CXCL16-driven CD4+ T cells orchestrate immunosurveillance against MHC-I-deficient hepatocellular tumors

Yunyi Zhou; Ping Chen; Zhixue Wang; Liming Gui; Zhengheng Zheng; Wei-Qiang Gao; Bin Ma

doi:10.1136/jitc-2025-012347

. 2026 May 12;14(5):e012347. doi: 10.1136/jitc-2025-012347

CXCL16-driven CD4+ T cells orchestrate immunosurveillance against MHC-I-deficient hepatocellular tumors

Yunyi Zhou ^1,^2,⁰, Ping Chen ^1,⁰, Zhixue Wang ^1,², Liming Gui ¹, Zhengheng Zheng ^1,², Wei-Qiang Gao ^2,³, Bin Ma ^1,^4,^✉

PMCID: PMC13182305 PMID: 42120180

Abstract

Background

Major histocompatibility complex class I (MHC)-I loss is a prevalent mechanism for immune evasion and resistance to immunotherapy. However, how MHC-I loss shapes the tumor microenvironment and influences immune cell interactions, ultimately affecting tumor growth, remains largely unknown.

Methods

We established B2m knockout (MHC-I-deficient) tumor cells using CRISPR/Cas9 and evaluated their growth in subcutaneous and orthotopic mouse models. Immune profiling was performed using flow cytometry and single-cell RNA sequencing. Antibody-mediated cell depletion was used to assess the functional contributions of specific immune subsets. Chemokine expression was analyzed by bulk RNA sequencing, quantitative PCR, ELISA and western blotting, and its functional relevance was determined using knockout or overexpression in tumor cells implanted in vivo. Signaling pathways were interrogated using pharmacological inhibition, RNA interference and western blotting.

Results

MHC-I loss promoted tumor growth in MC38, AKR, and LLC1 models, but unexpectedly suppressed Hepa1-6 and orthotopic MYC;Trp53^−/− hepatocarcinoma growth. This differential effect correlated with changes in immune infiltrates. CD4⁺ T cells, natural killer (NK) cells, and macrophages were required for suppression of MHC-I-deficient Hepa1-6 tumors. CD4⁺ T cells were essential for recruiting NK cells and monocytes/macrophages and for inducing their tumoricidal phenotypes, including iNOS (inducible nitric oxide synthase) expression in macrophages. The differential infiltration of CD4⁺ T cells was driven by opposite regulation of CXCL16 on B2m knockout: upregulation in Hepa1-6 cells and downregulation in other models. CXCL16 exerted potent antitumor effects by recruiting CD4⁺ T cells. Mechanistically, B2M loss regulated CXCL16 via suppression of Akt in MC38 and AKR cells, but via activation of NF-κB in Hepa1-6 cells.

Conclusion

CXCL16-driven CD4⁺ T cells are central regulators of antitumor immunity against MHC-I-deficient tumors. The context-dependent regulation of CXCL16 by MHC-I loss determines the immune landscape and tumor outcome, highlighting a potential therapeutic avenue for targeting MHC-I-deficient cancers.

Keywords: Major histocompatibility complex - MHC, Solid tumor, T cell, Tumor microenvironment - TME, Escape/evasion

WHAT IS ALREADY KNOWN ON THIS TOPIC

Major histocompatibility complex class I (MHC)-I loss is a common mechanism by which tumor cells evade CD8⁺ T cell-mediated killing and resist immunotherapy. Despite their increased vulnerability to natural killer (NK) cells and macrophages, MHC-I-deficient tumors often remain aggressive in vivo due to additional immunosuppressive mechanisms.
CD4⁺ T cells have been shown to reject immune-evasive, MHC-I-deficient tumors in some contexts, but their role and regulation within the tumor microenvironment remain unclear.

WHAT THIS STUDY ADDS

MHC-I loss exerts opposing effects on tumor growth across different tumor types: it promotes growth in non-hepatocellular carcinoma models but suppresses growth in hepatocellular carcinoma models, correlating with differential immune infiltration.
CD4⁺ T cells orchestrate antitumor immunity against MHC-I-deficient tumors, recruiting NK cells and reprogramming monocytes/macrophages toward a tumoricidal phenotype.
CXCL16 is revealed as a key chemokine differentially regulated by MHC-I loss via tumor type-specific signaling pathways, governing CD4⁺ T-cell infiltration and tumor suppression.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

This study challenges the conventional view that MHC-I loss uniformly promotes tumor progression, highlighting the need to consider tumor-specific immune microenvironmental responses.
This study provides a mechanistic rationale for targeting the CXCL16-CXCR6 axis to enhance CD4⁺ T cell-mediated antitumor immunity, particularly in MHC-I-deficient or immunotherapy-resistant cancers, supporting the development of CXCL16-based therapies such as local or targeted delivery as a potential strategy to overcome immune evasion and improve patient outcomes.

Background

Antigens presented through major histocompatibility complex class I (MHC-I) are essential for the recognition and killing of tumor cells by CD8⁺ T cells. Tumor cells frequently downregulate or even completely lose MHC-I during tumor progression or following immunotherapy, to escape immunosurveillance and/or resist immunotherapy.1,3 This downregulation is driven by diverse tumor-intrinsic alterations and extrinsic pressures from the immunosuppressive microenvironment. Since MHC-I serves as an inhibitory ligand for natural killer (NK) cells and macrophages, its loss could render tumor cells more vulnerable to them.^{4 5} However, MHC-I-deficient tumors often remain aggressive in vivo, likely due to additional immunosuppressive mechanisms.^{5 6} For instance, blocking IGF8 on tumors or NKG2A on NK cells enhances NK cell-mediated killing of MHC-I-deficient tumors.^{7 8}

Various oncogenic pathways contribute to an immune exclusion and/or suppression during tumor progression.⁹ Beyond protecting tumor cells from CD8⁺ T cells, how MHC-I loss shapes the immune microenvironment in different tumors and thereby affects tumor growth remains unclear. Notably, CD4⁺ T cells can reject immune-evasive, MHC-I-deficient tumors.10,13 Mechanistically, they induce an inducible nitric oxide synthase (iNOS)-expressing tumoricidal phenotype in myeloid cells, promoting inflammatory tumor cell death.¹³ Given the complexity of the tumor microenvironment (TME), whether CD4⁺ T cells are essential for the immunosurveillance of MHC-I-deficient tumors and how they are influenced by MHC-I loss remain elusive.

Here, we observed that the difference in growth between MHC-I-competent and MHC-I-deficient tumors correlated with immune infiltrates across models. CD4⁺ T cells regulated the infiltration and/or activity of NK cells and macrophages/monocytes, influencing MHC-I-deficient tumor growth. MHC-I-regulated expression of tumorous CXCL16 was crucial for CD4⁺ T cell-mediated immunosurveillance.

Results

MHC-I loss differentially affects tumor growth and immune microenvironment

To explore the effects of MHC-I loss, we generated MHC-I-deficient tumor cell lines via CRISPR/Cas9-mediated B2m knockout (KO) and established subcutaneous tumor models. MHC-I and programmed death-ligand 1 (PD-L1) were constitutively expressed in control (single-guide non-targeting (sgNT)) MC38, AKR, LLC1, and Hepa1-6 cells and were upregulated by interferon (IFN)-γ, but were nearly undetectable in B2m-KO (sgB2m) lines (online supplemental figure S1A). All cell lines were negative for MHC-II and the non-classical MHC-I molecules including Qa-1 and Qa-2 (online supplemental figure S1A). B2m KO did not alter cell proliferation in vitro, excluding a cell-autonomous effect (online supplemental figure S1B), but increased sensitivity to NK-mediated cytotoxicity (online supplemental figure S1C–F).

In vivo, MHC-I loss promoted tumor overgrowth in MC38, AKR, and LLC1 models (figure 1A–C). Surprisingly, in the Hepa1-6 tumor model, MHC-I deficiency inhibited tumor growth (figure 1D). Accordingly, CD4⁺ T and NK cell infiltration decreased on B2m KO in MC38, AKR, and LLC1 tumors, but increased in Hepa1-6 tumors (figure 1E–H, online supplemental figure S2A). Proportions of CD45⁺ cells or conventional dendritic cells among CD45⁺ cells were unchanged across models (figure 1E–H). B2m KO did not affect regulatory T cell or neutrophil proportions in LLC1 or Hepa1-6 models (online supplemental figure S2B,C). Thus, the effect of MHC-I loss on tumor growth correlated with the changes in immune infiltrates, particularly conventional CD4⁺ T cells, NK cells, and possibly macrophages.

To validate that B2m KO can also inhibit tumor growth, we used an orthotopic MYC;Trp53^−/− hepatocellular carcinoma (HCC) model induced by MYC overexpression and Trp53 knockout (figure 2A).^{14 15} B2m KO significantly suppressed tumor progression (figure 2B–D). B2m messenger RNA (mRNA) expression was reduced in control sgNT HCC versus healthy liver (figure 2E), possibly due to the inhibitory effect of MYC.¹⁶ A further reduction in B2m mRNA level correlated with decelerated HCC growth. Importantly, B2m KO increased conventional CD4⁺ T cells among live cells, and NK cells and monocytes among CD45⁺ cells (figure 2F). These results align with the Hepa1-6 model, suggesting that this effect of MHC-I/B2m loss may be HCC-specific.

CD4⁺ T cells are crucial for MHC-I-loss-induced Hepa1-6 tumor regression

To identify critical immune subsets, nude mice lacking T cells were first employed. In this model, Hepa1-6-sgB2m tumor growth recovered to control levels (figure 3A), indicating that T cells are indispensable for controlling MHC-I-deficient Hepa1-6 tumor growth. We then depleted specific immune populations with antibodies (figure 3B). Anti-CD8a treatment did not affect the inhibitory effect of MHC-I loss (figure 3C), indicating that CD8⁺ T cells are not responsible. In contrast, anti-CD4 treatment caused the largest increase in growth for both Hepa1-6-sgNT and Hepa1-6-sgB2m tumors, eliminating their growth difference, highlighting an essential role for CD4⁺ T cells (figure 3C). Depleting NK cells (anti-NK1.1) and macrophages (anti-CSF1R) in Hepa1-6-sgB2m tumors promoted overgrowth, indicating their suppressive roles. However, depletion of NK cells and macrophages in Hepa1-6-sgNT tumors caused drastic regression, unraveling their protumor roles in this context (figure 3C), consistent with recently reported immunosuppressive function of NK cells^{17 18} and well-recognized complex role of macrophages.⁶ Depletion efficiency was verified by flow cytometry (figure 3D–G, online supplemental figure S2D). Neutrophil depletion (anti-Ly6G) did not reverse the tumor-suppressive effect of B2m knockout on Hepa1-6 tumor growth (online supplemental figure S2E,F). Notably, anti-CD4 treatment significantly decreased tumor-infiltrating NK cells and total CD45⁺ cells in Hepa1-6-sgB2m tumors, suggesting that CD4⁺ T cells regulate NK cell infiltration and the overall immune microenvironment (figure 3D,G). Although macrophage proportions among CD45⁺ cells were unchanged, their proportion among live cells decreased after CD4⁺ T-cell depletion (figure 3H,I), implying a positive effect on macrophage accumulation. Thus, MHC-I loss-induced Hepa1-6 tumor regression depends on CD4⁺ T, NK cells, and macrophages, with CD4⁺ T cells being most crucial.

To investigate phenotypic changes in CD4⁺ T, NK cells, and macrophages from Hepa1-6-sgNT and Hepa1-6-sgB2m tumors, we performed single-cell RNA sequencing (scRNA-seq) on CD45⁺ populations (online supplemental figure S3A). Consistent with flow cytometric results (figure 1H), macrophage and NK cell infiltration increased in Hepa1-6-sgB2m tumors, as conventional CD4⁺ T cells were also substantially elevated in Hepa1-6-sgB2m tumors (online supplemental figure S3B). Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of CD4⁺ T cells identified the cytokine–cytokine receptor interaction pathway as the top enriched pathway in Hepa1-6-sgB2m tumors (online supplemental figure S3C), indicating a more active effector state. NK cells from these tumors showed upregulated cytotoxic genes (Gzma, Gzmb, Prf1) and enrichment of cytotoxicity pathways (online supplemental figure S3D,E), indicating enhanced tumor-killing activity. Monocytes/macrophages exhibited a cytokine-activated phenotype (online supplemental figure S3F,G). These phenotypic changes of NK cells and monocytes/macrophages may explain their opposing roles in Hepa1-6-sgNT versus Hepa1-6-sgB2m tumors.

CD4⁺ T cells are key mediators of NK cells and macrophages

Having established the essential role of CD4⁺ T cells, we next dissected how they orchestrate the immune microenvironment. scRNA-seq analysis of Hepa1-6-sgB2m tumors after anti-CD4 treatment showed a marked decline in monocytes and NK cells, while the macrophage proportions among CD45⁺ cells remained unchanged (figure 4A), consistent with flow cytometric data (figure 3G,H). CD4⁺ T cells were effectively depleted (figure 4B). CellChat analysis identified the CCL3/4/5–CCR5 axis as a key pathway for CD4⁺ T cell-mediated NK cell recruitment (figure 4C). Accordingly, CCR5-expressing NK cells were elevated in Hepa1-6-sgB2m tumors and decreased after CD4⁺ T-cell depletion (figure 4D). CCL4-driven NK cell migration was completely blocked by the CCR5 inhibitor Maraviroc, confirming that this chemotactic axis mediates NK cell recruitment by CD4⁺ T cells (online supplemental figure S4A–C). The cytotoxic profile of NK cells was largely preserved, except a moderate decline in Prf1 expression (figure 4E). Thus, CD4⁺ T cells primarily enhance NK cell infiltration via chemotactic attraction.

Gene Ontology (GO) analysis of monocytes and macrophages indicated downregulated cytokine-mediated responses and tumoricidal capacity after anti-CD4 treatment (figure 4G). Accordingly, CD4⁺ T-cell depletion reduced Cxcl9 and increased Apoe expression in macrophages (figure 4F). CXCL9 recruits NK and T cells and marks an antitumor phenotype linked to better patient prognosis,^{19 20} while Apoe promotes protumor M2-like polarization.²¹ CellChat and expression profiling further identified key interactions between conventional CD4⁺ T cells and macrophages, including: (1) CSF2-CSF2R, activating macrophages²²; (2) MIF-(CD74+CD44/CXCR4), supporting monocytes/macrophage recruitment, survival and activation^{23 24}; and (3) adhesion molecules like Thy1-(Itgam/Itgax+Itgb2) that facilitate cell–cell contact (figure 4C,H). Notably, mRNA levels of the critical tumoricidal gene Nos2 (encoding iNOS)¹³ rose in monocytes/macrophages from MHC-I-deficient Hepa1-6 tumors and fell on CD4⁺ T-cell depletion (figure 4I). Supporting this, Granulocyte-macrophage colony-stimulating factor (GM-CSF, encoded by Csf2) potently boosted iNOS expression in macrophages (online supplemental figure S4D). These data demonstrated that CD4⁺ T cells are essential for recruiting monocytes/macrophages and sustaining their effector functions.

CXCL16 promotes CD4⁺ T-cell infiltration to suppress MHC-I-deficient tumors

To elucidate factors underlying altered CD4⁺ T-cell infiltration, we performed bulk RNA-seq on tumor cell lines. Comparing genes oppositely regulated in MC38/AKR/LLC1 versus Hepa1-6 cells, the chemokine Cxcl16 was the only gene showing an opposite pattern (figure 5A,B), while a few genes changed similarly (online supplemental figure S5A). ELISA validated the altered secretion of CXCL16 in both MC38 and Hepa1-6 cells following MHC-I loss (online supplemental figure S5B). B2m knockout also increased Cxcl16 mRNA in orthotopic MYC;Trp53^−/− HCC (figure 5C). Consistently, overall CXCL16 protein was higher in the Hepa1-6-sgB2m tumors (online supplemental figure S5C), further confirming its upregulation at the protein level in vivo. Analysis of human tumor bulk RNA-seq datasets showed a positive correlation between CXCL16 and B2M expression, in colorectal and esophageal adenocarcinoma (online supplemental figure S5D,E). Analysis of human HCC scRNA-seq datasets revealed an inverse correlation between CXCL16 and B2M or HLA-A/B/C expression in malignant hepatocytes (online supplemental figure S5F,G), supporting the context-dependent regulation of CXCL16 by B2M across tumor types. To test whether the opposite correlation patterns might be confounded by differences in baseline B2M expression between tumor types, we compared B2M levels in liver versus colorectal cancer. Notably, B2M mRNA and protein levels were comparable between the two (online supplemental figure S5H), ruling out this possibility. Cxcr6, the only known receptor for CXCL16,^{25 26} was highly expressed in T cells and was upregulated in CD4⁺ T as well as CD8⁺ T cells from MHC-I-deficient Hepa1-6 tumors (figure 5D). This indicates that the CXCL16-CXCR6 axis is important for recruiting CD4⁺ T cells, whereas CXCR6⁺ CD8⁺ T-cell activation is limited due to the lack of MHC-I-restricted antigen presentation, consistent with the observation that CD8⁺ T-cell depletion did not affect tumor growth (figure 3C).

We next tested whether CXCL16 was required for tumor suppression. We knocked out Cxcl16 in MHC-I-deficient Hepa1-6 cells (online supplemental figure S6A). Cxcl16 knockout did not affect proliferation in vitro but markedly enhanced tumor growth in vivo (figure 5E, online supplemental figure S6B). Consistently, Cxcl16 knockout reduced infiltration by CD4⁺ T, CD8⁺ T, and NK cells (figure 5F). Although macrophage proportions were unchanged, iNOS expression in macrophages and the total monocyte-macrophage population were decreased (figure 5F). Depleting CD4⁺ T cells accelerated tumor growth and abolished the difference between Cxcl16-knockout and control tumors (figure 5E), indicating that the antitumor effect of CXCL16 depends on CD4⁺ T cells. Conversely, CXCL16 overexpression suppressed in vivo growth of both control and MHC-I-deficient MC38 cells, and MHC-I-deficient AKR cells, without affecting their proliferation in vitro (figure 5G,H, online supplemental figure S6C–F). This tumor-suppressive effect of CXCL16 in MHC-I-deficient tumors was reversed by anti-CD4 treatment (figure 5I), underscoring its essential role in regulating CD4⁺ T cell-mediated antitumor activity. Flow cytometry confirmed that CXCL16 promoted CD4⁺ T-cell infiltration and activation, and induced a tumoricidal monocytes/macrophage phenotype that depended on CD4⁺ T cells (figure 5J).

To directly test whether CXCL16 acts on CD4⁺ T cells via CXCR6, we performed a series of in vitro experiments. In a transwell assay with recombinant mouse CXCL16 placed in the lower chamber (online supplemental figure S7A), the proportion of CXCR6⁺ cells among CD4⁺ T cells significantly decreased in the upper chamber and increased in the lower chamber (online supplemental figure S7B), indicating specific migration of CXCR6⁺ CD4⁺ T cells toward CXCL16. Moreover, the migrated CXCR6⁺ cells in the lower chamber exhibited an activated phenotype with upregulated CD69 expression, whereas CXCR6⁻ cells showed minimal changes (online supplemental figure S7B). In a separate direct stimulation experiment, CXCL16 treatment elevated the overall CXCR6⁺ ratio within the CD4⁺ T cell population, likely attributable to increased proliferation (Ki67 upregulation) of CXCR6⁺ CD4⁺ T cells (online supplemental figure S7C). Although CXCR6⁻ CD4⁺ T cells showed modest increases in Ki67 expression on CXCL16 stimulation (online supplemental figure S7C), possibly due to secondary or bystander effects, their activation levels remained much lower than those of the CXCR6⁺ compartment. Together, these findings demonstrate that CXCL16 directly promotes the recruitment, proliferation, and activation of CXCR6⁺ CD4⁺ T cells.

To extend these findings to human cancers, we analyzed transcriptomic datasets. Analysis of human tumor bulk RNA-seq datasets revealed a positive correlation between CXCL16 and CD3E/CD4 expression in colorectal, esophageal, and HCCs (online supplemental figure S7D–F), supporting a conserved role for CXCL16 in recruiting CD4⁺ T cells across diverse tumor types. Furthermore, scRNA-seq analysis of human HCC showed that CXCR6-expressing CD4⁺ T cells are enriched for gene signatures associated with immune effector and cytotoxic functions (online supplemental figure S7G), consistent with the activated phenotype observed in our in vitro experiments. Collectively, these findings demonstrate that CXCL16 exerts its antitumor effects through CXCR6 on CD4⁺ T cells and serves as a critical determinant of the tumor phenotype driven by MHC-I loss.

B2M loss oppositely regulates CXCL16 via Akt suppression or NF-κB activation

The divergent regulation of CXCL16 prompted an investigation of the upstream signaling pathways responsible. KEGG pathway analysis revealed decreased PI3K/Akt signaling in MC38 and AKR cells—but not in Hepa1-6 cells—following B2M loss, whereas NF-κB signaling was activated only in Hepa1-6 cells (figure 6A–C). Based on reported roles of PI3K/Akt and NF-κB in regulating CXCL16 in other systems,²⁷ we explored their involvement. Western blotting confirmed reduced activity-associated phosphorylation of Akt in B2m knockout MC38 and AKR cells, but not in Hepa1-6 cells (figure 6D). Treatment with two distinct Akt inhibitors reduced Cxcl16 mRNA expression in MC38 and AKR cells (figure 6E), supporting a positive role for the PI3K/Akt pathway in these lines. In Hepa1-6 cells, B2m knockout promoted IκBα phosphorylation and subsequent degradation, leading to nuclear translocation and phosphorylation/activation of NF-κB p65 in (figure 6F). Both basal and TNF-α-induced Cxcl16 expression was decreased by knockdown of Rela (encoding NF-κB p65) (figure 6G), confirming NF-κB-dependent upregulation in Hepa1-6 cells. Together, these data reveal a tissue-specific mechanism whereby MHC-I/B2M loss divergently modulates CXCL16 through Akt suppression in some tumors and NF-κB activation in others.

Discussion

MHC-I-loss-mediated immune evasion typically allows tumors to escape recognition and killing by CD8⁺ T cells, promoting tumor progression and immunotherapy resistance. Here, we uncovered a tumor-dependent role for MHC-I loss in regulating tumor growth and the immune microenvironment. The divergent effects of MHC-I deficiency correlated with altered tumor-infiltrating immune cells, particularly CD4⁺ T and NK cells. Notably, MHC-I-deficient Hepa1-6 tumors and orthotopic HCC exhibited reduced growth, contrasting with accelerated progression in MC38, AKR and LLC1 models. We demonstrated that conventional CD4⁺ T cells are essential for an integrated immune response against MHC-I-deficient tumors (figure 7). Effective immunosurveillance requires the coordinated action of CD4⁺ T cells, NK cells, and monocytes/macrophages, as depletion of any subset abrogated tumor suppression (figure 3C).

Our data further reveal a pivotal instructional role for CD4⁺ T cells in determining the functional output of innate immune cells. Although NK cells and macrophages are present in both settings, their depletion yielded opposite effects on tumor growth (figure 3C). This functional reversal correlates with CD4⁺ T cell-dependent phenotypic reprogramming. In MHC-I-deficient tumors, while the loss of MHC-I per se heightens tumor cell susceptibility to NK-mediated killing, CD4⁺ T cells recruit CCR5⁺ NK cells (figure 4D) and enhance specific aspects of their cytotoxic potential, such as Prf1 expression (figure 4E). Moreover, CD4⁺ T cells drive monocytes/macrophages toward an iNOS⁺, Cxcl9-expressing, tumoricidal phenotype (figure 4F–I), contrasting their state in MHC-I-proficient settings. Thus, beyond mere recruitment, CD4⁺ T cells actively reshape the innate immune landscape, converting these cells into essential effectors of tumor suppression on MHC-I loss.

CD8⁺ T cells have been the main focus in cancer immunology and immunotherapy due to their direct cytotoxic ability. In contrast, the antitumor function of conventional CD4⁺ T cells is often underestimated. Consistent with reports that CD4⁺ T-cell transfer can be more efficient at tumor rejection than CD8⁺ T cells in various mouse tumor models,²⁸ our results also support a superior antitumor role of CD4⁺ T cells, active in both MHC-I-deficient and MHC-I-expressing tumors. In the Hepa1-6 model, CD4⁺ T-cell depletion accelerated tumor growth regardless of B2m status (figure 3C), indicating that these cells exert a baseline antitumor function even without CXCL16 upregulation. However, their activity is markedly amplified on MHC-I loss. This amplification is driven by the CXCL16-CXCR6 axis: B2M deficiency induces tumor-derived CXCL16 (figure 5A–C), which specifically recruits and activates CXCR6⁺ CD4⁺ T cells (figure 5D and online supplemental figure S7A). Consequently, CD4⁺ T-cell infiltration and effector activation are substantially increased in B2m-deficient tumors (figure 1H and online supplemental figure S3A), enabling them to more potently orchestrate the innate immune reprogramming described above. Thus, while CD4⁺ T cells possess intrinsic antitumor capacity independent of MHC-I expression, their potency is greatly enhanced by the CXCL16-driven recruitment and activation cascade that operates specifically in MHC-I-deficient hepatocellular tumors.

Although CD4⁺ T cells may directly recognize MHC-II-positive tumor cells, most solid tumors lack intrinsic MHC-II as do our models. Therefore, CD4⁺ T cell-mediated rejection predominantly involves indirect mobilization of other effector cells. Our study provides novel evidence that NK cell infiltration into tumors is at least partially regulated by CD4⁺ T cells, complementing the interplay between monocytes/macrophages and CD4⁺ T cells reported here or elsewhere.^{13 29} Although CD4⁺ T-cell priming requires antigen presentation by MHC-II-expressing antigen presenting cells (APCs), their indirect killing of tumor cells is not antigen-specific; such bystander effects are crucial for eliminating antigen-negative cells. Given their versatile role in assisting CD8⁺ T cells, NK cells and macrophages, immunotherapies using or targeting CD4⁺ T cells hold great promise.

Recent studies have reported alterations in the TME associated with MHC-I loss, but underlying mechanisms remain unclear.^{8 12} Our study provides direct molecular evidence that MHC-I/B2M-regulated CXCL16 governs CD4⁺ T-cell infiltration. Loss of MHC-I/B2M exerted opposite effects on CXCL16 expression, correspondingly enhancing or diminishing CD4⁺ T-cell recruitment. Notably, the positive expression correlation of CXCL16 and B2M expression in colorectal and esophageal tumors, alongside their negative correlation in HCC, were conserved between mouse models and human cancers. The precise upstream mechanisms are not fully resolved, but we show that divergent activity of PI3K/Akt and NF-κB pathways underlies the distinct outcomes. In MC38 and AKR cells, B2M knockout leads to complete loss of surface MHC-I heterodimers, which appears to disrupt a constitutive survival signal that maintains basal PI3K-Akt activity. Consistent with this observation, previous reports have shown that MHC-I can physically and functionally interact with integrins and growth factor receptors to sustain Akt signaling,^{30 31} as well as through its interaction with PIP5K1A.³² It is therefore plausible that similar mechanisms operate in our models, although direct evidence is lacking. Consequently, B2M knockout in these cells reduces Akt phosphorylation and downregulates Cxcl16 transcription. In contrast, in Hepa1-6 cells, B2M knockout does not affect Akt activity but instead triggers NF-κB activation. Notably, two independent studies have demonstrated that constitutively expressed membrane MHC-I molecules negatively regulate TLR-triggered NF-κB activation via the Fps-SHP-2 pathway, which targets TRAF6.33,35 In this model, intact MHC-I molecules suppress NF-κB signaling; their loss (as in B2M deficiency) removes this inhibitory brake, leading to enhanced NF-κB activity. While this mechanism has been established in other systems, whether it accounts for the NF-κB activation observed in our Hepa1-6 cells remains speculative and requires further investigation. Nevertheless, the loss of surface MHC-I in our Hepa1-6 cells results in enhanced NF-κB activity and subsequent CXCL16 upregulation.

Beyond the context-dependent reverse signaling described above,³⁵ B2M may also act as a secreted soluble factor,³⁶ indicating that B2M—or possibly other MHC-I components—possesses biological functions extending beyond antigen presentation. We acknowledge that the current study does not provide direct experimental evidence for these proposed mechanisms, and the precise molecular events linking B2M loss to the activation of NF-κB in Hepa1-6 cells remain to be fully defined. Future studies involving detailed molecular dissection of the signaling pathways downstream of MHC-I will be required to fully resolve how B2M loss differentially engages PI3K-Akt versus NF-κB signaling in distinct tumor types. Nevertheless, CXCL16 consistently demonstrated CD4⁺ T cell-recruiting and antitumor effects across multiple mouse models. Given the reported benefits of the CXCL16-CXCR6 axis on cytotoxic T cells,³⁷ delivering CXCL16 locally or via cellular carriers like mesenchymal stem cells²⁰ may offer a novel therapeutic strategy.

Materials and methods

Cell lines

The MC38 murine colon adenocarcinoma cell line was originally derived from colon adenocarcinoma in C57BL/6 mice and was purchased from Nanjing Cobioer Biosciences (Nanjing, China). The AKR murine esophageal squamous cell carcinoma cell line was established from the ED-L2-cyclin-D1;p53^−/− mice and was from the Cell Bank of Chinese Academy of Sciences (Shanghai, China). The LLC1 (Lewis lung carcinoma) cell line, originating from a spontaneous lung carcinoma in C57BL/6 mice and the Hepa1-6 murine hepatoma cell line, derived from a BW7756 hepatoma in C57L mice, were both purchased from Pricella Biotechnology (Wuhan, China). Parental and genetically modified tumor cell lines were cultured in Dulbecco's Modified Eagle Medium (DMEM, Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin/amphotericin B (Beyotime), and incubated at 37°C and 5% CO₂. All cell lines were tested negative for Mycoplasma (InvivoGen).

B2m-deficient tumor cell lines were generated using CRISPR/Cas9-mediated knockout. Three different target sequences for single-guide RNA (sgRNAs) were chosen from the Mouse CRISPR Knockout Pooled Library (Brie)³⁸ and were used in combination for each gene: B2m (sgB2m), #1: ATTTGGATTTCAATGTGAGG, #2: ACTCACTCTGGATAGCATAC, #3: TGAGTATACTTGAATTTGAG; Cxcl16 (sgCxcl16), #1: TCTGGCACCCAGATACCGCA, #2: CCATTAGGCGATGGCAACCA, #3: CCAGTGGGTCCGTGAACTAG. The sgNT sequence was CTGAAAAAGGAAGGAGTTGA.³⁹ The sgRNAs were cloned to the lentiGuide-Puro plasmid (Addgene #52963) or lentiGuide-Hygro (Addgene #139462). lentiCas9-Blast (Addgene #52962) was used to express Cas9. Mouse Cxcl16 complementary DNA (cDNA) was synthesized and cloned into a Green Fluorescent Protein (GFP)-expressing lentivector for overexpression.

Hepa1-6 cells stably expressing scramble shRNA (Hepa1-6-shScr) or Rela-targeting shRNA (Hepa1-6-shRela) were generated using a lentiviral shRNA system. The scramble shRNA construct was obtained from Addgene (scramble shRNA, #1864). The shRNA sequence targeting mouse Rela was 5′-GCATGCGATTCCGCTATAAAT-3′ and was cloned into the pLKO.1 lentiviral vector (Addgene #10878).

Lentiviral transduction

Lentiviral particles were produced in 293T cells using the packaging plasmids pMD2.G (Addgene #12259) and psPAX2 (Addgene #12260). Target cells were transduced with lentiviral supernatants in the presence of 6 µg/mL polybrene, and the medium was replaced with fresh culture medium after 18–24 hours. After an additional 48 hours of incubation, antibiotic selection was performed as follows: cells transduced with lentiCas9-Blast were selected with 10–15 µg/mL blasticidin S; cells transduced with lentiGuide-Puro carrying the sgB2m constructs and pLKO.1 carrying shRela constructs were selected with 3 µg/mL puromycin; and cells transduced with lentiGuide-Hygro carrying the sgCxcl16 construct were selected with 400 µg/mL hygromycin for more than 1 week. For CXCL16-overexpressing cell lines, GFP-positive cells were sorted by fluorescence-activated cell sorting.

Flow cytometric analysis of MHC-I, MHC-II, Qa-1, Qa-2, and PD-L1 expression on tumor cell lines

The indicated cancer cell lines were stimulated with 20 ng/mL IFN-γ (BioLegend) for 24 hours and then harvested using TrypLE Express (Thermo Fisher Scientific). Cells were stained with BV510 anti-mouse MHC-I, BUV395 anti-mouse MHC-II, PE anti-mouse Qa-1(b), APC anti-mouse Qa-2 and PE-Cy7 anti-mouse PD-L1. Samples were analyzed using the BD LSRFortessa (BD Biosciences) and FlowJo V.10.4 software.

Cell proliferation assay

Tumor cell lines were seeded in 96-well plates at a density of 3,000 cells per well. Cell proliferation was detected using the CCK-8 reagent (TargetMol). When cells adhered, this was designated as the 0 hours time point. Measurements were then taken every 24 hours for a total of 72 hours using a microplate reader (Tecan).

Isolation and culture of murine NK cells and macrophages

For NK cell isolation, 8-week-old mice were euthanized, and the spleens were mechanically ground and filtered through 40 µm cell strainers (BioFil). Cell suspensions were treated with red blood cell lysis solution and then stained with biotin-coupled anti-NK1.1 antibody (BioLegend). Anti-biotin microbeads (Miltenyi Biotec) were added to bind antibody-labeled cells. NK1.1⁺ cells were then isolated using magnetic-activated cell sorting (MACS) LS columns and a MidiMACS separator (Miltenyi Biotec). NK cells were cultured in Roswell Park Memorial Institute (RPMI-1640), supplemented with 10% FBS, 1% antibiotics, 1% sodium pyruvate, 1% non-essential amino acid (NEAA), and 2,000 U/mL recombinant mouse interleukin-2 (R&D Systems). NK cell purity (>80%) was confirmed by flow cytometry.

Murine macrophages were isolated from peritoneal lavage fluid. To increase macrophage yield, mice were intraperitoneally injected with 1–2 mL of sterile 3% thioglycollate medium (MedChemExpress) 5 days before cell harvest. After euthanasia, the abdominal skin was retracted to expose the intact peritoneal wall. Then, 10 mL of ice-cold phosphate-buffered saline (PBS) was injected into the peritoneal cavity, and the abdomen was gently massaged to dislodge resident and elicited cells. The peritoneal fluid was slowly aspirated using a syringe, transferred to centrifuge tubes, and centrifuged at 400×g for 10 min at 4°C. The collected cells were seeded into culture dishes and cultured in RPMI-1640 medium supplemented with 10% FBS, 1% antibiotics, 1% sodium pyruvate, and 1% NEAA. After 1 hour of incubation at 37°C, non-adherent cells were removed by medium replacement, and the adherent macrophages were further cultured for subsequent experiments.

Co-culture of NK and tumor cells

Tumor cells were labeled using carboxyfluorescein diacetate succinimidyl ester (CFSE, Thermo Fisher Scientific). NK cells were then plated in 24-well plates (50,000 cells in 500 µL). NK cells were co-cultured with tumor cells at a 1:1 ratio for 24–48 hours. The apoptotic cell death of the tumor cells was evaluated by Annexin V and Zombie Violet (BioLegend) staining.

Murine tumor models

C57BL/6J mice were obtained from Shanghai Model Organisms. All animal studies were carried out at Shanghai Jiao Tong University. All animal procedures were approved by the Institutional Animal Care and Use Committee of Shanghai Jiao Tong University, Shanghai, China (Approval No. 2023013). The reporting of this study was conducted in accordance with ARRIVE V.2.0 guidelines. Mice were randomly assigned to experimental groups before tumor inoculation. The random sequence was generated by R software according to their ear tag ID. Genetically modified MC38, AKR, LLC1, and Hepa1-6 cells (1–2×10⁶) were subcutaneously injected into the lower flank of female C57BL/6 mice (6–8 weeks) in 100 µL PBS. Tumor growth was measured every 3 days using a caliper, and tumor volume was calculated using the formula: tumor volume (mm³)=length×(width)²/2. Animals were measured in a randomized sequence rather than group by group and were performed by a researcher who was blinded to the treatment groups. Tumors or peripheral blood were harvested for flow cytometric analysis.

Hydrodynamic tail-vein injection

To establish an orthotopic hepatocellular carcinoma model with MHC-I deficiency in mice, we performed hydrodynamic tail-vein injection.^{14 15} The plasmid mixture was prepared by dissolving the following components in sterile 0.9% NaCl solution (total volume 2 mL): 12 µg of pT3-EF1a-MYC-IRES-luciferase (MYC-luc) (Addgene #129775), 10 µg of PX330-sg-p53 (sg-p53) (Addgene #59910), 12 µg of a mixture of two lentiGuide-sgB2m plasmids (1:1) or 12 µg of control lentiGuide-sgNT plasmid, and 6 µg of the transposon SB13 transposase-encoding plasmid. The plasmid/NaCl solution was rapidly injected into the tail vein of mice at a volume equivalent to 10% of their body weight (eg, 2 mL for a 20 g mouse) within 5–7 s.

Tumor burden was monitored through bioluminescent imaging with an IVIS Lumina Spectrum camera Series III (PerkinElmer). Mice were intraperitoneally injected with 150 mg/kg D-Luciferin Potassium (15 mg/mL, MedChemExpress) 10 min prior to imaging. Mice were euthanized on reaching humane endpoints, including severe abdominal distension (ascites), significant weight loss (>20%), or lethargy. Animals were excluded from this study if they failed to develop visible tumors.

Real-time quantitative PCR analysis

Total RNA was isolated from tissue samples using the Total RNA Isolation Kit (Vazyme). The cDNA was synthesized from 500 ng total RNA using the PrimeScript RT Reagent Kit (Takara). Quantitative PCR amplification was performed in triplicate using TB Green Premix Ex Taq II (Takara) on an ABI 7900HT Fast Real-Time PCR System (Applied Biosystems). The relative mRNA expression levels of target genes were normalized to GAPDH and calculated using the comparative ΔΔCt method.

Depleting antibody treatments

A total of 200 µg anti-CD8a (clone 2.43), 200 µg anti-CD4 (clone GK1.5), 200 µg anti-NK1.1 (clone PK136), 400 µg anti-CSF1R (Clone AFS98), and 200 µg anti-Ly6G (clone 1A8) InVivoMAb antibodies (Bio X Cell) or PBS were administrated intraperitoneally on the days as indicated in figure 2. Rat IgG2b (clone LTF-2) was used as control for anti-CD4 treatment in figures3 4. Rat IgG2a (clone 2A3) was used as control for anti-Ly6G treatment.

Flow cytometric analysis of immune infiltrates in tumors

Tumor dissociation solution was prepared by dissolving 0.05 mg/mL DNase I (Sigma-Aldrich) and 0.1 mg/mL Liberase TL (Roche) in serum-free RPMI-1640. Tumors were harvested at the experimental endpoint, cut into small pieces, and digested in the tumor dissociation solution using the gentleMACS Octo Dissociator (Miltenyi Biotec). Cell suspensions were then filtered through 40 µm cell strainers (BioFil), followed by red blood cell lysis using red blood cell (RBC) lysis buffer (BioLegend) for 5 min at room temperature. After washing, cells were stained with Zombie Fixable Viability Dye (BioLegend) for 10 min at room temperature in the dark. Cells were then washed and stained with fluorochrome-conjugated surface antibodies for 30 min on ice. For intracellular cytokine analysis, cells were stimulated with Cell Activation Cocktail (with Brefeldin A; BioLegend) for 3–4 hours to allow intracellular protein accumulation. Following surface staining, cells were fixed and permeabilized using the Foxp3/Transcription Factor Fixation/Permeabilization Kit (eBioscience) for IFN-γ and Foxp3 detection, followed by intracellular staining with the corresponding antibodies. In parallel, for iNOS analysis, surface-stained cells were fixed and permeabilized using the Cyto-Fast Fix/Perm Buffer Set (BioLegend), followed by intracellular staining for iNOS. Samples were acquired on a BD Fortessa flow cytometer and analyzed using FlowJo V.10.4. Gating strategies are shown in online supplemental figure S8.

Bulk RNA sequencing

Control (sgNT) and MHC-I-deficient (sgB2m) MC38, AKR, LLC1, and Hepa1-6 tumor cells were collected in three biological replicates for RNA extraction using the RNA Easy Fast Tissue/Cell Kit (TIANGEN). Library preparation and sequencing were conducted by Neo-Bio (Shanghai, China) using the Illumina NovaSeq 6000 sequencing platform. The sequencing reads were aligned to the reference genome using STAR software. Differential expression analysis between sgB2m and sgNT was conducted using DESeq2, while pathway functional analysis was performed using TopGO. The raw and processed sequencing data have been deposited in the Gene Expression Omnibus (GEO) database under the accession number GSE303349.

Single-cell RNA sequencing

Tumors were harvested at day 24 after inoculation, and immune cells were sorted using anti-CD45 microbeads (Miltenyi). To account for intertumoral heterogeneity, CD45⁺ cells isolated from three individual tumors of the same experimental group were pooled to constitute one biological sample for scRNA-seq. Single-cell libraries were prepared using 10x Chromium Single Cell Platform (10x Genomics). Samples were sequenced on an Illumina NovaSeq 6000 platform using 150 bp paired-end reads (Genergy Bio-Technology). Raw sequencing data were processed with Cell Ranger software and analyzed in R V.4.4.0. Quality control was performed on the cells based on the following metrics: cells with fewer than 1,000 Unique Molecular Identifiers (UMIs), fewer than 500 detected genes, more than 20% mitochondrial UMIs, and more than 1% red blood cell UMIs were filtered out. Following the initial quality control steps, gene expression matrices were normalized by a scaling factor (10,000), and log-transformed. Highly variable features were identified using the variance-stabilizing transformation method. The data were scaled, and principal component analysis was performed to reduce dimensionality. Uniform Manifold Approximation and Projection was employed for the visualization of the clusters. Cell cycle scores were calculated and regressed out to mitigate the impact of cell cycle heterogeneity on the analysis; doublets were identified and removed using the DoubletFinder package. To integrate multiple samples and correct for batch effects, the Harmony package was used. SingleR was employed to assign cell-type annotations of clusters. Differential gene expression analysis was conducted using the FindMarkers function, and functional analysis was performed based on differential gene expression, employing the enrichGO and enrichKEGG functions. Detailed KEGG and GO enrichment profiles are provided in online supplemental file 2. CellChat was used to identify the interactions between immune cell clusters via ligand-receptor interactions. The single-cell RNA sequencing data generated in this study have been deposited in the Open Archive for Miscellaneous Data (OMIX) database (National Genomics Data Center), under the accession number OMIX016604.

Transwell migration assay

Murine NK cells treated with or without the CCR5 antagonist Maraviroc (MedChemExpress) were placed in the upper chambers of Transwell inserts, while the lower chambers were filled with RPMI medium in the presence or absence of 50 ng/mL recombinant mouse CCL4 (ABclonal). Following 18 hours of incubation, migrated cells in the lower chambers were harvested and quantified. To investigate the CXCL16–CXCR6 axis in T cells, murine splenocytes were seeded into the upper chambers of Transwell inserts, and the lower chambers were filled with RPMI medium containing either 200 ng/mL recombinant mouse CXCL16 (ABclonal) or vehicle control. After 6 hours of incubation, cells from both the upper and lower chambers were collected for flow cytometric analysis.

Macrophage stimulation assay

Murine peritoneal macrophages were treated with recombinant GM-CSF (ABclonal) for 2.5 days. Cells were then detached using EDTA and analyzed using flow cytometry.

ELISA for mouse CXCL16

Tumor cells were cultured for 24 hours, and the supernatants were subsequently collected. The cell number was determined using a cell counter (Countstar). The concentration of CXCL16 protein in the supernatant was then quantified using the Mouse CXCL16 ELISA Kit (Multi Sciences) according to the manufacturer’s instructions.

Western blotting

Hepa1-6-sgNT and Hepa1-6-sgB2m tumors were harvested at day 24 after inoculation, and total protein was isolated with Radio Immunoprecipitation Assay (RIPA) Lysis Buffer (Thermo Fisher) supplemented with protease and phosphatase inhibitors (Roche). Total protein from MC38, AKR, and Hepa1-6 tumor cells was also collected under the same conditions. Nuclear and cytoplasmic fractions from Hepa1-6 tumor cells were prepared with a Nuclear Protein Extraction Kit (Beyotime). A total of 20–30 µg of protein was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting. The protein bands were visualized by ultraviolet (UV) imaging, and band intensities were quantified using ImageJ software. Uncropped western blot images are shown in online supplemental figure S9.

Analysis of CXCL16 and B2M co-expression in scRNA-seq datasets

HCC datasets CNP0000650 (18 patients) and GSE156625 (14 patients) are from CNGBdb and GEO, respectively. scRNA-seq data were analyzed using Seurat (V.4.4.0) in R. Malignant hepatocytes were identified using established markers (ALB, ALDOB, KRT18). Normalized gene expression values for CXCL16, B2M were extracted from tumor cells using the FetchData function. Pearson and Spearman’s correlation coefficients were calculated to evaluate linear and monotonic associations, respectively, between CXCL16 and B2M expression across individual tumor cells. Statistical significance was assessed using the Wilcoxon test. In addition, the coefficient of determination (R²) was used to estimate the proportion of variance explained. Correlation results were visualized using scatter plots with fitted linear regression lines generated by the ggplot2 package. All statistical analyses were conducted in R, and p<0.05 was considered statistically significant.

Correlation analysis of CXCL16 with B2M and T-cell markers in human cancer databases

The cBioPortal (https://www.cbioportal.org/) was used to analyze the correlation between CXCL16 mRNA expression and the expression of classical HLA-I components and CD4⁺ T cell markers. For this analysis, datasets of colorectal adenocarcinoma (592 samples), esophageal adenocarcinoma (181 samples), lung adenocarcinoma (510 samples), and hepatocellular carcinoma (366 samples) with mRNA data (RNA Seq V2) from The Cancer Genome Atlas (TCGA) and PanCancer Atlas were employed. Tumor sample expression z-scores were compared with the log-transformed mRNA expression distribution of adjacent normal samples within the cohort (log RNA Seq V2 RNA-Seq by Expectation-Maximization [RSEM]). A z-score threshold of ±2.0 was established for this analysis. The analysis incorporated both Spearman’s rank correlation and Pearson correlation, with their correlation coefficients and corresponding p values reported in the results. To compare B2M expression levels between HCC and colorectal adenocarcinoma, TCGA RNA-seq datasets and mass spectrometry-based protein abundance data were retrieved from The Human Protein Atlas (https://www.proteinatlas.org/) and analyzed at both the mRNA and protein levels.

Statistical analysis

Statistical analysis was performed in R V.4.4.0 or GraphPad Prism V.10. For two-group comparison, unpaired Student’s t-test with Welch’s correction was used. For multiple-group comparison, one-way analysis of variance (ANOVA) was used. For tumor growth curves, two-way ANOVA was used to compare tumor volumes at different time points. Wilcoxon test was used to compare gene expression in single-cell RNA analysis. Results were presented as mean values±SE of the measurement, and p<0.05 was considered statistically significant. The 95% CIs for the primary outcome were calculated.

Supplementary material

online supplemental file 1

jitc-14-5-s001.pdf^{(13.5MB, pdf)}

DOI: 10.1136/jitc-2025-012347

online supplemental file 2

jitc-14-5-s002.xlsx^{(159.1KB, xlsx)}

DOI: 10.1136/jitc-2025-012347

Footnotes

Funding: This work was supported by grants from the Ministry of Science and Technology of the People’s Republic of China (2023YFC3404101), National Natural Science Foundation of China (W2431055), Science and Technology Commission of Shanghai Municipality (25ZR1401313), Natural Science Foundation of Chongqing Municipality, China (CSTB2024NSCQ-KJFZMSX0044), the National 111 Project of China (B21024), Shanghai Jiading District Health Commission (2025-KY-ZD-05), and Peak Disciplines (Type IV) of Institutions of Higher Learning in Shanghai.

Provenance and peer review: Not commissioned; externally peer reviewed.

Patient consent for publication: Not applicable.

Ethics approval: Not applicable.

Data availability statement

Data are available upon reasonable request.

References

1.Dersh D, Hollý J, Yewdell JW. A few good peptides: MHC class I-based cancer immunosurveillance and immunoevasion. Nat Rev Immunol. 2021;21:116–28. doi: 10.1038/s41577-020-0390-6. [DOI] [PubMed] [Google Scholar]
2.Hazini A, Fisher K, Seymour L. Deregulation of HLA-I in cancer and its central importance for immunotherapy. J Immunother Cancer. 2021;9:e002899. doi: 10.1136/jitc-2021-002899. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Yang K, Halima A, Chan TA. Antigen presentation in cancer - mechanisms and clinical implications for immunotherapy. Nat Rev Clin Oncol. 2023;20:604–23. doi: 10.1038/s41571-023-00789-4. [DOI] [PubMed] [Google Scholar]
4.Barkal AA, Weiskopf K, Kao KS, et al. Engagement of MHC class I by the inhibitory receptor LILRB1 suppresses macrophages and is a target of cancer immunotherapy. Nat Immunol. 2018;19:76–84. doi: 10.1038/s41590-017-0004-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Wolf NK, Kissiov DU, Raulet DH. Roles of natural killer cells in immunity to cancer, and applications to immunotherapy. Nat Rev Immunol. 2023;23:90–105. doi: 10.1038/s41577-022-00732-1. [DOI] [PubMed] [Google Scholar]
6.Christofides A, Strauss L, Yeo A, et al. The complex role of tumor-infiltrating macrophages. Nat Immunol. 2022;23:1148–56. doi: 10.1038/s41590-022-01267-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Li Y, Wu X, Sheng C, et al. IGSF8 is an innate immune checkpoint and cancer immunotherapy target. Cell. 2024;187:2703–16. doi: 10.1016/j.cell.2024.03.039. [DOI] [PubMed] [Google Scholar]
8.Taylor BC, Sun X, Gonzalez-Ericsson PI, et al. NKG2A Is a Therapeutic Vulnerability in Immunotherapy Resistant MHC-I Heterogeneous Triple-Negative Breast Cancer. Cancer Discov. 2024;14:290–307. doi: 10.1158/2159-8290.CD-23-0519. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Spranger S, Gajewski TF. Impact of oncogenic pathways on evasion of antitumour immune responses. Nat Rev Cancer. 2018;18:139–47. doi: 10.1038/nrc.2017.117. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Torrejon DY, Abril-Rodriguez G, Champhekar AS, et al. Overcoming Genetically Based Resistance Mechanisms to PD-1 Blockade. Cancer Discov. 2020;10:1140–57. doi: 10.1158/2159-8290.CD-19-1409. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Germano G, Lu S, Rospo G, et al. CD4 T Cell-Dependent Rejection of Beta-2 Microglobulin Null Mismatch Repair-Deficient Tumors. Cancer Discov. 2021;11:1844–59. doi: 10.1158/2159-8290.CD-20-0987. [DOI] [PubMed] [Google Scholar]
12.Torrejon DY, Galvez M, Abril-Rodriguez G, et al. Antitumor Immune Responses in B2M-Deficient Cancers. Cancer Immunol Res. 2023;11:1642–55. doi: 10.1158/2326-6066.CIR-23-0139. [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Kruse B, Buzzai AC, Shridhar N, et al. CD4+ T cell-induced inflammatory cell death controls immune-evasive tumours. Nature New Biol. 2023;618:1033–40. doi: 10.1038/s41586-023-06199-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Li Q, Guo W, Qian Y, et al. Protein O-fucosyltransferase 1 promotes PD-L1 stability to drive immune evasion and directs liver cancer to immunotherapy. J Immunother Cancer. 2024;12:e008917. doi: 10.1136/jitc-2024-008917. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Ruiz de Galarreta M, Bresnahan E, Molina-Sánchez P, et al. β-Catenin Activation Promotes Immune Escape and Resistance to Anti–PD-1 Therapy in Hepatocellular Carcinoma. Cancer Discov. 2019;9:1124–41. doi: 10.1158/2159-8290.CD-19-0074. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Versteeg R, Noordermeer IA, Krüse-Wolters M, et al. c-myc down-regulates class I HLA expression in human melanomas. EMBO J. 1988;7:1023–9. doi: 10.1002/j.1460-2075.1988.tb02909.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Guan X, Lu Y, Zhang Y, et al. Tumor-associated NK cells facilitate tumor growth via NKp46 in immunocompetent murine hepatocellular carcinoma. Immunol Lett. 2023;258:8–19. doi: 10.1016/j.imlet.2023.04.009. [DOI] [PubMed] [Google Scholar]
18.Neo SY, Tong L, Chong J, et al. Tumor-associated NK cells drive MDSC-mediated tumor immune tolerance through the IL-6/STAT3 axis. Sci Transl Med. 2024;16:eadi2952. doi: 10.1126/scitranslmed.adi2952. [DOI] [PubMed] [Google Scholar]
19.Bill R, Wirapati P, Messemaker M, et al. CXCL9:SPP1 macrophage polarity identifies a network of cellular programs that control human cancers. Science. 2023;381:515–24. doi: 10.1126/science.ade2292. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Yin P, Gui L, Wang C, et al. Targeted Delivery of CXCL9 and OX40L by Mesenchymal Stem Cells Elicits Potent Antitumor Immunity. Mol Ther. 2020;28:2553–63. doi: 10.1016/j.ymthe.2020.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Zheng P, Luo Q, Wang W, et al. Tumor-associated macrophages-derived exosomes promote the migration of gastric cancer cells by transfer of functional Apolipoprotein E. Cell Death Dis. 2018;9:434. doi: 10.1038/s41419-018-0465-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Kaufman HL, Ruby CE, Hughes T, et al. Current status of granulocyte–macrophage colony-stimulating factor in the immunotherapy of melanoma. J Immunotherapy Cancer. 2014;2:11. doi: 10.1186/2051-1426-2-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Fan H, Hall P, Santos LL, et al. Macrophage Migration Inhibitory Factor and CD74 Regulate Macrophage Chemotactic Responses via MAPK and Rho GTPase. The Journal of Immunology. 2011;186:4915–24. doi: 10.4049/jimmunol.1003713. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Kang I, Bucala R. The immunobiology of MIF: function, genetics and prospects for precision medicine. Nat Rev Rheumatol. 2019;15:427–37. doi: 10.1038/s41584-019-0238-2. [DOI] [PubMed] [Google Scholar]
25.Ozga AJ, Chow MT, Luster AD. Chemokines and the immune response to cancer. Immunity. 2021;54:859–74. doi: 10.1016/j.immuni.2021.01.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Mabrouk N, Tran T, Sam I, et al. CXCR6 expressing T cells: Functions and role in the control of tumors. Front Immunol. 2022;13:1022136. doi: 10.3389/fimmu.2022.1022136. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Korbecki J, Bajdak-Rusinek K, Kupnicka P, et al. The Role of CXCL16 in the Pathogenesis of Cancer and Other Diseases. Int J Mol Sci. 2021;22:3490. doi: 10.3390/ijms22073490. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Perez-Diez A, Joncker NT, Choi K, et al. CD4 cells can be more efficient at tumor rejection than CD8 cells. Blood. 2007;109:5346–54. doi: 10.1182/blood-2006-10-051318. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Bogen B, Fauskanger M, Haabeth OA, et al. CD4+ T cells indirectly kill tumor cells via induction of cytotoxic macrophages in mouse models. Cancer Immunol Immunother. 2019;68:1865–73. doi: 10.1007/s00262-019-02374-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Jin YP, Fishbein MC, Said JW, et al. Anti-HLA class I antibody-mediated activation of the PI3K/Akt signaling pathway and induction of Bcl-2 and Bcl-xL expression in endothelial cells. Hum Immunol. 2004;65:291–302. doi: 10.1016/j.humimm.2004.01.002. [DOI] [PubMed] [Google Scholar]
31.Valenzuela NM, Reed EF. The link between major histocompatibility complex antibodies and cell proliferation. Transplant Rev (Orlando) 2011;25:154–66. doi: 10.1016/j.trre.2011.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Li D, Zhang Q, Li L, et al. β2-Microglobulin Maintains Glioblastoma Stem Cells and Induces M2-like Polarization of Tumor-Associated Macrophages. Cancer Res. 2022;82:3321–34. doi: 10.1158/0008-5472.CAN-22-0507. [DOI] [PubMed] [Google Scholar]
33.Xu S, Liu X, Bao Y, et al. Constitutive MHC class I molecules negatively regulate TLR-triggered inflammatory responses via the Fps-SHP-2 pathway. Nat Immunol. 2012;13:551–9. doi: 10.1038/ni.2283. [DOI] [PubMed] [Google Scholar]
34.Wu H, Jiang K, Guo S, et al. IFN-τ Mediated Control of Bovine Major Histocompatibility Complex Class I Expression and Function via the Regulation of bta-miR-148b/152 in Bovine Endometrial Epithelial Cells. Front Immunol. 2018;9:167. doi: 10.3389/fimmu.2018.00167. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Muntjewerff EM, Meesters LD, van den Bogaart G, et al. Reverse Signaling by MHC-I Molecules in Immune and Non-Immune Cell Types. Front Immunol. 2020;11:605958. doi: 10.3389/fimmu.2020.605958. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Gao Y, Hong Y, Huang L, et al. β2-microglobulin functions as an endogenous NMDAR antagonist to impair synaptic function. Cell. 2023;186:1026–38. doi: 10.1016/j.cell.2023.01.021. [DOI] [PubMed] [Google Scholar]
37.Di Pilato M, Kfuri-Rubens R, Pruessmann JN, et al. CXCR6 positions cytotoxic T cells to receive critical survival signals in the tumor microenvironment. Cell. 2021;184:4512–30. doi: 10.1016/j.cell.2021.07.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Doench JG, Fusi N, Sullender M, et al. Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9. Nat Biotechnol. 2016;34:184–91. doi: 10.1038/nbt.3437. [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Joung J, Konermann S, Gootenberg JS, et al. Genome-scale CRISPR-Cas9 knockout and transcriptional activation screening. Nat Protoc. 2017;12:828–63. doi: 10.1038/nprot.2017.016. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

online supplemental file 1

jitc-14-5-s001.pdf^{(13.5MB, pdf)}

DOI: 10.1136/jitc-2025-012347

online supplemental file 2

jitc-14-5-s002.xlsx^{(159.1KB, xlsx)}

DOI: 10.1136/jitc-2025-012347

Data Availability Statement

Data are available upon reasonable request.

[R1] 1.Dersh D, Hollý J, Yewdell JW. A few good peptides: MHC class I-based cancer immunosurveillance and immunoevasion. Nat Rev Immunol. 2021;21:116–28. doi: 10.1038/s41577-020-0390-6. [DOI] [PubMed] [Google Scholar]

[R2] 2.Hazini A, Fisher K, Seymour L. Deregulation of HLA-I in cancer and its central importance for immunotherapy. J Immunother Cancer. 2021;9:e002899. doi: 10.1136/jitc-2021-002899. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R3] 3.Yang K, Halima A, Chan TA. Antigen presentation in cancer - mechanisms and clinical implications for immunotherapy. Nat Rev Clin Oncol. 2023;20:604–23. doi: 10.1038/s41571-023-00789-4. [DOI] [PubMed] [Google Scholar]

[R4] 4.Barkal AA, Weiskopf K, Kao KS, et al. Engagement of MHC class I by the inhibitory receptor LILRB1 suppresses macrophages and is a target of cancer immunotherapy. Nat Immunol. 2018;19:76–84. doi: 10.1038/s41590-017-0004-z. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] 5.Wolf NK, Kissiov DU, Raulet DH. Roles of natural killer cells in immunity to cancer, and applications to immunotherapy. Nat Rev Immunol. 2023;23:90–105. doi: 10.1038/s41577-022-00732-1. [DOI] [PubMed] [Google Scholar]

[R6] 6.Christofides A, Strauss L, Yeo A, et al. The complex role of tumor-infiltrating macrophages. Nat Immunol. 2022;23:1148–56. doi: 10.1038/s41590-022-01267-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Li Y, Wu X, Sheng C, et al. IGSF8 is an innate immune checkpoint and cancer immunotherapy target. Cell. 2024;187:2703–16. doi: 10.1016/j.cell.2024.03.039. [DOI] [PubMed] [Google Scholar]

[R8] 8.Taylor BC, Sun X, Gonzalez-Ericsson PI, et al. NKG2A Is a Therapeutic Vulnerability in Immunotherapy Resistant MHC-I Heterogeneous Triple-Negative Breast Cancer. Cancer Discov. 2024;14:290–307. doi: 10.1158/2159-8290.CD-23-0519. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R9] 9.Spranger S, Gajewski TF. Impact of oncogenic pathways on evasion of antitumour immune responses. Nat Rev Cancer. 2018;18:139–47. doi: 10.1038/nrc.2017.117. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] 10.Torrejon DY, Abril-Rodriguez G, Champhekar AS, et al. Overcoming Genetically Based Resistance Mechanisms to PD-1 Blockade. Cancer Discov. 2020;10:1140–57. doi: 10.1158/2159-8290.CD-19-1409. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R11] 11.Germano G, Lu S, Rospo G, et al. CD4 T Cell-Dependent Rejection of Beta-2 Microglobulin Null Mismatch Repair-Deficient Tumors. Cancer Discov. 2021;11:1844–59. doi: 10.1158/2159-8290.CD-20-0987. [DOI] [PubMed] [Google Scholar]

[R12] 12.Torrejon DY, Galvez M, Abril-Rodriguez G, et al. Antitumor Immune Responses in B2M-Deficient Cancers. Cancer Immunol Res. 2023;11:1642–55. doi: 10.1158/2326-6066.CIR-23-0139. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] 13.Kruse B, Buzzai AC, Shridhar N, et al. CD4+ T cell-induced inflammatory cell death controls immune-evasive tumours. Nature New Biol. 2023;618:1033–40. doi: 10.1038/s41586-023-06199-x. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Li Q, Guo W, Qian Y, et al. Protein O-fucosyltransferase 1 promotes PD-L1 stability to drive immune evasion and directs liver cancer to immunotherapy. J Immunother Cancer. 2024;12:e008917. doi: 10.1136/jitc-2024-008917. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Ruiz de Galarreta M, Bresnahan E, Molina-Sánchez P, et al. β-Catenin Activation Promotes Immune Escape and Resistance to Anti–PD-1 Therapy in Hepatocellular Carcinoma. Cancer Discov. 2019;9:1124–41. doi: 10.1158/2159-8290.CD-19-0074. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] 16.Versteeg R, Noordermeer IA, Krüse-Wolters M, et al. c-myc down-regulates class I HLA expression in human melanomas. EMBO J. 1988;7:1023–9. doi: 10.1002/j.1460-2075.1988.tb02909.x. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Guan X, Lu Y, Zhang Y, et al. Tumor-associated NK cells facilitate tumor growth via NKp46 in immunocompetent murine hepatocellular carcinoma. Immunol Lett. 2023;258:8–19. doi: 10.1016/j.imlet.2023.04.009. [DOI] [PubMed] [Google Scholar]

[R18] 18.Neo SY, Tong L, Chong J, et al. Tumor-associated NK cells drive MDSC-mediated tumor immune tolerance through the IL-6/STAT3 axis. Sci Transl Med. 2024;16:eadi2952. doi: 10.1126/scitranslmed.adi2952. [DOI] [PubMed] [Google Scholar]

[R19] 19.Bill R, Wirapati P, Messemaker M, et al. CXCL9:SPP1 macrophage polarity identifies a network of cellular programs that control human cancers. Science. 2023;381:515–24. doi: 10.1126/science.ade2292. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Yin P, Gui L, Wang C, et al. Targeted Delivery of CXCL9 and OX40L by Mesenchymal Stem Cells Elicits Potent Antitumor Immunity. Mol Ther. 2020;28:2553–63. doi: 10.1016/j.ymthe.2020.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Zheng P, Luo Q, Wang W, et al. Tumor-associated macrophages-derived exosomes promote the migration of gastric cancer cells by transfer of functional Apolipoprotein E. Cell Death Dis. 2018;9:434. doi: 10.1038/s41419-018-0465-5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Kaufman HL, Ruby CE, Hughes T, et al. Current status of granulocyte–macrophage colony-stimulating factor in the immunotherapy of melanoma. J Immunotherapy Cancer. 2014;2:11. doi: 10.1186/2051-1426-2-11. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R23] 23.Fan H, Hall P, Santos LL, et al. Macrophage Migration Inhibitory Factor and CD74 Regulate Macrophage Chemotactic Responses via MAPK and Rho GTPase. The Journal of Immunology. 2011;186:4915–24. doi: 10.4049/jimmunol.1003713. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] 24.Kang I, Bucala R. The immunobiology of MIF: function, genetics and prospects for precision medicine. Nat Rev Rheumatol. 2019;15:427–37. doi: 10.1038/s41584-019-0238-2. [DOI] [PubMed] [Google Scholar]

[R25] 25.Ozga AJ, Chow MT, Luster AD. Chemokines and the immune response to cancer. Immunity. 2021;54:859–74. doi: 10.1016/j.immuni.2021.01.012. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R26] 26.Mabrouk N, Tran T, Sam I, et al. CXCR6 expressing T cells: Functions and role in the control of tumors. Front Immunol. 2022;13:1022136. doi: 10.3389/fimmu.2022.1022136. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R27] 27.Korbecki J, Bajdak-Rusinek K, Kupnicka P, et al. The Role of CXCL16 in the Pathogenesis of Cancer and Other Diseases. Int J Mol Sci. 2021;22:3490. doi: 10.3390/ijms22073490. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] 28.Perez-Diez A, Joncker NT, Choi K, et al. CD4 cells can be more efficient at tumor rejection than CD8 cells. Blood. 2007;109:5346–54. doi: 10.1182/blood-2006-10-051318. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R29] 29.Bogen B, Fauskanger M, Haabeth OA, et al. CD4+ T cells indirectly kill tumor cells via induction of cytotoxic macrophages in mouse models. Cancer Immunol Immunother. 2019;68:1865–73. doi: 10.1007/s00262-019-02374-0. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] 30.Jin YP, Fishbein MC, Said JW, et al. Anti-HLA class I antibody-mediated activation of the PI3K/Akt signaling pathway and induction of Bcl-2 and Bcl-xL expression in endothelial cells. Hum Immunol. 2004;65:291–302. doi: 10.1016/j.humimm.2004.01.002. [DOI] [PubMed] [Google Scholar]

[R31] 31.Valenzuela NM, Reed EF. The link between major histocompatibility complex antibodies and cell proliferation. Transplant Rev (Orlando) 2011;25:154–66. doi: 10.1016/j.trre.2011.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] 32.Li D, Zhang Q, Li L, et al. β2-Microglobulin Maintains Glioblastoma Stem Cells and Induces M2-like Polarization of Tumor-Associated Macrophages. Cancer Res. 2022;82:3321–34. doi: 10.1158/0008-5472.CAN-22-0507. [DOI] [PubMed] [Google Scholar]

[R33] 33.Xu S, Liu X, Bao Y, et al. Constitutive MHC class I molecules negatively regulate TLR-triggered inflammatory responses via the Fps-SHP-2 pathway. Nat Immunol. 2012;13:551–9. doi: 10.1038/ni.2283. [DOI] [PubMed] [Google Scholar]

[R34] 34.Wu H, Jiang K, Guo S, et al. IFN-τ Mediated Control of Bovine Major Histocompatibility Complex Class I Expression and Function via the Regulation of bta-miR-148b/152 in Bovine Endometrial Epithelial Cells. Front Immunol. 2018;9:167. doi: 10.3389/fimmu.2018.00167. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] 35.Muntjewerff EM, Meesters LD, van den Bogaart G, et al. Reverse Signaling by MHC-I Molecules in Immune and Non-Immune Cell Types. Front Immunol. 2020;11:605958. doi: 10.3389/fimmu.2020.605958. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] 36.Gao Y, Hong Y, Huang L, et al. β2-microglobulin functions as an endogenous NMDAR antagonist to impair synaptic function. Cell. 2023;186:1026–38. doi: 10.1016/j.cell.2023.01.021. [DOI] [PubMed] [Google Scholar]

[R37] 37.Di Pilato M, Kfuri-Rubens R, Pruessmann JN, et al. CXCR6 positions cytotoxic T cells to receive critical survival signals in the tumor microenvironment. Cell. 2021;184:4512–30. doi: 10.1016/j.cell.2021.07.015. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] 38.Doench JG, Fusi N, Sullender M, et al. Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9. Nat Biotechnol. 2016;34:184–91. doi: 10.1038/nbt.3437. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R39] 39.Joung J, Konermann S, Gootenberg JS, et al. Genome-scale CRISPR-Cas9 knockout and transcriptional activation screening. Nat Protoc. 2017;12:828–63. doi: 10.1038/nprot.2017.016. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

CXCL16-driven CD4+ T cells orchestrate immunosurveillance against MHC-I-deficient hepatocellular tumors

Yunyi Zhou

Ping Chen

Zhixue Wang

Liming Gui

Zhengheng Zheng

Wei-Qiang Gao

Bin Ma

Abstract

Background

Methods

Results

Conclusion

WHAT IS ALREADY KNOWN ON THIS TOPIC

WHAT THIS STUDY ADDS

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

Background

Results

MHC-I loss differentially affects tumor growth and immune microenvironment

CD4+ T cells are crucial for MHC-I-loss-induced Hepa1-6 tumor regression

CD4+ T cells are key mediators of NK cells and macrophages

CXCL16 promotes CD4+ T-cell infiltration to suppress MHC-I-deficient tumors

B2M loss oppositely regulates CXCL16 via Akt suppression or NF-κB activation

Discussion

Materials and methods

Cell lines

Lentiviral transduction

Flow cytometric analysis of MHC-I, MHC-II, Qa-1, Qa-2, and PD-L1 expression on tumor cell lines

Cell proliferation assay

Isolation and culture of murine NK cells and macrophages

Co-culture of NK and tumor cells

Murine tumor models

Hydrodynamic tail-vein injection

Real-time quantitative PCR analysis

Depleting antibody treatments

Flow cytometric analysis of immune infiltrates in tumors

Bulk RNA sequencing

Single-cell RNA sequencing

Transwell migration assay

Macrophage stimulation assay

ELISA for mouse CXCL16

Western blotting

Analysis of CXCL16 and B2M co-expression in scRNA-seq datasets

Correlation analysis of CXCL16 with B2M and T-cell markers in human cancer databases

Statistical analysis

Supplementary material

Footnotes

Data availability statement

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

CD4⁺ T cells are crucial for MHC-I-loss-induced Hepa1-6 tumor regression

CD4⁺ T cells are key mediators of NK cells and macrophages

CXCL16 promotes CD4⁺ T-cell infiltration to suppress MHC-I-deficient tumors