27a/24 − 2 cluster promotes NSCLC immune evasion and PD-1/PD-L1 blockade resistance. Our data revealed that the miR-23a/27a/24 − 2 cluster upregulated PD-L1 expression in NSCLC by directly targeting the PD-L1 negative regulator CBLB and downregulated MHC-I expression through the upregulation of the MHC-I negative regulator eIF3B by directly targeting MITF. The promoting effect of miRNAs in the miR-23a/27a/24 − 2 cluster on tumor immune evasion has also been reported in other tumors. A previous study revealed that miR-27a inhibits MHC-I expression and CD8+ T-cell infiltration in colorectal cancer [28]. Lin et al. reported that miR-23a is upregulated in CD8+ cytotoxic T lymphocytes and that targeting miR-23a in CD8+ cytotoxic T lymphocytes prevents tumor-dependent immunosuppression in lung cancer [32]. In addition, studies have shown that upregulated expression of miR-23a and miR-27a in macrophages contributes to cancer immune evasion by upregulating PD-L1 expression through the PTEN/PIK3 pathway in liver cancer [33] and breast cancer [34], respectively. Taken together, these findings indicate that the upregulated expression of miRNAs in the miR-23a/27a/24 − 2 cluster promotes tumor immune evasion by upregulating PD-L1 and downregulating MHC-I. Our study revealed a novel mechanism by which PD-L1 is upregulated and MHC-I expression is downregulated in NSCLC.

In this study, we further elucidated the molecular mechanism by which miRNAs in the miR-23a/27a/24 − 2 cluster maintain high expression in NSCLC. Abnormally upregulated expression of miRNAs in the miR-23a/27a/24 − 2 cluster has been identified in approximately 50% of patients with early-stage NSCLC and is closely correlated with disease progression [9]; however, the mechanism of high expression in the miR-23a/27a/24 − 2 cluster is not clear. In this study, we demonstrated that TCF4 increases miR-23a/27a/24 − 2 cluster expression by directly binding to its promoter region and that this interaction is promoted by β-catenin. Importantly, our previous study revealed that miRNAs in the miR-23a/27a/24 − 2 cluster promote TCF/LEF complex-mediated gene expression through activating Wnt/ β-catenin signaling by targeting negative regulators of Wnt/ β-catenin signaling in NSCLC [9]. Collectively, these findings suggest that the miR-23a/27a/24 − 2 cluster maintains its expression levels by activating the β-catenin/TCF4 axis in NSCLC. In addition, these results suggest that blocking this feedback loop may be a useful strategy for treating NSCLC with high expression in the miR-23a/27a/24 − 2 cluster. In fact, our previous study revealed that downregulation of the miR-23a/27a/24 − 2 cluster or silencing of β-catenin can inhibit tumor progression in NSCLC with high expression in the miR-23a/27a/24 − 2 cluster [9].

Finally, we propose a novel strategy for treating NSCLC with high expression in the miR-23a/27a/24 − 2 cluster. Our studies revealed that the miR-23a/27a/24 − 2 cluster activates β-catenin signaling [9] and the eIF3B pathway and that both of these pathways contribute to PD-1/PD-L1 blockade resistance [23, 35]. In this study, we demonstrated that both targeting the β-catenin/TCF4 axis and the eIF3B pathway can significantly increase PD-L1 mAb therapeutic efficacy. However, targeting the eIF3B pathway more effectively enhances PD-L1 mAb therapeutic efficacy than does targeting the β-catenin/TCF4 axis. These results may be caused by the differential regulation of PD-L1 and MHC-I expression by the β-catenin/TCF4 and eIF3B pathways. Clinical studies have shown that the success of PD-1/PD-L1 blockade is closely correlated with the expression level of PD-L1 in tumor cells [36, 37]. Zhang et al. reported that upregulation of PD-L1 expression can enhance the therapeutic effects of PD-1/PD-L1 blockade [38]. Our data showed that targeting the β-catenin/TCF4 axis upregulated MHC-I expression but decreased the expression of PD-L1. In contrast, targeting the eIF3B pathway did not inhibit the miR-23a/27a/24 − 2 cluster-induced increase in PD-L1 expression but significantly upregulated MHC-I expression.

Conclusion

In summary, miRNAs in the miR-23a/27a/24 − 2 cluster maintain their expression through activating the β-catenin/TCF4 axis in NSCLC. High expression of miRNAs in the miR-23a/27a/24 − 2 cluster upregulated PD-L1 expression by targeting the PD-L1 negative regulator CBLB and downregulated MHC-I expression through upregulating eIF3B expression by targeting MITF (Fig. 8 left). In addition, this study provides a therapeutic strategy in which targeting the eIF3B pathway can significantly increase PD-1/PD-L1 blockade therapeutic efficacy in lung cancer with high expression of miRNAs in the miR-23a/27a/24 − 2 cluster, by restoring MHC-I expression and maintaining high PD-L1 expression induced by the miR-23a/27a/24 − 2 cluster (Fig. 8 right).

Fig. 8
figure 8

Working models for miRNAs of miR-23a/27a/24 − 2 cluster maintains their expression and promotes tumor immune evasion (left), and for targeting eIF3B sensitizes tumors to PD-1/PD-L1 blockade in miR-23a/27a/24 − 2 cluster high expressing tumors (Right)

Full size image

Data availability

Data is provided within the manuscript or supplementary information files.

References

  1. Bray F, Laversanne M, Sung HYA, Ferlay J, Siegel RL, Soerjomataram I, Jemal A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. Ca-Cancer J Clin. 2024;74(3):229–63.

    Article  PubMed  Google Scholar 

  2. Cronin KA, Lake AJ, Scott S, Sherman RL, Noone AM, Howlader N, Henley SJ, Anderson RN, Firth AU, Ma JM et al. Annual Report to the Nation on the Status of Cancer, part I: National cancer statistics. Cancer-Am Cancer Soc. 2018;124(13):2785–2800.

  3. Zavitsanou AM, Pillai R, Hao Y, Wu WL, Bartnicki E, Karakousi T, Rajalingam S, Herrera A, Karatza A, Rashidfarrokhi A, et al. KEAP1 mutation in lung adenocarcinoma promotes immune evasion and immunotherapy resistance. Cell Rep. 2023;42(11):113295.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Vinay DS, Ryan EP, Pawelec G, Talib WH, Stagg J, Elkord E, Lichtor T, Decker WK, Whelan RL, Kumara HMCS, et al. Immune evasion in cancer: mechanistic basis and therapeutic strategies. Semin Cancer Biol. 2015;35:S185–98.

    Article  PubMed  Google Scholar 

  5. Cable J, Greenbaum B, Pe’er D, Bollard CM, Bruni S, Griffin ME, Allison JP, Wu CJ, Subudhi SK, Mardis ER et al. Frontiers in cancer immunotherapy-a symposium report. Ann Ny Acad Sci: 2021, 1489(1):30–47.

  6. Sharma P, Hu-Lieskovan S, Wargo JA, Ribas A. Primary, adaptive, and Acquired Resistance to Cancer Immunotherapy. Cell. 2017;168(4):707–23.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Cioffi M, Trabulo SM, Vallespinos M, Raj D, Kheir TB, Lin ML, Begum J, Baker AM, Amgheib A, Saif J et al. The miR-25-93-106b cluster regulates tumor metastasis and immune evasion via modulation of CXCL12 and PD-L1. Oncotarget. 2017;8(13):21609–21625.

  8. Kundu ST, Rodriguez BL, Gibson LA, Warner AN, Perez MG, Bajaj R, Fradette JJ, Class CA, Solis LM, Alvarez FRR, et al. The microRNA-183/96/182 cluster inhibits lung cancer progression and metastasis by inducing an interleukin-2-mediated antitumor CD8 cytotoxic T-cell response. Gene Dev. 2022;36(9–10):582–600.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Fan XQ, Tao SL, Li Q, Deng B, Tan QY, Jin H. The miR-23a/27a/24 – 2 cluster promotes postoperative progression of early-stage non-small cell lung cancer. Mol Ther-Oncolytics. 2022;24:205–17.

    Article  PubMed  Google Scholar 

  10. Muto S, Enta A, Maruya Y, Inomata S, Yamaguchi H, Mine H, Takagi H, Ozaki Y, Watanabe M, Inoue T, et al. Wnt/β-Catenin signaling and resistance to Immune Checkpoint inhibitors: from non-small-cell lung Cancer to other cancers. Biomedicines. 2023;11(1):190.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Chandran PA, Keller A, Weinmann L, Seida AA, Braun M, Andreev K, Fischer B, Horn E, Schwinn S, Junker M, et al. The TGF-β-inducible miR-23a cluster attenuates IFN-γ levels and antigen-specific cytotoxicity in human CD8 T cells. J Leukoc Biol. 2014;96(4):633–45.

    Article  PubMed  Google Scholar 

  12. Li Q, Zhou ZW, Lu J, Luo H, Wang SN, Peng Y, Deng MS, Song GB, Wang JM, Wei X, et al. PD-L1P146R is prognostic and a negative predictor of response to immunotherapy in gastric cancer. Mol Ther. 2022;30(2):621–31.

    Article  PubMed  Google Scholar 

  13. Shang S, Yang YW, Chen F, Yu L, Shen SH, Li K, Cui B, Lv XX, Zhang C, Yang C, et al. TRIB3 reduces CD8 T cell infiltration and induces immune evasion by repressing the STAT1-CXCL10 axis in colorectal cancer. Sci Transl Med. 2022;14(626):eabf0992.

    Article  PubMed  Google Scholar 

  14. Li Q, Zhou ZW, Duan W, Qian CY, Wang SN, Deng MS, Zi D, Wang JM, Mao CY, Song GB, et al. Inhibiting the redox function of APE1 suppresses cervical cancer metastasis via disengagement of ZEB1 from E-cadherin in EMT. J Exp Clin Canc Res. 2021;40(1):220.

    Article  Google Scholar 

  15. Nuovo GJ. In situ detection of microRNAs in paraffin embedded, Formalin fixed tissues and the co-localization of their putative targets. Methods. 2010;52(4):307–15.

    Article  PubMed  Google Scholar 

  16. Guo ZY, Hardin H, Montemayor-Garcia C, Asioli S, Righi A, Maletta F, Sapino A, Lloyd RV. In situ hybridization analysis of miR-146b-5p and miR-21 in thyroid nodules: diagnostic implications. Endocr Pathol. 2015;26(2):157–63.

    Article  PubMed  Google Scholar 

  17. Li JW, Jiang HM, Lv ZY, Sun ZY, Cheng CQ, Tan GH, Wang MC, Liu AL, Sun H, Guo H, et al. Articular fibrocartilage-targeted therapy by microtubule stabilization. Sci Adv. 2022;8(46):eabn8420.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Dai FQ, Li CR, Fan XQ, Tan L, Wang RT, Jin H. Mir-150-5p inhibits non-small-cell Lung Cancer Metastasis and recurrence by targeting HMGA2 and β-Catenin signaling. Mol Ther-Nucl Acids. 2019;16:675–85.

    Article  Google Scholar 

  19. Zhang Q, Liu W, Zhang HM, Xie GY, Miao YR, Xia MX, Guo AY. hTFtarget: a comprehensive database for regulations of human transcription factors and their targets. Genom Proteom Bioinf. 2020;18(2):120–8.

    Article  Google Scholar 

  20. Vlachos IS, Zagganas K, Paraskevopoulou MD, Georgakilas G, Karagkouni D, Vergoulis T, Dalamagas T, Hatzigeorgiou AG. DIANA-miRPath v3.0: deciphering microRNA function with experimental support. Nucleic Acids Res. 2015;43(W1):W460–6.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Liu ZC, Wang TT, She YL, Wu KQ, Gu SR, Li L, Dong CL, Chen C, Zhou YX. N-methyladenosine-modified circIGF2BP3 inhibits CD8 T-cell responses to facilitate tumor immune evasion by promoting the deubiquitination of PD-L1 in non-small cell lung cancer. Mol Cancer. 2021;20(1):105.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Wang S, Xu L, Che XF, Li C, Xu L, Hou KZ, Fan YB, Wen T, Qu XJ, Liu YP. E3 ubiquitin ligases Cbl-b and c-Cbl downregulate PD-L1 in wild-type non-small cell lung cancer. Febs Lett. 2018;592(4):621–30.

    Article  PubMed  Google Scholar 

  23. Santasusagna S, Zhu S, Jawalagatti V, Carceles-Cordon M, Ertel A, Garcia-Longarte S, Song WM, Fujiwara N, Li P, Mendizabal I, et al. Master transcription factor reprogramming unleashes selective translation promoting Castration Resistance and Immune Evasion in Lethal prostate Cancer. Cancer Discov. 2023;13(12):2584–609.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Rambow F, Bechadergue A, Luciani F, Gros G, Domingues M, Bonaventure J, Meurice G, Marine JC, Larue L. Regulation of Melanoma Progression through the TCF4/miR-125b/NEDD9 Cascade. J Invest Dermatol. 2016;136(6):1229–37.

    Article  PubMed  Google Scholar 

  25. de Galarreta MR, Bresnahan E, Molina-Sánchez P, Lindblad KE, Maier B, Sia D, Puigvehi M, Miguela V, Casanova-Acebes M, Dhainaut M, et al. β-Catenin activation promotes Immune escape and resistance to Anti-PD-1 therapy in Hepatocellular Carcinoma. Cancer Discov. 2019;9(8):1124–41.

    Article  Google Scholar 

  26. Moerke NJ, Aktas H, Chen H, Cantel S, Reibarkh MY, Fahmy A, Gross JD, Degterev A, Yuan JY, Chorev M, et al. Small-molecule inhibition of the interaction between the translation initiation factors eIF4E and eIF4G. Cell. 2007;128(2):257–67.

    Article  PubMed  Google Scholar 

  27. Wang J, Ge JS, Wang YA, Xiong F, Guo JY, Jiang XJ, Zhang LS, Deng XY, Gong ZJ, Zhang SS, et al. EBV miRNAs BART11 and BART17-3p promote immune escape through the enhancer-mediated transcription of PD-L1. Nat Commun. 2022;13(1):866.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Colangelo T, Polcaro G, Ziccardi P, Pucci B, Muccillo L, Galgani M, Fucci A, Milone MR, Budillon A, Santopaolo M, et al. Proteomic screening identifies calreticulin as a miR-27a direct target repressing MHC class I cell surface exposure in colorectal cancer. Cell Death Dis. 2016;7(2):e2120.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Wu XY, Li TH, Jiang R, Yang X, Guo HQ, Yang R. Targeting MHC-I molecules for cancer: function, mechanism, and therapeutic prospects. Mol Cancer. 2023;22(1):194.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Korkolopoulou P, Kaklamanis L, Pezzella F, Harris AL, Gatter KC. Loss of antigen-presenting molecules (MHC class I and TAP-1) in lung cancer. Brit J Cancer. 1996;73(2):148–53.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Dhatchinamoorthy K, Colbert JD, Rock KL. Cancer Immune Evasion through loss of MHC Class I Antigen Presentation. Front Immunol. 2021;12:636568.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Lin R, Chen L, Chen G, Hu CY, Jiang S, Sevilla J, Wan Y, Sampson JH, Zhu B, Li QJ. Targeting miR-23a in CD8 cytotoxic T lymphocytes prevents tumor-dependent immunosuppression. J Clin Invest. 2014;124(12):5352–67.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Liu JT, Fan LL, Yu HQ, Zhang J, He Y, Feng DC, Wang F, Li XQ, Liu QQ, Li YH, et al. Endoplasmic reticulum stress causes Liver Cancer cells to Release Exosomal miR-23a-3p and Up-regulate programmed death Ligand 1 expression in macrophages. Hepatology. 2019;70(1):241–58.

    Article  PubMed  Google Scholar 

  34. Yao XL, Tu Y, Xu YL, Guo YY, Yao F, Zhang XH. Endoplasmic reticulum stress-induced exosomal miR-27a-3p promotes immune escape in breast cancer via regulating PD-L1 expression in macrophages. J Cell Mol Med. 2020;24(17):9560–73.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Luke JJ, Bao RY, Sweis RF, Spranger S, Gajewski TF. WNT/β-catenin pathway activation correlates with Immune Exclusion across Human cancers. Clin Cancer Res. 2019;25(10):3074–83.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Herbst RS, Soria JC, Kowanetz M, Fine GD, Hamid O, Gordon MS, Sosman JA, McDermott DF, Powderly JD, Gettinger SN, et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature. 2014;515(7528):563–7.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Iwai Y, Ishida M, Tanaka Y, Okazaki T, Honjo T, Minato N. Involvement of PD-L1 on tumor cells in the escape from host immune system and tumor immunotherapy by PD-L1 blockade. P Natl Acad Sci USA. 2002;99(19):12293–7.

    Article  Google Scholar 

  38. Zhang JF, Bu X, Wang HZ, Zhu YS, Geng Y, Nihira NT, Tan YY, Ci YP, Wu F, Dai XP, et al. Cyclin D-CDK4 kinase destabilizes PD-L1 via cullin 3-SPOP to control cancer immune surveillance. Nature. 2018;553(7686):91–5.

    Article  PubMed  Google Scholar 

Download references

Funding

This work was supported by the National Natural Science Foundation of China (81672283, to H.J.), the Chongqing Natural Science Foundation (cstc2021jcyj-msxmX0350, to H.J.), and the Beijing Natural Science Foundation (7232168, to R.-T.W.).

Author information

Author notes

  1. Hao Luo, Bin Hu, Xiang-Rong Gu and Jing Chen contributed equally to this work.

Authors and Affiliations

  1. Department of Thoracic Surgery, Daping Hospital, Army Medical University, Chongqing, 400042, China

    Hao Luo, Xiao-Qing Fan, Xian-Dong He, Wei Guo & Hua Jin

  2. School of Medicine, Chongqing University, Chongqing, 400030, China

    Hao Luo, Jing Chen & Cheng-Xiong Xu

  3. Department of Medical Oncology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, Medicine School of University of Electronic Science and Technology, Chengdu, China

    Bin Hu & Wei Zhang

  4. Department of Radiology, Daping Hospital, Army Military Medical University, Chongqing, 400042, China

    Xiang-Rong Gu

  5. College of Pulmonary and Critical Care Medicine, Chinese PLA General Hospital, Beijing, China

    Ren-Tao Wang

  6. Cancer Center, Daping Hospital, Army Medical University, Chongqing, 400042, China

    Nan Dai & Dan Jian

  7. The Shapingba Hospital, Chongqing University, Chongqing, 400030, China

    Qing Li

Authors
  1. Hao Luo
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Bin Hu
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Xiang-Rong Gu
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Jing Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Xiao-Qing Fan
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Wei Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Ren-Tao Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Xian-Dong He
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Wei Guo
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Nan Dai
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Dan Jian
    View author publications

    You can also search for this author inPubMed Google Scholar

  12. Qing Li
    View author publications

    You can also search for this author inPubMed Google Scholar

  13. Cheng-Xiong Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

  14. Hua Jin
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

HJ, QL and CXX designed experiments and prepared the manuscript. HJ, CXX, HL, XRG, JC and BH performed in vitro and in vivo data analysis. JC performed bioinformatics analysis and draw a graphic abstract. HL, BH, QL, JC, and XF performed vitro and animal experiments. ND, DJ, WZ, RTW, XDH, and WG participated in human sample collection and clinical data analysis.

Corresponding authors

Correspondence to Qing Li, Cheng-Xiong Xu or Hua Jin.

Ethics declarations

Ethical approval

This study was approved by the Ethics Committee of Daping Hospital of Army Medical University (Medical Research Ethics Review No.120), and all patients obtained informed consent before enrollment. All mouse procedures were approved by the Laboratory Animal Welfare and Ethics Committee of Army Medical University (AMUWEC20211547).

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Luo, H., Hu, B., Gu, XR. et al. The miR-23a/27a/24 − 2 cluster drives immune evasion and resistance to PD-1/PD-L1 blockade in non-small cell lung cancer. Mol Cancer 23, 285 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12943-024-02201-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12943-024-02201-w

Molecular Cancer

ISSN: 1476-4598