Evidences of early senescence in multiple myeloma bone marrow mesenchymal stromal cells

Abstract

Background: In multiple myeloma, bone marrow mesenchymal stromal cells support myeloma cell growth. Previous studies have suggested that direct and indirect interactions between malignant cells and bone marrow mesenchymal stromal cells result in constitutive abnormalities in the bone marrow mesenchymal stromal cells.

Design and methods: The aims of this study were to investigate the constitutive abnormalities in myeloma bone marrow mesenchymal stromal cells and to evaluate the impact of new treatments.

Results: We demonstrated that myeloma bone marrow mesenchymal stromal cells have an increased expression of senescence-associated β-galactosidase, increased cell size, reduced proliferation capacity and characteristic expression of senescence-associated secretory profile members. We also observed a reduction in osteoblastogenic capacity and immunomodulatory activity and an increase in hematopoietic support capacity. Finally, we determined that current treatments were able to partially reduce some abnormalities in secreted factors, proliferation and osteoblastogenesis.

Conclusions: We showed that myeloma bone marrow mesenchymal stromal cells have an early senescent profile with profound alterations in their characteristics. This senescent state most likely participates in disease progression and relapse by altering the tumor microenvironment.

Figure 1. RNA expression profile in MM BM-MSCs is highly similar to senescent BM-MSCs.

(A) Differential gene expression in 3 HD BM MSC samples and 4 untreated MM BM MSC samples was analyzed by Affymetrix GeneChip technology. In all, 419 ESTs were more than 2-fold up-regulated (green), and 537 were more than 2-fold down-regulated (red). Analysis demonstrated high changes in the global gene expression pattern in MM BM-MSCs. (B and C) Gene Ontology analysis (DAVID) was performed for the subsets of genes that were 2-fold up-regulated or 2-fold down-regulated, and the profile obtained was highly similar to the one observed by Wagner et al. in senescent BM-MSCs. The percentages of genes that contributed to representative categories are depicted (p<0.0001).

Figure 2. Reduced proliferation capacity of MM BM-MSCs.

(A) Number of BM-MSC cultures in a growth arrested state at each passage. BM-MSCs from untreated MM patients stop at the first to third passage, whereas the BM-MSCs from healthy donors never stop before the third passage and could last until the seventh passage. BM-MSCs from treated MM patients have a mixed profile. (B) Doubling time of BM-MSC cultures from healthy donors (n = 17), MGUS patients (n = 5), untreated MM patients (n = 10) and patients treated with Lenalidomide (n = 5), Thalidomide (n = 7) or Bortezomib (n = 8). (C) Cumulative number of CFU-F at P1 for HD BM-MSCs (n = 6), untreated MM BM-MSCs (n = 5) and treated MM BM-MSCs (n = 6). Green bars represent the mean ± SEM. *p<0.05 and **p<0.01 compared to HD BM-MSCs.

Figure 3. Early cellular senescence in MM BM-MSCs.

(A - left) BM-MSCs were stained for senescence-associated β-Galactosidase (SA β-Gal) between passage 1 and 3. The mean percentage of SA β-Gal positive cells is higher in untreated (n = 9) and treated (n = 12) MM BM-MSCs compared to HD BM-MSCs (n = 6). (A – right) Representation of the SA β-Gal staining in HD BM-MSCs and untreated MM BM-MSCs. (B) Cell cycle distribution of BM-MSCs was determined by flow cytometric analysis after propidium iodide DNA staining. The graph indicates the percentages of cells in G₀–G₁, G₂-M, S and Sub-G₀ phases of the cell cycle. We observed a reduction in G₀–G₁ phase and an increase in S phase for BM-MSCs of untreated (n = 7) and treated (n = 12) MM patients compared to HD BM-MSCs (n = 9). (C) mRNA expression of p53 and p21 by MGUS patients (n = 4), untreated MM BM-MSCs (n = 6) and MM BM-MSCs treated by Lenalidomide (n = 6), Thalidomide (n = 8) and Bortezomib (n = 6), compared to HD BM-MSCs (n = 12). *p<0.05 compared to HD BM-MSCs.

Figure 4. Reduced osteoblastogenesis in MM BM-MSCs.

(A - left) Relative calcium production (mg of calcium/µg of total proteins) by HD BM-MSCs (n = 6) as well as BM-MSCs from MGUS patients (n = 3), untreated (n = 12) and MM patients treated by Lenalidomide (n = 5), Thalidomide (n = 6) and Bortezomib (n = 8) after 7, 14 and 21 days of differentiation into osteoblasts. (A – right) Alizarin Red staining of HD BM-MSCs and BM-MSCs from untreated MM patients after 21 days of differentiation. (B) Relative alkaline phosphatase (ALP - Units/µg of total proteins) activity in HD BM-MSCs (n = 9) as well as BM-MSCs from untreated (n = 11) and MM patients treated by Lenalidomide (n = 5), Thalidomide (n = 6) and Bortezomib (n = 5) after 7, 14 and 21 days of differentiation into osteoblasts. *p<0.05 compared to HD BM-MSCs.

Figure 5. BM-MSC cytokine and chemokine expression profile.

(A) Representative cytokine/chemokine antibody array associated with conditioned media (CM) from myeloma and healthy BM-MSCs. (+) and (−) represent the internal positive and negative controls and numbers represents the targeted cytokine: 1. BDNF/2. EGF/3. GCSF/4. GMCSF/5. HGF/6. IFN-γ/7. IGF-I/8. IGF-II/9. IL-10/10. IL-1β/11. IL-6/12. IL-7/13. IL-8/14. MCP-1/15. MCP-3/16. MIP-1α/17. MIP-1β/18. MMP-2/19. MMP-9/20. OPG/21. PDGF-BB/22. RANTES/23. SCF/24. SDF-1/25. TGF-β2/26. TIMP-1/27. TIMP-2/28. TIMP-4/29. TNF-α/30. VEGF. We observed an upregulation in MM BM-MSC CM of BDNF, HGF, IGF-II, IL-6, IL-8, MCP-1, MIP-1a, MIP-1b MMP2, MMP9, OPG, RANTES, SCF, TIMP-1, TIMP-2, and VEGF. (B, C, E and F) Levels of IL-6, VEGF, GDF-15, DKK1 and TGF-β1, measured by ELISA, in conditioned media obtained from healthy donors (n = 14), MGUS patients (n = 5), untreated MM (n = 11) and MM patients treated by Lenalidomide (n = 9), Thalidomide (n = 12) or Bortezomib (n = 8). *p<0.05; **p<0.01; ***p<0.001 compared to HD BM-MSCs.

Figure 6. Altered MM BM-MSC hematopoietic support and immunomodulatory capacity.

(A) Relative number of colonies per 10⁴ cord blood CD34⁺ cells cultured on BM-MSC confluent layers for 7 days. BM-MSCs from untreated (n = 6; Kruskal-Wallis test, statistic = 12.88; *P<0.05; Dunn’s multiple comparison test Δ rank sum = 14.00) and treated (n = 10) MM patients support the growth of CD34⁺ cells more efficiently compared to HD BM-MSCs (n = 6). (B) Number of secondary colonies formed by CD34⁺ cells cultured for 7 days in semi-solid medium with cytokines and erythropoietin, after 5 weeks of co-culture with HD or MM BM-MSCs. We observed an increased number of colonies in untreated (n = 7) and treated (n = 14) MM patients compared to HD (n = 8). Green bars represent the mean ± SEM. (C) Different CD3⁺/PBMC ratios were used to perform MLR in the presence of HD BM-MSCs (n = 5), untreated MM BM-MSCs (n = 5) and treated MM BM-MSCs (n = 6). T cell proliferation was assessed by BrdU incorporation assay after 5 days of co-culture. For each ratio, MM BM-MSCs had decreased inhibition capacity. *p<0.05 and **p<0.01 compared to HD BM-MSCs.