Doxorubicin upregulates CXCR4 via miR-200c/ZEB1-dependent mechanism in human cardiac mesenchymal progenitor cells

Abstract

Doxorubicin (DOXO) treatment is limited by its cardiotoxicity, since it causes cardiac-progenitor-cell depletion. Although the cardioprotective role of the stromal cell-derived factor-1/C-X-C chemokine receptor type 4 (SDF1/CXCR4) axis is well established, its involvement during DOXO-induced cardiotoxicity has never been investigated. We showed that in a mouse model of DOXO-induced cardiomyopathy, CXCR4⁺ cells were increased in response to DOXO, mainly in human cardiac mesenchymal progenitor cells (CmPC), a subpopulation with regenerative potential. Our in vitro results showed a CXCR4 induction after 24 h of DOXO exposure in CmPC. SDF1 administration protected from DOXO-induced cell death and promoted CmPC migration. CXCR4 promoter analysis revealed zinc finger E-box binding homeobox 1 (ZEB1) binding sites. Upon DOXO treatment, ZEB1 binding decreased and RNA-polymerase-II increased, suggesting a DOXO-mediated transcriptional increase in CXCR4. Indeed, DOXO induced the upregulation of miR-200c, that directly targets ZEB1. SDF1 administration in DOXO-treated mice partially reverted the adverse remodeling, decreasing left ventricular (LV) end diastolic volume, LV ejection fraction and LV anterior wall thickness in diastole, recovering LV end systolic pressure and reducing±dP/dt. Moreover, in vivo administration of SDF1 partially reverted DOXO-induced miR-200c and p53 protein upregulation in mouse hearts. In addition, downmodulation of ZEB1 mRNA and protein by DOXO was significantly increased by SDF1. In keeping, p21 mRNA, that is induced by p53 and inhibited by ZEB1, is induced by DOXO treatment and is decreased by SDF1 administration. This study showed new players of the DOXO-induced cardiotoxicity, that can be exploited to ameliorate DOXO-associated cardiomyopathy.

Figure 1

CXCR4 is increased in vivo in response to DOXO. (a) CXCR4⁺ cell quantitative analysis. The number of CXCR4⁺ cells from 9 randomly selected fields was counted for each tissue section of DOXO and untreated mice (CTR). The CXCR4⁺ cells were calculated as a percentage on the total number of nuclei in the same fields (9 different fields per mouse). The number of CXCR4⁺ cells were significantly higher in DOXO compared to CTR mice (**P<0.005). Data were representative of five independent experiments. Results are presented as mean±S.E.M. (b) Representative images of heart sections of CTR and DOXO mice, immunostained with CXCR4 (green) and CD44 (red) antibody. Nuclei are stained with Hoechst 33258 (blue). Scale bar: 20 μm and 50 μm in the inset

Figure 2

CXCR4 is induced in response to DOXO in CmPC. (a) qPCR analysis showing CXCR4 upregulation in CmPC treated with 1 μM DOXO for 24 h as compared to untreated cells (n=3, *P<0.05) Results are presented as mean±S.D.; (b) Representative western blot showing an increase in CXCR4 expression in CmPC upon DOXO treatment. β-actin was used as a loading control. Actin indicate β-actin; (c) Expression levels of CXCR4 protein were evaluated by densitometric analysis and normalized by α-tubulin protein levels (n=3, *P<0.05)

Figure 3

SDF1 protects and induces migration of DOXO treated CmPC. (a) Cell death analysis of CmPC upon DOXO treatment for increasing periods of time starting from 6 h up to 48 h as indicated. Cell death increased significantly 48 h after DOXO treatment (n=4, ***P<0.001). (b) Cell death analysis of CmPC exposed or not to 1 μM DOXO. DOXO was added to CmPC for 24 h and then removed. SDF1, AMD3100 and α-CXCR4 were added after 24 h at the following dosages: SDF-1 at 100 ng/ml, AMD3100 at 3.2 μg/ml and α-CXCR4 at 10 μg/ml. Cell death was determined at 48 h. Mean values±SEM of three independent experiments run in triplicate (*P<0.05; ***P<0.0001). (c) Chemotatic responses of CmPC exposed or not to 1 μM DOXO for 24 h in response to 100 ng/ml SDF1 or 10% FBS. Migration efficiency of CmPC was estimated using a Transwell assay. After 16 h of incubation transmigrated cells were quantified using crystal violet. The results are expressed as fold change of the untreated control cells exposed or not to DOXO. DOXO—cells treated with 1 μM DOXO for 24 h; AMD3100—serum starved CmPC treated AMD3100 25 μg/ml during migration; DOXO+AMD3100—cells treated with 1 μM DOXO for 24 h and AMD3100 25 μg/ml during migration (n=6, *P<0.05; **P<0.005; ***P<0.0001)

Figure 4

DOXO treatment upregulates miR-200c and downregulates ZEB1. (a) miR-200c expression upon 24 h DOXO treatment. miR-200c was significantly induced by DOXO treatment (n=4, *P<0.05). Results are presented as mean±S.E.M. (b) ZEB1 mRNA expression was quantified by qPCR; ZEB1 mRNA decreased upon DOXO treatment at 24 h (n=4, **P<0.01). Results are presented as mean±S.E.M. (c) CmPC were treated with 1 μM DOXO for 24 h. ZEB1 and p53 proteins were evaluated by Western blot analysis. A representative western blot showing DOXO treatment induced p53 protein expression and ZEB1 downregulation. β-actin was used as a loading control. Actin indicates β-actin (d) Expression levels of ZEB1 and p53 protein were evaluated by densitometric analysis and normalized by β-actin protein levels (n=3, *P<0.05)

Figure 5

ZEB1 knockdown elicits CXCR4 increase. CmPC were infected either with the lentivirus carrying ZEB1-specific shRNA or with the control virus. After 24 h, cells were selected with puromycin. (a) Representative western blot demonstrating a 60% knockdown of ZEB1 expression in CmPC infected with a lentivirus encoding a ZEB1-specific shRNA sequence. (b) Expression levels of ZEB1 protein were evaluated by densitometric analysis and normalized by β-actin protein levels (n=3, *P<0.02). (c) CmPC transduced with a ZEB1-specific shRNA showed an upregulation of CXCR4 mRNA compared with control. (d) CmPC transduced with a ZEB1-specific shRNA displayed CXCR4 protein expression on the cell surface of CmPC, and increased both in absence of DOXO (upper and bottom left panels) and in presence of DOXO (upper and bottom right panels)

Figure 6

CXCR4 is a ZEB1 target gene. (a) Schematic figure showing the E-box binding sites of ZEB1 in the promoter and in the intronic region of the human CXCR4 gene. Five E-box sites are present at position – 2029 bp, −1644 bp, −983 bp, −554 bp, −262 bp upstream the first exon and four in the intronic region, at position +881 bp, +916 bp, +1238 bp, +1454 bp. (b) ZEB1 consensus sites revealed that the E-box sites at +881 bp and +916 bp are highly conserved across species of humans, mice, rats, rabbits, bos taurus and pan troglodytes. (c) ChIP assay of CmPC in the absence or presence of 1 μM DOXO treatment for 24 h was performed with a ZEB1 antibody, followed by quantitative real-time PCR (ChIP-qPCR), using specific primers encompassing ZEB1 consensus sequences. (d) ChIP-qPCR CmPC in the absence or presence of 1 μM DOXO treatment for 24 h was performed with a Pol II-specific antibody

Figure 7

SDF1 partially rescues cardiac dysfunction induced by DOXO. Echocardiographic examination was performed on mice before (Day 0) and 21 days after the administration of DOXO (n=13), DOXO+SDF1 (n=13) or saline (n=13); (a) LV end-systolic volume (LVESV), (b) LV end-diastolic volume (LVEDV), (c) LV ejection fraction (LVEF), (d) LV anterior wall thickness at diastole (LVAWd). Evaluation of LV hemodynamic function with a Millar micro-tip catheter was performed at day 21; (e) LV systolic pressure (LVSP), (f) maximal rate of pressure development (+dP/dt) and maximal rate of pressure relaxation (-dP/dt). Saline, n=10; DOXO, n=10; DOXO+SDF1, n=13. Results are presented as mean±SD. (*P<0.05; **P<0.01; ***P<0.001)

Figure 8

SDF1 partially rescues DOXO-dependent cardiac dysfunction via a miR-200c/ZEB1/p53 pathway modulation. Mice were treated with DOXO, DOXO+SDF1 or saline for 21 days. (a) Then mRNA was extracted from LV and analyzed for miR-200c, p53, ZEB1 and p21 mRNA expression levels. p53 mRNA was not modulated either by DOXO or DOXO+SDF1. miR-200c, and p21 mRNA were induced by DOXO and were all significantly decreased by SDF1 treatment. ZEB1 mRNA was downregulated by DOXO and returned to control levels by SDF1 treatment (Saline, n=5; DOXO, n=5; DOXO+SDF1, n=5; *P<0.05; **P<0.01; ***P<0.001). (b) Representative western blot with anti-p53 and ZEB1 antibodies showed that p53 upregulation by DOXO was decreased by SDF1 treatment and ZEB1 demise was reverted by SDF1 (Saline, n=5; DOXO, n=5; DOXO+SDF1, n=5). (c) Expression levels of p53 and ZEB1 protein were evaluated by densitometric analysis and normalized by β-actin protein levels (n=5; *P<0.05; ***P<0.001)