Targeting BCL9/BCL9L enhances antigen presentation by promoting conventional type 1 dendritic cell (cDC1) activation and tumor infiltration

Conventional type 1 dendritic cells (cDC1) are the essential antigen-presenting DC subset in antitumor immunity. Suppressing B-cell lymphoma 9 and B-cell lymphoma 9-like (BCL9/BCL9L) inhibits tumor growth and boosts immune responses against cancer. However, whether oncogenic BCL9/BCL9L impairs antigen presentation in tumors is still not completely understood. Here, we show that targeting BCL9/BCL9L enhanced antigen presentation by stimulating cDC1 activation and infiltration into tumor. Pharmacological inhibition of BCL9/BCL9L with a novel inhibitor hsBCL9z96 or Bcl9/Bcl9l knockout mice markedly delayed tumor growth and promoted antitumor CD8+ T cell responses. Mechanistically, targeting BCL9/BCL9L promoted antigen presentation in tumors. This is due to the increase of cDC1 activation and tumor infiltration by the XCL1-XCR1 axis. Importantly, using single-cell transcriptomics analysis, we found that Bcl9/Bcl9l deficient cDC1 were superior to wild-type (WT) cDC1 at activation and antigen presentation via NF-κB/IRF1 signaling. Together, we demonstrate that targeting BCL9/BCL9L plays a crucial role in cDC1-modulated antigen presentation of tumor-derived antigens, as well as CD8+ T cell activation and tumor infiltration. Targeting BCL9/BCL9L to regulate cDC1 function and directly orchestrate a positive feedback loop necessary for optimal antitumor immunity could serve as a potential strategy to counter immune suppression and enhance cancer immunotherapy.


INTRODUCTION
Antigen presentation is indispensable to antitumor responses.In order to elicit effective antitumor CD8 + T cell responses, antigen presentation must be successful in two major events: First, tumor antigens are captured by antigen-presenting cells (APCs), processed into peptide fragments and presented on APCs with major histocompatibility complex class I (MHC-I) or human leukocyte antigen class I (HLA-I) to prime CD8 + T cells.Second, the activated CD8 + T cells recognize the tumor antigens that directly presented by APCs and then kill the tumor cells. 1 Admittedly, conventional type 1 dendritic cells (cDC1) are considered the superior DC subset to present tumor antigens onto MHC-I to prime CD8 + T cells. 2,38][9] However, DCs are often dysfunctional, immature or even immunosuppressive, absent antigen-presenting capability for cross-priming within the tumor microenvironment (TME).Moreover, cDC1 are rare because TME excludes cDC1 from tumors through various mechanisms. 10Thus, approaches aimed to enhance antigen presentation, including increasing cDC1 activation and tumor infiltration may therefore promote antitumor immunity and improve cancer immunotherapy.
The Wnt/β-catenin signaling is not only critical in the regulation of tumor carcinogenesis, but also favors immune invasion in many malignancies. 11,12An increasing number of studies have reported that Wnt/β-catenin signaling pathway regulates DC functions. 13Loss of β-catenin in CD11c + DCs during priming attenuated tumor-induced suppression of memory CD8 + T cell responses. 14Deletion of LDL receptor-related protein 5/6 (LRP5/6) in CD11c + DCs abrogated tumor growth with elevated antitumor responses. 15Downregulation of the β-catenin or constitutively active glycogen synthase kinase 3 beta (GSK3β) in DCs also enhanced DCs functionality. 16,17Besides, Wnt1 fostered transcriptional silencing of CC/CXC chemokines in cDCs, inducing immune resistance in lung adenocarcinoma. 18Furthermore, active Wnt/β-catenin signaling pathway dampens CD103 + DCs recruitment to the tumors by CCL4 and CCL5. 19,20-cell lymphoma 9 and B-cell lymphoma 9-like (BCL9/BCL9L), functioning as transcriptional coactivators of β-catenin, are crucial in the Wnt/β-catenin signaling pathway.21 BCL9/BCL9L expression is elevated in various human cancers and correlates with poor prognosis in cancer patients.[22][23][24] Our previous studies reported that inhibition of BCL9/BCL9L promotes antitumor responses via various mechanisms, involving regulatory T cells (Treg) infiltration, CD8 + T cell function, DC infiltration and tumor-associated macrophages.[25][26][27][28] cDC1 have recognized to play an essential role in cross-presentation for CD8 + T cell priming and cancer immunotherapy.However, whether targeting BCL9/BCL9L governs the antigen presentation, especially antigen presentation of cDC1 in antitumor immunity is still not completely clear.
Here, we report that inhibition or depletion of BCL9/BCL9L markedly restrained tumor growth and enhanced antitumor responses.Interestingly, targeting BCL9/BCL9L to augment the antigen presentation capacity of cDC1 and boost the infiltration of cDC1 and CD8 + T cells into tumors underscores a pivotal role of this approach in bolstering cDC1-mediated antitumor immunity.Therapies aiming at improving antigen presentation might benefit from combination with BCL9 inhibitors.

BCL9 expression is negatively associated with antigen presentation in cancers
Absenting effective antigen presentation is an important reason for immune evasion.Notably, high antigen processing and presentation and HLA-I expression were associated with prolonged overall survival in cancer patients (Supplementary Fig. 1a).Especially, immunotherapy resistance often occurs in microsatellite-stable/microsatellite instability-low (MSS/MSI-L) tumors that have deficient antigen processing and presentation, HLA-I expression and immune infiltration (Supplementary Fig. 1b).We wondered whether oncogenic BCL9 highly expressed in colon cancer correlates with antigen presentation (Fig. 1a).We analyzed recently published gene expression datasets in tumors. 29Interestingly, tumors with high BCL9 expression lacked antigen processing and presentation, HLA-I expression, and immune infiltration (Fig. 1b and Supplementary Fig. 1c).Similar results were also observed in patients with melanoma and breast cancer (Supplementary Fig. 2a-d).Moreover, cancer patients with high BCL9 expression had significantly worse overall survival than those with low BCL9 expression among the tumors exhibiting low immune infiltration, especially among those also displaying Wnt activity (Supplementary Fig. 2e).Collectively, such data indicate that BCL9 expression is negatively associated with antigen presentation and immune infiltration in cancers.Targeting BCL9 might have beneficial effects in improving resistance to immunotherapy by enhancing antigen presentation.
Pharmacological inhibition or depletion of BCL9/BCL9L markedly delays tumor growth As our results showed that elevated BCL9 expression is associated with diminished antigen presentation in cancers, we explored this supposition in tumor models.To investigate tumor growth driven by oncogenic BCL9, we developed a novel hsBCL9 z96 that targets the BCL9/β-catenin interaction to inhibit the Wnt/β-catenin signaling pathway.hsBCL9 z96 was similar to hsBCL9 CT -24, a BCL9/ β-catenin inhibitor reported in our previous study, but had a superior cell penetration ability than hsBCL9 CT -24. 25 As a tool of pharmacological inhibition of BCL9, the kinetic constants for the interaction of the protein β-catenin with hsBCL9 z96 were analyzed by Biacore assay.The kinetic constants were shown in a sensorgram (Equilibrium dissociation constant, K D = 3.104 × 10 −8 M; association rate constant, K a = 1.329 × 10 5 M −1 s −1 ; dissociation rate constant, K d = 0.004126 s −1 ), indicating that hsBCL9 z96 could excellently interact with β-catenin (Supplementary Fig. 3a).The half-maximal inhibitory concentration (IC 50 ) of hsBCL9 z96 trifluoroacetic acid (TFA) salts for inhibition of the T-cell factor/lymphoid enhancer-binding factor (LEF/TCF) pathway, the major end point mediators of Wnt/ Wingless signaling, was only 142.1 nM, indicating that hsBCL9 z96 could inhibit Wnt/β-catenin signaling in a low dose (Supplementary Fig. 3b).Moreover, the IC 50 of hsBCL9 z96 for inhibiting the proliferation of Colo320DM cells was 0.87 μM, which lowed 0.5 μM than that of doxorubicin (Supplementary Fig. 3c).In addition, the cell permeability of hsBCL9 z96 was approximately 2 times higher than that of hsBCL9 CT -24 (Supplementary Fig. 3d, e).Collectively, hsBCL9 z96 exhibits robust inhibition against the interaction between β-catenin and BCL9, as well as shows a favorable profile in vitro.
Furthermore, hsBCL9 z96 treatment remarkably reduced tumor volumes in subcutaneous implanted CT26 model, which is Wnt/ β-catenin signaling-dependent colon carcinoma, with the tumor growth inhibition rate (TGI) of 48.0%, but the body weight also modestly decreased in hsBCL9 z96 -treated tumor-bearing mice (Fig. 1c and Supplementary Fig. 3f).In contrast, histological examination of hearts, livers, lungs, kidneys and spleens from hsBCL9 z96 -treated CT26 tumor-bearing mice indicated that there is no significant toxicity in these organs compared to those in the vehicle control group, except for some intestinal adhesion (Supplementary Fig. 2g).Similarly, hsBCL9 z96 also demonstrated tumor regression in MC38, B16F10 and 4T1 models, with the TGIs of 56.1%, 45.7% and 33.1%, respectively (Supplementary Fig. 3h-j).In summary, hsBCL9 z96 demonstrates robust antitumor activity in colon cancer, melanoma and breast cancer models with improved feature of cell permeability, suggesting hsBCL9 z96 is a good research tool of pharmacological inhibition of BCL9/β-catenin driven Wnt transcriptional activity.
Pharmacological inhibition and genetic depletion are usually used to uncover functional roles of a target protein.To investigate Bcl9/Bcl9l genes in mice, we used a Bcl9 f/f Bcl9l f/f Cre-ERT2 mouse model, in which Bcl9/Bcl9l deletion was induced with tamoxifen in adults.We found that tumor growth was notably decreased in MC38 tumor-bearing Bcl9/Bcl9l deficiency mice compared with control mice, with a TGI of 65.6% (Fig. 1d).This result indicates that targeting BCL9/BCL9L impairs the development of Wnt/β-catenin signaling-dependent cancers in vivo.
Targeting BCL9/BCL9L enhances antigen presentation and facilitates antigen-specific CD8 + T cell responses CD4 + T cells and CD8 + T cells have been implicated in antitumor responses. 30hsBCL9 z96 treatment attenuated the antitumor response in CD8 + T cells-depleted group, but not in CD4 + T cells-depleted group of CT26 tumor-bearing mice, suggesting that the antitumor responses of hsBCL9 z96 treatment depends on CD8 + T cells (Supplementary Fig. 4a, b).To further investigate the key role of CD8 + T cells, we evaluated the frequencies of effector CD8 + T cells in hsBCL9 z96 -treated tumors by flow cytometry.Our study revealed that treatment with hsBCL9 z96 significantly enhanced the proportions of IFN-γ + CD8 + , Granzyme B + CD8 + and Ki67 + CD8 + T cells in tumor-infiltrating lymphocytes (TILs), as well as in tumordraining lymph nodes (TdLN) and spleens from hsBCL9 z96 -treated CT26 tumor-bearing mice (Supplementary Fig. 4c-f).Similarly, RNA sequencing (RNA-seq) suggested that TCR signaling pathway was upregulated in hsBCL9 z96 -treated CT26 tumors (Supplementary Fig. 4g).Together, these data indicate that targeting BCL9/BCL9L increases CD8 + T cell activation and proliferation.
Antigen presentation is essential for activation and proliferation of CD8 + T cells.Our previous bioinformatics analysis indicated that targeting BCL9 might be beneficial for antigen presentation.Consistently, the expression of genes related to antigen processing and presentation was upregulated in hsBCL9 z96 -treated CT26 tumors shown by RNA-seq data (Fig. 1e).Similarly, the expression of genes related to antigen processing and presentation such as Tap1, Tap2, B2m, and Psmb9, was also upregulated in tumors from hsBCL9 z96 -treated CT26 tumor-bearing mice and MC38 tumorbearing Bcl9/Bcl9l deficiency mice (Fig. 1f, g).Such data indicate that targeting BCL9/BCL9L enhances antigen presentation in tumors.
In addition, staining results of SIINFEKL-H2Kb complexes on cDC1 of TDLNs isolated from OVA-expressing tumor-bearing mice showed an increase in cross-presentation among hsBCL9 z96treated MC38-OVA tumor-bearing mice or Bcl9 f/f Bcl9l f/f Cre-ERT2 MC38-OVA tumor-bearing mice compared with vehicle or Bcl9 f/f Bcl9l f/f mice, respectively (Supplementary Fig. 5b, c).
To assess the impact of this activation directly, we then sorted naïve CD8 + T cells from OT-I mice, labeled them with CFSE and injected intravenously into mice that had been implanted with MC38-OVA tumors.Enhanced proliferation of OT-I CD8 + T cells was observed in TdLNs from hsBCL9 z96 -treated MC38-OVA  Tumor tumor-bearing mice (Fig. 2e-g).Similar results were also obtained in TdLNs from MC38-OVA tumor-bearing Bcl9/Bcl9l deficiency mice (Fig. 2h-j).
To determine the cross-presentation activity upon BCL9 inhibition, we used in vitro and in vivo approaches.CD8 + T cell in vitro proliferation was increased in cross-presentation of CD11c + DC-CD8 + T cells in both hsBCL9 Z96 tumor model and Bcl9/Bcl9l KO tumor model compared with vehicle and wild-type (WT) respectively (Supplementary Fig. 5d, e).In ELISPOT assay, in vitro co-culture of CD11c + DCs from TDLNs and naïve CD8 + T derived from OT-I mice, IFN-γ secretion was increased in CD11c + DCs from hsBCL9z96-treated MC38-OVA-expressing tumors compared with vehicle treatment; IFN-γ secretion was also increased in CD11c + DCs from MC38-OVA tumor-bearing Bcl9 f/f Bcl9l f/f Cre-ERT2 mice compared with Bcl9 f/f Bcl9l f/f mice (Supplementary Fig. 5f, g).We then purified CD8 + T cells from TDLNs and stimulated them with of 10 μg/ml of OVA-I peptide for 24 h to determine IFN-γ-producing cells by ELISPOT assay.Upon OVA-I peptide stimulation, IFN-γ-producing cells were increased in CD8 + T cells from hsBCL9z96-treated MC38-OVA tumor-bearing mice.Similar results were obtained in Cre-ERT2 MC38-OVA tumor-bearing mice (Supplementary Fig. 5h, i).In summary, BCL9 inhibition promotes CD8 + T cell proliferation and IFN-γ secretion as well as cross-presentation in MC38-OVA tumor-bearing models.
Collectively, the data indicates that inhibition of BCL9/BCL9L promotes cDC1 activation and facilitates cross-priming of CD8 + T cells.
Single-cell transcriptional profiling of CD8 + T cells and cDC1 of tumors and TdLNs from B16-OVA tumor-bearing Bcl9/Bcl9l deficiency mice To precisely investigate how BCL9/BCL9L governs the functions of cDC1, we performed single-cell sequencing (scRNA-seq) of tumors and TdLNs from B16-OVA tumor-bearing Bcl9/Bcl9l deficiency mice (Fig. 3a).All cells were clustered according to t-SNE (t-distributed Stochastic Neighbor Embedding), and then myeloid cells (marked by Cd68) and T cells (marked by Cd3e) were extracted for further clustering (Fig. 3b and Supplementary Fig. 6a).T cells were mainly divided into 2 clusters, including CD8 + T cells (marked by Cd8a) and CD4 + T cells (marked by Cd4) (Fig. 3b, and Supplementary Fig. 6a).Additionally, DCs (marked by Zbtb46) were separated from the myeloid cells and then clustered into cDC1 (marked by Xcr1) and cDC2 (marked by Clec10a) (Fig. 3c-e and Supplementary Fig. 6b-d).The expression profiles of DC-related genes and differentially expressed genes in the tumors and TdLNs of DC subclusters were shown (Supplementary Fig. 7a-d).In summary, we successfully identified the CD8 + T cells and cDC1 populations in tumors and TdLNs from B16-OVA tumor-bearing Bcl9/Bcl9l deficiency mice using scRNA-seq.
Bcl9/Bcl9l deficient cDC1 are superior to WT cDC1 in activation, antigen presentation and cross-priming of CD8 + T cells To further investigate how BCL9/BCL9L regulates the functions of cDC1, we explored the transcriptional differences of CD8 + T cells and cDC1 in TdLNs and tumors from B16-OVA tumor-bearing Bcl9/ Bcl9l deficiency mice using single-cell transcriptomics analysis.We analyzed transcriptional changes and signaling enrichment of cDC1 in tumors and TdLNs from B16-OVA tumor-bearing Bcl9/Bcl9l deficiency mice.Enhancement of activation and antigen presentation was observed in Bcl9/Bcl9l deficiency cDC1 of tumors and TdLNs, consistently with the results that inhibition of BCL9/ BCL9L induced cDC1 activation (Fig. 4a-c).Interacting pairs of HLAE_HUMAN-KLRK1 showed stronger signals in TdLNs from Bcl9/ Bcl9l deficiency mice than those from the control mice, suggesting enhanced antigen presentation and interaction between cDC1 and CD8 + T cells (Fig. 4d).Importantly, T cell activation was enhanced in tumors and TdLNs from B16-OVA tumor-bearing Bcl9/Bcl9l deficiency mice (Fig. 4e and Supplementary Fig. 8a).Consistently, positive regulation of T cell proliferation in cDC1 was also upregulated in tumors of B16-OVA tumor-bearing Bcl9/Bcl9l deficiency mice, supporting that cDC1 primed CD8 + T cells (Fig. 4f and Supplementary Fig. 8b).In line with this result, CD8 + T cells infiltration was markedly increased in B16-OVA tumor-bearing Bcl9/Bcl9l deficiency mice compared with control mice (Fig. 4g).Together, these data indicate that Bcl9/Bcl9l -deficient cDC1 outperform WT cDC1 in activation, antigen presentation and cross-priming of CD8 + T cells.
XCL1 is mainly produced by activated CD8 + T cells and NK cells. 9o identify the source of XCL1, we analyzed the XCL1 expression in these cells.We found that XCL1 expression by intratumoral CD8 + T cells was upregulated in hsBCL9 z96 -treated CT26 tumor-bearing mice and MC38 tumor-bearing Bcl9/Bcl9l deficiency mice (Fig. 6g, h).These data indicate that targeting BCL9/BCL9L enhances XCL1 production by activating CD8 + T cells and subsequently promotes the recruitment of XCR1 + cDC1 into tumor sites.
Targeting BCL9/BCL9L results in CD8 + T cell accumulation in tumors through CXCL9-CXCR3 axis Having established that targeting BCL9/BCL9L facilitates cDC1 activation and tumor infiltration, we next examined how cDC1 contribute to CD8 + T cell responses in tumors after the inhibition of BCL9.Using RNA-seq data, we found that IFN-γ response was cDC1 t-score IRAK4 deficiency TLR2/4 TICAM1/TRAF6-dependent induction of TAK1 complex TRAF6 mediated IRF7 activation in TLR7 8 or 9 signaling Activated TAK1 mediates p38-MAPK activation TRAF6-mediated induction of TAK1 complex within TLR4 complex MYD88-independent TLR4 cascade IRAK2-mediated activation of TAK1 complex Toll-like receptor TLR1/TLR2 cascade Toll-like receptor 9/TLR9 cascade TAK1 activates NF-κB by phosphorylation and activation of IKKS complex   upregulated in hsBCL9 z96 -treated CT26 tumors compared with vehicle control (Fig. 7a).Similarly, increased Ifng mRNA and IFN-γ protein levels were observed in tumors from hsBCL9 z96 -treated CT26 tumor-bearing mice and MC38 tumor-bearing Bcl9/Bcl9l deficiency mice (Fig. 7b, c).The mRNA and protein levels of the IFN-γ-induced chemokine CXCL9 were also upregulated in tumors from hsBCL9 z96 -treated CT26 tumor-bearing mice and MC38 tumorbearing Bcl9/Bcl9l deficiency mice (Fig. 7d, e).cDC1 could produce CXCL9 to recruit effector CXCR3 + CD8 + T cells into the TME through CXCL9-CXCR3 axis (Fig. 7f). 43We wondered if CXCL9 produced from intratumoral cDC1 promotes CD8 + T cells infiltration.We thus evaluated CXCL9 expression of cDC1 and CXCR3 expression of CD8 + T cells in tumors by flow cytometry.Consistently, the CXCL9 expression by cDC1 and CXCR3 expression by CD8 + T cells were upregulated in tumors from hsBCL9 z96 -treated CT26 tumor-bearing mice, as well as from MC38 tumor-bearing Bcl9/Bcl9l deficiency mice (Fig. 7g-i).In support of these findings, we characterized the datasets of cancer patients from TCGA with  either low or high BCL9 mRNA expression.We found that both Ifng signature and CD8 + T cells infiltration were increased in low BCL9 expression patients with COAD, SKCM and BRCA from TCGA datasets compared to those from high BCL9 expression group (Supplementary Fig. 13a, b).In summary, targeting BCL9/BCL9L promotes the infiltration of CD8 + T cells into tumors via the CXCL9-CXCR3 axis.
Targeting BCL9/BCL9L sensitizes tumors to immune checkpoint blockade therapy Our studies showed that targeting BCL9/BCL9L facilitates antigen presentation and tumor infiltration of cDC1, as well as boosts tumor infiltration of CD8 + T cells, indicating a critical role of targeting BCL9/BCL9L in overcoming immunotherapy resistance.To evaluate this point, we combined hsBCL9 z96 treatment with  Results are presented as the mean ± standard deviation (SD) for each group, derived from three independent experiments; "n" denotes the number of biological replicates; Unpaired Student's t test (b-e, g-i); One-way ANOVA followed by Bonferroni test (f) anti-PD-1 treatment in established CT26 tumor-bearing mice.The CT26 tumors showed a stronger response to combination therapy than to hsBCL9 z96 monotherapy as well as increased the survival of treated mice (TGI = 78.1%,Fig. 8a, b).Similarly, established MC38 tumors also showed a stronger response to anti-PD-1 combination therapy than to hsBCL9 z96 monotherapy as well as increased the survival of treated mice (TGI = 80.3%, Supplementary Fig. 14a, b).Comparable results were obtained after combined hsBCL9 z96 and anti-CTLA-4 treatment was administered to established CT26 tumor-bearing mice (TGI = 74.8%,Supplementary Fig. 14c, d) and to established MC38 tumor-bearing mice (TGI = 78.4%,Supplementary Fig. 14e, f).In addition, anti-PD-1 therapy achieved a greater decrease in MC38 tumor growth of Bcl9/Bcl9l deficiency mice than those of Bcl9 f/f Bcl9l f/f mice as well as increased the survival in Bcl9 f/f Bcl9l f/f Cre-ERT2 mice (TGI = 92.3%,Fig. 8c, d).Together, these results indicate that therapeutic approaches aiming at overcoming immunotherapy resistance might benefit from combination with hsBCL9 z96 to enhance antigen presentation.

DISCUSSION
Although immunotherapy has a promising clinical response in a small number of patients with mismatch repair deficient (dMMR) or MSI-H solid tumors, most patients with MSS or MSI-L solid tumors still do not benefit. 44,45To an effective antitumor response, tumor antigens have to be captured by DCs, processed into peptide fragments, and presented on DCs with MHC-I to prime CD8 + T cells. 1 However, tumors exploit multiple escape mechanisms to decrease antigen presentation within TME.Amazingly, our findings uncovered that targeting BCL9/BCL9L can excellently enhance antigen presentation in cancers.We demonstrated that targeting BCL9/BCL9L facilitates the antigen presentation and tumor infiltration of cDC1, as well as boosts the tumor infiltration of CD8 + T cells, thereby improving the responses to cancer immunotherapy.The following aspects summarize the detailed mechanisms and importance of our findings.
First, Wnt/β-catenin signaling is recognized as a key factor to dampen tumor immune cell infiltration in many human cancers.Antigen presentation is indispensable for immune cell infiltration,  especially for the tumor infiltration of CD8 + T cells.It was reported that inhibition of Wnt/β-catenin signaling promotes antitumor immunity by enhancing CD8 + T cell responses, while the mechanism of this effect is still not completely clear.cDC1 are thought to perform cross-presentation for CD8 + T cell priming. 3,46ur study demonstrated that targeting BCL9/BCL9L facilitates antigen presentation by promoting cDC1 activation and tumor infiltration.Therefore, our data may provide an excellent mechanism for inhibiting Wnt/β-catenin signaling to promotes CD8 + T cell-mediated antitumor responses.Second, cDC1 are often dysfunctional and absent from antigen-presenting capability for cross-priming within TME.NF-κB/IRF1 axis governs the activation and antigen presentation of cDC1. 35At present, the role of Wnt/β-catenin signaling in cDC1 is still elusive.Our study demonstrated that BCL9/BCL9L mediates the activation and antigen presentation of cDC1 via NF-κB/IRF1 axis.This is the first work to show that Wnt/β-catenin signaling plays a negative role in NF-κB-dependent antigen presentation of cDC1.Therefore, our study may also provide a novel mechanism for targeting Wnt/ β-catenin signaling to promote antitumor immunity by overcoming cDC1 dysfunction in TME, suggesting that targeting BCL9/ BCL9L to improve NF-κB-dependent antigen presentation in tumors is a potential approach to cancer therapy.Third, it is well documented that intratumoral cDC1 numbers are associated with the improved prognosis and responses to cancer immunotherapy. 9,47CCL4, CCL5 and XCL1, regulate migration of cDC1. 6,9,19e also uncovered that BCL9 and β-catenin drive Wnt signaling to promote XCR1 + cDC1 migration via XCL1, thereby offering a novel mechanism for targeting Wnt/β-catenin signaling to promote cDC1 tumor infiltration.Finally, DCs have been found to increase the production of CXCL9 and CXCL10 in an IFN-γ-dependent manner, which facilitates the infiltration of CD8 + T cells into tumors. 43Consistently, we also demonstrated that cDC1 in the TME produced more CXCL9 to recruit effector CD8 + T cells into TME through CXCL9-CXCR3 axis after inhibition of BCL9, supporting that targeting Wnt/β-catenin signaling promotes the tumor infiltration of CD8 + T cells.Admittedly, manipulation of cellular microenvironment is crucial for analyzing the roles of various immune cells in checkpoint blockade outcome evaluation and prognosis prediction.In addition, we used both pharmacological inhibition and genetic depletion on the same target in our study.The key findings and mechanisms were all confirmed by these two approaches, credibly illustrating that targeting BCL9/BCL9L facilitates antitumor immunity by enhancing antigen presentation.We observed metabolic pathway changes in the tumor microenvironment which are a result of both intrinsic tumor cell characteristics and interactions with the surrounding stromal and immune cells.It appears that DC overaction may also be regulated by additional pathways rather than directly modulated by BCL9/9L targeting chemicals alone.The metabolic changes in cDC1 are interesting and further study is required to reveal the underlying mechanisms.
In the current study, we reveal a novel mechanism of positive feedback loop.Our findings indicate that targeting BCL9/BCL9L facilitates the antigen presentation and infiltration of cDC1, eliciting robust antitumor CD8 + T responses, further advancing our understanding of targeting Wnt/β-catenin pathway to promote antitumor immunity.Our work also provides a potential combination therapy of hsBCL9 z96 and immune checkpoint blockade for cancer immunotherapy.However, unintended side effects should be carefully and systematically evaluated.In addition, these applications in other cancer types with survival related to BCL9, such as hepatocellular carcinoma, melanoma and triple negative breast cancer, need to be investigated.
In summary, our data uncover that targeting BCL9/BCL9L plays a key role in cDC1-regulated presentation of tumor-derived antigens and subsequently in cDC1-triggered activation and tumor infiltration of CD8 + T cells, constituting a positive feedback loop required for optimal antitumor immunity.Importantly, our study provides an excellent mechanism for inhibiting Wnt/ β-catenin signaling to facilitate CD8 + T cell-mediated antitumor responses by promoting antigen presentation.Thus, our findings offer some insights into enhancing the susceptibility of tumors with high BCL9/BCL9L expression to cancer immunotherapy.

Treatment of tumor tissues
Tumors were excised at the end of the experiment.The weight of the tumors was measured using a precision microscale.For further analysis, the tumors were sectioned and enzymatically broken down with collagenase IV (1 mg/mL; C5138, Sigma) and DNase I (20 μg/mL; DN25, Sigma), suspended in RPMI 1640 medium, and incubated at 37 °C for 1-2 h.The single cell suspensions from tumors were passed through a 70 μm cell strainer (352350, BD Biosciences) and washed with PBS 2 times.For intracellular staining by flow cytometry and chemokine analysis by ELISA, the single cell suspensions from tumors were cultured and stimulated with cell stimulation cocktail (including protein transport inhibitors) (00-4975-03, eBioscience) and incubated for 4-6 h at 37 °C in RPMI 1640 medium.After that the cells and the supernatants were collected.The single-cell suspensions from tumors were washed with PBS for staining by flow cytometry and the supernatants were collected for chemokine analysis by ELISA.For tumor RNA isolation, tumor tissues were extracted immediately after homogenization following the manufacturer's instructions.
Real-time PCR Total RNA was isolated from cells using TRIZOL reagent (15596026, Invitrogen) and subsequently transcribed into cDNA using a reverse transcription kit (RR036A, TaKaRa), adhering to the instructions provided by the manufacturer.Real-time PCR was performed employing SYBR Premix Ex Taq (RR420, TaKaRa) and specific primers, utilizing the Applied Biosystems StepOne Plus Real-Time PCR Systems.Data analysis was carried out using the 2 −ΔΔCt method.The sequences of the primers are detailed in Supplementary Table 2. ELISA Supernatants were harvested from single-cell suspensions of tumor tissues, cultured and stimulated with a cell stimulation cocktail (including protein transport inhibitors) (00-4975-03, eBioscience) for 4-6 h at 37 °C in RPMI 1640 medium, followed by centrifugation at 2000 rpm for 10 min for chemokine analysis using ELISA kits (Mouse CXCL9/MIG ELISA Kit, 70-EK2143/2-48, MULTI SCIENCE; Mouse IFN-gamma ELISA Kit, 70-EK280/3-96, MULTI SCIENCE; Mouse lymphotactin/XCL1 ELISA Kit, SEK50677, Sino Biological) according to the manufacturer's instructions.

Flow cytometry
For surface staining, cells were directly stained with surface markers according to the manufacturer's instructions.For intracellular staining, cells were first cultured and stimulated with cell stimulation cocktail (plus protein transport inhibitors) (00-4975-03, eBioscience).Following surface marker staining, the cells were fixed and permeabilized using intracellular fixation & permeabilization buffer (88-8824-00, eBioscience) following the manufacturer's instructions.For transcription factor staining, the cells underwent a similar fixation and permeabilization process using Foxp3/Transcription factor staining buffer (00-5523-00, eBioscience) as outlined by the manufacturer, before being stained for specific transcription factors.Definitions of cell types were as follows: CD8 + T cells were identified as CD45 + CD3 + CD8 + or CD45 + CD8 + T cells, NK cells as CD45 + CD3 − NK1.1 + cells, and cDC1 as CD45 + CD11b − CD11c + MHC − II + CD103 + cells.The specific monoclonal antibodies used are detailed in Supplementary Table 3.

DC generation
To generate murine iCD103 DCs, bone marrow mononuclear cells after depletion of red cells were cultured in 10 mL volume of RPMI 1640 medium enriched with 10% heat-inactivated FBS (10099, Gibco), 40 ng/mL recombinant murine FLT3L (250-31 L, Peprotech) and 20 ng/mL recombinant murine GM-CSF (315-03, Peprotech) for 7 days, and subsequently replaced with the same combination of cytokines and collected at days 10 to 14 for subsequent experiments.

RNA-seq analysis
Tumor RNA was extracted and reverse transcribed into cDNA-to-cDNA libraries, followed by sequencing by TIANGEN BIOTECH (Beijing, China) using Illumina Hiseq platform.Raw reads for each sample were aligned to the mouse reference genome (GRCm38) using Hisat2 v2.0.5.The expression of individual gene was normalized to fragments per kilobase of transcript per million mapped reads (FPKM), factoring in the gene's length and the mapped read counts.We identified significantly differentially expressed genes (DEGs) when they exhibited a fold change greater than 2 (an absolute value of Log2 ratio exceeding 1) and an adjusted p-value under 0.05, employing the DESeq2 analytical approach.For the DEGs, Gene Ontology (GO) enrichment was analyzed using the topGO package in R, considering GO terms with p-values less than 0.05 as significantly impacted.Additionally, the extent of DEG enrichment was assessed through the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways using the clusterProfiler package in R.
10× library preparation and sequencing 10× library preparations created from single cells carrying distinct barcodes were pooled for sequencing on a Chromium Single-Cell Platform (10×Genomics Chromium TM ) using 10× Genomics Single Cell 3′ Reagent Kits v2 (10× Genomics Chromium TM ) following the guidelines provided by the manufacturer.The quality of the library was evaluated using a Qubit fluorometer.Sample clustering was carried out on a cBot Cluster Generation System using the TruSeq PE Cluster Kit v3.The final library was sequenced with 150 base pair paired-end reads on an Illumina HiSeq 2000.

PCA analysis and linear dimension reduction
Genes that were detected in more than 3 cells and cells that had detective genes ranging from 200 to 2500 and the percentage of all the counts belonging to mitochondrial gene lower than 5% were kept to form high-quality data.Data were normalized by LogNormalize, an algorithm of Package Seurat before further analysis.Principle component analysis (PCA) calculates the similarity and variability of cells.For visualization, the Barnes-Hut t-Distributed Stochastic Neighbor embedding (t-SNE) was performed to reduce the dimension.The analysis above was conducted using the functions sourcing from Package Seurat version 4.0.3.

Statistical analysis
Statistical analyses were conducted using GraphPad Prism 8.0 software.Data are presented as mean ± standard deviation (SD).To compare differences between two groups, an unpaired Student's t test was utilized.Differences among multiple groups were analyzed using either one-way or two-way ANOVA, depending on the data structure.Survival rates were evaluated with the logrank test.Pearson and Spearman Correlation Coefficients were employed to assess the relationships between genes and signaling pathways.A p-value of less than 0.05 was considered statistically significant.

Fig. 2
Fig. 2 Inhibition of BCL9/BCL9L enhances cDC1 activation and facilitates cross-priming of CD8 + T cells.a, b CD40 (left) and CD86 (right) expression by CD103 + cDC1 of TdLNs (a) and tumors (b) from 30 mg/kg hsBCL9 z96 -treated CT26 tumor-bearing mice analyzed by flow cytometry (n = 3-4).c, d CD40 (left) and CD86 (right) expression by CD103 + cDC1 of TdLNs (c) and tumors (d) from MC38 tumor-bearing Bcl9 f/f Bcl9l f/f Cre-ERT2 mice treated i.p. with tamoxifen (1 mg/100 μL) in olive oil on days −7, −6, −5, +1, +6, and +11 post inoculation analyzed by flow cytometry (n = 3-4).e The representative plot of OT-I CD8 + T cells in TdLNs from hsBCL9 z96 -treated MC38-OVA tumorbearing mice analyzed by flow cytometry.f and g Quantitative analysis of the percentage of OT-I CD8 + T cells (f) and CFSE dilution of OT-I CD8 + T cells (mean fluorescent intensity, MFI) (g) based on the result of (e) (n = 3).h The representative plot of OT-I CD8 + T cells in TdLNs from MC38-OVA tumor-bearing Bcl9 f/f Bcl9l f/f Cre-ERT2 mice treated i.p. with tamoxifen (1 mg/100 μL) in olive oil on days −7, −6, −5, +1, and +6 post inoculation analyzed by flow cytometry.i, j Quantitative analysis of the percentage of OT-I CD8 + T cells (i) and CFSE dilution of OT-I CD8 + T cells (j) based on the result of (h) (n = 3).These data are representative values expressed as the mean ± standard deviation (SD) for each group, derived from three independent experiments; "n" denotes the number of biological replicates.An unpaired Student's t test was used for statistical analysis of the data in groups a-d, f, g, i, and j

Fig. 3
Fig. 3 Single-cell transcriptional profiling of CD8 + T cells and cDC1 in tumors and TdLNs from B16-OVA tumor-bearing Bcl9/Bcl9l deficiency mice.a Illustration of experiment and analysis process of single-cell transcriptional analysis.b TSNE plots of clustering process and marker genes (Zbtb46 for DCs, Cd68 for myeloid cells, Mlana for B16-OVA tumor cells, Cd3e for T cells, Cd4 for CD4 + T cells and Cd8a for CD8 + T cells) in tumors from B16-OVA tumor-bearing Bcl9 f/f Bcl9l f/f mice and Bcl9 f/f Bcl9l f/f Cre-ERT2 mice treated i.p. with tamoxifen (1 mg/100 μL) in olive oil on days −7, −6, −5, +1, +6 and +11 post inoculation.c-e TSNE plots of DC reclustering (c, e) and marker genes (Xcr1 for cDC1 and Clec10a for cDC2) (d) in tumors from B16-OVA tumor-bearing Bcl9 f/f Bcl9l f/f mice and Bcl9 f/f Bcl9l f/f Cre-ERT2 mice treated i.p. with tamoxifen (1 mg/100 μL) in olive oil on days −7, −6, −5, +1, +6, and +11 post inoculation presentation APP of peptide antigen via MHC class I APP of exogenous peptide antigen via MHC class I APP of peptide antigen APP via MHC class Ib(Bcl9 f/f Bcl9l f/f Cre-ERT2 vs Bcl9 f/f Bcl9l f/f ) Bcl9 f/f Bcl9l f/f Cre-ERT2 vs Bcl9 f/f Bcl9l f/f + αβ Treg cell differentiation T cell activation via TCR-MHC on APC T cell homeostasis NK T cell differentiation T cell differentiation T cell migration T cell proliferationT cell activation T cell apoptotic process t-score (Bcl9 f/f Bcl9l f/f Cre-ERT2 vs Bcl9 f/f Bcl9l f/f ) cell mediated immunity Positive regulation of activated T cell proliferation t-score (Bcl9 f/f Bcl9l f/f Cre-ERT2 vs Bcl9 f/f Bcl9l f/f ) Bcl9 f/f Bcl9l f/f Cre-ERT2 vs Bcl9 f/f Bcl9l f/f b