You are here

  • Home
  • Online First
  • Nuclear translocation of plasma membrane protein ADCY7 potentiates T cell-mediated antitumour immunity in HCC

Article Text

Article menu
Hepatology
Original research
Nuclear translocation of plasma membrane protein ADCY7 potentiates T cell-mediated antitumour immunity in HCC
  1. Jianan Chen1,
  2. Youhai Jiang2,
  3. Minghui Hou3,
  4. Chunliang Liu1,
  5. Erdong Liu4,
  6. Yali Zong4,
  7. Xiang Wang5,
  8. Zhengyuan Meng6,
  9. Mingye Gu6,
  10. Yu Su1,
  11. Hongyang Wang1,
  12. Jing Fu1
  1. 1International Cooperation Laboratory on Signal Transduction, National Center for Liver Cancer, Ministry of Education Key Laboratory on Signaling Regulation and Targeting Therapy of Liver Cancer, Shanghai Key Laboratory of Hepato-biliary Tumor Biology, Eastern Hepatobiliary Surgery Hospital, Naval Medical University, Shanghai, China
  2. 2Cancer Research Center, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology, Anhui, China
  3. 3Research Center for Organoids, Department of Translational Medicine Center, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China
  4. 4Institute of Metabolism & Integrative Biology, Fudan University, Shanghai, China
  5. 5Second Department of Biliary Surgery, Eastern Hepatobiliary Surgery Hospital, Naval Medical University, Shanghai, China
  6. 6State Key Laboratory of Oncogenes and Related Genes, Shanghai Cancer Institute, Renji Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
  1. Correspondence to Jing Fu; fujing-724{at}163.com

Abstract

Background The potency of T cell-mediated responses is a determinant of immunotherapy effectiveness in treating malignancies; however, the clinical efficacy of T-cell therapies has been limited in hepatocellular carcinoma (HCC) owing to the extensive immunosuppressive microenvironment.

Objective Here, we aimed to investigate the key genes contributing to immune escape in HCC and raise a new therapeutic strategy for remoulding the HCC microenvironment.

Design The genome-wide in vivo clustered regularly interspaced short palindromic repeats (CRISPR) screen library was conducted to identify the key genes associated with immune tolerance. Single-cell RNA-seq (scRNA-seq), flow cytometry, HCC mouse models, chromatin immunoprecipitation and coimmunoprecipitation were used to explore the function and mechanism of adenylate cyclase 7 (ADCY7) in HCC immune surveillance.

Results Here, a genome-wide in vivo CRISPR screen identified a novel immune modulator-ADCY7. The transmembrane protein ADCY7 undergoes subcellular translocation via caveolae-mediated endocytosis and then translocates to the nucleus with the help of leucine-rich repeat-containing protein 59 (LRRC59) and karyopherin subunit beta 1 (KPNB1). In the nucleus, it functions as a transcription cofactor of CCAAT/enhancer binding protein alpha (CEBPA) to induce CCL5 transcription, thereby increasing CD8+ T cell infiltration to restrain HCC progression. Furthermore, ADCY7 can be secreted as exosomes and enter neighbouring tumour cells to promote CCL5 induction. Exosomes with high ADCY7 levels promote intratumoural infiltration of CD8+ T cells and suppress HCC tumour growth.

Conclusion We delineate the unconventional function and subcellular location of ADCY7, highlighting its pivotal role in T cell-mediated immunity in HCC and its potential as a promising treatment target.

  • IMMUNE RESPONSE
  • HEPATOCELLULAR CARCINOMA

Data availability statement

Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information.

http://creativecommons.org/licenses/by-nc/4.0/

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/.

https://doi.org/10.1136/gutjnl-2024-332902

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Immunotherapy, especially T-cell therapies, demonstrates excellent clinical efficacy across various tumours and holds great promise for clinical applications.

    • The extensive immunosuppressive microenvironment prevalent in hepatocellular carcinoma (HCC) has constrained the effectiveness of T-cell therapies.

    WHAT THIS STUDY ADDS

    • Adenylate cyclase 7 (ADCY7), a transmembrane protein belonging to the adenylate cyclase family, induces CD8+ T cells infiltration and cytotoxicity to retard HCC progression.

    • Mechanistically, ADCY7 translocates into the nucleus and serves as a transcriptional cofactor for CEBPA, binding to the promoter of CCL5 to promote its transcription, thereby facilitating CD8+ T cell infiltration.

    • ADCY7 sheds from the plasma membrane through caveolae-mediated endocytosis and then translocates to the nucleus with the help of leucine-rich repeat-containing protein 59 (LRRC59) and karyopherin subunit beta 1 (KPNB1).

    • Exosomal ADCY7 reshapes the immunosuppressive microenvironment and inhibits HCC progression.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • ADCY7 could be a viable therapeutic target for HCC patients by regulating T cell-mediated immune response.

    Introduction

    Hepatocellular carcinoma (HCC) is the third-leading cause of cancer-related death worldwide, and its incidence and mortality rates are increasing.1 Although innovative and efficient therapeutic strategies and advances in surgical techniques and precision medicine have been developed,2–6 the majority of HCC patients are ineligible for curative radical resection, and the prognosis is still unsatisfactory; this is ascribable to the late stage on diagnosis, early recurrence, metastasis and drug resistance.7

    A growing amount of studies have highlighted the strong efficacy of immunotherapy in halting the spread of a range of human cancers by harnessing the immune system.8–12 Numerous regimens targeting inhibitory receptors such as programmed death-1 (PD-1), programmed death-ligand 1 (PD-L1) and cytotoxic T-lymphocyte associated antigen-4 (CTLA-4) have been developed and/or approved as first-line or second-line medications in the clinic, either as single agents or in combination with traditional therapeutics.13 14 Regretfully, only a fraction of HCC patients have been shown to benefit clearly from these immune checkpoint inhibitors (ICIs). Successful immunotherapy is contingent on T cell-mediated antitumour immunity, whereas the major barrier is the paucity of T cell infiltration and gradual dysfunction of antitumour T cells in the HCC tumour microenvironment (TME), resulting in a thorough immunosuppressive microenvironment.15 These findings emphasise the need for novel targets to improve the immune microenvironment to enhance immunotherapy efficacy in HCC patients.

    Genome-wide, targeted loss-of-function pooled screens via the clustered regularly interspaced short palindromic repeats (CRISPR)-associated nuclease Cas9 (CRISPR-Cas9) system are powerful reverse-genetics tools capable of inducing complete gene knockout in cells.16 Due to the advantages of greater efficacy, ease of candidate identification and minimal off-target effects, managing genome-wide CRISPR screening techniques in oncology is gaining popularity, enabling comprehensive exploration of the mechanisms driving malignancy and potential drug targets.17–19 Genome-scale CRISPR screens have been successfully employed in vitro and/or in vivo to find common essential genes and other cancer type-dependent survival genes in diverse types of tumours.20–23 Furthermore, in conjunction with antitumour immune cells and existing treatments, such as CD8+ T cells, sorafenib and PD-1 antibodies, CRISPR screens have enabled the identification of key genes that indicate resistance or sensitivity to these treatments; thus, this approach has great significance in clinical decision-making for the treatment of cancers.24–29 However, an unbiased in vivo CRISPR screen to assess the immune response of HCC has not yet been carried out. Here, we leveraged the exceptional advantages of this technique to explore the crux of directing the T cell-mediated immune response in HCC to open the door for the astonishing efficacy of immunotherapy. Through systematic analysis, we identified the transmembrane protein ADCY7 as a robust regulator of T cell-mediated immunity and a predictor of the ICI response. Mechanistically, we revealed the previously undescribed subcellular nuclear localisation of ADCY7 and the role of ADCY7 in inducing CCL5 expression by acting as a transcription cofactor. In addition, ADCY7 can be secreted in the form of exosomes, and exosome-transmitted ADCY7 promotes CCL5 transcription in HCC cells to facilitate CD8+ T cell infiltration and tumour inhibition. Our findings highlight ADCY7 as a regulator of the T cell-mediated immune response in HCC and suggest that ADCY7 may be a promising target for HCC treatment.

    Materials and methods

    Additional materials and methods are included in online supplemental material and online supplemental tables.

    Supplemental material

    Results

    Identification of putative regulators of T cell-mediated antitumour immunity via a genome-wide in vivo CRISPR screen

    To decipher potential genes that regulate T cell-mediated immunity in HCC, genome-wide in vivo CRISPR knockout screens were performed in three mouse models under different immune selective pressures, including immunocompetent (wild-type (WT)), immunocompromised (Rag1null (T-cell-deficient) and severe combined immunodeficient (NSG)) models (figure 1A). The mouse GeCKO v2 library, a pooled genome-wide CRISPR knockout library containing 130 209 single-guide RNAs (sgRNAs) targeting 20 611 genes, was introduced into Hepa1-6 cells at a low multiplicity of infection (<0.3). After puromycin screening, the infected Hepa1-6 cells were subcutaneously inoculated into the aforementioned mice (figure 1A). To determine the abundance of single-guide RNAs (sgRNAs) isolated from xenografts, the MAGeCKFlute algorithm30 was used to calculate β scores (an indicator of fold change between the Rag1null, NSG and WT groups) after PCR amplification of the genomic DNA.

    Figure 1
    Figure 1

    In vivo CRISPR/Cas9 screen identifies potential candidates as determinants of T cell-mediated antitumour immunity. (A, B) The schematic of the genome-wide in vivo CRISPR/Cas9 knockout library screen. (C) The detailed workflow of potential candidate selection. (D) 118 homologous genes between human and mouse species were obtained according to the selection criteria. (E) 2 genes (ADCY7 and FCER1G) strongly associated with T cell-mediated antitumour responses were remained among the above-mentioned 118 homogenes. (F) Pearson correlation of ADCY7 or FCER1G with T cell-mediated antitumour responses. (G) Expression of ADCY7 and FCER1G in HCC patients responding to ICIs in GSE202069 cohort. (H, I) The expression status of ADCY7 and FCER1G in HCC from GEO-combat (H) and TCGA-LIHC and GTEX (I) data sets. (J) Patients with low ADCY7 expression have worse prognosis using the online KM-Plotter tool based on TCGA database. (K) HCC TMA cohort shows a low ADCY7 protein expression in HCC tissues compared with adjacent non-tumour tissues. (L) Kaplan-Meier analysis demonstrates that low ADCY7protein expression correlated with worse prognosis. (M) Multivariate analysis shows ADCY7, tumour size and AFP level are independent risk factor for HCC. Error bars in (G, K) indicate the mean±SD. The p value and the R value in (F) were calculated using Pearson’s correlation analysis. P values in (G–I) were calculated by two-tailed Student’s t-test. P values in (J, L) were calculated by log-rank test. P value in (K) was calculated by Mann-Whitney U test. P value in (M) was calculated by Cox regression analysis. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. HCC, hepatocellular carcinoma; TMA, tissue microarray.

    With a Δβ score±SD (Δβ score signifies β score of the Rag1null or NSG group minus β score of the WT group), |β score of the WT group|>2 and |β score of the Rag1null or NSG group|<1 as the criteria, four regions with significant enrichment of sgRNAs were identified (figure 1B). Among them, genes enrichment in regions 2 and 4 indicated that gene knockout may affect tumour cell proliferative ability, while those enrichment in regions 1 and 3 signified sensitivity or resistance to immune pressure, respectively. Because sufficient genes are needed for candidates selection, we focused on region 3 (figure 1B, online supplemental tables 1 and 2). As shown by the detailed workflow of candidate selection in figure 1C,D, we obtained 521 common genes between the two regions three of which indicated that their deletion would confer resistance to T cell-mediated immunity. Subsequently, the following criteria were performed to narrow down these genes: difference of absolute β scores between the Rag1null and NSG groups<0.2 (147 genes), which further supported the possibility that the candidates were involved in modulating adaptive immune system-mediated responses rather than affecting tumour cell viability (online supplemental table 3). After filtering, 118 genes homologous between human and mouse species remained (figure 1D, online supplemental table 4). We next used the GEO-combat cohort (a pooled cohort combining 5 HCC GEO datasets generated via the R package ‘sva’, online supplemental figure 1A,B) to calculate the correlations between these genes and T cell-mediated antitumour response scores (seven gene sets that have been identified as indicators of elevated T-cell infiltration and antitumour response) (online supplemental table 5).31–34

    By integrating the above results, two genes (ADCY7 and Fc fragment of IgE receptor 1 g, FCER1G) that were strongly correlated with T cell-mediated immunity (Pearson r>0.5 and p<0.05) were preliminarily selected (figure 1E,F, online supplemental table 6). Subsequent analysis demonstrated that ADCY7 and FCER1G were highly expressed in HCC patients who were sensitive to ICIs treatment (figure 1G). Although both were downregulated in tumour tissues, only ADCY7 was associated with a better prognosis in HCC patients in the GEO-combat and TCGA-LIHC&GTEx datasets (downloaded and processed through the UCSC Xena database, https://xenabrowser.net/datapages/) (figure 1H–J). More importantly, to our knowledge, the involvement of ADCY7 in immunoregulation has not yet been well defined. Therefore, we selected this gene for further exploration. The proteomics data from the CPTAC database and western blot results from paired tumour and adjacent non-tumour tissues as well as different HCC cell lines, revealed lower ADCY7 protein levels in HCC tissues and cell lines (online supplemental figure 1C,D). Intriguingly, IHC staining showed that ADCY7 expression was relatively low in both normal liver tissues and HCC tissues but was high in cirrhosis tissues and adjacent non-tumour ones (online supplemental figure 1E). Furthermore, our HCC tissue microarray (HCC-TMA) cohort corroborated that ADCY7 protein expression was downregulated in HCC tissues compared with adjacent ones (figure 1K). The ADCY7 protein levels in HCC tissues vary greatly among individuals and patients with high ADCY7 had a better prognosis, underscoring the remarkable heterogeneity of ADCY7 expression in HCC (figure 1L). Moreover, univariate and multivariate Cox proportional hazards regression analyses indicated that ADCY7 was an independent favourable factor for HCC prognosis (online supplemental figure 1F, figure 1M). Collectively, ADCY7 may play an important role in HCC development.

    Single-cell transcriptome analysis confirmed the association of CD8+ T-cell infiltration with ADCY7 expression in HCC

    To systematically explore the association between ADCY7 expression and T-cell infiltration in the HCC microenvironment, single-cell RNA-seq (scRNA-seq) data from 10 primary HCC tissues (GSE149614) from the GEO database were downloaded and analysed. Following quality control and batch effect removal, 34 204 single cells were obtained and partitioned into seven subclusters (malignant cells, myeloid cells, T cells, endothelial cells, mesenchymal cells, epithelial cells and B cells) (online supplemental figure 2A.B). Based on the median value of ADCY7 mRNA expression in malignant cells, HCC samples were categorised into high-ADCY7 and low-ADCY7 groups (online supplemental figure 2C). Subsequently, OR analysis, which has been employed as a distribution preference indicator (an OR>1.5 denotes the distribution preference of this tissue, while an OR<0.5 signifies the opposite), highlighted the enrichment of T cells in the high-ADCY7 expression group (online supplemental figure 2D). T cells were further grouped into 10 subclusters (CD8-C1-IFNG, CD8-C2-FCGR3A, CD8-C3-MKI67, CD8-C4-PDCD1, MAIT, CD4 Tn, CD4-C1-TIGIT, CD4-C2-FOXP3, CD4-C3-S100A4 and NKT cells) according to their specific markers, with the CD8-C1-IFNG and CD4 Tn clusters emerging as two primary populations (online supplemental figure 2E,F).

    We then assessed the enrichment of publicly available gene sets for naïve, cytotoxic, exhausted and regulatory T cells in the subclusters and derived a transcriptional score model using the AddModuleScore.35 Among the CD8+ T cells, the CD8-C1-IFNG and CD8-C2-FCGR3A cells had high cytotoxicity scores, while the CD8-C4-PDCD1 cells had the highest exhaustion score (online supplemental figure 2G). Among the CD4+ T cells, CD4 Tn cells had a high cytotoxicity score, while CD4-C1-TIGIT and CD4-C2-FOXP3 cells had high exhaustion and Treg scores (online supplemental figure 2H). OR analysis indicated that cytotoxic subclusters, such as CD8-C1-IFNG, CD8-C2-FCGR3A and CD4 Tn, were preferentially distributed in the high-ADCY7 subgroup, while immunosuppressive subclusters, such as CD8-C4-PDCD1 and CD4-C2-FOXP3, were preferentially distributed in the low-ADCY7 subgroup (online supplemental figure 2I,J). On the other hand, we investigated interactions between malignant cells and T cells via CellChat. Multiple T-cell subsets preferentially interacted with the high-ADCY7 subgroup compared with the low-ADCY7 subgroup (online supplemental figure 2K). Analysis of the ligand and receptor pairs revealed that of all the T-cell clusters, IFNG-(IFNGR1+IFNGR2) occurred only between the CD8+ T subsets and malignant cells in the high-ADCY7 subgroup; CD8-C1-IFNG and CD8-C2-FCGR3A were the main sources of IFNG; and malignant cells with high ADCY7 expression were the main cells receiving the signals (online supplemental figure 2L–O). Collectively, these data support the idea that ADCY7 is closely associated with the infiltration of cytotoxic CD8+ T cells.

    Re-expression of ADCY7 promotes the CD8+ T cell-mediated antitumour response in HCC

    Having established the potential association of ADCY7 with CD8+ T cells, we performed in vitro and in vivo experiments to further verify the function of ADCY7 in the T cell-mediated antitumour response during HCC development. Notably, ADCY7 re-expression had no significant effect on the proliferative capacity of HCC cells in vitro or tumour growth in immunocompromised NSG mice (online supplemental figure 3A–C). Moreover, tumours from the control and ADCY7 re-expression groups showed equivalent levels of Ki-67 and cleaved-caspase three staining (online supplemental figure 3D,E). Similar results were obtained in T cell-deficient Rag1null mice (online supplemental figure 3F–I). However, we noticed that ADCY7 re-expression drastically suppressed tumour growth compared with control group in immunocompetent C57BL/6J mice (figure 2A–C). The flow cytometry (FCM) results revealed increased infiltration of CD3+ T, CD8+ T and IFN-γ+CD8+ T cells but no significant changes in CD4+ T and NK cells in tumours from the ADCY7 re-expression group (figure 2D,E), which was consistent with the IHC results (figure 2F,G).

    Figure 2
    Figure 2

    ADCY7 re-expression positively correlates CD8+ T cell-mediated antitumour immunity in HCC. (A) Western blot confirms the stable re-expression of ADCY7 in Hepa1-6 cells. (B, C) The images of tumour xenografts (B) and the tumour growth curves (C) of C57BL/6J mice; n=9 mice for NC group and n=10 mice for ADCY7 re-expression group. (D, E) Flow cytometry results exhibit a higher infiltration of CD3+ T, CD8+ T cells after ADCY7 re-expression, but not CD4+ T and NK cells (n=4). (F, G) Similar results are observed in IHC staining of CD3+ T, CD8+ T, CD4+ T and NK cells (n=5). (H, I) IHC staining of ADCY7, CD3+ T cells and CD8+ T cells in HCC TMA cohort confirms the higher infiltration in patients with high ADCY7 expression. (J) Western blot confirms the stable re-expression of ADCY7 in HCC cells. (K, L) T cell migration assay on the coincubation of HCC cells with healthy donor CD8+ T cells (n=3). Error bars in (C, E, G, I, K, L) indicate the mean±SD. P value in (C) was calculated by two-way analysis of variance. P values in (E, G, I, K, L) were calculated by two-tailed Student’s t-test. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. HCC, hepatocellular carcinoma; NS, not significant; TMA, tissue microarray.

    Furthermore, in our own HCC TMA cohort, we observed significant infiltration of CD3+ T and CD8+ T cells in patients with high ADCY7 expression (figure 2H,I). Remarkably, the T-cell migration assay revealed that the human HCC cell lines, HCCLM3 and HepG2, attracted more CD8+ T cells after ADCY7 re-expression (figure 2J–L). These findings rule out the possibility that ADCY7 mainly regulates adaptive immunity rather than HCC cell proliferation to suppress HCC progression.

    ADCY7 induces CCL5 expression to attract CD8+ T cells in HCC

    To explore the mechanism by which ADCY7 regulates CD8+ T cells recruitment, KEGG and GO analyses were performed on the differentially expressed genes (|fold change|>1.5 fold or <0.585 fold, p<0.05) between HCC patients with high and low ADCY7 mRNA levels in the GEO-combat cohort. The results showed that ADCY7 was positively associated with the chemokine signalling pathway and T-cell activation (figure 3A, online supplemental figure 4A,B). Furthermore, gene set enrichment analysis indicated significant upregulation of the chemokine signalling pathway in the high ADCY7 group (figure 3B). To elucidate such chemokines governing CD8+ T cells recruitment in ADCY7 re-expression HCC cells, the human chemokine antibody array was conducted using supernatants of HepG2 and HCCLM3 cells. Among them, chemokine (C-C motif) ligand 5 (CCL5), CCL26, (C-X-C motif) ligand 12 (CXCL12) and Midkine (MDK) were upregulated in ADCY7 high group (figure 3C, online supplemental figure 4C). Subsequent T cell migration analysis showed that silencing CCL5, rather than the other three chemokines, could obviously attenuate the CD8+ T cells attraction induced by ADCY7 re-expression (online supplemental figure 4D–G). Meanwhile, multiple prior articles have also reported its indispensable role in T-cell infiltration in diverse types of tumours.36–38 Therefore, CCL5 was focused for further exploration and we verified the upregulation of CCL5 mRNA and protein expression after ADCY7 re-expression in HCC cell lines (figure 3D,E).

    Figure 3
    Figure 3

    ADCY7 hampers HCC progression in a CCL5-dependent manner. (A, B) Enrichment analysis of KEGG (A) and GSEA (B) showed that ADCY7 is significantly associated with the chemokine signalling pathway based on GEO-combat data set. (C) The human chemokine protein array was used to determine chemokine levels in supernatants of HepG2 cells. (D, E) qPCR (D) and western blotting (E) show that CCL5 is upregulated in the ADCY7 re-expression group. (F) T cell migration assay showed CCL5 knockdown drastically suppressed CD8+ T cells migration induced by ADCY7 re-expression (n=3). (G, H) The images (G) and tumour growth curves (H) of C57BL/6J mice inoculated with negative control (NC) cells, ADCY7 re-expression cells, ADCY7 re-expression and shCCL5 cells and shCCL5 cells, respectively; n=8 mice for ADCY7 re-expression group and n=6 for other groups. (I) The general images of orthotopic tumours derived from C57BL/6J mice. (J) The longest diameters of orthotopic tumours (n=6). (K, L) Flow cytometry results exhibit a higher infiltration of CD3+ T, CD8+ T and IFN-γ+CD8+ T cells with ADCY7 re-expression but reduced after CCL5 knockdown (n=4). (M, N) Similar results are observed in IHC staining of CD8+ T and CD4+ T cells (n=5). Error bars in (D, F, H, J, L, N) indicate the mean±SD. P values in (D, F, J, L, N) were calculated by two-tailed Student’s t-test. P values in (H) were calculated by two-way analysis of variance. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. HCC, hepatocellular carcinoma; NS, not significant.

    To clarify the role of CCL5 in ADCY7-mediated antitumour immunity, we exposed ADCY7 re-expressing HCC cells to lentiviruses containing either a control or a CCL5-targeted short hairpin RNA (shRNA). CCL5-sh3 was chosen for subsequent experiments due to its best knockdown efficacy (online supplemental figure 4H). Similar results of T cell migration were obtained in control and ADCY7 re-expression HCC cell lines with CCL5 knockdown using shRNA (figure 3F). An in vivo xenograft model in C57BL/6J mice demonstrated that restoring ADCY7 expression failed to halt tumour growth in the context of CCL5 knockdown (figure 3G,H). IHC results showed a significant reduction in CD3+ T and CD8+ T-cell infiltration in ADCY7&sh-CCL5 group, suggesting that CCL5 is needed for the ADCY7-mediated immune response in HCC (online supplemental figure 4I,J). Consistently, the orthotopic HCC model corroborated that ADCY7 significantly inhibited tumour development, whereas knockdown of CCL5 was capable of mitigating this suppression effect (figure 3I,J). FCM analysis showed a higher infiltration of CD3+ T, CD8+ T cells and IFN-γ+CD8+ T cells in ADCY7 re-expression group (figure 3K,L). IHC results further ascertained these findings (figure 3M,N). Subsequently, we looked more closely at the function of ADCY7 in CD34+ humanised mice with a relatively intact human immune system.39 As anticipated, re-expression of ADCY7 in human HCCLM3 cells decreased tumour volume, whereas these effects could be counteracted by CCL5 knockdown (online supplemental figure 4K). Taken together, these results indicate that CCL5 is a principal molecule involved in the ADCY7-mediated antitumour response in HCC.

    Nuclear ADCY7 functions as a transcription cofactor to induce CCL5 expression

    We then attempted to determine how ADCY7 induces CCL5 expression. ADCY7 is a transmembrane protein that catalyses the conversion of ATP to cyclic adenosine 3′,5′-monophosphate (cAMP) to activate specific signalling pathways.40 Therefore, we first speculated that the ability of ADCY7 to retard HCC progression was dependent on its enzyme activity. Intriguingly, no differences in intracellular cAMP levels or expression of key downstream proteins (CREB, p-CREB, AMPK or p-AMPK) were observed between ADCY7 re-expressing and control cells (online supplemental figure 5A,B). Then we used forskolin, a direct activator of adenylate cyclases (ACs) to activate its enzyme activity; however, a comparable or slightly reduced CCL5 protein expression was observed between ADCY7 re-expressing and control cells (online supplemental figure 5A and C). These indicated that CCL5 elevation is induced by other ADCY7-related mechanisms rather than its enzyme activity in HCC.

    Unexpectedly, we found that the ADCY7 protein was distributed in the nucleus in some HCC tissues (online supplemental figure 5D), which was not previously reported. Western blot analysis and confocal microscopy confirmed that ADCY7 was distributed in the nuclei of HCC cells (figure 4A,B, online supplemental figure 5E). This made us consider that ADCY7 might regulate CCL5 expression by binding to its promoter sequence. To verify this, we extracted the promoter sequence of the CCL5 gene (2000 bp) and designed 11 primers every 200 bp from the transcription start site (figure 4C). Chromatin immunoprecipitation assays of ADCY7 re-expressing HCCLM3 cells revealed that ADCY7 was able to bind to the CCL5 promoter sequence, mainly at the region −1100 to−900 bp from the transcription start site (figure 4D). However, ADCY7 cannot bind to the DNA sequence predicted by the iDRBP_MMC and DNAbinder databases, suggesting that ADCY7 indirectly regulates CCL5 transcription. After searching and selecting potential transcription factors in online public database, CCAAT/enhancer-binding protein alpha (CEBPA) and MYB were obtained as the potential transcription factors of CCL5 (figure 4E). qPCR showed that CEBPA silencing significantly reduced CCL5 mRNA expression compared with MYB knockdown (figure 4F,G). Moreover, coimmunoprecipitation (co-IP) and confocal microscopy demonstrated an interaction and colocalisation between ADCY7 and CEBPA, in contrast to MYB (figure 4H, online supplemental figure 5F–H). ADCY7 re-expression in HCC cells had no apparent effect on CEBPA expression but significantly enhanced the capacity of CEBPA to bind to the CCL5 promoter (figure 4I,J). We also found that CEBPA silencing obviously attenuated the ADCY7-induced increase in CCL5 transcription and protein expression (figure 4K,L). Furthermore, a series truncated plasmids of ADCY7 were constructed to verify the key domains for its nuclear translocation and interaction with CEBPA (figure 4M). As a result, all truncations were still able to translocate into the nucleus and induced CCL5 mRNA expression except the N-tail deleted ADCY7 mutant (figure 4N,O). Co-IP results further confirmed the interaction of N-tail with CEBPA (figure 4P). These results together demonstrated that nuclear ADCY7 through the N-tail assisted the transcription factor CEBPA in promoting CCL5 transcription.

    Figure 4
    Figure 4

    Nuclear ADCY7 upregulates CCL5 expression by binding to its promoter sequence. (A, B) Subcellular distribution of ADCY7 in ADCY7 re-expressing HCC cells by western blot (A) and STED (B) analysis. LMNB1 as a nuclear internal reference and GAPDH as the cytoplasm reference, indicating the purity of the extracted nuclear and cytoplasm proteins. (C, D) ChIP-qPCR indicated ADCY7 could bind to CCL5 gene promoter sequence (n=3). (E) Prediction of putative transcription factors of CCL5 via different kinds of public databases. ‘All’ indicates the full 2000 bp sequence of CCL5 promoter; ‘P6’ indicates the −1100 to −900 sequence. (F, G) CEBPA silence more significantly reduced the CCL5 mRNA expression than MYB in HCCLM3 cells (n=3). The mRNA expression of ADCY7 and CCL5 in NC and ADCY7 groups are from the same wells in figure 4F,G. (H) Co-IP assay showed the interaction of ADCY7 with CEBPA in HCC cells. (I) Western blot showed no significant difference of CEBPA expression in indicated HCC cells. (J) ChIP-qPCR showed stronger binding capacity of CEBPA to CCL5 promoter in ADCY7 re-expression cells than that in control cells (n=3). (K) ChIP-qPCR suggested after CEBPA silence, ADCY7 failed to enhance CCL5 transcription (n=3). (L) Western blot analysis of ADCY7, CEBPA and CCL5 levels after CEBPA silence. (M) Schematic of ADCY7 domain structure. (N) The subcellular distribution of ADCY7 truncations identified by cytoplasmic and nuclear western blot. (O) qPCR analysis of CCL5 mRNA expression in HepG2 cells transfected with empty vector and ADCY7 truncations (n=5). (P) The interactions between ADCY7 fragments and CEBPA were evaluated via Co-IP in HepG2 cells. Error bars in (D, F, G, J, K, O) indicate the mean±SD, and p values were calculated by two-tailed Student’s t-test. **p<0.01, ***p<0.001, ****p<0.0001. ChIP, chromatin immunoprecipitation; Co-IP, coimmunoprecipitation; EV, extracellular vesicle; HCC, hepatocellular carcinoma; NS, not significant; WT, wild-type.

    Caveolae-mediated endocytosis of ADCY7 and translocation to the nucleus require its interaction with LRRC59 and KPNB1

    The above-described function of ADCY7 encouraged us to explore the mechanism of its nuclear translocation. In terms of transmembrane proteins, the first step of translocation is shedding from the membrane into the cytoplasm. Clathrin-dependent or caveolae-mediated endocytosis is a crucial route by which proteins, such as PD-L1 and EGFR, enter the cytoplasm.41 42 Fortunately, we discovered that though its N-tail, ADCY7 can interact with caveolin-1 (CAV-1), an important component that results in the formation of caveolae (figure 5A, online supplemental figure 6A). Confocal microscopy also confirmed the colocalisation of the two proteins in HCC cells (figure 5B, online supplemental figure 6B). Moreover, filipin III,43 an inhibitor of caveolae-mediated endocytosis that had no obvious effect on ADCY7 and CCL5 expression, could elevate the plasma localisation of ADCY7 and diminish its nuclear distribution along with CCL5 mRNA and protein reduction in ADCY7 re-expressing cells (figure 5C, online supplemental figure 6C,D). Meanwhile, several studies have shown a lack of or low expression of CAV-1 in Hepa1-6 and HepG2 cells,44 45 which appears to contradict our results that ADCY7 facilitates its protein translocation by binding to CAV-1. However, we observed that CAV-1 levels were elevated in the ADCY7 re-expression group. Moreover, concurrent results were observed with CAV-1 knockdown and filipin III treatment, indicating that caveolae-mediated endocytosis might signify an initial step in ADCY7 translocation (figure 5D, online supplemental figure 6E–H).

    Figure 5
    Figure 5

    ADCY7 accesses the nuclei under the help of caveolae-mediated endocytosis and LRRC59 and KPNB1. (A, B) Co-IP (A) and STED (B) assays showed the nuclear interaction and colocalisation of ADCY7 with CAV-1 in HCC cells. (C) The distribution of ADCY7 in subcellular organisation after Filipin III treatment (15 µM used for 12 hours). (D) Schematic model of ADCY7 nuclear translocation. (E–G) STED (E, F) and Co-IP (G) assays showed the nuclear colocalisation and interaction of ADCY7 with LRRC59 and KPNB1 in HCC cells. (H, I) Co-IP assay indicated ADCY7 was unable to interact with KPNB1 after LRRC59 silence. (J) Western blot analysis showed reduced ADCY7 expression in the nucleus after LRRC59 silence. (K–M) Western blot analysis showed total ADCY7 and cytoplasmic or nuclear ADCY7 levels by importazole (IPZ) treatment in indicated HCC cells (0, 6 or 8 µM used for 48 hours). Co-IP, coimmunoprecipitation; HCC, hepatocellular carcinoma.

    Subsequently, we determined how ADCY7 reached the nucleus from the cytoplasm. Our above results indicated that the N-tail of ADCY7 was important for ADCY7 nuclear translocation (figure 4N), however, its N-tail domain does not contain potential nuclear location sequence (NLS) predicted by the NLStradamus website tool (http://www.moseslab.csb.utoronto.ca/NLStradamus/),46 suggesting that ADCY7 might enter the nucleus by other means. Some proteins without an NLS can enter the nucleus with the assistance of importin alpha/beta subunits, which may bind to other molecules containing NLSs and then form compounds to move into the nucleus.47–49 On the basis of our mass spectrum analysis, we found that ADCY7 might interact with the endoplasmic reticulum (ER) protein, leucine-rich repeat containing 59 (LRRC59), which has been reported to play an important role in protein nuclear translocation in cooperation with karyopherin subunit beta 1 (KPNB1), a member of the importin beta family (online supplemental figure 6I-J). As expected, our confocal microscopy experiments revealed the colocalisation of ADCY7 with LRRC59 and KPNB1 (figure 5E,F, online supplemental figure 6K). Co-IP further confirmed that ADCY7 interacted with LRRC59 and KPNB1 via the N-tail (figure 5G, online supplemental figure 6A). Additionally, when LRRC59 was silenced, the interaction between ADCY7 and KPNB1 was significantly decreased and the distribution of ADCY7 in the nucleus was also reduced (figure 5H–J). Moreover, to confirm the above findings, we used importazole (IPZ), a selective inhibitor known to block importin beta-mediated nuclear translocation, to assess its effect on the subcellular distribution of ADCY7.50 After treatment with IPZ at a concentration near the IC50 in HCC cells (online supplemental figure 6L), the total expression of ADCY7 remained stable, while nuclear ADCY7 distribution was obviously decreased (figure 5K–M). Furthermore, such effects of filipin III on CCL5 expression were also observed in HCC cell lines under LRRC59 silence and IPZ treatment, respectively (online supplemental figure 6M–P). Collectively, these results indicate that ADCY7 is taken into the nucleus via caveolae-mediated endocytosis and LRRC59 and KPNB1.

    Exosomal ADCY7 suppresses HCC progression by attracting CD8+ T cells

    To improve the clinical utility of our findings, we investigated whether ADCY7 is present in exosomes because membrane proteins that are released into the cytoplasm can be packaged into exosomes and secreted.42 According to the public database Vesiclepedia, ADCY7 has been detected in extracellular vesicles (EVs) from some human cancer cell lines.51 To test this phenomenon in HCC cells, we first purified and validated exosomes from the culture supernatant of HCC cells by transmission electron microscopy and analysed the exosomal markers (figure 6A,B). Western blotting and LC-MS/MS confirmed the existence of ADCY7 protein in exosomes derived from the ADCY7 re-expressing HCCLM3 cells (figure 6B,C). To investigate the function of exosomal ADCY7 in vitro, HCCLM3 cells were co-cultured with exosomes from control and ADCY7 re-expressing cells, respectively. Confocal microscopy revealed that exosomal ADCY7 moved to HCCLM3 cells and correspondingly upregulated CCL5 expression (figure 6D). Consistently, ADCY7 was also detected in exosomes from ADCY7 re-expressing Hepa1-6 and HepG2 cells (figure 6E, online supplemental figure 7A,B). Furthermore, in C57BL/6J mice bearing subcutaneous xenografted Hepa1-6 cells, intratumoural injection of exosomes from the ADCY7 re-expressing Hepa1-6 cells significantly inhibited tumour growth (figure 6F–H). FCM and IHC analysis revealed relatively high infiltration of CD3+ T cells (online supplemental figure 7C–F). Importantly, the proportion of CD8+ T cells and IFN-γ+CD8+ T cells was elevated in the tumours treated with ADCY7-expressing exosomes (figure 6I,J). Multiplex IHC confirmed the presence of exosomal ADCY7 and increased CCL5 expression in tumour cells (figure 6K). Taken together, these results suggest that exosomal ADCY7 can suppress HCC progression by reshaping the immune microenvironment.

    Figure 6
    Figure 6

    Exosomal ADCY7 retards HCC progression. (A, B) Identification of exosomes by TEM (A) and western blot (B). (C) Identification of ADCY7 exists in exosomes by LC-MS/MS in HCCLM3 (ADCY7) cells. (D) Confocal microscopy analysis showed exosomal ADCY7 is able to enter into HCCLM3 cells. (E) Identification of exosomes from hepa1-6 ADCY7-reexpression cells by TEM and western blot. (F) Treatment schedule of exosomes derived from control and ADCY7 re-expression Hepa1-6 cells in the C57BL/6J mice bearing HCC tumours. (G, H) The images of tumour xenografts (G) and tumour growth curves (H) of C57BL/6J mice treated with exosomes derived from control and ADCY7 re-expression Hepa1-6 cells. n=9 mice for NC-exo group and n=7 for ADCY7-exo group. (I, J) FCM revealed higher percentage of CD8+ T cells (I) and IFN-γ+CD8+ T cells (J) in ADCY7-exo groups (n=4). (K) mIHC analysis displays the staining of CCL5 and ADCY7 in NC-exo and ADCY7-exo groups. (L) A schematic model showing that ADCY7 penetrates the nucleus directly bounding to CCL5 gene promoter in order to draw CD8+ T cells into HCC. On the other hand, ADCY7 is packaged by exosomes to enhance CCL5 levels by entering into HCC cells with low ADCY7 expression. Error bars in (H, I, J) indicate the mean±SD. P value in (H) was calculated by two-way analysis of variance. P values in (I, J) were calculated by two-tailed Student’s t-test. *p<0.05, **p<0.01, ****p<0.0001. HCC, hepatocellular carcinoma; TEM, transmission electron microscopy.

    Discussion

    Harnessing the immune system against tumours, including HCC, has been identified as an effective therapeutic approach. However, a tremendous barrier to HCC treatment is the comprehensive immunosuppressive microenvironment,52 which confers a dismal outcome for HCC patients and needs finding out the significant regulatory mechanisms, as altering the state of immunosuppression is a desirable treatment strategy. In this study, we identified a new key gene (ADCY7) involved in adaptive immune regulation via an in vivo genome-wide CRISPR screen. ADCY7 is a transmembrane protein belonging to the AC family, whose main function is to generate the second messenger cAMP from ATP. To date, one study has demonstrated that ADCY7 exerts an oncogenic effect on leukaemia by inducing c-MYC expression,53 and sporadic studies have briefly characterised the function of ADCY7 in immune surveillance.54 Here, we discovered that ADCY7 was downregulated in HCC and correlated with poor prognosis. Moreover, its expression was also low in normal liver tissues compared with cirrhosis tissues and adjacent non-tumour ones at both mRNA and protein levels. This expression pattern leads us to speculate that ADCY7 levels might be linked to inflammation, given the inflammatory nature of chronic liver diseases. Meanwhile, a prior study has delineated the unique characteristics of adjacent non-tumour tissues, representing an intermediate state between healthy and tumour tissues. It is proposed that chronic inflammation, through the secretion of proinflammatory signals, induces specific alterations in gene expression, thereby modulating the surrounding microenvironment and establishing a distinct tissue composition conducive to tumour progression or suppression.55 Accordingly, ADCY7 might be one of these specifically elevated genes, with its expression being inducible under chronic inflammation rather than constitutive in liver tissues. However, its reduced expression in HCC may be attributed to high promoter methylation, highlighting the need for further detailed investigation.

    Subsequent scRNA-seq analysis and multiple experiments demonstrated that ADCY7 high expression was not only associated with more CD8+ T cells infiltration but had a closer relationship with various CD8+ T cytotoxic subclusters, underscoring the potential significant value of ADCY7 in regulating HCC TME. Notably, we identified an unexpected nuclear location and function of ADCY7. Nuclear ADCY7 served as the transcriptional cofactor of CEBPA, inducing CCL5 expression and then attracting a considerable number of CD8+ T cells to impede HCC development. However, despite evidence that CCL5 could attract CD4+ T cells and Treg cells,56 57 no significant association between ADCY7 expression and CD4+ T cells has been observed in our study, which was inconsistent with previous reports.58 59 This discrepancy might stem from variations in receptor expression. Literature indicates that CCR5, the receptor with the highest affinity for CCL5, is the major receptor on CD8+ T cells, while most CD4+ T cell subsets mainly express CCR4 and CXCR3.60 Moreover, ADCY7 might affect CD4+ T cell infiltration through other potential mechanisms, requiring further exploration. Additionally, ADCY7 can also be packaged into exosomes, and exosomal ADCY7 enters tumour cells with low ADCY7 expression to induce CCL5 transcription (figure 6L), highlighting the pivotal role of ADCY7 in reshaping the TME.

    Previous studies have reported that the localisation of many proteins can shift in tumours with corresponding consequences. Transmembrane proteins are generally shed into the cytoplasm via clathrin-mediated/caveolae-mediated endocytosis or alternative mechanisms, and then move to a special subcellular location to perform their corresponding functions. For example, the notable membrane protein PD-L1 enters the cytoplasm through clathrin-mediated endocytosis and then moves to the nucleus to exert its function with the help of the cytoskeletal protein vimentin and the nuclear transport protein KPNB1.42 Exogenous fibroblast growth factor 1 (FGF1) moves into cells’ nucleus by interacting with leucine-rich repeat containing 59 (LRRC59), an ER-bound protein that assisted proteins nuclear translocation,61 where FGF1 exerts its growth regulatory activity.62 Moreover, a recent study revealed that TSPAN8, a membrane protein, can translocate into the nucleus after binding to cholesterol and subsequently interact with STAT3 to induce the transcription of downstream oncogenes and promote breast cancer progression.63 We here discovered successful ADCY7 nuclear translocation initiated with caveolae-mediated endocytosis followed by interacting with LRRC59 and KPNB1. But rather than promoting tumour development, nuclear ADCY7 induced CCL5 expression by enhancing the ability of CEBPA to bind to the CCL5 gene promoter, which in turn attracted considerable CD8+ T cells to impede HCC progression. Additionally, the nuclear location of ADCY7 is also observed in certain para-tumourous hepatocytes but not in liver tissues of healthy individuals. However, ADCY7 is still predominantly localised to the plasma membrane and cytoplasm in those non-tumour tissues. This may be attributed to relatively low expression levels of those molecules (such as CAV-1 and KPNB1),64 65 which guide the nuclear translocation of ADCY7 or other mechanisms waiting for exploration.

    Exosomes, as small EV ranging in size from 50 to 160 nm, are significant components of cell–cell communication and regulate recipient cells’ physiological activities.66 67 Growing evidence has proven the importance of exosomes in tumour development and clinical treatment and exosomes have been seen as a promising new tool of immunotherapy to combat cancer in a cell-free system.68–70 Moreover, it is well known that membrane proteins entering the cell mediated by endocytosis could be secreted by exosomes.42 71 In this study, we confirmed that ADCY7 was present in exosomes isolated from HCC cells re-expressing ADCY7 and exosomal ADCY7 facilitated CD8+ T cells infiltration and showed strong antitumour immunity in vivo, suggesting great potential for clinical use of ADCY7 in the future.

    In summary, we identified a novel gene, ADCY7, as an immune regulator of HCC progression by means of an in vivo genome-wide CRISPR screen. ADCY7 penetrates the nucleus and assists CEBPA in binding to the CCL5 gene promoter to recruit CD8+ T cells to HCC. On the other hand, ADCY7 is packaged into exosomes to boost CD8+ T-cell infiltration by enhancing CCL5 transcription. Collectively, our findings highlight that ADCY7 is a viable therapeutic target for the clinical treatment of HCC patients.

    Data availability statement

    Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study involves human participants and was approved by the Ethical Committee of Eastern Hepatobiliary Surgery Hospital (No. EHBHKY2018-1-001). Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    We are grateful to patients and doctors in Eastern Hepatobiliary Hospital for HCC tissues collection.

    References

    1. ↵
      1. Sung H,
      2. Ferlay J,
      3. Siegel RL, et al
      . Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209–49. doi:10.3322/caac.21660
      OpenUrlCrossRefPubMed
    2. ↵
      1. Duan H,
      2. Liu Y,
      3. Gao Z, et al
      . Recent advances in drug delivery systems for targeting cancer stem cells. Acta Pharm Sin B 2021;11:55–70. doi:10.1016/j.apsb.2020.09.016
      OpenUrl
    3. ↵
      1. Yadav D,
      2. Kwak M,
      3. Chauhan PS, et al
      . Cancer immunotherapy by immune checkpoint blockade and its advanced application using bio-nanomaterials. Semin Cancer Biol 2022;86:909–22. doi:10.1016/j.semcancer.2022.02.016
      OpenUrl
    4. ↵
      1. Wang S,
      2. Yang Y,
      3. Ma P, et al
      . CAR-macrophage: An extensive immune enhancer to fight cancer. EBioMedicine 2022;76:103873. doi:10.1016/j.ebiom.2022.103873
    5. ↵
      1. Dhir M,
      2. Melin AA,
      3. Douaiher J, et al
      . A Review and Update of Treatment Options and Controversies in the Management of Hepatocellular Carcinoma. Ann Surg 2016;263:1112–25. doi:10.1097/SLA.0000000000001556
      OpenUrl
    6. ↵
      1. Vibert E,
      2. Schwartz M,
      3. Olthoff KM
      . Advances in resection and transplantation for hepatocellular carcinoma. J Hepatol 2020;72:262–76. doi:10.1016/j.jhep.2019.11.017
      OpenUrlCrossRef
    7. ↵
      1. Villanueva A
      . Hepatocellular Carcinoma. N Engl J Med 2019;380:1450–62. doi:10.1056/NEJMra1713263
      OpenUrlCrossRefPubMed
    8. ↵
      1. Yang Y
      . Cancer immunotherapy: harnessing the immune system to battle cancer. J Clin Invest 2015;125:3335–7. doi:10.1172/JCI83871
      OpenUrlCrossRefPubMed
    9. ↵
      1. Morrison AH,
      2. Byrne KT,
      3. Vonderheide RH
      . Immunotherapy and Prevention of Pancreatic Cancer. Trends Cancer 2018;4:418–28. doi:10.1016/j.trecan.2018.04.001
      OpenUrl
    10. ↵
      1. Li K,
      2. Zhang A,
      3. Li X, et al
      . Advances in clinical immunotherapy for gastric cancer. Biochim Biophys Acta Rev Cancer 2021;1876:188615. doi:10.1016/j.bbcan.2021.188615
      OpenUrlCrossRef
    11. ↵
      1. Sangro B,
      2. Sarobe P,
      3. Hervás-Stubbs S, et al
      . Advances in immunotherapy for hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol 2021;18:525–43. doi:10.1038/s41575-021-00438-0
      OpenUrlCrossRefPubMed
    12. ↵
      1. Afshari AR,
      2. Sanati M,
      3. Ahmadi SS, et al
      . Harnessing the capacity of phytochemicals to enhance immune checkpoint inhibitor therapy of cancers: A focus on brain malignancies. Cancer Lett 2024;593:216955. doi:10.1016/j.canlet.2024.216955
      OpenUrl
    13. ↵
      1. Gong J,
      2. Chehrazi-Raffle A,
      3. Reddi S, et al
      . Development of PD-1 and PD-L1 inhibitors as a form of cancer immunotherapy: a comprehensive review of registration trials and future considerations. J Immunother Cancer 2018;6:8. doi:10.1186/s40425-018-0316-z
    14. ↵
      1. Chauvin JM,
      2. Zarour HM
      . TIGIT in cancer immunotherapy. J Immunother Cancer 2020;8:e000957. doi:10.1136/jitc-2020-000957
    15. ↵
      1. Zhang Q,
      2. He Y,
      3. Luo N, et al
      . Landscape and Dynamics of Single Immune Cells in Hepatocellular Carcinoma. Cell 2019;179:829–45. doi:10.1016/j.cell.2019.10.003
      OpenUrlCrossRefPubMed
    16. ↵
      1. Doudna JA,
      2. Charpentier E
      . Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science 2014;346:1258096. doi:10.1126/science.1258096
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Wang T,
      2. Wei JJ,
      3. Sabatini DM, et al
      . Genetic Screens in Human Cells Using the CRISPR-Cas9 System. Science 2014;343:80–4. doi:10.1126/science.1246981
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Shalem O,
      2. Sanjana NE,
      3. Hartenian E, et al
      . Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 2014;343:84–7. doi:10.1126/science.1247005
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Chen H,
      2. Dong K,
      3. Ding J, et al
      . CRISPR genome-wide screening identifies PAK1 as a critical driver of ARSI cross-resistance in prostate cancer progression. Cancer Lett 2024;587:216725. doi:10.1016/j.canlet.2024.216725
      OpenUrl
    20. ↵
      1. Marcotte R,
      2. Sayad A,
      3. Brown KR, et al
      . Functional Genomic Landscape of Human Breast Cancer Drivers, Vulnerabilities, and Resistance. Cell 2016;164:293–309. doi:10.1016/j.cell.2015.11.062
      OpenUrlCrossRefPubMed
    21. ↵
      1. Meyers RM,
      2. Bryan JG,
      3. McFarland JM, et al
      . Computational correction of copy number effect improves specificity of CRISPR-Cas9 essentiality screens in cancer cells. Nat Genet 2017;49:1779–84. doi:10.1038/ng.3984
      OpenUrlCrossRefPubMed
    22. ↵
      1. Chen S,
      2. Sanjana NE,
      3. Zheng K, et al
      . Genome-wide CRISPR screen in a mouse model of tumor growth and metastasis. Cell 2015;160:1246–60. doi:10.1016/j.cell.2015.02.038
      OpenUrlCrossRefPubMed
    23. ↵
      1. Song C-Q,
      2. Li Y,
      3. Mou H, et al
      . Genome-Wide CRISPR Screen Identifies Regulators of Mitogen-Activated Protein Kinase as Suppressors of Liver Tumors in Mice. Gastroenterology 2017;152:1161–73. doi:10.1053/j.gastro.2016.12.002
      OpenUrl
    24. ↵
      1. Wang X,
      2. Yu J,
      3. Liu X, et al
      . PSMG2-controlled proteasome-autophagy balance mediates the tolerance for MEK-targeted therapy in triple-negative breast cancer. Cell Rep Med 2022;3:100741. doi:10.1016/j.xcrm.2022.100741
      OpenUrl
    25. ↵
      1. Zhang Z,
      2. Kong X,
      3. Ligtenberg MA, et al
      . RNF31 inhibition sensitizes tumors to bystander killing by innate and adaptive immune cells. Cell Rep Med 2022;3:100655. doi:10.1016/j.xcrm.2022.100655
    26. ↵
      1. Dong MB,
      2. Wang G,
      3. Chow RD, et al
      . Systematic Immunotherapy Target Discovery Using Genome-Scale In Vivo CRISPR Screens in CD8 T Cells. Cell 2019;178:1189–204. doi:10.1016/j.cell.2019.07.044
      OpenUrlCrossRefPubMed
    27. ↵
      1. Feng J,
      2. Lu H,
      3. Ma W, et al
      . Genome-wide CRISPR screen identifies synthetic lethality between DOCK1 inhibition and metformin in liver cancer. Protein Cell 2022;13:825–41. doi:10.1007/s13238-022-00906-6
      OpenUrl
    28. ↵
      1. Wei L,
      2. Lee D,
      3. Law C-T, et al
      . Genome-wide CRISPR/Cas9 library screening identified PHGDH as a critical driver for Sorafenib resistance in HCC. Nat Commun 2019;10:4681. doi:10.1038/s41467-019-12606-7
    29. ↵
      1. Li F,
      2. Huang Q,
      3. Luster TA, et al
      . In Vivo Epigenetic CRISPR Screen Identifies Asf1a as an Immunotherapeutic Target in Kras-Mutant Lung Adenocarcinoma. Cancer Discov 2020;10:270–87. doi:10.1158/2159-8290.cd-19-0780
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Wang B,
      2. Wang M,
      3. Zhang W, et al
      . Integrative analysis of pooled CRISPR genetic screens using MAGeCKFlute. Nat Protoc 2019;14:756–80. doi:10.1038/s41596-018-0113-7
      OpenUrlCrossRefPubMed
    31. ↵
      1. Cheng W-C,
      2. Tsui Y-C,
      3. Ragusa S, et al
      . Uncoupling protein 2 reprograms the tumor microenvironment to support the anti-tumor immune cycle. Nat Immunol 2019;20:206–17. doi:10.1038/s41590-018-0290-0
      OpenUrlPubMed
    32. ↵
      1. Gatza ML,
      2. Silva GO,
      3. Parker JS, et al
      . An integrated genomics approach identifies drivers of proliferation in luminal-subtype human breast cancer. Nat Genet 2014;46:1051–9. doi:10.1038/ng.3073
      OpenUrlCrossRefPubMed
    33. ↵
      1. Parikh JR,
      2. Klinger B,
      3. Xia Y, et al
      . Discovering causal signaling pathways through gene-expression patterns. Nucleic Acids Res 2010;38:W109–17. doi:10.1093/nar/gkq424
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Harlin H,
      2. Meng Y,
      3. Peterson AC, et al
      . Chemokine expression in melanoma metastases associated with CD8+ T-cell recruitment. Cancer Res 2009;69:3077–85. doi:10.1158/0008-5472.CAN-08-2281
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Tirosh I,
      2. Izar B,
      3. Prakadan SM, et al
      . Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq. Science 2016;352:189–96. doi:10.1126/science.aad0501
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Wang X,
      2. Li X,
      3. Wei X, et al
      . PD-L1 is a direct target of cancer-FOXP3 in pancreatic ductal adenocarcinoma (PDAC), and combined immunotherapy with antibodies against PD-L1 and CCL5 is effective in the treatment of PDAC. Sig Transduct Target Ther 2020;5:38. doi:10.1038/s41392-020-0144-8
      OpenUrl
    37. ↵
      1. Zhang X-N,
      2. Yang K-D,
      3. Chen C, et al
      . Pericytes augment glioblastoma cell resistance to temozolomide through CCL5-CCR5 paracrine signaling. Cell Res 2021;31:1072–87. doi:10.1038/s41422-021-00528-3
      OpenUrlCrossRef
    38. ↵
      1. Luo J,
      2. Cheng K,
      3. Ji X, et al
      . Anlotinib enhanced CD8+ T cell infiltration via induction of CCL5 improves the efficacy of PD-1/PD-L1 blockade therapy in lung cancer. Cancer Lett 2024;591:216892. doi:10.1016/j.canlet.2024.216892
      OpenUrl
    39. ↵
      1. De La Rochere P,
      2. Guil-Luna S,
      3. Decaudin D, et al
      . Humanized Mice for the Study of Immuno-Oncology. Trends Immunol 2018;39:748–63. doi:10.1016/j.it.2018.07.001
      OpenUrlCrossRefPubMed
    40. ↵
      1. Hanoune J,
      2. Defer N
      . Regulation and role of adenylyl cyclase isoforms. Annu Rev Pharmacol Toxicol 2001;41:145–74. doi:10.1146/annurev.pharmtox.41.1.145
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Doherty GJ,
      2. McMahon HT
      . Mechanisms of endocytosis. Annu Rev Biochem 2009;78:857–902. doi:10.1146/annurev.biochem.78.081307.110540
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. Gao Y,
      2. Nihira NT,
      3. Bu X, et al
      . Acetylation-dependent regulation of PD-L1 nuclear translocation dictates the efficacy of anti-PD-1 immunotherapy. Nat Cell Biol 2020;22:1064–75. doi:10.1038/s41556-020-0562-4
      OpenUrlCrossRefPubMed
    43. ↵
      1. Schnitzer JE,
      2. Oh P,
      3. Pinney E, et al
      . Filipin-sensitive caveolae-mediated transport in endothelium: reduced transcytosis, scavenger endocytosis, and capillary permeability of select macromolecules. J Cell Biol 1994;127:1217–32. doi:10.1083/jcb.127.5.1217
      OpenUrlAbstract/FREE Full Text
    44. ↵
      1. Zhang C,
      2. Wu Q,
      3. Huang H, et al
      . Caveolin-1 promotes Rfng expression via Erk-Jnk-p38 signaling pathway in mouse hepatocarcinoma cells. J Physiol Biochem 2019;75:549–59. doi:10.1007/s13105-019-00703-6
      OpenUrl
    45. ↵
      1. Tang Y,
      2. Zeng X,
      3. He F, et al
      . Caveolin-1 is related to invasion, survival, and poor prognosis in hepatocellular cancer. Med Oncol 2012;29:977–84. doi:10.1007/s12032-011-9900-5
      OpenUrlPubMed
    46. ↵
      1. Nguyen Ba AN,
      2. Pogoutse A,
      3. Provart N, et al
      . NLStradamus: a simple Hidden Markov Model for nuclear localization signal prediction. BMC Bioinformatics 2009;10:202. doi:10.1186/1471-2105-10-202
    47. ↵
      1. Timney BL,
      2. Raveh B,
      3. Mironska R, et al
      . Simple rules for passive diffusion through the nuclear pore complex. J Cell Biol 2016;215:57–76. doi:10.1083/jcb.201601004
      OpenUrlAbstract/FREE Full Text
    48. ↵
      1. Conti E,
      2. Uy M,
      3. Leighton L, et al
      . Crystallographic analysis of the recognition of a nuclear localization signal by the nuclear import factor karyopherin alpha. Cell 1998;94:193–204. doi:10.1016/s0092-8674(00)81419-1
      OpenUrlCrossRefPubMedWeb of Science
    49. ↵
      1. Enenkel C,
      2. Blobel G,
      3. Rexach M
      . Identification of a yeast karyopherin heterodimer that targets import substrate to mammalian nuclear pore complexes. J Biol Chem 1995;270:16499–502. doi:10.1074/jbc.270.28.16499
      OpenUrlAbstract/FREE Full Text
    50. ↵
      1. Soderholm JF,
      2. Bird SL,
      3. Kalab P, et al
      . Importazole, a small molecule inhibitor of the transport receptor importin-β. ACS Chem Biol 2011;6:700–8. doi:10.1021/cb2000296
      OpenUrlCrossRefPubMedWeb of Science
    51. ↵
      1. Hurwitz SN,
      2. Rider MA,
      3. Bundy JL, et al
      . Proteomic profiling of NCI-60 extracellular vesicles uncovers common protein cargo and cancer type-specific biomarkers. Oncotarget 2016;7:86999–7015. doi:10.18632/oncotarget.13569
      OpenUrlCrossRefPubMed
    52. ↵
      1. Buoncervello M,
      2. Gabriele L,
      3. Toschi E
      . The Janus Face of Tumor Microenvironment Targeted by Immunotherapy. Int J Mol Sci 2019;20:4320. doi:10.3390/ijms20174320
    53. ↵
      1. Li C,
      2. Xie J,
      3. Lu Z, et al
      . ADCY7 supports development of acute myeloid leukemia. Biochem Biophys Res Commun 2015;465:47–52. doi:10.1016/j.bbrc.2015.07.123
      OpenUrl
    54. ↵
      1. Jiang LI,
      2. Sternweis PC,
      3. Wang JE
      . Zymosan activates protein kinase A via adenylyl cyclase VII to modulate innate immune responses during inflammation. Mol Immunol 2013;54:14–22. doi:10.1016/j.molimm.2012.10.027
      OpenUrlCrossRefPubMed
    55. ↵
      1. Aran D,
      2. Camarda R,
      3. Odegaard J, et al
      . Comprehensive analysis of normal adjacent to tumor transcriptomes. Nat Commun 2017;8:1077. doi:10.1038/s41467-017-01027-z
    56. ↵
      1. Wang X,
      2. Lang M,
      3. Zhao T, et al
      . Cancer-FOXP3 directly activated CCL5 to recruit FOXP3+Treg cells in pancreatic ductal adenocarcinoma. Oncogene 2017;36:3048–58. doi:10.1038/onc.2016.458
      OpenUrl
    57. ↵
      1. Sun Y-F,
      2. Wu L,
      3. Liu S-P, et al
      . Dissecting spatial heterogeneity and the immune-evasion mechanism of CTCs by single-cell RNA-seq in hepatocellular carcinoma. Nat Commun 2021;12:4091. doi:10.1038/s41467-021-24386-0
    58. ↵
      1. Dangaj D,
      2. Bruand M,
      3. Grimm AJ, et al
      . Cooperation between Constitutive and Inducible Chemokines Enables T Cell Engraftment and Immune Attack in Solid Tumors. Cancer Cell 2019;35:885–900. doi:10.1016/j.ccell.2019.05.004
      OpenUrlCrossRefPubMed
    59. ↵
      1. Yang Z,
      2. Feng J,
      3. Xiao L, et al
      . Tumor‐Derived Peptidoglycan Recognition Protein 2 Predicts Survival and Antitumor Immune Responses in Hepatocellular Carcinoma. Hepatology 2020;71:1626–42. doi:10.1002/hep.30924
      OpenUrl
    60. ↵
      1. Matsushima K,
      2. Terashima Y,
      3. Toda E, et al
      . Chemokines in inflammatory and immune diseases. Inflamm Regen 2011;31:11–22. doi:10.2492/inflammregen.31.11
      OpenUrl
    61. ↵
      1. Kim J,
      2. Kang J,
      3. Kang Y-L, et al
      . Ketohexokinase-A acts as a nuclear protein kinase that mediates fructose-induced metastasis in breast cancer. Nat Commun 2020;11:5436. doi:10.1038/s41467-020-19263-1
    62. ↵
      1. Zhen Y,
      2. Sørensen V,
      3. Skjerpen CS, et al
      . Nuclear import of exogenous FGF1 requires the ER-protein LRRC59 and the importins Kpnα1 and Kpnβ1. Traffic 2012;13:650–64. doi:10.1111/j.1600-0854.2012.01341.x
      OpenUrlCrossRefPubMed
    63. ↵
      1. Lu X,
      2. An L,
      3. Fan G, et al
      . EGFR signaling promotes nuclear translocation of plasma membrane protein TSPAN8 to enhance tumor progression via STAT3-mediated transcription. Cell Res 2022;32:359–74. doi:10.1038/s41422-022-00628-8
      OpenUrl
    64. ↵
      1. Chai F,
      2. Li Y,
      3. Liu K, et al
      . Caveolin enhances hepatocellular carcinoma cell metabolism, migration, and invasion in vitro via a hexokinase 2-dependent mechanism. J Cell Physiol 2019;234:1937–46. doi:10.1002/jcp.27074
      OpenUrl
    65. ↵
      1. Yang L,
      2. Hu B,
      3. Zhang Y, et al
      . Suppression of the nuclear transporter-KPNβ1 expression inhibits tumor proliferation in hepatocellular carcinoma. Med Oncol 2015;32:128. doi:10.1007/s12032-015-0559-1
      OpenUrl
    66. ↵
      1. Han Q-F,
      2. Li W-J,
      3. Hu K-S, et al
      . Exosome biogenesis: machinery, regulation, and therapeutic implications in cancer. Mol Cancer 2022;21:207. doi:10.1186/s12943-022-01671-0
    67. ↵
      1. Kalluri R,
      2. LeBleu VS
      . The biology, function, and biomedical applications of exosomes. Science 2020;367. doi:10.1126/science.aau6977
    68. ↵
      1. Zhang M,
      2. Hu S,
      3. Liu L, et al
      . Engineered exosomes from different sources for cancer-targeted therapy. Sig Transduct Target Ther 2023;8:124. doi:10.1038/s41392-023-01382-y
      OpenUrl
    69. ↵
      1. Roy A,
      2. Girija As S,
      3. Sankar Ganesh P, et al
      . Exosome Mediated Cancer Therapeutic Approach:Present Status and Future Prospectives. Asian Pac J Cancer Prev 2023;24:363–73. doi:10.31557/APJCP.2023.24.2.363
      OpenUrl
    70. ↵
      1. Wen J,
      2. Chen Y,
      3. Liao C, et al
      . Engineered mesenchymal stem cell exosomes loaded with miR-34c-5p selectively promote eradication of acute myeloid leukemia stem cells. Cancer Lett 2023;575:216407. doi:10.1016/j.canlet.2023.216407
      OpenUrl
    71. ↵
      1. Wang Y-N,
      2. Wang H,
      3. Yamaguchi H, et al
      . COPI-mediated retrograde trafficking from the Golgi to the ER regulates EGFR nuclear transport. Biochem Biophys Res Commun 2010;399:498–504. doi:10.1016/j.bbrc.2010.07.096
      OpenUrlCrossRefPubMed

    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

    • Contributors JC, YJ and MH are joint first authors. HW and JF conceived and designed the experiments; JC, YJ, MH and CL performed the experiments and bioinformatics analysis; EL, YZ, XW, ZM, MG and YS collected the human specimens; JF, JC, YJ and MH analysed the data; JF, JC and YJ wrote the manuscript. All authors have read, revised and approved the final manuscript. JF and HW are responsible for the overall content as the guarantor.

    • Funding This study is supported by the National Research Program of China (2022YFC3400902), the National Natural Science Foundation of China (81988101, 82073411, 82203209), Shanghai Discipline Leaders Program (2022XD058), Shanghai Sailing Program (21YF1458500), China Postdoctoral Science Foundation (2022M720786).

    • 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.