Background:

Systemic light chain (AL) amyloidosis is caused by clonal expansion of plasma cells (PCs) secreting misfolded monoclonal immunoglobulin that aggregate in tissues, leading to organ dysfunction. While recent studies have begun to reveal the transcriptional heterogeneity and immune dysregulation in AL amyloidosis, the specific transcriptomic features of amyloidogenic malignant plasma cells (mPCs), their effects on non-malignant polyclonal PCs (nmPCs), and interactions with the bone marrow microenvironment (BME) remain poorly understood.Methods: We performed 5' scRNA-seq (10x Genomics) on bone marrow aspirates from 13 newly diagnosed AL patients (whole BM n=6; CD138+ and CD138- sorted n=7) and 10 healthy donors. Reads were processed in Cell Ranger and analyzed in Seurat. mPCs were defined by CD138, immunoglobulin light-chain restriction, MZB1 and Cyclin-D; nmPCs and healthy donor PCs (hPCs) by CD138, MZB1 and Azimuth mapping. To address inter-patient heterogeneity, differential gene expression (DGE) between mPCs and nmPCs was done within-patient, reporting genes conserved across ≥7 patients. DGE relative to hPCs was done with edgeR. Gene ontology (GO) enrichment was done with enrichR. Cell-cell communication analyses were performed using CellChat.

Results:

We analyzed a total of 62,241 PCs, (29,671 mPCs, 32,354 nmPCs, 216 hPCs). Compared to hPC, mPC harbored 359 upregulated and 106 downregulated genes (FDR<0.05). Among those enriched were proteostasis genes (HSPA1A, HSPA6, DNAJB1, BAG3), Wnt antagonists (DKK1, FRZB), and cytokine regulators (DUSP2, CCL5). Analysis of mPCs relative to nmPCs showed increased expression of 418 genes (FDR < 0.05). These included known targets in multiple myeloma (DKK1, GPRC5D, and XBP1), as well as several proteasome subunits (PSMB1, PSMB3, PSMB6, PSMB8, PSMB9). Genes linked to non-AL amyloid formation (ITM2A, ITM2B) were also upregulated. Potential new therapeutic targets were identified, including strong overexpression of Glutathione S-Transferase Pi 1 (GSTP1) in mPCs. Treating ALMC-1 cell line with the GSTP1 inhibitor ezatiostat reduced GSTP1 expression, triggered apoptosis (caspase cleavage and Annexin V staining), and significantly decreased cell viability in a dose-dependent way within 24 hours. These results suggest that AL plasma cells may be sensitive to GSTP1-targeted therapy.

Further comparison between nmPCs and hPCs (421 genes up, 74 down, FDR<0.05) revealed transcriptional differences in pathways commonly associated with plasma cell malignancy. Specifically, nmPCs exhibited increased activity of NF-κB, unfolded protein response (UPR), and mTORC1 signaling, along with decreased MHC class II and PD-1 signaling (FDR < 0.01). These findings suggest that even polyclonal plasma cells are significantly influenced by the diseased BME. Compared to nmPCs using GO enrichment, mPCs were characterized by dysregulation of multiple proteostatic, metabolic, and immune-modulatory programs. mPCs activated proteasome, ERAD, UPR, oxidative phosphorylation and survival signaling (Hedgehog, non-canonical NF-κB, NOTCH2), alongside enhanced MHC-I presentation; B-cell receptor, interferon and MAPK/TGF-β pathways were downregulated (FDR<0.01). This pattern suggests an integrated strategy of immune evasion, enhanced proteostasis, and BME-dependent survival in mPCs.

Finally, relative to both nmPCs and hPCs, CellChat in mPCs predicted stronger macrophage migration inhibitory factor (MIF) and midkine (MK) signaling from mPCs to B, T, NK and myeloid cells, and increased ITGB2 contact, implying altered lymphocyte adhesion. MK signaling was also present in nmPC, but not in hPC. Together, these findings suggest that mPCs contribute to pathogenic remodeling of the immune microenvironment including alteration of nmPC, promoting chronic inflammation and immune dysfunction.Conclusion: Our scRNA-seq study in AL amyloidosis delineates the transcriptomic landscape of mPCs, comparing them to nmPCs and hPCs. We identify novel therapeutic targets, including GSTP1, and reveal that even the polyclonal plasma cells in AL patients exhibit an altered phenotype. Additionally, we uncover immune communication pathways that may drive disease progression. Together, these findings provide mechanistic insight into plasma cell biology in AL amyloidosis and highlight new opportunities for therapeutic intervention. Ongoing preclinical studies are underway to functionally validate these targets.

This content is only available as a PDF.
Sign in via your Institution