Effects of azacitidine on matrix metalloproteinase-9 in acute myeloid leukemia and myelodysplasia

Experimental Hematology 2013;41:172–179

Effects of azacitidine on matrix metalloproteinase-9 in acute myeloid leukemia and myelodysplasia  Teresa Bernala, Angela Moncada-Pazosb, Clara Soria-Vallesb, and Ana Gutierrez-Fernandezb a

Servicio de Hematologıa, Hospital Universitario Central de Asturias, Oviedo, Spain; bDepartamento de Bioquımica y Biologıa Molecular, Instituto Universitario de Oncologıa, IUOPA, Universidad de Oviedo, Oviedo, Spain (Received 24 August 2011; revised 22 September 2012; accepted 10 October 2012)

Matrix metalloprotease-9 (MMP9) plays a critical role in acute myeloid leukemia (AML) by increasing the invasive properties of malignant myeloblasts. The role of this enzyme in highrisk myelodysplastic diseases (MDS) and the effect of azacitidine on its expression in MDS and AML have not been studied in detail. In this work, we have analyzed the effect of different concentrations of azacitidine in two well-established, MDS-derived, acute myeloid leukemic cell lines: MOLM-13 and SKM-1. We have demonstrated that 1 mmol/L azacitidine decreases MMP9 DNA methylation levels and that this is correlated with a significant increase in messenger RNA expression in both cell lines. Surprisingly, changes in protein levels were minor. This paradoxic effect is explained by the drug-dependent induction of apoptosis that reduces the amount of active secreting cells. A balance between induced expression and apoptosis was established at an azacitidine concentration of 0.2 mmol/L in MOLM-13 cells. This dose significantly increased the invasive capacity of viable cells, as measured in the Matrigel assay. To evaluate the clinical relevance of this observation, we have examined the effect of azacitidine on MMP9 expression in bone marrow from five patients with MDS, with the finding that this drug significantly increased MMP9 protein levels in all analyzed patients after six cycles of treatment. Based on these results, we conclude that azacitidine increases MMP9 expression and may enhance invasiveness in vitro. Because all five patients relapsed, these findings might explain, at least partially, the clinical failure of the drug and the progression to a more aggressive disease. Ó 2013 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc.

Myelodysplastic syndromes (MDSs) are a group of neoplastic disorders characterized by peripheral blood cytopenia and a variable risk of leukemic evolution. The probability of survival and leukemic evolution is categorized by the International Prognostic Scoring System (IPSS) [1]. The leukemia-free survival of patients with an IPSS score of 1.5–2.5 is less than 1 year, and the mean survival is in the range of 4–12 months. These results are similar to those of untreated acute leukemia. Because of these dramatic features, the delay of evolution to acute leukemia is a major therapeutic goal in these diseases [2]. New drugs, such as lenalidomide and the hypomethylating agents azacitidine and decitabine, have emerged recently for the treatment of MDS. In Europe, azacitidine is the only drug approved for the treatment of high-risk Offprint requests to: Dr. Teresa Bernal, Servicio de Hematologıa, Hospital Universitario Central de Asturias, Celestino Villamil s/n, 33006, Oviedo, Spain; E-mail: [email protected]

MDS that is not eligible for stem cell transplantation. This agent delays leukemic evolution, improves hematopoiesis, and prolongs survival compared with conventional treatment [3]. The rationale for the use of methylation inhibitors to treat high-risk MDS is that during the progression from the myelodysplastic phase to acute leukemia, key tumor-suppressor genes become silenced by hypermethylation [4,5]. Supporting this model, an association between reversal of the hypermethylated status of key genes and clinical response has been observed in some studies [6]. Other groups have not found such an association [7], probably because there are several processes involved in the progression from MDS to acute myeloid leukemia (AML) and other mechanisms of action of the drug [8–11]. The rate of response to azacitidine monotherapy rises to 45–50%, but the complete remission rate is low (17%) and eventually all patients experience some kind of response relapse [3]. The patients who relapse after an initial response or fail to respond to primary treatment with

0301-472X/$ - see front matter. Copyright Ó 2013 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.exphem.2012.10.005

T. Bernal et al./ Experimental Hematology 2013;41:172–179

hypomethylating agents have a poor prognosis, because the rate of response to rescue therapies is low and of short duration [12]. Moreover, the failure of previous treatment with hypomethylating agents is an adverse prognostic factor for the response to clofarabine and subsequent therapies [13,14], which suggests acquired resistance. Nevertheless, the factors causing a loss of response to azacitidine and resistance are only partially known [11,15]. Given that the demethylation effect of 5-azacitidine is nonspecific, there are concerns regarding the potential stimulation of protumoral genes, such as some matrix metalloproteinases [16]. MMPs are a family of zinc-dependent proteolytic enzymes with a key role in cancer, including acute leukemia [17,18]. Through cleavage of a wide variety of substrates, including virtually all components of the extracellular matrix and different cytokines and chemokines, these enzymes regulate migration of cells across the extracellular matrix, cell growth, angiogenesis, and apoptosis. All these processes are essential for the dissemination of neoplastic cells. Interestingly, and similar to the case of solid tumors, expression of MMPs is also dysregulated in MDS [19,20]. Specifically, MMP9, an essential protein that regulates extramedullary disease in acute leukemia, is overexpressed in bone marrow mononuclear cells (MNCs) from MDS patients [21]. In this work, we have studied the effects of azacitidine on MMP9 expression and the consequences for cell invasiveness in two human leukemic cell lines derived from MDS patients. We have investigated the mechanisms that account for these changes. Finally, MMP9 has been evaluated in a cohort of high-risk MDS patients before and during azacitidine therapy until leukemic transformation. Methods Cell lines and culture MOLM-13 and SKM-1, two human cell lines established from patients with acute myeloid leukemia evolved from MDS, were obtained from the DSZM collection (DSZM, Braunschweig, Germany). Cells were cultured in RPMI 1640 medium (GIBCO, Life Technologies, Paisley, UK) containing 10% heat-inactivated fetal bovine serum (FBS), penicillin (100 U/mL), streptomycin 50 (mg/mL), and glutamine (2 mM) and kept in a 5% CO2 incubator at 37
C. Exponentially growing cells were plated at a density of 106 cells/mL in serum-free medium in a 12-well culture plate. Cells were treated with azacitidine at 0.1, 0.2, 0.5, or 1 mmol/L or with solvent alone (phosphate-buffered-saline) as control for 48 hours at 37
C. Quantitative reverse transcriptase polymerase chain reaction analysis RNA was extracted from cultured cells using Trizol reagent (Invitrogen). Complementary DNA (cDNA) was synthesized using 1 mg of total RNA, using the Thermoscript reverse transcriptase polymerase chain reaction (RT-PCR) system with random hexamers, following the manufacturer’s instructions (Invitrogen). Quantitative RT-PCR was carried out in triplicates for each sample

173

using 40 ng of cDNA, TaqMan Universal PCR master mix and 0.5 mL of the specific TaqMan custom gene expression assay for MMP9 (Hs00234579-m1, Applied Biosystems, Foster City, CA, USA). To quantify gene expression, PCR was performed at 95
C for 10 min and at 50
C for 2 min, followed by 40 cycles at 95
C for 15 s and 60
C for 1 min using an Applied Biosystems 7300 real-time PCR system. As an internal control for the amount of template cDNA used, we used the human b-actin endogenous control TaqMan gene expression assay (Applied Biosystems). Relative expression of the analyzed genes was normalized to b-actin using the following formula: the mean values of 2(DCT gene of interest DCT bactin) for at least three different samples. DNA extraction and processing DNA was extracted from cell lines using the Qiamp DNA purification kit (Qiagen, Hilden, Germany). Bisulfite treatment of genomic DNA was performed using the Epitect Plus Bisulfite Conversion Kit (Qiagen) according to the manufacturer’s instructions. Methylation-specific PCR Methylation status of the MMP9 promoter was evaluated by methylation-specific PCR (MSP) using a previously described set of primers [22]: methylated-MMP9: 50 -GAAGTTCGAAATTAGTTTGGTTAAC-30 (sense) and 50 -TCCCGAATAACTAATAT TATAAACGTA-30 (antisense); unmethylated-MMP9: 50 -AGTTTGAAATTAGTTTGGTTAATGT-30 (sense) and 50 -CCTCCCAAATAACTAATATTATAAACATA-30 (antisense). These primers amplify a region from 1654 to 1544 bp, considering the transcription start site as position 0 bp. Samples were incubated for 2 min at 94
C, followed by 40 cycles as follows: 30 sec at 95
C, 30 sec at 56
C, and 30 sec at 72
C, and a final extension at 72
C for 2 min. Products were separated by gel electrophoresis. Relative methylation levels were quantified by densitometry, as the ratio of the density of the methylated band to the total density of the unmethylated and methylated bands. Bisulfite sequencing of DNA Using the Methyl Primer Express version 1 software (Applied Biosystems), we identified a region of 384 bp between 90 and þ294 in the MMP9 gene 5 CpG dinucleotides at positions 36, þ89, þ139, þ196, and þ273. This region was amplified using two bisulfite-sequencing primers (BSP): 50 -TATTTGTTTGTTAAGGAGGGGT-30 (sense) and 50 -CCCCACTACCTAACCCTAAAC-30 (antisense). PCR was performed by mixing bisulfite-treated DNA with 1.5 mmol/L of each primer, 0.2 mmol/L of each dNTP, 1.5 mmol/L MgCl2, 1U/50 mL Taq platinum polymerase and buffer. Samples were incubated for 2 min at 94
C, followed by 37 cycles as follows: 30 sec at 94
C, 30 sec at 58
C, and 30 sec at 72
C, and a final extension at 72
C for 2 min. PCR products were cloned in a Bluescript II SK vector and, after transformation, 8–12 colonies per sample were sequenced. Gelatin zymography For gelatin zymography, 9 mL of MNCs in conditioned medium were loaded on an 8% SDS-polyacrylamide gel containing 0.2% gelatin. An electrophoresis was performed, and the gel was washed twice in 2.5% Triton X-100 and incubated overnight at 37
C in a buffer containing 20 mmol/L Tris/HCl and 5 mmol/L CaCl2 (pH 7.4). Afterward, it was stained using Coomassie Blue

174

T. Bernal et al./ Experimental Hematology 2013;41:172–179

and destained with a mixture of acetic acid and methanol. Gels were scanned, and the intensity of bands was quantified (in arbitrary density units) using Image J software (National Institutes of Health). The experiment was performed in triplicate for each cell line and condition. Apoptosis analysis We studied apoptosis in each cell line treated or not with azacitidine at 0.1 to 1 mmol/L for 48 hours. Apoptosis was quantified using a standard annexin V–propidium iodine assay (Sigma-Aldrich, St. Louis, MO, USA). Briefly, 500 000 cells were washed with phosphate-buffered saline, centrifuged (1200 rpm, 10 min), resuspended in 500 mL of binding buffer, and incubated in the presence of annexin V and propidium iodine for 10 min. Samples were then analyzed in an FC500 flow cytometer (Beckman Coulter, Brea, CA, USA) and studied using RXP software (Beckman Coulter). Viable cells were those negative for both markers. Early apoptotic cells were defined as annexin V–positive and propidium iodide negative, and late apoptotic cells were defined as annexin V–positive and propidium iodide–positive. The experiment was performed in triplicate for each cell line and condition. Invasion assay MOLM-13 cells were cultured alone or in the presence of azacitidine 0.2 mmol/L or azacitidine plus doxycycline 25 mmol/L. We chose 0.2 mmol/L, because it was the concentration that induced a significant increment of MMP9 protein levels in this cell line, as shown in the Results section; 2  105 MOLM-13 cells were resuspended in 500 mL of RPMI and added to each insert coated with extracellular matrix (BD BioCoat Matrigel Invasion Chamber, Becton Dickinson, Franklin Lakes, NJ, USA); and 500 mL of medium containing 10% FBS was placed into lower chambers as a chemoattractant. After 48 h of incubation at 37
C in 5% CO2, cells that crossed the Matrigel-coated inserts were recovered from the lower compartments and counted in a hematocytometer. Leukemic cells that had remained in the upper compartment were also recovered and counted. The percentage of invasion was calculated as the ratio of the number of cells recovered from the lowered compartment to the total number of cells placed in the upper compartment. In addition, the percentage of viable cells in cell suspensions recovered from both compartments was measured by trypan blue staining. Each experiment was performed in triplicate. Patient samples Fresh bone marrow samples were obtained from patients with de novo or secondary high-risk MDS diagnosed at the Hematology Department of the Hospital Universitario Central de Asturias after informed consent. Patients were treated with azacitidine if the IPSS score was 1.5 or higher. Bone marrow aspirate was performed at diagnosis, after six azacitidine cycles, and whenever the physician considered necessary for clinical management. Cytologic MDS diagnosis was made according to the FrenchAmerican-British [23] and World Health Organization classification [24]. The IPSS was used to define prognosis. Clinical response was evaluated after six cycles of azacitidine according to International Working Group criteria [25]. MNCs were isolated by Ficoll-Hypaque density-gradient centrifugation, washed with RPMI medium, and cultured in RPMI without FBS. Seeding

density was kept constant within each patient. After incubation for 48 hours, supernatant was used for zymography. Statistical analysis Data are expressed as mean 6 SEM. Differences between groups were evaluated using a Kruskal-Wallis test. When appropriate, post hoc tests were done using a Mann-Whitney U test and Bonferroni’s correction; p ! 0.05 was considered significant. All p values are two-sided.

Results Azacitidine increases MMP9 expression To evaluate changes in the expression of MMP9 during azacitidine treatment, MOLM-13 and SKM-1 cells were cultured with increasing azacitidine concentrations as described previously, and MMP9 gene expression measured by quantitative RT-PCR. As shown in Figure 1, azacitidine increased MMP9 messenger RNA (mRNA) in a dosedependent fashion in both MOLM-13 (Fig. 1A) and SKM cells (Fig. 1B). These increments were significant with doses greater than 0.2 mmol/L and 1 mmol/L, respectively, for each cell line (p ! 0.05 in post hoc tests). Azacitidine decreases MMP9 methylation levels To determine whether the expression of MMP9 could be epigenetically regulated by azacitidine, MOLM-13 cells were exposed to 1 mmol/L azacitidine and compared with the same cells without treatment. Exposure to azacitidine decreased MMP9 methylation levels as measured by methylation-specific PCR (Fig. 1C, 1D). In addition, bisulfite-sequencing analysis revealed how methylation of the CpG site at 36 bp decreased from 27% to 9% after azacitidine treatment. There were no significant changes in the methylation status of the other CpG sites studied. Azacitidine modulates MMP9 proteolytic activity The effect of azacitidine at the protein level was studied by gelatin zymography in both cell lines cultured in the same conditions and conditioned medium. Figure 2A shows that, in MOLM-13 cells, MMP9 activity was significantly increased at 0.2 mmol/L when compared with baseline. Higher concentrations resulted in a decreased, but not significant, activity of MMP9. In SKM cells, treatment with azacitidine decreased MMP9 activity, but only reached statistical significance at 1 mmol/L (Fig. 2B). Representative zymograms are shown in Figure 2C and 2D. With the aim to explain the differences between MMP9 gene expression and the levels of secreted protein, the effect on apoptosis and cell viability was analyzed after culture in the same conditions. Azacitidine significantly decreased viability of both leukemic cell lines as a result of an increase in apoptosis (Fig. 3A, 3B). Specifically, high doses of azacitidine significantly increased the percentage of cells in late apoptosis (Fig. 3C). Representative flow cytometry plots are shown in Figure 3D.

T. Bernal et al./ Experimental Hematology 2013;41:172–179

Effect of MMP9 activity on invasiveness of leukemic cells To explore whether the effect of increased MMP9 activity could affect the invasive capacity of cells, an invasion assay was performed using MOLM-13 cell line in the presence of azacitidine at 0.2 mmol/L, which was the concentration that produced a significant increment in MMP9 activity. The experiment was performed in the presence or absence of doxycycline as an MMP9 inhibitor. Overall, no increase in the percentage of invading cells was observed (data not shown). However, when only viable cells were taken into account, 0.2 mmol/L azacitidine significantly increased the ratio of invasive-to-noninvasive cells (Fig. 4). MMP9 inhibition with doxycycline reverted this phenomenon. Effect of azacitidine on MMP9 activity in patients To try to confirm the in vitro results, we analyzed MMP9 activity in bone marrow MNCs from patients receiving azacitidine therapy. Characteristics of patients and response to therapy are detailed in Table 1. After six cycles, one patient reached a complete remission, one reached partial remis-

175

sion, and two reached stable diseases. These four patients relapsed after a median of eight cycles. Acute leukemia developed after four cycles in the last patient, and azacitidine therapy was stopped. MMP9 activity significantly increased after six cycles in all patients (Fig. 5). Protein levels after four cycles were more variable, increasing in three patients and decreasing in two; therefore, the average was not different from baseline. Discussion In this work we have studied the effect of azacitidine on MMP9 and its potential consequences on azacitidine resistance in MDS and AML. We have found that azacitidine therapy increases MMP9 gene expression in two different acute leukemia cell lines. This increment was associated with decreased DNA methylation levels in the promoter region and resulted in a higher invasiveness in vitro at the lower dose. Our results in patients also support these findings, as MMP9 levels increased in bone marrow MNCs after six cycles of azacitidine, preceding clinical relapse.

Figure 1. Changes in MMP9 expression induced by azacitidine (Aza) in cultured cell lines MOLM-13 (A) and SKM-1 (B). There was a significant change in MMP9 gene expression (p ! 0.05 in Kruskal-Wallis test in both cell lines). Methylation-specific PCR shows a significant decrease in MMP9 gene methylation levels after azacitidine treatment (C). (D) A representative MSP. Data are means 6 SEM of three experiments. *p ! 0.05 compared with baseline in post hoc tests.

176

T. Bernal et al./ Experimental Hematology 2013;41:172–179

Figure 2. MMP9 activity measured in conditioned medium by gelatin zymography: MOLM-13 (A) and SKM-1 (B). In MOLM-13 cells, MMP9 activity increased at 0.1 and 0.2 mmol/L, and decreased at higher doses (p ! 0.01 in Kruskal-Wallis test). In SKM-1 cells, MMP9 activity significantly decreased at 1 mmol/L. (C and D) Representative zymographs.

We have focused on MMPs because there is evidence that these enzymes have a role in normal and malignant hematopoiesis. MMP9 is a key protein that regulates the stem cell niche by altering the quiescent state of normal hematopoietic precursors to a proliferative state [26]. Regarding malignant hematopoiesis, MMPs have a role similar to that described in solid tumors, and they have an essential role in angiogenesis and metastasis. Leukemic cells disseminate from bone marrow to peripheral blood and may also infiltrate tissues; therefore, acute leukemia could be considered as a model of metastasis [27]. MMP9 is secreted by blast cells of patients with acute myeloid leukemia [28] and is essential for the invasion of extramedullary tissues characteristic of monocytic leukemias [29,30]. There is also evidence that MMP9 is expressed by the MNCs of MDS patients [21], but its role in these entities is not well studied. For example, some authors have found a positive correlation between low plasmatic MMP9 levels at diagnosis and better survival and response to chemotherapy in acute myeloid leukemia [31]. These results contradict those by Travaglino et al. [32], who described a better survival in MDS patients with high MMP9 levels measured by immunocytochemistry. Of note, immunohistochemical methods cannot measure secreted MMP9, but only the cell-bound fraction. In contrast, gelatin zymography is a reliable technique to study the MMP subfamily of gelatinases, because it assesses both levels and activity status of these enzymes [33].

MMPs expression is regulated at the transcriptional level by epigenetic mechanisms among others [18,34]. These epigenetic mechanisms include histone-modifying enzymes, chromatin-remodeling complexes and methylation of CpG sites in the promoter region. All these mechanisms cooperate with each other to maintain a repressed status for some genes. Regarding MMP9, there is substantial evidence supporting the role of methylation in the transcriptional regulation of this gene in diverse cancer and other models. Using methylation-specific PCR analysis, Sato et al. [16] observed an almost complete methylation of CpG sites in the MMP9 gene promoter in five cancer cell lines. These cell lines showed little or no basal expression of MMP9. Azacitidine treatment led to partial demethylation of MMP9 and increased expression of mRNA in four of the five cell lines. Treated cells acquired an invasive phenotype compared with untreated cells as determined in an invasion assay. This invasive phenotype was observed despite a dose-dependent inhibition of growth and cell proliferation. Chicoine et al. [35] observed a high frequency of methylation in the Mmp9 promoter region in two murine lymphoma cell lines not expressing constitutive Mmp9. Conversely, low methylation at CpG sites of the promoter were observed in two other cell lines that produced high levels of MMP9 protein and mRNA. When cells with high levels of promoter methylation were treated with azacitidine, an increase in both Mmp9 mRNA and protein levels was observed, and this was correlated with the hypomethylation of the promoter region. A similar effect was observed in the leukemia cell

T. Bernal et al./ Experimental Hematology 2013;41:172–179

177

Figure 3. Azacitidine induced apoptosis in a dose-dependent fashion. The percentage of viable MOLM-13 (A) and SKM-1 (B) cells decreased after treatment with azacitidine. There was a progressive increase in the percentage of late apoptotic MOLM-13 cells with higher doses of azacitidine (C). (D) Representative flow cytometry plots. Data are means 6 SEM of at least three experiments. *p ! 0.05 compared with baseline in post hoc tests.

line HL-60 [36], although the authors did not study the underlying mechanism. Further evidence exists in noncancer models. Using a primary human aortic smooth muscle cell line, Chen et al. [22], found that DNA methyltransferase-1 is downregulated by miR-29b, which resulted in demethylation of MMP9 promoter and increased mRNA and protein.

Figure 4. Effects of MMP9 expression on invasiveness. MOLM-13 cells were cultured in FBS-free medium. Azacitidine (0.2 mmol/L) and doxycycline (25 mmol/L) were added in parallel experiments (n 5 3 per group). Azacitidine treatment increased the invasive potential of viable cells. This phenomenon was reverted by the treatment with the MMP inhibitor doxycycline. Data are means 6 SEM of three experiments. *p ! 0.05 compared with baseline in post hoc tests.

Our work with two cell lines established from MDS in the leukemic phase provides further evidence of the epigenetic regulation of MMP9 and demonstrates that MMP9 expression enhances with increasing azacitidine doses in the MOLM-13 cell line. The effect of the drug in SKM cells was less pronounced and was observed only at higher doses. Baseline methylation levels of MMP9 promoter between cell lines might account for this difference. Surprisingly, a positive correlation between the increased gene expression induced by the drug and protein activity was not observed. Instead, the highest dose translated into the lower protein activity in both cell lines. These findings could be explained as follows. First, the increased apoptotic rate with higher doses of azacitidine is mainly due to a higher count of cells in the late stage of this process. At this point, cells are unable to synthesize and secrete proteins. This dose-effect relationship on apoptosis may explain the in vitro results and has been previously observed by Khan et al. [37,38] in other cell lines. Nevertheless, this hypothesis does not explain the in vivo results. When a group of patients treated with this drug was studied, we found that MMP9 activity changed throughout subsequent cycles, increasing in all of them after six cycles. In patient 4, MMP9 activity increased even when a clinical complete remission (CR) had been established; in patient 1, MMP9 increased while the patient was categorized as having stable disease. Any of these states define a bone marrow failure where apoptosis is predominant; therefore, we can only speculate that the increase in MMP9 in these

178

T. Bernal et al./ Experimental Hematology 2013;41:172–179

Table 1. Clinical characteristics of patients

Patient 1 2 3 4 5

Dx (FAB/WHO)

Age/Gender

Gender

Response after six cycles

Relapse

Response to salvage chemotherapy

RAEB/RAEB-2 RAEB/RAEB-1 RA/RCMD (complex karyotype) RAEB/RAEB-1(complex karyotype) RAEB-T/AML

68 69 50

Female Male Female

Stable disease Stable disease Progression after 4 cycles

Yes, after 8 cycles Yes, after 8 cycles N/A

No No No

74 87

Female Male

CR PR

Yes Yes

N/A N/A

CR 5 complete remission; N/A 5 not applicable; PR 5 partial remission; RA 5 refractory anemia; RAEB 5 refractory anemia with excess of blasts; RCMD 5 refractory cytopenia with multilineage dysplasia.

patients could be due to the residual neoplastic cells in which the MMP9 gene is maximally expressed. Second, we cannot rule out a role of microRNAs, which could interfere the translation of the MMP9 transcript into protein. Accordingly, it has been described that microRNAs are subjected to epigenetic regulation [39], and azacitidine can increase their expression [40]. Unfortunately, our results cannot clarify the mechanisms responsible for this result. Another limitation of the study is that the number of patients studied was low. Nevertheless, we think this sample fairly reflects the pattern of response to azacitidine

with a global response rate of 50%, a low complete response, and a 100% relapse rate with an aggressive behavior of disease. We are also aware that other mechanisms can contribute to azacitidine resistance [11,15], but we think they are not exclusive and our results might explain some aspects of the disease, especially the high resistance to therapy and aggressiveness once relapse has occurred. In this regard, we observed increased activity in MMP9 activity in three patients who were resistant to further chemotherapy, and all but one died of progressive disease. Two patients developed extramedullary disease, which might be explained by our in vitro results demonstrating that activation of MMP9 is associated with an aggressive cell behavior measured in the invasion assay. Despite these limitations, the present work shows that MMP9 expression and cell invasiveness increase during azacitidine therapy, suggesting that this gelatinase could be involved in the development of resistance to azacitidine and further therapy. Therefore, the role of MMP9 in the therapeutic management of MDS [41,42] or as a monitoring tool should be investigated in more detail. Acknowledgments This work was supported by a grant from Ministerio de Ciencia e Innovaci on-Spain.

Conflict of interest disclosure No financial interest or relationships with financial interest relating to the topic of this article have been declared.

References Figure 5. Azacitidine increased MMP9 activity in high-risk myelodysplastic patients. (A) MNCs of patients were isolated after the fourth and sixth cycle of azacitidine and MMP9 activity was measured in conditioned medium by gelatin zymography. MMP9 activity significantly increased after the sixth cycle. (B) Representative zymography of patient 1 showing increased MMP9 activity after the fourth and sixth cycles. Response was classified as stable disease after the sixth cycle of treatment according to International Working Group criteria. Disease progression was documented after eight cycles. *p ! 0.05 compared with baseline.

1. Greenberg P, Cox C, LeBeau MM, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood. 1997; 89:2079–2088. 2. Nimer SD. Myelodysplastic syndromes. Blood. 2008;111:4841–4851. 3. Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al. Efficacy of azacitidine compared with that of conventional care regimens in the treatment of higher-risk myelodysplastic syndromes: a randomised, open-label, phase iii study. Lancet Oncol. 2009;10:223–232. 4. Jiang Y, Dunbar A, Gondek LP, et al. Aberrant DNA methylation is a dominant mechanism in MDS progression to AML. Blood. 2009; 113:1315–1325.

T. Bernal et al./ Experimental Hematology 2013;41:172–179 5. Quesnel B, Guillerm G, Vereecque R, et al. Methylation of the p15(ink4b) gene in myelodysplastic syndromes is frequent and acquired during disease progression. Blood. 1998;91:2985–2990. 6. Follo MY, Finelli C, Mongiorgi S, et al. Reduction of phosphoinositide-phospholipase C beta1 methylation predicts the responsiveness to azacitidine in high-risk MDS. Proc Natl Acad Sci U S A. 2009;106:16811–16816. 7. Fandy TE, Herman JG, Kerns P, et al. Early epigenetic changes and DNA damage do not predict clinical response in an overlapping schedule of 5-azacytidine and entinostat in patients with myeloid malignancies. Blood. 2009;114:2764–2773. 8. Braun T, Carvalho G, Coquelle A, et al. Nf-kappab constitutes a potential therapeutic target in high-risk myelodysplastic syndrome. Blood. 2006;107:1156–1165. 9. Fabre C, Grosjean J, Tailler M, et al. A novel effect of DNA methyltransferase and histone deacetylase inhibitors: Nfkappab inhibition in malignant myeloblasts. Cell Cycle. 2008;7:2139–2145. 10. Komashko VM, Farnham PJ. 5-azacytidine treatment reorganizes genomic histone modification patterns. Epigenetics. 2010;5:3. 11. Simo-Riudalbas L, Melo SA, Esteller M. Dnmt3b gene amplification predicts resistance to DNA demethylating drugs. Genes Chromosomes Cancer. 2011;50:527–534. 12. Jabbour E, Garcia-Manero G, Batty N, et al. Outcome of patients with myelodysplastic syndrome after failure of decitabine therapy. Cancer. 2010;116:3830–3834. 13. Prebet T, Gore SD, Esterni B, et al. Outcome of high-risk myelodysplastic syndrome after azacitidine treatment failure. J Clin Oncol. 2011;29:3322–3327. 14. Faderl S, Garcia-Manero G, Estrov Z, et al. Oral clofarabine in the treatment of patients with higher-risk myelodysplastic syndrome. J Clin Oncol. 2011;28:2755–2760. 15. Qin T, Castoro R, El Ahdab S, et al. Mechanisms of resistance to decitabine in the myelodysplastic syndrome. PloS One. 2011;6:e23372. 16. Sato N, Maehara N, Su GH, Goggins M. Effects of 5-aza-2’-deoxycytidine on matrix metalloproteinase expression and pancreatic cancer cell invasiveness. J Natl Cancer Inst. 2003;95:327–330. 17. Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression. Nat Rev Cancer. 2002;2:161–174. 18. Fanjul-Fernandez M, Folgueras AR, Cabrera S, Lopez-Otin C. Matrix metalloproteinases: evolution, gene regulation and functional analysis in mouse models. Biochim Biophys Acta. 2010;1803:3–19. 19. Moncada-Pazos A, Obaya AJ, Fraga MF, et al. The adamts12 metalloprotease gene is epigenetically silenced in tumor cells and transcriptionally activated in the stroma during progression of colon cancer. J Cell Sci. 2009;122:2906–2913. 20. Viloria CG, Obaya AJ, Moncada-Pazos A, et al. Genetic inactivation of adamts15 metalloprotease in human colorectal cancer. Cancer Res. 2009;69:4926–4934. 21. Ries C, Loher F, Zang C, Ismair MG, Petrides PE. Matrix metalloproteinase production by bone marrow mononuclear cells from normal individuals and patients with acute and chronic myeloid leukemia or myelodysplastic syndromes. Clin Cancer Res. 1999;5:1115–1124. 22. Chen KC, Wang YS, Hu CY, et al. Oxldl up-regulates microrna-29b, leading to epigenetic modifications of mmp-2/mmp-9 genes: a novel mechanism for cardiovascular diseases. FASEB J. 2011;25:1718–1728. 23. Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the acute leukaemias. French-American-British (FAB) co-operative group. Br J Haematol. 1976;33:451–458. 24. Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO) classification of the myeloid neoplasms. Blood. 2002;100: 2292–2302.

179

25. Cheson BD, Greenberg PL, Bennett JM, et al. Clinical application and proposal for modification of the International Working Group (IWG) response criteria in myelodysplasia. Blood. 2006;108:419– 425. 26. Heissig B, Hattori K, Dias S, et al. Recruitment of stem and progenitor cells from the bone marrow niche requires mmp-9 mediated release of kit-ligand. Cell. 2002;109:625–637. 27. Hatfield KJ, Reikvam H, Bruserud O. The crosstalk between the matrix metalloprotease system and the chemokine network in acute myeloid leukemia. Curr Med Chem. 2010;17:4448–4461. 28. Janowska-Wieczorek A, Marquez LA, Matsuzaki A, et al. Expression of matrix metalloproteinases (mmp-2 and -9) and tissue inhibitors of metalloproteinases (timp-1 and -2) in acute myelogenous leukaemia blasts: comparison with normal bone marrow cells. Br J Haematol. 1999;105:402–411. 29. Paupert J, Mansat-De Mas V, Demur C, Salles B, Muller C. Cell-surface mmp-9 regulates the invasive capacity of leukemia blast cells with monocytic features. Cell Cycle. 2008;7:1047– 1053. 30. Stefanidakis M, Karjalainen K, Jaalouk DE, et al. Role of leukemia cell invadosome in extramedullary infiltration. Blood. 2009;114: 3008–3017. 31. Lin LI, Lin DT, Chang CJ, Lee CY, Tang JL, Tien HF. Marrow matrix metalloproteinases (mmps) and tissue inhibitors of mmp in acute leukaemia: potential role of mmp-9 as a surrogate marker to monitor leukaemic status in patients with acute myelogenous leukaemia. Br J Haematol. 2002;117:835–841. 32. Travaglino E, Benatti C, Malcovati L, et al. Biological and clinical relevance of matrix metalloproteinases 2 and 9 in acute myeloid leukaemias and myelodysplastic syndromes. Eur J Haematol. 2008; 80:216–226. 33. Snoek-van Beurden PA, Von den Hoff JW. Zymographic techniques for the analysis of matrix metalloproteinases and their inhibitors. Biotechniques. 2005;38:73–83. 34. Yan C, Boyd DD. Regulation of matrix metalloproteinase gene expression. J Cell Physiol. 2007;211:19–26. 35. Chicoine E, Esteve PO, Robledo O, Van Themsche C, Potworowski EF, St-Pierre Y. Evidence for the role of promoter methylation in the regulation of mmp-9 gene expression. Biochem Biophys Res Commun. 2002;297:765–772. 36. Liu HB, Voso MT, Gumiero D, Duong J, McKendrick JJ, Dear AE. The anti-leukemic effect of a novel histone deacetylase inhibitor mct-1 and 5-aza-cytidine involves augmentation of nur77 and inhibition of mmp-9 expression. Int J Oncol. 2009; 34:573–579. 37. Khan R, Aggerholm A, Hokland P, Hassan M, Hellstrom-Lindberg E. A pharmacodynamic study of 5-azacytidine in the p39 cell line. Exp Hematol. 2006;34:35–43. 38. Khan R, Schmidt-Mende J, Karimi M, et al. Hypomethylation and apoptosis in 5-azacytidine-treated myeloid cells. Exp Hematol. 2008;36:149–157. 39. Choudhry H, Catto JW. Epigenetic regulation of microrna expression in cancer. Methods Mol Biol. 2011;676:165–184. 40. Wong KY, So CC, Loong F, et al. Epigenetic inactivation of the mir-124-1 in haematological malignancies. PloS One. 2011;6: e19027. 41. Overall CM, Lopez-Otin C. Strategies for mmp inhibition in cancer: innovations for the post-trial era. Nat Rev Cancer. 2002; 2:657–672. 42. Lopez-Otin C, Hunter T. The regulatory crosstalk between kinases and proteases in cancer. Nat Rev Cancer. 2010;10:278–292.