Targeting the Inflammatory Niche in MDS By Tasquinimod Restores Hematopoietic Support and Suppresses Immune-Checkpoint Expression in Vitro

Introduction

Myelodysplastic syndromes (MDS) belong to the most common hematological neoplasms in the elderly population, characterized by ineffective hematopoiesis, peripheral cytopenia and the risk of transformation into acute myeloid leukemia. There is increasing evidence that an aberrant innate immune response and a proinflammatory bone marrow (BM) microenvironment play a critical role in the pathogenesis of MDS. The alarmin S100A9, a key player for regulation of inflammatory responses, has been shown to be elevated in MDS patients. It directs an inflammatory cell death (pyroptosis) by increased NF-kB mediated transcription and secretion of proinflammatory, hematopoiesis-inhibitory cytokines and production of reactive oxygen species.

Tasquinimod (TASQ, Active Biotech) is a novel, oral small molecular drug with S100A9 inhibitory activity and it is currently investigated in a phase Ib/IIa trial in relapsed/refractory multiple myeloma (NCT04405167). TASQ has demonstrated anti-angiogenic, antitumor and immunomodulatory properties in a broad range of preclinical solid tumor models; however, little is known about its effects in myeloid malignancies.

Aim

We investigated the role of S100A9 in cellular models of MDS and the potential of TASQ to target S100A9 within the MDS stroma in vitro.

Methods

Immunohistochemical staining of S100A9, CD271+ mesenchymal stromal cells (MSCs), CD68+ macrophages and CD66b+ neutrophils in BM tissues from MDS patients and healthy donors was performed with multiplex immunohistochemistry and analyzed with the VECTRA imaging system.

MSCs from patients with either low-risk MDS, CMML or age-adjusted healthy donors were exposed to S100A9 (1.5µg/ml) in the presence or absence of TASQ (10µM). Subsequently, TLR4 downstreaming molecules such as IRAK1, gasdermin and NF-kB-p65 were analyzed by Western blot. Moreover, the mRNA expression of further proinflammatory molecules (IL-1b, IL-18, caspase1) and PD-L1 was quantified by real-time PCR. To study the impact on the hematopoietic support, MSCs were pre-treated for one week with S100A9 ± TASQ before CD34+ hematopoietic stem and progenitor cells (HSPCs) were seeded on the stromal layer. The colony formation (CAF-C) was analyzed weekly followed by a CFU-GEMM assay in methylcellulose medium. Additionally, PD-1 mRNA expression was quantified in cocultured HSPCs.

Results

Immunohistochemical staining of BM tissue demonstrated S100A9 expression mainly by CD66b+ neutrophils and with less extent by CD68+ macrophages. In line with this, we could not detect relevant S100A9 mRNA expression in cultured MDS or healthy MSCs in vitro.

Exposure of MDS and healthy MSCs with S100A9 induced TLR4 downstream signalling as demonstrated by increased expression of IRAK1 and NF-kB-p65. We further detected a higher expression of gasdermin, an inductor of pyroptosis, in S100A9 exposed MSCs. Addition of TASQ abolished these effects and inhibited the expression of the mentioned proteins, indicating an alleviation of inflammation.

Furthermore, we detected a 2-fold increase of mRNA expression of the proinflammatory cytokines IL-1b and IL-18 as well as a 5-fold increase of their activator caspase 1 in MSCs after treatment with S100A9, which could be prevented by TASQ. Interestingly, PD-L1 as a potential downstream target was induced by S100A9 by 2.5-fold and could be suppressed by TASQ to about 50%.

To evaluate the impact on the hematopoietic support of MSCs, we analysed MSC/HSPC cocultures after treatment with S100A9. We observed a decreased number of cobblestone area forming cells (CAF-C) as well as reduced numbers of colonies (CFU) in a subsequent clonogenic assay, indicating a disturbed hematopoietic support by S100A9 treated MSCs. Interestingly, both the number of CAF-C and CFU could be increased by TASQ pre-treatment. Finally, the PD-1 expression in co-cultured HSPCs was regulated in the same way as its ligand in treated MSCs, nominating this interaction as a potential target of S100A9/TASQ in the MDS BM.

Conclusion

In summary, we provide evidence that the pathological inflammasome activation in the myelodysplastic bone marrow can be rescued by TASQ at least in part by inhibition of the S100A9 mediated TLR4 downstream signalling including NF-kB-p65 transcription and PD-L1 expression. These effects result in an improved hematopoietic support by MSCs, suggesting a potential efficacy to improve cytopenia in low-risk MDS patients.