The follicular lymphoma and chronic lymphocytic leukemia proliferative microenvironment at single-cell resolution

Antonio Ferreira, Shumei Wang, Larysa Poluben, Joshua D. Brandstadter, Eric Perkey, Steven Sotirakos, Mai Drew, Pan Li, Madeleine E. Lemieux, David M. Dorfman, Brent Shoji, Ivan Maillard, Stephen C. Blacklow, Jon C. Aster

Abstract

Lymph nodes (LNs) serve as the primary site for the growth of follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL), two B-cell neoplasms that remodel the stromal architecture in stereotypic ways. To characterize these malignancies within their native environments, we developed a pipeline for single-cell RNA sequencing (scRNA-seq) of all resident LN cells. Our analysis reveals that while FL and CLL occupy distinct architectural niches—defined by follicular dendritic cells and CCL19-expressing fibroblasts, respectively—they share a common proliferative program and a conserved set of ligand-receptor interactions. These findings provide a high-resolution roadmap for identifying vulnerabilities in the microenvironment-dependent survival of indolent lymphomas.

Introduction

Follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL) are characterized as "indolent" malignancies. While they often progress slowly, they remain difficult to eradicate due to their profound dependence on the lymph node microenvironment. These cancer cells do not exist in isolation; they rely on a complex support system of immune cells and structural stromal cells (fibroblasts) to survive and proliferate.

Historically, research has focused primarily on the malignant B cells themselves, often overlooking the critical "cross-talk" between the tumor and its surrounding architecture. This study addresses that gap by mapping the cellular neighborhoods of FL and CLL at single-cell resolution, identifying the specific molecular "handshakes" that sustain these cancers.

Holistic Single-Cell Mapping

To capture the full complexity of the tumor microenvironment, we utilized a specialized scRNA-seq pipeline on nine human lymph nodes (3 reactive, 3 FL, and 3 CLL). Unlike traditional studies that often focus exclusively on immune cells (CD45+), our approach sequenced both immune and structural/stromal cells (CD45-). This resulted in a comprehensive atlas of 82,596 individual cells.

One technical challenge addressed was the "N_Mito" filter. Standard pipelines often discard cells with high mitochondrial gene expression, assuming they are dying. However, we discovered that actively proliferating CLL cells naturally exhibit high mitochondrial content. By adjusting our filters, we preserved these critical populations, revealing that mitochondrial reprogramming is a hallmark of the CLL proliferative state and a potential contributor to drug resistance.

The Transient Proliferative Program

A key insight from our data is the nature of the proliferative cycle in these indolent cancers. We observed that proliferation commences with a transient upregulation of the MYC gene. Interestingly, MYC is quickly downregulated as cells progress through the cell cycle. This "on-off" switch likely acts as a natural brake, explaining why these neoplasms remain indolent rather than exhibiting the explosive growth seen in aggressive lymphomas.

Distinct Niches, Shared Survival Strategies

While FL and CLL share similar proliferative programs, they reside in different "neighborhoods" within the lymph node:

  • Follicular Lymphoma (FL): Proliferating cells co-localize with Follicular Dendritic Cells (FDCs) within neoplastic follicles. We found that the increase in FDCs in FL is driven by the enhanced differentiation of precursor stromal cells, effectively expanding the "scaffolding" available for tumor growth.
  • Chronic Lymphocytic Leukemia (CLL): Proliferating cells are spatially associated with a distinct set of CCL19-high fibroblasts located in specialized proliferation centers.

Despite these different architectural preferences, both cancers utilize a remarkably similar "survival kit" of molecular interactions. Using computational tools like CellPhoneDB and NicheNet, validated by multiplex confocal microscopy, we identified several conserved ligand-receptor pairs:

  • CD74-MIF: Involved in growth and survival signaling.
  • CD55-CD97: Linked to cell motility.
  • Galectin-9-CD44: Associated with immune evasion.

Conclusion and Therapeutic Implications

By mapping the interactions between malignant B cells, immune cells, and stromal fibroblasts, this study provides a roadmap for new therapeutic strategies. The identification of conserved interactions like CD74-MIF and CD55-CD97 suggests that we may be able to treat these cancers by disrupting their support systems. As several inhibitors for these pathways are already in clinical development for other indications, these findings offer a promising path toward starving indolent lymphomas of the environmental signals they need to persist.

Summary updated

Abstract

Abstract Lymph nodes (LNs) are secondary lymphoid organs where lymphocytes interact with antigen presenting cells to initiate adaptive immune responses within microenvironments established by resident stromal cells. LNs are also the major site of growth of follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL), B cell neoplasms that alter the stromal architecture of LNs in highly stereotypic ways. To characterize FL and CLL cells within the LN microenvironment, we developed a pipeline for single-cell RNA sequencing of all resident LN cells. We observed that proliferation of FL and CLL cells within specialized niches commences with transient upregulation of MYC, subsequent downregulation of which may act to limit the growth potential of these indolent neoplasms. Proliferating FL cells within neoplastic follicles co-localized with follicular dendritic cells, whereas proliferating CLL cells were spatially associated with a distinct set of fibroblasts expressing CCL19 that localized to proliferation centers. We used informatic analyses and microscopy to identify and validate interacting sets of ligand-receptor pairs between proliferating neoplastic B cells, immune cells and stromal fibroblasts, including interactions involving CD74-MIF, TNFRSF13C (BAFF receptor), immunomodulatory factors such as CD55 and Galectin-9 (Gal9), and adhesion molecules. Our analyses highlight common features of these two microenvironment-dependent neoplasms and provide a roadmap for identifying vulnerabilities and new therapeutic strategies. Key points Nodal FL and CLL cells have similar proliferative programs and predicted ligand-receptor interactions with immune and stromal cells. Proliferation of FL and CLL cells occurs in distinct lymph node niches defined by different types of fibroblasts.
BibTeX
@inproceedings{ferreira2024follicular,
  author = {Antonio Ferreira and Shumei Wang and Larysa Poluben and Joshua D. Brandstadter and Eric Perkey and Steven Sotirakos and Mai Drew and Pan Li and Madeleine E. Lemieux and David M. Dorfman and Brent Shoji and Ivan Maillard and Stephen C. Blacklow and Jon C. Aster},
  title = {{The follicular lymphoma and chronic lymphocytic leukemia proliferative microenvironment at single-cell resolution}},
  booktitle = {Blood},
  year = {2024},
  doi = {10.1182/blood-2024-199241},
}
Figure