Author + information
- Received May 25, 2013
- Revision received August 1, 2013
- Accepted August 26, 2013
- Published online March 18, 2014.
- Otmar Pfister, MD∗,†,
- Vera Lorenz, MSc∗,
- Angelos Oikonomopoulos, PhD‡,
- Lifen Xu, PhD∗,
- Stéphanie P. Häuselmann, MSc∗,
- Christopher Mbah, MD, MA‡,
- Beat A. Kaufmann, MD∗,†,
- Ronglih Liao, PhD‡,
- Aleksandra Wodnar-Filipowicz, PhD∗ and
- Gabriela M. Kuster, MD∗,†∗ ()
- ∗Department of Biomedicine, University Hospital Basel and University of Basel, Basel, Switzerland
- †Division of Cardiology, University Hospital Basel, Basel, Switzerland
- ‡Cardiovascular Division, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts
- ↵∗Reprint requests and correspondence:
Dr. Gabriela M. Kuster, Myocardial Research, Department of Biomedicine and Division of Cardiology, University Hospital Basel, Hebelstrasse 20, 4031 Basel, Switzerland.
Objectives The goal of this study was to define the role of FMS-like tyrosine kinase 3 (FLT3) in the heart.
Background FLT3 is a prominent target of receptor tyrosine kinase inhibitors (TKIs) used for anticancer therapy. TKIs can cause cardiomyopathy but understanding of the mechanisms is incomplete, partly because the roles of specific TKI target receptors in the heart are still obscure.
Methods Myocardial infarction was induced in mice by permanent ligation of the left anterior descending coronary artery followed by intramyocardial injection of FLT3 ligand (FL) or vehicle into the infarct border zone. Cardiac morphology and function were assessed by echocardiography and histological analysis 1 week after infarction. In addition, FLT3 expression and regulation, as well as molecular mechanisms of FLT3 action, were examined in cardiomyocytes in vitro.
Results The intramyocardial injection of FL into the infarct border zone decreased infarct size and ameliorated post-myocardial infarction remodeling and function in mice. This beneficial effect was associated with reduced apoptosis, including myocytes in the infarct border zone. Cardiomyocytes expressed functional FLT3, and FLT3 messenger ribonucleic acid and protein were up-regulated under oxidative stress, identifying cardiomyocytes as FL target cells. FLT3 activation with FL protected cardiomyocytes from oxidative stress–induced apoptosis via an Akt-dependent mechanism involving Bcl-2 family protein regulation and inhibition of the mitochondrial death pathway.
Conclusions FLT3 is a cytoprotective system in the heart and a potential therapeutic target in ischemic cardiac injury. The protective mechanisms uncovered here may be further explored in view of potential cardiotoxic effects of FLT3-targeting anticancer therapy, particularly in patients with ischemic heart disease.
- heart failure
- hematopoietic cytokines
- myocardial infarction
- receptor tyrosine kinase
FMS-like tyrosine kinase 3 (FLT3) ligand (FL) belongs to the family of early-acting hematopoietic cytokines, which signal through class III tyrosine kinase receptors. FLT3 is expressed on the surface of hematopoietic progenitor cells residing adjacently to FL-expressing stromal fibroblasts in the hematopoietic niche of the bone marrow (1). Upon FLT3 triggering, signal transduction pathways involving STAT5, Ras-Raf-MEK-ERK, and phosphatidylinositol 3 kinase (PI3K)/Akt are activated (2), which promote proliferation (3) and survival (4) of hematopoietic progenitor cells.
In addition to the hematopoietic system, FLT3 (5) and FL (6) are also expressed in nonhematopoietic organs. Importantly, FLT3 gene and FL gene and protein expression have been identified in whole heart homogenates (5–8), but the functional relevance of this expression is unknown. Because gene expression of FLT3 and FL are increased after myocardial infarction (8), we hypothesized that FLT3/FL play a role in the ischemic myocardium. It is now well established that hematopoietic growth factors participate in the infarct-related inflammatory response and may contribute to the preservation and regeneration of myocardial tissue. Granulocyte-colony stimulating factor (G-CSF) (9), erythropoietin (EPO) (10), and stem cell factor (SCF) (8) are markedly increased after myocardial infarction, and their receptors have been identified on cardiomyocytes (G-CSFR; EPOR) (11,12), endothelial cells (G-CSFR; EPOR) (13,14), and cardiac progenitor cells (SCF receptor c-kit) (15). Notably, a direct cytoprotective effect has been demonstrated for G-CSF, which decreases infarct size and attenuates cardiac dysfunction via activation of G-CSFR–coupled pro-survival signaling in cardiomyocytes (11). However, whether cardiomyocytes express FLT3 is not known, and the functional relevance of FLT3/FL regulation in the infarcted heart remains to be elucidated.
FLT3 is part of the so-called “cancer kinome.” Because of its high expression levels and the activity-enhancing mutations associated with acute myelogenous leukemia, FLT3 has become a prominent drug target. Several receptor tyrosine kinase inhibitors (TKIs) targeting FLT3 exist, some of which are already in clinical use (16). Recent evidence suggests that TKIs inhibiting multiple receptors, including FLT3, impair cardiac function (17–19). However, the mechanisms of TKI-associated cardiotoxicity are not well understood, and identification of specific TKI target receptors in the heart and knowledge of their roles are incomplete. In particular, to date, FLT3 was not considered to play a role in the heart and, whereas experimental evidence predicts cardiotoxic effects of inhibition for numerous kinases in the heart (20), no data exist for FLT3. In the present study, using a mouse model of myocardial infarction and cultured primary cardiomyocytes, we demonstrate that FLT3 exerts cardioprotective effects in the ischemic heart.
A detailed Methods section is available in the Online Appendix.
Myocardial infarction and intramyocardial FL injection
Myocardial infarction was induced in female C57BL/6J mice (age 8 to 10 weeks; N = 30) by permanent ligation of the left anterior descending (LAD) coronary artery as previously described (22). Mice were randomly assigned to recombinant mouse FL (MI-FL; n = 15) or vehicle (MI-V; n = 15) to be injected into the infarct border zone immediately after LAD ligation. Sham-operated mice (n = 4) served as controls. Three animals died during surgery, 2 of which were assigned to MI-FL and 1 to MI-V, respectively. The Online Appendix presents the technical details on animal surgery and drug injection.
Echocardiography was performed 1 day before (baseline) and 1 week after myocardial infarction induction by using a Vevo 2100 high-resolution, small animal digital ultrasound system (VisualSonics Inc., Toronto, Ontario, Canada) equipped with a linear-array transducer operating at a centerline frequency of 30 MHz. The procedural details are presented in the Online Appendix.
Data are presented as mean ± SEM. Statistical analyses were performed with GraphPad Prism software version 6 (GraphPad Prism Software Inc., La Jolla, California). The unpaired, 2-tailed Student t test was used for comparison of 2 groups, 1-way analysis of variance followed by Newman-Keuls posttesting for comparison of all means of 3 groups, or Bonferroni for comparison of selected, prespecified pairs of columns of ≥4 groups. A p value <0.05 was considered statistically significant.
All animal procedures and handling were in accordance with the guidelines of Harvard Medical School, the Longwood Medical Area Institutional Animal Care and Use Committee, and the National Society for Medical Research.
FL improves post-myocardial infarction remodeling and function in mice
We first examined whether activation of FLT3 affects myocardial remodeling and function after myocardial infarction in mice. From the 27 mice who survived surgery, 1 died at day 6 (MI-V) and 1 at day 7 (MI-FL) post-myocardial infarction, leaving 13 (MI-V) and 12 (MI-FL) mice for final analysis, respectively. Echocardiography at baseline showed no differences in terms of wall thickness and left ventricular dimensions and function between groups (Online Table 1). One week after myocardial infarction, left ventricular end-diastolic and end-systolic diameters were significantly increased in both myocardial infarction groups compared with the sham-operated group (Table 1). In addition, systolic wall thickness and systolic function were decreased; this outcome was more pronounced in the MI-V group, in which left ventricular ejection fraction was significantly lower compared with the MI-FL group. Myocardial remodeling was further examined by using histological analysis. Infarct size was significantly reduced in MI-FL hearts compared with MI-V hearts (Figs. 1A and 1B). The beneficial effect of FL on infarct size was associated with a significantly lower amount of apoptotic nuclei in the infarct border zone of MI-FL–treated mice compared with MI-V–treated mice (Figs. 1C and 1D). These histological findings are consistent with the echocardiographically documented decrease in total wall thickness in MI-V hearts as opposed to MI-FL hearts, in which total wall thickness was maintained compared with baseline (Fig. 1E). The preservation of total wall thickness in MI-FL hearts was associated with an increase in left ventricular end-diastolic diameter comparable to MI-V hearts (p = NS; data not shown). Taken together, these findings imply increased cellularity of the infarcted MI-FL–treated hearts compared with the untreated (MI-V) hearts, which may be due, at least in part, to improved cell survival in the infarct border zone.
FLT3 is expressed by cardiomyocytes and up-regulated under oxidative stress
Cardiomyocytes account for >90% of the myocardial mass and represent the structural substrate of the contractile function of the heart. Because FL preserved myocardial tissue and improved postinfarct function, we sought to determine whether cardiomyocytes express FLT3 and qualify as FL target cells. Isolated primary cardiomyocytes were cultured, and FLT3 gene and protein expression were examined. FLT3 gene was robustly expressed in whole heart homogenates from neonatal and adult rats and in cultured cardiomyocytes (Fig. 2A). Using quantitative real-time reverse transcription polymerase chain reaction, we found that FLT3 messenger ribonucleic acid (mRNA) was time dependently up-regulated in response to endogenous (hypoxia) (Fig. 2B) or exogenous (hydrogen peroxide [H2O2]) (Fig. 2C) oxidative stress. We further assessed FLT3 protein expression in cardiomyocytes by flow cytometry using a phycoerythrin-conjugated monoclonal antibody against FLT3. These experiments demonstrated solid immunoreactivity for FLT3 protein, which was absent in the cardiomyocytes incubated with isotype control antibody, thus supporting FLT3 expression in cardiomyocytes. Similar to FLT3 mRNA, oxidative stress also increased the amount of FLT3 protein after 24 h, which was most pronounced in cardiomyocytes exposed to hypoxia (Figs. 2D and 2E). Our findings are consistent with the increase in FLT3 mRNA described in the post-myocardial infarction mouse heart (8) and substantiate the presence of FLT3 in cardiomyocytes. Its up-regulation under oxidative stress implies a specific role of FLT3/FL in the ischemic or otherwise injured heart.
FLT3 stimulation by recombinant FL induces pro-survival signaling in cardiomyocytes
FLT3 activation in early hematopoietic progenitor cells induces Akt and ERK signaling (2), which improve cell survival by inhibiting apoptosis (16). Our goal was to determine if similar pro-survival pathways are activated in cardiomyocytes in response to FL. Stimulation of cardiomyocytes with recombinant mouse FL (mFL) induced phosphorylation of MEK and ERK, which was transient and peaked at 5 min (Figs. 3A to 3C). FL also led to a significant time- and dose-dependent increase in Akt-phosphorylation (Figs. 3D and 3E), which was sustained over 90 min. FL-induced Akt-phosphorylation was FLT3 mediated because it was attenuated in cardiomyocytes, in which FLT3 was down-regulated by FLT3 targeting small interfering ribonucleic acid (Online Fig. 1). These findings support the expression of functional FLT3 on cardiomyocytes and the FL-induced activation of Akt- and transient MEK/ERK signaling.
FL protects cardiomyocytes from oxidative stress–induced apoptosis by inhibiting the mitochondrial death pathway
Activation of FLT3 improves survival of hematopoietic progenitor cells (4). We therefore tested whether FL exerts antiapoptotic effects in cardiomyocytes under oxidative stress. Exposure to H2O2 significantly increased the amount of TUNEL-positive cardiomyocytes, and this increase was inhibited in the presence of FL (Figs. 4A and 4B). To examine the molecular mechanism of this antiapoptotic effect, specifically the role of the mitochondria, expression of the pro-apoptotic Bax and the antiapoptotic Bcl-2, cytochrome c, and cleaved caspase-3 were measured. H2O2 increased Bax while decreasing Bcl-2 expression, and these H2O2-evoked changes were inhibited in cardiomyocytes pre-treated with FL (Fig. 4C). Similarly, H2O2 enhanced cytosolic cytochrome c and increased cleaved caspase-3, which were markedly inhibited by FL (Figs. 4D to 4F).
Akt has been shown to prevent apoptosis via inhibition of mitochondrial cytochrome c release (23). Because FL produced sustained Akt-phosphorylation in cardiomyocytes, we tested whether Akt mediates the antiapoptotic effect of FL by using the Akt inhibitor VIII (Akti). Akti completely suppressed basal and FL-induced Akt-phosphorylation, whereas it accentuated both basal and FL-induced ERK phosphorylation (Fig. 5A). Consistent with our hypothesis, FL failed to prevent the H2O2-associated release of cytochrome c in the presence of Akti (Fig. 5B), suggesting that Akt is necessary for the FL-induced cytoprotective effect. Together, our in vitro findings identify cardiomyocytes as FL target cells and demonstrate that FL protects cardiomyocytes from oxidative stress–induced apoptosis via prevention of the Bax/Bcl-2 increase and Akt-dependent inhibition of cytochrome c release.
This report provides the first evidence that FLT3 represents an intrinsic cytoprotective system in the heart. We demonstrated that cardiomyocytes express FLT3 and serve as FL target cells. In cultured cardiomyocytes exposed to oxidative stress, FLT3 expression is up-regulated, and its activation restores the Bax/Bcl-2 ratio, thus inhibiting cytochrome c release and apoptosis. FLT3 stimulation with FL also improves post-myocardial infarction remodeling and function in vivo, and this outcome is associated with decreased apoptosis in the infarct border zone. These findings identify FLT3 as a potential therapeutic target in ischemic heart disease. Importantly, they may also provide a novel mechanistic rationale for the potential contribution of FLT3 inhibition to the cardiotoxic effects of multitargeting TKIs.
FLT3 mRNA has previously been identified in whole heart homogenates (5,8) and shown to be up-regulated after myocardial infarction (8). However, whether intrinsic cardiac cells or circulating and/or homing extracardiac cells account for cardiac FLT3 expression was not known. We found FLT3 mRNA and protein expression in isolated, cultured cardiomyocytes. Activation of FLT3 with FL induced a time- and dose-dependent signal and protected cardiomyocytes against apoptosis. Our findings, therefore, strongly support that cardiomyocytes express functional FLT3 and react to FLT3 stimulation.
FLT3 activation with FL protected cardiomyocytes from H2O2-induced apoptosis via Akt-dependent inhibition of cytochrome c release and regulation of Bax/Bcl-2. PI3K/Akt is an important pathway of cell survival (24), and its activation is required for survival of acute myeloid leukemia cells (25). Although a variety of upstream regulators may activate Akt, constitutive activation of PI3K and Akt is found in myeloid cells harboring activity-enhancing mutations of FLT3 such as internal tandem duplications (26). We observed sustained phosphorylation of Akt in response to FL and failure of FL to suppress the H2O2-induced cytochrome c release in the presence of the Akt inhibitor VIII, supporting the theory that Akt is necessary for the antiapoptotic effect of FL in cardiomyocytes.
Bcl-2 and in particular the ratio of Bax to Bcl-2 is an upstream regulator of mitochondrial cytochrome c release and an important determinant of the cell's susceptibility to undergo apoptosis (27). Interestingly, the antiapoptotic effect of FL has previously been linked to its ability to inhibit the up-regulation of Bax in acute myeloid leukemia cells (28). We found a similar effect in cardiomyocytes exposed to oxidative stress, in which the H2O2-induced increase of Bax was abolished in the presence of FL. FL also rescued the decreased expression of Bcl-2 in response to H2O2 in the pooled cardiomyocyte population, hence restoring the ratio of Bax to Bcl-2.
Due to their ability to mobilize bone marrow–derived progenitor cells, combinations of hematopoietic cytokines, including G-CSF, SCF, and FL, have previously been used to promote cardiac regeneration in various in vivo models of ischemic injury (29–31). In particular, the systemic co-administration of G-CSF and FL over a period of 10 days improved left ventricular remodeling and function at 5 weeks after ischemia-reperfusion injury in mice via a mechanism that involves bone marrow–derived progenitor cell-mediated cardiac regeneration (31). According to the distinct hypotheses (i.e., enhanced cardiac regeneration as opposed to cardioprotection), there are 2 important differences in design between this study and ours: in the previous study, FL was administered systemically and over several days, and used in combination with other cytokines, namely, G-CSF; we administered FL as a single-dose monotherapy and also directly into the myocardium adjacent to the area at risk as demarcated after LAD ligation. This temporally and spatially restricted application was intended to study the effects of FL on intrinsic cardiac cells independently from its mobilizing properties on bone marrow–derived progenitor cells. Our strategy of FL administration improved cardiac morphology and function as early as 1 week post-myocardial infarction and also reduced the infarct size, which was not seen in the earlier study (31), in which improved remodeling was observed 5 weeks after myocardial infarction and was mostly based on enhanced cardiac regeneration. Although we cannot rule out that FL affected heart-resident progenitor cells in our model, our data support the hypothesis that FL per se acts as a cytoprotective factor in ischemic cardiac injury and that its method of action includes a direct protective effect on cardiomyocytes. Interestingly, similar to FLT3 shown in our study, G-CSFR has recently been identified on cardiomyocytes and a direct cardioprotective effect of G-CSF involving protection of cardiomyocytes from apoptosis was uncovered (11), which was sufficient to improve myocardial remodeling and function in the absence of peripheral cell mobilization.
FLT3 is one of the targets of TKIs used for anticancer therapy (2), and cardiotoxic effects (including decline in left ventricular ejection fraction, congestive heart failure, acute coronary syndrome, and hypertension) occurred in solid tumor patients treated with TKIs targeting multiple receptors to inhibit tumor angiogenesis and tumor cell survival (17–19,32–34). Cardiac dysfunction is among the most frequent cardiac adverse effects, and it is related to impaired cardiomyocyte survival due to the suppression of pro-survival (antiapoptotic) signaling pathways, which takes effect not only in the cancer cells but also in other tissues (35). This may include the mitochondrial release of cytochrome c and apoptotic cardiomyocyte death, which have been postulated, among others, in response to the FLT3-targeting TKIs sunitinib and sorafenib (32). Both sunitinib and sorafenib have multiple targets, and the upstream mechanisms of enhanced pro-apoptotic signaling are incompletely understood. Because of the established roles of vascular endothelial growth factor (36) and platelet-derived growth factor (37) in the myocardium, inhibition of their receptors, as well as off-target effects (18,19), have primarily been held responsible for the cardiotoxicity of these drugs. The cardioprotective role of FLT3 uncovered in the present study raises the possibility that FLT3 inhibition may be involved in cardiomyocyte death, particularly in patients with ischemic heart disease. However, we would like to emphasize that although we demonstrated an antiapoptotic effect of FLT3 activation by exogenous FL in cardiomyocytes, further studies are needed to establish the precise roles of endogenous FLT3 signaling in the heart and to determine whether its inhibition indeed contributes to cardiotoxicity.
Regarding a potential therapeutic use of FL, it is of note that FL has previously been applied to humans and has been well tolerated (38,39). Importantly, neither human nor animal studies have reported tumorigenic effects of systemic FL application.
As already pointed out, this study did not address the role of endogenous FLT3 signaling or the consequences of its inhibition in the heart. Results from this pre-clinical study conducted in mice cannot necessarily be extrapolated to humans.
Our data demonstrate that cardiomyocytes express FLT3 and provide a functional relevance to this expression. We identified FLT3/FL as a cardioprotective system and a potential therapeutic target that is up-regulated and the activation thereof is beneficial in cardiac ischemia. The recognition of FLT3/FL as an innate cytoprotective system in the heart may offer a novel therapeutic strategy to mitigate ischemic cardiac injury but also raises concerns about the potential cardiotoxicity of FLT3-targeting anticancer therapy. Further mechanistic and clinical studies are needed to explore putative adverse effects of FLT3 inhibition on the heart, paying particular attention to subsets of patients with ischemic heart disease and/or multitargeted combination chemotherapy.
The authors thank the Cardiovascular Physiology Core at the Brigham and Women's Hospital and Harvard Medical School for providing animal surgery and echocardiographic imaging, especially Mr. Soeun Ngoy for performing mouse surgery.
For an expanded methods section and a supplemental table and figure, please see the online version of this article.
This study was supported by the Mach-Gaensslen Foundation Switzerland and the Swiss Heart Foundation (both to Dr. Pfister), as well as in part by the Novartis Foundation for Medical-Biological Research Switzerland and the Research Foundation of the University of Basel, Switzerland (Spezialprogramm Nachwuchsförderung in der klinischen Medizin) (both to Dr. Kuster). Dr. Kuster was supported by a SCORE career development grant from the Swiss National Science Foundation (32323B-111352), and Drs. Pfister and Kuster received a grant-in-aid from the University Hospital Basel (VFWAWF) and the Kardiovaskuläre Stiftung Basel. Dr. Kaufmann is supported by SCORE grants (32323B-123819 and 32323B-141603) from the Swiss National Science Foundation. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- extracellular signal-regulated kinase
- granulocyte-colony stimulating factor
- FMS-like tyrosine kinase 3 ligand
- FMS-like tyrosine kinase 3
- hydrogen peroxide
- left anterior descending
- mitogen-activated protein kinase
- myocardial infarction group treated with recombinant mouse FMS-like tyrosine kinase 3 ligand
- myocardial infarction group treated with vehicle
- messenger ribonucleic acid
- phosphatidylinositol 3 kinase
- stem cell factor
- receptor tyrosine kinase inhibitor
- Received May 25, 2013.
- Revision received August 1, 2013.
- Accepted August 26, 2013.
- American College of Cardiology Foundation
- Wodnar-Filipowicz A.
- Gilliland D.G.,
- Griffin J.D.
- Veiby O.P.,
- Jacobsen F.W.,
- Cui L.,
- Lyman S.D.,
- Jacobsen S.E.
- Lyman S.D.,
- James L.,
- Johnson L.,
- et al.
- Ayach B.B.,
- Yoshimitsu M.,
- Dawood F.,
- et al.
- Namiuchi S.,
- Kagaya Y.,
- Ohta J.,
- et al.
- Wright G.L.,
- Hanlon P.,
- Amin K.,
- Steenbergen C.,
- Murphy E.,
- Arcasoy M.O.
- Anagnostou A.,
- Lee E.S.,
- Kessimian N.,
- Levinson R.,
- Steiner M.
- Lal H.,
- Kolaja K.L.,
- Force T.
- Mouquet F.,
- Pfister O.,
- Jain M.,
- et al.
- Kennedy S.G.,
- Kandel E.S.,
- Cross T.K.,
- Hay N.
- Dudek H.,
- Datta S.R.,
- Franke T.F.,
- et al.
- Xu Q.,
- Simpson S.E.,
- Scialla T.J.,
- Bagg A.,
- Carroll M.
- Brandts C.H.,
- Sargin B.,
- Rode M.,
- et al.
- Lisovsky M.,
- Estrov Z.,
- Zhang X.,
- et al.
- Orlic D.,
- Kajstura J.,
- Chimenti S.,
- et al.
- Kawada H.,
- Fujita J.,
- Kinjo K.,
- et al.
- Dawn B.,
- Guo Y.,
- Rezazadeh A.,
- et al.
- Schmidinger M.,
- Zielinski C.C.,
- Vogl U.M.,
- et al.
- Chen M.H.,
- Kerkela R.,
- Force T.
- May D.,
- Gilon D.,
- Djonov V.,
- et al.
- Maraskovsky E.,
- Daro E.,
- Roux E.,
- et al.
- Higano C.S.,
- Vogelzang N.J.,
- Sosman J.A.,
- Feng A.,
- Caron D.,
- Small E.J.