Nuclear factor-erythroid 2 related factor (NRF2) is involved in cell defense and survival against endogenous and exogenous stress. Constitutively active Nrf2 in malignant cells increases the expression of cytoprotective genes and consequently, enhances proliferation via metabolic reprograming and inhibition of apoptosis (Leinonem, Advances in cancer Research, 2014). NRF2 is persistently activated in many human tumors, including acute myelogenous leukemia thus, inhibition of NRF2 activity may be a promising strategy for leukemia therapy. Flavonoids, present in vegetables, fruit and propolis, might exert antitumoral effects through induction of apoptosis and chromatin remodeling (Link, Biochem Pharmacol., 2010). A previous study from our group showed that Quercetin (Qu), a natural polyphenolic flavonoid compound, induced apoptosis, partly due to its DNA demethylating activity, through HDAC inhibition and by the enrichment of H3ac and H4ac in the promoter regions of genes involved in the apoptosis pathway, leading to their transcription activation (Alvarez, Clinical epigenetics 2018). In the present study, we evaluated the effect of Qu as a modulator of NRF2. This study was performed in vivo in human xenograft acute myeloid leukemia (AML) models, and in vitro using leukemia cell lines. Qu treatment (50 µM Qu) for 48h downregulated HDAC4, NRF2 and p-NRF2 at protein levels (p<0.05; p<0.005; p<0.005 respectively). Imaging Flow Cytometry (AMNIS, ImageStream ISX mkIITM) and Confocal Microscopy evidenced a decrease in NRF2 nuclear localization. Furthermore, combined treatment with the proteasome inhibitor MG132 prevented degradation of NRF2, indicating that treatment increased proteasomal degradation. Loss of NRF2 decreases HDAC4, a redox sensitive histone deacetylase, resulting in an increased expression of miR-1 and miR-206 (Singh, J Clin Invest. 2013). Herein, expression profile of 84 miRNAs (Apoptosis miRNA PCR array) were performed in samples from human xenograft model. Treatment up-regulated the expression profile of 5% (n=4) of the 84 miRNAs evaluated, corresponding exclusively to miRNAs that target anti-apoptotic genes and to miRNAs that have been demonstrated to have pro-apoptotic functions. Furthermore, expression levels of miR-1, miR-133a/b, which target anti-apoptotic genes and miR-206, a pro apoptotic miR, were validated in xenograft model samples, resulting in a significant up-regulation of the expression levels in treated animals compared to controls (p<0.05). In addition, lentiviral sh down regulation of NRF2, led to an increased apoptosis, decreased cell survival and an up regulation of miRNA 206 expression in Qu treated cells. In summary, Qu might induce programmed cell death in part, by decreasing NRF2 nuclear localization, by inducing NRF2 proteasomal degradation and down regulation of HDAC4 which led to up-regulation of pro apoptotic miRNAs. Disclosures No relevant conflicts of interest to declare.

The mechanism underlying quiescence and/or mobilization of hematopoietic stem cells and their bone marrow progenitors (HSPC) into circulation are tightly regulated for the continuous supply of peripheral blood cells; however, non-physiological or stress conditions, such as infections, can accelerate these mechanisms. Our results have shown that polyphenols modulate quiescence/mobilization of HSPC, but do not affect mature populations. Thrombin has been reported to induce the rapid HSPC mobilization through coagulation thrombin/PAR-1 axis, and quiescence is maintained across the APC/EPCR/PAR-1 axis (Nat. Med. 2015, 21:1307-17). Our objective was to investigate the effect of polyphenols on thrombin/PAR-1 and APC/EPCR/PAR-1 axis. C57BL/6J mice (6-8 weeks old) were treated with polyphenols from green tea extract (250 mg/kg body weight) orally (gavage) once every seven days and injected (i.p.) at day 7 with lipopolysaccharide (LPS) (100μg;Sigma) (n=6). The control group received vehicle and was injected with LPS (n=6). After 24h of LPS injection, mice were anesthetized for blood collection, and then sacrificed for bone marrow collection. PAR-1 and EPCR expression, nitric oxide (NO) production and endothelial nitric oxide synthase (eNOS) phosphorylation were evaluated in HSPC by flow cytometry. The functional ability of HSC was assessed by competitive repopulation assay. Vascular permeability was studied using Evans blue. After LPS injection, mice showed reduced expression of EPCR in bone marrow LSK parallel to an increase of PAR-1 expression in circulating immature and mature cells. Treatment of these mice with polyphenols partially prevented the reduced expression of EPCR in bone marrow LSK (13±3 vs 54±12; p<0.05), but did not affect the increased PAR-1 expression in circulating immature and mature cells. Evans blue assay revealed a reduction in the vascular permeability of the bone marrow of LPS-injected mice treated with polyphenols (3.9±0.5 vs 2.1±0.1; p<0.05). To assess whether polyphenols altered NO production, we measured NO levels and eNOS phosphorylation in immature LSK EPCRhigh (or LT-HSC) cells. NO production is activated by eNOS phosphorylation at Ser1177 and negatively regulated by eNOS phosphorylation at Thr495. LPS injection rapidly increased NO levels and eNOS phosphorylation at Ser1177 in bone marrow LSK of mice. Treatment of these mice with polyphenols reduced the percentage of bone marrow LSK EPCRhigh cells with higher intracellular NO (52±2.8 vs 28±5.6; p<0.01) and increased eNOS phosphorylation at Thr495 in immature LSK. In order to evaluate the action of polyphenols on the functional ability of HSC, a competitive bone marrow repopulation assay was performed. Donor mice (C57BL/6J) received or not polyphenols followed by LPS injection (treated group: Polyphenols+LPS; control group: LPS), and bone marrow cells were transplanted (1:1) together with bone marrow cells of competitors (B6.SJL-PtprcaPepcb/BoyJ) in lethally irradiated recipients (B6.SJL-PtprcaPepcb/BoyJ). Mice were followed for 16 weeks and hematological analysis revealed no difference in circulating leukocytes, platelets or hemoglobin levels. Transplanted mice (recipients) presented a higher percentage of CD45+ cells from Polyphenols+LPS donors (33.7±13 vs 78.6±0.9; p<0.05) in the peripheral blood, as well as increased number of T lymphocytes (6.7±4.5 vs 37.2±2.9; p<0.05) and myeloid cells (68.5±1.7 vs 82.5±3.5; p<0.05) from Polyphenols+LPS group. After 16 weeks, mice were euthanized and a higher percentage of LSK (or HSC) and LSK EPCRhigh (or LT-HSC) cells from Polyphenols+LPS donors were detected in the bone marrow, although only the percentage of LSK EPCRhigh was statistically different (0.0014±0.0001 vs 0.0032±0.001; p<0.05). Taken together, our results indicate that polyphenols increased the functional ability of HSC in LPS-injected mice showing increased percentage of bone marrow LSK EPCRhigh cells, which are the most quiescent stem cells with strong self-renewal ability. Polyphenols reduced EPCR expression and NO production in immature cells of LPS-injected mice, exhibiting an anti-inflammatory effect that leads to the maintenance of barrier integrity and quiescence of cells, which was corroborated by reducing vascular permeability in the bone marrow. Thus, polyphenols appear to modulate quiescence/mobilization of HSPC through APC/EPCR/PAR-1 axis. Disclosures No relevant conflicts of interest to declare.

No abstract

In recent years, the role of tumor microenvironment in neoplasm initiation and malignant evolution has been increasingly recognized. However, the bone marrow mesenchymal stem/stromal cell (BMMSC) contribution to disease progression remains poorly explored. We had previously performed a microarray analysis of myelodysplastic syndrome (MDS) patient-derived BMMSC (MDS-BMMSC) and found an underexpression of HAI-2/SPINT2, an endogenous inhibitor of the hepatocyte growth factor (HGF) activator. This gene has been described as methylated in various cancer types and has been associated with disease progression. Despite of being related to the pathogenesis of several neoplasms, the role of HAI-2/SPINT2 has not yet been fully elucidated in hematological diseases, such as MDS and acute myeloid leukemia (AML). Thus, the aim of this study was to evaluate HAI-2/SPINT2 expression in derived BMMSC and total bone marrow (BM) of healthy donors (HD), MDS and AML patients as well as in BMMSC treated with 5-Azacitidine (Aza), a DNA methyltransferase (DNMT) inhibitor. To achieve this, we collected BM hematopoietic cells and plastic-adherent BMMSC from aspirates of HD, MDS and AML patients. BMMSC were expanded to passage 4 and defined as CD73+/CD90+/CD105+/CD45-/CD34-/CD31-/HLA-DR-. A total of 29 HD and 121 patients at diagnosis (MDS=72 [low-risk=46, high-risk=26], AML with myelodysplastic related changes (AML-MRC)=17 and de novo AML=32) were included. HAI-2/SPINT2 mRNA was significantly decreased in MDS- (0.34[0.01-2.06];P <.01) compared to HD-BMMSC (0.89[0.46-1.59]). When patients were stratified according to WHO classification, HAI-2/SPINT2 expression was lower in both low-risk (0.31[0.01-1.33]) and high-risk (0.43[0.01-2.06]) MDS-BMMSC. Similar results were found in total BM: HAI-2/SPINT2 transcripts were significantly decreased in MDS (0.41[0.01-2.53];P <.01), AML-MRC (0.38[0.014-0.84];P <.01) and AML patients (0.33[0.01-2.07];P <.001) compared to HD (0.91[0.19-4.79]). To investigate whether this loss of expression was due to HAI-2/SPINT2 methylation, BMMSC were treated with Aza (1µM or IC50 value) for 48h. In MDS- and AML-BMMSC, Aza treatment resulted in a pronounced upregulation of HAI-2/SPINT2 mRNA and protein levels. Moreover, Aza treatment of HD-BMMSC did not improve the HAI-2/SPINT2 mRNA and protein levels as much as the observed in MDS- and AML-BMMSC. To better understand the role of HAI-2/SPINT2 downregulation in BMMSC physiology, its expression was inhibited in a BM stromal cell line (HS5). As previously reported, HAI-2/SPINT2 silencing resulted in an increased secretion of HGF, known to be overexpressed in plasma of MDS patients and considered a prognostic factor in MDS and AML patients (Matsuda et al., Leukemia, 2004). Moreover, after co-culture, HAI-2/SPINT2 knockdown improved survival of blasts isolated from AML-MRC and AML patients. We also observed an increased adhesion of CD34+ hematopoietic stem cells (HSC) to HAI-2/SPINT2 silenced HS5 cells. This prompted us to analyze the expression of cell adhesion molecules in MDS- and AML-BMMSC. We observed a significant augment in the expression of CD49b and CD49d integrins in MDS- and AML-compared to HD-BMMSC. Taken together, SPINT2 inhibition improves HGF secretion, consequently with alteration in molecule receptor adhesion, resulting in an increased expression of integrins (CD49b and CD49d) responsible for cell-to-cell adhesion. Thus, reactivation of HAI-2/SPINT2 levels after Aza treatment indicates that this gene is probably epigenetically silenced by methylation in MDS and AML, and is possibly a tumor suppressor gene. Interestingly, nowadays, epigenetic therapy by Aza is the first-line treatment for MDS patients, and induces prolonged survival and delayed AML evolution. Likewise, our results suggest that HAI-2/SPINT2 may play a role in deregulation of HGF cytokine secretion with consequently alteration in HSC adhesion and growth/survival. Tumor microenvironment niche is currently known to play a critical role in cancer initiation and progression, thus HAI-2/SPINT2 may contributes to functional and morphological abnormalities of microenvironment niche and with the stem/progenitor cancer cell progression. Hence, downregulation in HAI-2/SPINT2 gene expression, due to methylation in MDS- and AML-BMMSC, provides novel insights into the pathogenic role of the leukemic bone marrow microenvironment. Disclosures No relevant conflicts of interest to declare.

Coordinated arrangement of bone marrow (BM) microenvironment is essential for the maintenance of hematopoietic stem cells and their progenitors (HSPC), and for the continuous supply of peripheral blood (PB) cells. Quiescence and/or mobilization of these BM cells into the circulation are tightly regulated however non-physiological or stress conditions, such as infections, can accelerate them. Our previous findings have shown that polyphenols from green tea modulate the myeloid population (Gr-1+Mac-1+) of BM, spleen and PB of mice challenged by lipopolysaccharide (LPS). We have also observed a reduction in the circulating levels of thrombin-antithrombin complex (TAT) and soluble endothelial protein C receptor (sEPCR) of these mice. It has been reported that thrombin induces the rapid HSPC mobilization through coagulation thrombin/PAR-1 axis, and quiescence is maintained across the APC/EPCR/PAR-1 axis (Nat. Med. 2015, 21:1307-17). In this context, our goal was to investigate the effects of polyphenols on the Lin-Sca-1+c-Kit+ (LSK) primitive stem cell population and mature populations including: T and B lymphocytes, monocytes, granulocytes and erythrocytes (Lin+). The expression of PAR-1 and EPCR in these cells, and the levels of sEPCR in the BM fluid were also evaluated. The 250 mg/kg dose of polyphenol from green tea extract was given once every 7 days orally (gavage) to mice (n=6/group) challenged with i.p. injection of 100µg LPS. Control group received vehicle only. After 24h, mice were sacrificed; BM and PB were collected for the assays. LPS injection induced an increase in circulating LSK, without affecting the pool of these cells in BM. Treatment with polyphenols prevented recruitment of LSK into the PB, and did not alter the percentage of BM LSK. Corroborating these findings, polyphenols also reduced the number of immature progenitors (CFU-C) in the PB, evaluated by the clonal culture assay. In addition, polyphenols did not affect the increase in the number of Lin+ cells and circulating leukocytes (total WBC) induced by LPS injection. We also observed that LPS induced a rapid increase in the number of PB LSK and Lin+ cells expressing PAR-1, while reducing BM LSK expressing EPCR. Polyphenols did not affect PAR-1 expression in both populations, but partially recovered expression of EPCR in BM LSK. Polyphenols also reduced the increased levels of sEPCR induced by LPS in the BM niche. Taken together, our results demonstrate that polyphenols promote an effect on the quiescence/mobilization of both the most primitive hematopoietic cells and their progenitors, but does not affect mature populations. The reduction in vascular permeability observed in the BM of mice using Evans blue reinforces this action of polyphenols. It appears that the effect of polyphenols is thrombin dependent; although polyphenols did not affect the expression of PAR-1, they modulated EPCR, mainly in the medullary microenvironment with maintenance of barrier integrity and consequent cellular quiescence. In addition, polyphenols appear to have an anti-inflammatory effect, possibly triggered by their effect on the EPCR. Disclosures De Paula: Hematology and Transfusion Medicine Center, University of Campinas: Employment. Shiraishi:Hematology and Transfusion Medicine Center, University of Campinas: Employment. Queiroz:University of Campinas: Employment.