Proteomic Analysis Reveals Autophagy as Pro-Survival Pathway Elicited by Long-Term Exposure with 5-Azacitidine in High-Risk Myelodysplasia

Abstract

Azacytidine (5-AZA) is the standard first-choice treatment for high-risk myelodysplasia (MDS) patients. However, the clinical outcome for those patients who interrupt treatment or whose disease failed to respond is very poor. In order to identify the cellular pathways that are modified by long-term exposure to 5-AZA, we evaluated key proteins associated with the autophagy pathway by reverse-phase microarray (RPPA). Comparing bone marrow mononucleated cells (BMMCs) obtained from 20 newly-diagnosed patients and after four 5-AZA cycles we found an increased autophagy signaling. We then evaluated ex-vivo the effect of the combination of 5-AZA with autophagy inhibitors chloroquine (CQ) and leupeptin. Since 5-AZA and CQ showed synergism due to an increase of basal autophagy after 5-AZA exposure, we adopted a sequential treatment treating BMMCs with 5 μM 5-AZA for 72 h followed by 10 μM CQ for 24 h and found increased apoptosis, associated to a reduction of G2M phase and increase in G0-G1 phase. Long-term exposure to 5-AZA induced the reduction of the autophagic marker SQSTM1/p62, reversible by CQ or leupeptin exposure. In conclusion, we identified autophagy as a compensatory pathway occurring in MDS-BM after long-term exposure to 5-AZA and we provided evidences that a sequential treatment of 5-AZA followed by CQ could improve 5-AZA efficacy, providing novel insight for tailored therapy in MDS patients progressing after 5-AZA therapy.

Keywords: autophagy; azacytidine; myelodysplastic syndrome.

Figure 1

Proteomic profile of MDS-BMMCs after long-term exposure to 5-AZA in vivo. (A–R) Proteins obtained from BMMCs isolated from patients at diagnosis (gray bars) and after treatment (black bars) with 4 cycles of 5-AZA were analyzed using RPPA to as described in the Materials and Methods section. Results represent the mean of 20 observations in triplicate; error bars denote standard deviation. ^*p < 0.05, ^**p < 0.001, ^***p < 0.0001 (Mann-Whitney test). The label indicates the endpoint tested (A–C autophagy pathway, D–F proliferation, G–R intracellular signaling).

Figure 2

Effects of autophagy inhibition and 5-AZA treatment in MDS-BMMCs in vitro. Cell viability was assessed after 72 h of treatment with AZA or CQ or their combination (10 μM) using ATP-lite assay, showing a synergic effect (A). Pre-treatment with CQ (10 μM) for 24 h and addition of 10 μM AZA for 72 h did not exhibit a synergic effect of the two drugs (B). Standard deviation was determined from 8 independent experiments, including triplicate wells per experiment.

Figure 3

Effects of sequential treatment with 5-AZA followed by CQ on MDS-BMMCs in vitro. BMMCs were treated with 5 μM 5-AZA for 72 h, washed in PBS and exposed to 10 μM CQ (A), 10 μM leupeptin for further 24 h (B). When cells were treated with CQ after 72 h of treatment with 5-AZA, a change in cell cycle was appreciated (C). The effect on the autophagic marker SQSTM1/p62 after 72 h of treatment with 5-AZA followed or not by 10 μM CQ or leupeptin was evaluated by WB (D,E). Standard deviation was determined from 6 independent experiments, including triplicate wells per experiment.