Tumor-infiltrating CD27^-IgD^- regulatory B cells suppress cytotoxic CD8⁺T cell responses in renal cell carcinoma

Background

The function of B cells is highly dependent on the specific cancer type and its unique tumor microenvironment (TME). Typically, having a large number of B cells in a tumor is a good sign and leads to a higher chance of survival [1]. However, in renal cell carcinoma (RCC), the opposite seems to be true, with more B cells leading to worse outcomes.

Recent pan-cancer atlases have started to map out and investigate the multiple functions of tumor-infiltrating B cells (TIBs) by combining data from many different cancers [2] [3] [4]. However, these large-scale studies are often biased towards cancers with high B cell numbers, which can mask critical differences in tumors with low B-cell infiltration, like renal cell carcinoma (RCC). In this paper, Baig et al. use a novel single-cell method, combined with flow cytometry and high-plex imaging, to identify a previously unrecognized double negative 1 (DN1) B-cell subset (“double-negative” because these B cells lack the usual IgD and CD27 markers) that is enriched in RCC and is heavily associated with worse survival.

 

Key Highlights

Double Negative 1 (DN1) regulatory B cells (Breg) are enriched in RCC tumors

The authors used a unique clustering method of scRNA-seq data by first clustering B cells within each of four cancer types (breast, colon, lung, and RCC) before projecting them into a shared space for comparison. This method allowed them to identify B cell clusters in different cancers. Notably, they identified a new B cell population called double-negative 1 (DN1) B cells. Using a new integration model, known as the single-cell Variational Inference (scVI), the authors found that DN1 B cells are more common in RCC than other cancers. Specifically, DN1 B cells constitute 35.7% of all tumor-infiltrating B cells in RCC, while DN1 B cells constitute 1% of the other cancers.

Figure 1D from the preprint – Percentage of different B cell subtypes in each of the four cancers studied (BRCA, COAD, LC, and RCC).

DN1 B cell enrichment is associated with worse survival

The researchers collected tumor samples from patients with RCC undergoing surgery (nephrectomy). These patients had not yet received treatment prior to surgery. They ran flow cytometry on these samples, which confirmed a significant increase in DN1 B cells in tumor tissue.

The researchers linked B cell enrichment to worse survival by using a validated Leibovich score to predict the risk of cancer recurrence after nephrectomy. They found that high-risk patients exhibit increased levels of DN1 B cells, DN3 B cells, and plasma cells.

Additionally, researchers used an adjusted multivariable Cox proportional hazards model to analyze a large public database (TCGA). From this model, they showed that three gene signatures for these regulatory B cells (CD19, IL10, TGFB1) were associated with poorer patient survival, as shown in Figure 5I.

Figure 5I from the preprint – This figure is a forest plot showing that in a large database of RCC patients gathered from The Cancer Genome Atlas (TCGA), a high gene signature score for these regulatory B cells (CD19, IL10, TGFB1) was associated with worse survival.

 

DN1 Breg cells are found in tertiary lymphoid structures (TLSs) in RCC tumors

Researchers conducted high-plex spatial imaging of regions of interest (ROIs) to determine tertiary lymphoid structures (TLSs). They used cell segmentation and found that DN B cells are the most abundant B cell population within these structures. They found that the DN1 subset was the most frequent out of all DN B cells identified.

 

Naive B cells differentiate into DN1 Breg cells due to Toll-like receptor (TLR) signaling

The authors conducted a gene set enrichment analysis (GSEA) of the scRNA-seq data and found that Toll-like receptor (TLR) signaling pathways were upregulated. GSEA further revealed that the genes TLR7 and TLR9 specifically were enriched within the tumor’s DN1 B cell clusters. The authors then validated these findings using in vitro stimulation assays of TLR7 and TLR9 ligands. From these analyses, the authors were able to conclude TLR’s role in promoting B cell differentiation.

 

DN1 Breg cells promote reprogramming of CD8⁺ T cells into a T-regulatory (i.e. more tumor-friendly) phenotype

The authors conducted cell-cell interaction and found that B cells mostly engage with CD8⁺ T cells compared to other immune cell subsets. Spatial profiling further showed that Bregs are located close to CD8⁺ T cells. In vitro experiments revealed that tumor-derived DN B cells co-cultured with cytotoxic CD8⁺ T cells significantly reduced the T cells’ ability to produce the key anti-tumor cytokine IFN-γ.

 

Overall, Baig et al. used a variety of computational and experimental analysis to identify a new, enriched B cell population within RCC tumors. Further experimental analysis showed that these DN1 B cells are enriched in TLSs, and their abundance is associated with higher risk and poorer survival. They are developed from naïve B cells via TLR7/9-driven differentiation. Functionally, these DN1 Bregs engage nearby CD8+ T cells to suppress IFN-γ and bring these T cells toward a more tumor-permissive state.

 

Why this paper is important

One notable strength of this paper is the authors’ multi-modal approach to discover a new cell population (DN1 Bregs) responsible for immunosuppression in RCC. The authors first used a unique bioinformatic workflow to identify an immunosuppressive DN1 Breg population in RCC, then confirmed this finding’s significance using flow cytometry, spatial imaging, and in vitro functional assays.

            I thought the scRNA-seq pipeline that the authors took was especially unique and inventive. Their innovative pipeline of clustering and integrating B cells within different cancer types and implementing a deep-learning model for projection allowed them to have more detailed cell clusters compared to the regular scRNA-seq pipeline and even led them to discover a new and important B cell cluster.

 

References

 

[1]       Liu, H., Li, Z., Han, X., Li, Z., Zhao, Y., Liu, F., Zhu, Z., Lv, Y., Liu, Z., and Zhang, N. (2023). The prognostic impact of tumor-infiltrating B lymphocytes in patients with solid malignancies: A systematic review and meta-analysis. Critical Reviews in Oncology/Hematology 181, 103893. https://doi.org/10.1016/j.critrevonc.2022.103893.

[2]       Fitzsimons, E., Qian, D., Enica, A., Thakkar, K., Augustine, M., Gamble, S., Reading, J.L., and Litchfield, K. (2024). A pan-cancer single-cell RNA-seq atlas of intratumoral B cells. Cancer Cell 42, 1784-1797.e4. https://doi.org/10.1016/j.ccell.2024.09.011.

[3]       Yang, Y., Chen, X., Pan, J., Ning, H., Zhang, Y., Bo, Y., Ren, X., Li, J., Qin, S., Wang, D., et al. (2024). Pan-cancer single-cell dissection reveals phenotypically distinct B cell subtypes. Cell 187, 4790-4811.e22. https://doi.org/10.1016/j.cell.2024.06.038.

[4]       Ma, J., Wu, Y., Ma, L., Yang, X., Zhang, T., Song, G., Li, T., Gao, K., Shen, X., Lin, J., et al. (2024). A blueprint for tumor-infiltrating B cells across human cancers. Science 384, eadj4857. https://doi.org/10.1126/science.adj4857.

Tags: b cells, renal cell carcinoma, scrna-seq

doi: https://doi.org/10.1242/prelights.41204

Read preprint (No Ratings Yet)