Involvement of MAF/SPP1 axis in the development of bone marrow fibrosis in PMF patients

Abstract

Primary myelofibrosis (PMF) is a myeloproliferative neoplasm characterized by hyperplastic megakaryopoiesis and myelofibrosis. We recently described the upregulation of MAF (v-maf avian musculoaponeurotic fibrosarcoma oncogene homolog) in PMF CD34+ hematopoietic progenitor cells (HPCs) compared to healthy donor. Here we demonstrated that MAF is also upregulated in PMF compared with the essential thrombocytemia (ET) and polycytemia vera (PV) HPCs. MAF overexpression and knockdown experiments shed some light into the role of MAF in PMF pathogenesis, by demonstrating that MAF favors the megakaryocyte and monocyte/macrophage commitment of HPCs and leads to the increased expression of proinflammatory and profibrotic mediators. Among them, we focused our further studies on SPP1 and LGALS3. We assessed SPP1 and LGALS3 protein levels in 115 PMF, 47 ET and 24 PV patients plasma samples and we found that SPP1 plasma levels are significantly higher in PMF compared with ET and PV patients. Furthermore, in vitro assays demonstrated that SPP1 promotes fibroblasts and mesenchymal stromal cells proliferation and collagen production. Strikingly, clinical correlation analyses uncovered that higher SPP1 plasma levels in PMF patients correlate with a more severe fibrosis degree and a shorter overall survival. Collectively our data unveil that MAF overexpression contributes to PMF pathogenesis by driving the deranged production of the profibrotic mediator SPP1.

Figure 1

Expression levels of MAF in CD34+ cells from MPN patients. Expression levels of MAF in CD34+ cells from MPN patients and HDs. Gene expression levels were measured by microarray analysis performed by Affymetrix platform as detailed in Norfo et al.(please, see the ‘Materials and Methods’ section). MAF expression levels are reported as Robust Multiarray Analysis (RMA)-normalized log2 signals, which were obtained by using the Partek GS software. Boxes represent the interquartile range that contains 50% of the subjects, the horizontal line in the box marks the median, and bars show the range of values. Data in a are representative of 38 PMF, 27 ET, 24 PV and 30 HD samples (14 from BM and 16 from PB). Only PMF (n=38) and HD (n=30) samples are shown in b, where PMF samples are classified into JAK2-mutated (n=20), MPL-mutated (n=3), CALR-mutated (n=9) and triple-negative (n=6), based on the mutational status. Only JAK2V617F-positive PMF (n=20) and ET (n=16) samples are shown in c, where samples are classified based on the JAK2V617F allele burden. BM, bone marrow; ET, essential thrombocythemia; MPN, myeloproliferative neoplasm; PMF, primary myelofibrosis; PV, polycythemia vera; HD, healthy donor; PB, peripheral blood; n, number of samples.

Figure 2

Detection of a subset of secreted molecules in LMAFVAR1IDN-, LMAFVAR2IDN- and LXIDN-transduced cells and in PMF patients plasma by ELISA. Protein levels of seven secreted molecules (IL-8, a; CCL2, b; PLAUR, c; MMP9, d; LGALS3, e; SPP1, f; THBS1, g) were measured by ELISA in culture supernatants from LMAFVAR1IDN, LMAFVAR2IDN and LXIDN-transduced cells (subpanels i) and plasma from PMF patients and HDs (subpanels ii). Protein levels are expressed as ng/ml. Protein levels in culture supernatants from LMAFVAR1IDN-, LMAFVAR2IDN- and LXIDN-transduced cells (subpanels i) are reported as mean±standard error of the mean (s.e.m.). *P<0.05; **P<0.01 versus LXIDN. The results come from three independent experiments. Plasma levels in PMF patients and HDs (subpanels ii) are shown as dot plots. Boxes represent the interquartile range that contains 50% of the subjects, the horizontal line in the box marks the median, and the bars show the range of values. Data are representative of 30 PMF and 10 HD samples. Comparisons between HDs and PMF patients were performed by using the Mann–Whitney U test, and the chosen level of significance was P<0.05 with a two-tailed test. ELISA, enzyme-linked immunosorbent assay; HD, healthy donor; PMF, primary myelofibrosis.

Figure 3

Detection of SPP1 and LGALS3 in the plasma of MPN patients by ELISA. Protein levels of SPP1 (a) and LGALS3 (b) were measured by ELISA in plasma samples derived from PMF (n=115), ET (n=47) and PV (n=24) patients and HDs (n=60). Protein levels are expressed as ng/ml. Boxes represent the interquartile range that contains 50% of the subjects, the horizontal line in the box marks the median, and the bars show the range of values. Comparisons between two groups were performed by using the Mann–Whitney U test, and the chosen level of significance was P<0.05 with a two-tailed test. ELISA, enzyme-linked immunosorbent assay; ET, essential thrombocythemia; HDs, healthy donors; n, number of samples; PMF, primary myelofibrosis; PV, polycythemia vera.

Figure 4

Role of MAF in the regulation of SPP1 and LGALS3 expression. (a) MAFvar1 and MAFvar2-dependent transactivation of human SPP1 promoter-driven (i) and LGALS3 promoter-driven (ii) luciferase expression. The amounts (in ng) of co-transfected plasmids are reported. Renilla-normalized Firefly luciferase levels (mean±s.e.m.;n=5 for SPP1 promoter and n=5 for LGALS3 promoter, respectively) are normalized by setting as =1 the pXP1-SPP1(−853/+22)/LXIDN empty vector-transfected and the pXP1-LGALS3(−1720/+35)/LXIDN empty vector-transfected samples, respectively. Error bars represent s.e.m. *P<0.05 compared to pXP1-SPP1(−853/+22)/LXIDN empty vector-transfected (i) or pXP1- pXP1-LGALS3(−1720/+35)/LXIDN empty vector-transfected (ii) samples, respectively. (b) Expression levels of MAFvar1, MAFvar2, SPP1 and LGALS3 upon MAF knockdown in HD-derived (i) megakaryocytes (n=6) and (ii) monocytes (n=6) as well as (iii) in PMF patients-derived CD34+ cells (n=5) were measured by qRT-PCR 24 h post-nucleofection. Expression levels are reported as mean RQ±s.e.m. respect to the NegCTRsiRNA sample, set as calibrator. *P<0.05; **P<0.01; ***P<0.001 versus NegCTRsiRNA. (c) Expression levels of proinflammatory cytokines IL-8, IL-6, TNF-α and IL-1β in HDs monocytes are measured by qRT-PCR after 24 h of culture in the presence or absence of rhSPP1 or rhLGALS3 and are reported as RQ (mean±s.e.m.; n=6) respect to the untreated control samples, set as calibrator. *P<0.05 versus untreated control. IL-1β, interleukin-1β IL-6, interleukin-6; IL-8, interleukin-8; n, number of experiments; rh, recombinant human; RQ, relative quantity; s.e.m., standard error of the mean; TNF-α, tumor necrosis factor-alpha.

Figure 5

Effects of SPP1 and LGALS3 on normal human fibroblasts. Effects of rhSPP1 and rhLGALS3 on proliferation of NHDFs. NHDFs were cultured in the absence or presence of 500 ng/ml rhSPP1, 1000 ng/ml rhLGAL3 or 5 ng/ml rhTGF-β1 (as positive control) for 24 and 48 h. (a) Evaluation of cell growth by Trypan Blue exclusion assay after 24 and 48 h of treatment. Results are shown as number of cells, starting from 5000 cells per sample, as reported in ‘Materials and Methods’ section. Values are reported as mean±standard error of the mean (s.e.m.); n=6. *P<0.05 versus untreated control. (b) Effect of an anti-SPP1 (i) and anti-LGALS3 (ii) neutralizing antibody on SPP1-induced NHDFs cell growth. NHDFs were treated with rhSPP1 (i) or rhLGALS3 (ii) in the absence or presence of anti-SPP1 neutralizing antibody (2.5 μg/ml, bi) or anti-LGALS3 neutralizing antibody (10 μg/ml, bii) or an isotype-matched antibody for 24 h. Cell growth was evaluated by Trypan Blue exclusion assay. For each sample the fold of increase in cell counts after 24 h of culture is shown as normalized to the number of cells plated (t₀). Values are reported as mean±s.e.m. (n=6). (c) Expression levels of COL1A1, an extracellular matrix-related gene involved in fibrosis, measured by qRT-PCR after 24 h of culture in the presence or absence of rhSPP1, rhLGALS3 or rhTGF-β1 and are reported as RQ (mean±s.e.m.; n=3) respect to the untreated control samples, set as calibrator. COL1A1, collagen type 1 alpha 1; NHDFs, normal human dermal fibroblasts; n=number of experiments; rh, recombinant human.

Figure 6

Effects of SPP1 and LGALS3 on PMF mesenchymal stromal cells. PMF-derived MSCs were cultured in the absence or presence of 500 ng/ml rhSPP1, 1000 ng/ml rhLGAL3 or 5 ng/ml rhTGF-β1 (used as positive control) for 48 h. (a) Cell proliferation was monitored by Trypan Blue exclusion assay after 48 h of treatment. Results are shown as number of cells, starting from 1.6 × 10⁴ cells per sample. Values are reported as mean±standard error of the mean (s.e.m.). The results come from three independent experiments. *P<0.05; **P<0.01 versus untreated control. (b) Effect of an anti-SPP1 (i) and anti-LGALS3 (ii) neutralizing antibody on MSCs cell proliferation. PMF MSCs were treated with rhSPP1 (i) or rhLGALS3 (ii) in the absence or presence of anti-SPP1 neutralizing antibody (2.5 μg/ml, bi) or anti-LGALS3 neutralizing antibody (10 μg/ml, bii) or an isotype-matched antibody for 48 h. Cell proliferation was evaluated by Trypan Blue exclusion assay. For each sample the fold of increase in cell counts after 48 h of culture was normalized to the number of cells plated at t₀. Values are reported as mean±s.e.m. (n=5). *P<0.05. (c) Expression levels of the fibrotic marker COL1A1 in PMF MSCs were measured by qRT-PCR after 48 h of culture with rhSPP1, rhLGALS3 or rhTGF-β1 and are reported as RQ (mean±s.e.m.; n=5) respect to the untreated control sample, set as calibrator. *P<0.05; **P<0.01. COL1A1, collagen type 1 alpha 1; MSCs, mesenchymal stromal cells; rh, recombinant human; n, number of experiments.

Figure 7

Clinical features of PMF patients according to MAF expression levels in CD34+ cells and SPP1 plasma levels. (a) MAF expression levels in CD34+ cells isolated from PMF patients at diagnosis according to the BM fibrosis stage they presented. Patients were divided in prefibrotic (MF-0/1, n=7) and overt fibrotic (MF-2, n=9 and MF-3, n=14) based on the BM fibrosis grading. Gene expression levels were measured by microarray analysis performed by Affymetrix platform as described in Norfo et al. MAF expression levels are reported as Robust Multiarray Analysis (RMA)-normalized log2 signals, which were obtained by using the Partek GS software. Boxes represent the interquartile range that contains 50% of the subjects, the horizontal line in the box marks the median, and bars show the range of values. The comparison between MF-0/1 and MF-2/3 groups was performed by using the Mann–Whitney U test. (b) Protein levels of SPP1 in the plasma of PMF patients (n=115) according to BM fibrosis grade. PMF patients were classified based on grade of fibrosis in prefibrotic (MF-0/1, n=54) and overt fibrotic (MF-2, n=43 and MF-3, n=18). Protein levels were measured by ELISA and are expressed as ng/ml. Plasma levels of SPP1 are shown as dot plots. Boxes represent the interquartile range that contains 50% of the subjects, the horizontal line in the box marks the median, and the bars show the range of values. Comparisons between two groups were performed by using the Mann–Whitney U test. (c) Kaplan–Meier estimates of overall survival in 115 patients with PMF, stratified by the plasma levels of SPP1. Patients with SPP1 plasma levels greater than the median value observed in the plasma of HDs were considered patients with high SPP1 levels. BM, bone marrow; ELISA, enzyme-linked immunosorbent assay; HDs, healthy donors; MF, myelofibrosis; PMF, primary myelofibrosis; n, number of samples.