Targeting S100A9-mediated inflammation: a novel therapeutic approach for CLL

Abstract

Chronic lymphocytic leukemia (CLL) presents challenges in treatment despite advancements in targeted therapies, often facing resistance or relapse. Chronic inflammation plays a significant role in CLL biology, with heightened inflammatory responses and immune dysfunction. Elevated levels of inflammatory cytokines support this notion. Activating signaling pathways such as NF-κB, Phosphoinositide 3-kinase delta (PI3Kδ), and MAPK via B-cell receptors and CD40 confers advantages to leukemic lymphocytes. Our research focuses on the proinflammatory protein S100A9 in CLL progression. We previously described that patients with CLL release exosomes containing S100A9 during disease progression, correlating with NF-κB activation. S100A9, known for its role in autoimmune diseases and cancers, modulates the antitumor immune response by influencing myeloid-derived suppressor cells. Receptors for S100A9 include Toll-like receptor 4, receptor for advanced glycation end products, and extracellular matrix metalloproteinase inducer (EMMPRIN). We identified a novel molecular mechanism involving the S100A9-EMMPRIN interaction in CLL using primary cells and an in vivo CLL mouse model (Eμ-TCL1). Additionally, we developed an Eμ-TCL1/S100A9-/- mouse model and explored pharmacological targeting of S100A9 in a patient-derived xenograft model, highlighting S100A9 as a promising therapeutic target in CLL with potential clinical applications.

Graphical abstract

Figure 1.

S100A9 promotes the activation of proinflammatory pathways in primary B-CLL cells from patients with progressive disease. CLL PBMCs were incubated with and without rhS100A9 for 72 hours, and several parameters were assessed. (A-C) Phosphorylation levels of AKT1 (Ser473 and Thr308), IKK (Ser176/180), and JNK (Thr183/185) were measured in CD19⁺CD5⁺ cells by FC. (D) JUN and FOS mRNA expression was assessed by qPCR. (E) Expression of the antiapoptotic proteins MCL-1 and BCL-2 in CD19⁺CD5⁺ was also evaluated by FC. (F) A multiplex cytokine assay was performed in PBMCs from patients with CLL after S100A9 in vitro stimulation for 24 hours. ∗∗∗∗P < .0001; ∗∗∗P < .001; ∗∗P < .005; ∗P < .05 (paired t test). Ctrl, Control of unstimulated cells; MCL-1, myeloid cell leukemia-1; MFI, median fluorescence intensity; ns, not significant.

Figure 2.

EMMPRIN blocking prevents S100A9-mediated activation of proinflammatory pathways. (A-C) Expression levels of S100A9 receptors TLR4, RAGE, and EMMPRIN were compared between B cells from HDs and CD19⁺CD5⁺ cells from thawed PBMCs using FC. (D-G) CLL PBMCs were stimulated with rhS100A9 and treated with S100A9 inhibitors (TasQ [10 μM] or PaQ [10 μM]) or EMMPRIN-blocking antibody [10 μg/mL] for 72 hours. Phospho-AKT1^{(Ser473 and Thr308)}, phospo-IKK^(Ser176/180), and phospo-JNK^(Thr183/185) were assessed in CD19⁺CD5⁺ cells by FC. Cells without rhS100A9 stimulation were used as control. ∗∗∗∗P < .0001; ∗∗∗P < .001; ∗∗P < .005; ∗P < .05 (panels A-C, unpaired t test; panels D-G, 1-way analysis of variance). Ctrl, Control of unstimulated cells; HD, healthy donor; MFI, median fluorescence intensity; ns, not significant.

Figure 3.

HG-EMMPRIN is upregulated in patients with progressive CLL. (A) EMMPRIN mRNA expression levels were quantified by qPCR in HDs and M-IGHV and UM-IGHV patients with CLL. (B) EMMPRIN protein levels in B-CLL cells of progressive and indolent patients were assessed by FC. Relative expression of percent positive cells and MFI analysis demonstrated higher levels of EMMPRIN in B-CLL cells of progressive patients than indolent patients. (C) CLL PBMC protein lysates were treated with PNGase F, and EMMPRIN expression was analyzed by western blot. GAPDH detection (bottom) was performed as loading control. The HG and LG forms are visible in untreated samples. The protein core is exposed after PNGase F treatment, confirming N-type glycosylation. (D) Representative blot of EMMPRIN in patients with progressive and indolent CLL, and relative protein expression was normalized to GAPDH. (E) Representative western blot of EMMPRIN expression in 1 patient with progressive CLL and 1 with indolent CLL after PBMC stimulation with IgM, CPG⁺IL15, and CD40L⁺IL4. (F) FC analysis of EMMPRIN expression upon stimulation with IgM, CPG⁺IL-15, and CD40L⁺IL-4 (n = 10 per group). ∗∗∗∗P < .0001; ∗∗∗P < .001; ∗∗P < .005; ∗P < .05 panels A-E, unpaired t test; panel G, t test. Ctrl, control of unstimulated cells; LG, low glycosylated; MFI, median fluorescence intensity; NA, not activated; ns, not significant; PNGase F, peptide N-glycosidase F; WB, western blot.

Figure 4.

Genetic silencing of S100A9 in murine B-CLL cells delays disease progression. (A-B) Splenocytes from 10- to 12-month-old Eμ-TCL1 or C57BL6 mice were used to evaluate S100-A9, EMMPRIN, RAGE, and TLR4 expression in CD19⁺CD5⁺ leukemic lymphocytes and CD19⁺CD5^– normal murine B cells. (C) We created a novel Eμ-TCL1/S100A9^–/– mouse model, and leukemic infiltration of CD19⁺CD5⁺ cells in the spleen of 10-month-old mice was compared between Eμ-TCL1 and the Eμ-TCL1/S100A9^–/– mice. (D) Aged Eμ-TCL1/S100A9^–/– mice show longer survival than the aged Eμ-TCL1 mouse model. (E-K) CD19⁺/CD5⁺ B cells isolated from Eμ-TCL1 or Eμ-TCL1/S100A9^–/– mice were transferred via TVI into NSG mice. (E-F) The tumor burden in PB was assessed weekly; the percentage of CD19⁺CD5⁺ cells and the absolute count of B-CLL cells were measured by FC. (G) Spleen size from both groups after 5 weeks of adoptive transfer. (H) Malignant B-cell infiltration in the spleen and bone marrow at week 5 after adoptive transfer. (I) Longer survival was observed in the Eμ-TCL1/S100A9^–/– group (n = 4) compared to the Eμ-TCL1 recipient mice (n = 5). (J) B cells were isolated from NSG mice after 5 weeks of adoptive transfer (AT), and then the mRNA was used for NanoString analysis. Volcano plot showing upregulated (red) and downregulated (blue) genes in S100A9^–/– B cells vs B cells from NSG adoptive transfer Eμ-TCL1 mice. (K) Downregulated genes in S100A9^–/– B cells relative to B cells from NSG adoptive transfer Eμ-TCL1 mice, belonging to the TNF-α signaling via NF-κB from the MSigDB Hallmark genes. ∗∗∗∗P < .0001; ∗∗∗P < .001; ∗∗P < .005; ∗P < .05. MSigDB, Molecular Signatures Database; ns, not significant; TNF-α, tumor necrosis factor α.

Figure 5.

Pharmacological inhibition of S100A9 prolongs survival in adoptive transfer Eμ-TCL1 mice. Splenocytes from Eμ-TCL1 mice were transferred via TVI into C57BL/6 mice, and PaQ and TasQ were administered at 25 mg/kg in drinking water for 4 weeks. (A) Tumor burden assessment in the PB using FC. (B) Representative picture of the 3 mice groups after 4 weeks of treatment. (C-D) Spleen weight and infiltration after 4 weeks of treatment. (E) Mice treated with PaQ or TasQ show longer survival than the control group. (F) Volcano plot showing differential gene expression in B cells from PaQ vs vehicle using NanoString PanCancer Immune Profiling panel. (G) Downregulated genes in PaQ-treated B cells relative to B cells from vehicle-recipient mice, belonging to the TNF-α signaling via NF-κB from the MSigDB Hallmark genes. ∗∗∗∗P < .0001; ∗∗∗P < .001; ∗∗P < .005; ∗P < .05. ns, not significant.

Figure 6.

TasQ eliminates B-CLL cells in a PDX model. (A) Clinical information of patient samples from the 2 donors used in this experiment. (B) Schematic representation of the experiment design. (C-D) Representative spleen pictures and FC dot plots after 3 weeks of treatment. (E-F) Spleen weight and absolute count of splenic B cells at week 3 after treatment. ∗∗∗∗P < .0001; ∗∗∗P < .001 (paired t test). FISH, fluorescence in situ hybridization; hCD19, human CD19; mCD45, mouse CD45; WBC, white blood cells. Uriepero-Palma, A. (2025) https://BioRender.com/04mb8cm. Panel B created with BioRender.com.