Original Research Received: July 6, 2015 Accepted after revision: December 15, 2015 Published online: February 17, 2016
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
Differential Effects of Colchicine on Cardiac Cell Viability in an in vitro Model Simulating Myocardial Infarction Gilad Margolis a, c Einat Hertzberg-Bigelman b, c Ran Levy b Jeremy Ben-Shoshan b, c Gad Keren b, c Michal Entin-Meer b, c a
Department of Internal Medicine H and b Cardiovascular Research Laboratory, Department of Medicine, Tel Aviv Sourasky Medical Center, and c Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Key Words Apoptosis · Cardiomyocytes · Cardiomyoblasts · Cell viability · Colchicine · Heart failure · Myocardial infarction
ing AMI and reduced their adherence capability. HL-1 was not affected by colchicine; nevertheless, no salvage effect was observed. We thus conclude that colchicine may not inhibit myocardial apoptosis following AMI. © 2016 S. Karger AG, Basel
© 2016 S. Karger AG, Basel 0008–6312/16/1341–0057$39.50/0 E-Mail [email protected] www.karger.com/crd
Introduction
Congestive heart failure (HF) is a clinical presentation resulting from a common pathway of multiple pathophysiological processes, e.g. ischemia secondary to coronary artery disease [1]. Apoptosis, or planned cell death, has a significant role in the development of HF [2]. In particular, caspase-dependent apoptosis of cardiomyocytes has been demonstrated in animal models of cardiac injury as well as in patients with congestive HF or acute myocardial infarction (AMI) [3, 4]. Therefore, apoptosis has been proposed as an important process in cardiac remodeling and HF progression. Indeed, coronary ischemia exerts an apoptosis-promoting effect on myocardial cells [5]. During cardiac ischemia, several apoptosis-promoting factors are present in the circulation, among them are reactive oxygen species [6] and inflammatory cytokines such as tumor necrosis factor (TNF)α [7]. Michal Entin-Meer, PhD Cardiovascular Research Laboratory Tel Aviv Sourasky Medical Center, Weizmann 6 Street Tel Aviv 64239 (Israel) E-Mail michale @ tlvmc.gov.il
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
Abstract Objectives: We aimed to examine the effects of colchicine, currently in clinical trials for acute myocardial infarction (AMI), on the viability of cardiac cells using a cell line model of AMI. Methods: HL-1, a murine cardiomyocyte cell line, and H9C2, a rat cardiomyoblast cell line, were incubated with TNFα or sera derived from rats that underwent AMI or sham operation followed by addition of colchicine. In another experiment, HL-1/H9C2 cells were exposed to anoxia with or without subsequent addition of colchicine. Cell morphology and viability were assessed by light microscopy, flow cytometry and Western blot analyses for apoptotic markers. Results: Cellular viability was similar in both sera; however, exposing both cell lines to anoxia reduced their viability. Adding colchicine to anoxic H9C2, but not to anoxic HL-1, further increased their mortality, at least in part via enhanced apoptosis. Under any condition, colchicine induced detachment of H9C2 cells from their culture plates. This phenomenon did not apply to HL-1 cells. Conclusions: Colchicine enhanced cardiomyoblast mortality under in vitro conditions mimick-
TNFα has a significant role in the death receptor pathway. It activates the apoptosis cascade by binding to its receptor on the cardiomyocyte membrane. Then, through mediator proteins, caspase-8 is activated. One of the caspase-8 substrates is BID (BH3 interacting domain death agonist), a proapoptotic protein inducing the release of cytochrome c from the mitochondria, which eventually activates caspase-9. Both caspase-8 and caspase-9 activate other caspases, which in turn cause DNA fragmentation and apoptosis [8]. Microtubules are dynamic polymers which serve as key elements of the cytoskeleton; they are composed of the heterodimers α- and β-tubulin. A constant turnover of microtubules by polymerization and depolymerization occurs regularly. In cardiomyocytes, only 30% of total tubulin are present in the polymerized form as microtubules, whereas 70% present as nonpolymerized cytosolic protein. Previous studies have shown that in HF tubulin is hyperpolymerized, contributing to abnormalities in cellular contractility [9, 10]. Interestingly, it has been reported that exposure of cardiomyocytes from hypertrophied right ventricles to colchicine, a depolymerizing agent, reversed the contractility abnormalities fully back to normal [11]. Colchicine is an alkaloid approved by the Food and Drug Administration for treatment of inflammatory conditions such as familial Mediterranean fever and gout. Apart from its anti-inflammatory action, colchicine exerts a direct effect on the cell cycle by binding to microtubules. There are two main subtypes of microtubulebinding agents: polymerizing and depolymerizing agents. Given its high concentration, colchicine is considered a depolymerizing agent [12, 13]. A previous study has shown that exposure of rat-derived cardiomyocytes to paclitaxel, a microtubule polymerizing agent, resulted in apoptosis in vitro. In contrast, the same study reported that incubating cardiomyocytes with colchicine prevented apoptosis following exposure to proapoptotic agents such as TNFα [14]. Since colchicine is prescribed for inflammatory conditions and is currently in clinical trials for ST-elevation MI [15], we aimed to examine the potential effects of this drug on the viability of cardiac cells using a cell line model simulating AMI.
Materials and Methods
as the rat cardiomyoblast cell line H9C2 (ATCC® clone CRL1446TM; a generous gift from Gania Kessler-Icekson, Felsenstein Medical Research Center, Petach Tikva, Israel). Serum Effect/TNFα In the first set of experiments, HL-1 as well as H9C2 cells were incubated for 24 h with media containing 5% sera derived from rats that had surgically induced MI (lasting 3 or 30 days) or with 5% sera from sham-operated controls. This was done in order to simulate the inflammation-prone environment that occurs after MI. Another set of H9C2 dishes was treated with the proapoptotic agent TNFα for 24 h. Later, colchicine (10 μM) was added for 1 h to half of the wells, as described by others [16–18], followed by recovery with full media containing 10% FBS for 16 h (n = 4 for each experimental well). As a control, H9C2 cells were incubated with DMEM with or without colchicine (1 or 10 μM) [19, 20] for doseresponse analysis. Effect of Oxygen Deprivation In the second set of experiments, we exposed HL-1/H9C2 cells to anoxic conditions for 24 h to measure the effect of direct stress on cardiomyocytes. Later, colchicine (1 μM) was added for 1 h to half of the wells, and then recovery with full media containing 10% FBS was allowed for 16 h (n = 4 for each experimental well). Cell Morphology and Viability Analysis In both sets of experiments, cell morphology was assessed by light microscopy. Photos were taken using an Olympus DP73 camera and cellSense standard 1.6 software. The numbers of floating/ total cells were determined in two captures representing two plates/experimental group, each one divided into four fields. The cells were then stained with annexin V-propidium iodide (PI; Abcam, USA) and percent viable cells, i.e. cells which stained negative for both annexin V and PI, were determined by flow cytometry (BD Biosciences FACSCanto II). Western Blot Analyses Cells from experimental culture dishes were extracted using a commercial lysis buffer (Sigma). Equal protein amounts (50 μg) were loaded on a 4–20% acrylamide gel followed by electric transfer to nitrocellulose membranes (n = 4/arm). Following blocking with 5% low-fat milk diluted in TBS-Tween for 1 h, the membranes were incubated with the following antibodies: anti-PARP-1 (Santa Cruz, sc-7150), anti-cleaved caspase-3 (Sigma, C8487) or antiGAPDH (Abcam, clone 6C5) antibody used for validating equal loading. The primary antibodies were followed by blotting with HRP-conjugated secondary antibodies. After rapid incubation with an ECL substrate (Biological Industries, Israel), the membranes were exposed to an imaging film. Band intensity was determined using GelQuant.Net BiochemicalLabSolutions.com software. Statistical Analysis Groups were compared using one-way ANOVA. The Tukey post hoc correction was taken to account for multiple testing (IBM SPSS statistics 20). Results are expressed as means ± SE. p < 0.05 was accepted as statistically significant.
58
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
Margolis/Hertzberg-Bigelman/Levy/ Ben-Shoshan/Keren/Entin-Meer
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
The study was performed using the murine cardiomyocyte cell line HL-1 (a generous gift from the laboratory of Prof. W. Claycomb, LSU Health Sciences Center, New Orleans, La., USA) as well
40 20 0
a
Sham
MI 3 days MI 30 days
40 20 Sham
MI 3 days
Fig. 1. Colchicine (C) attenuates cellular viability of cardiomyoblasts (H9C2) but not of cardiomyocytes (HL-1) incubated with MI or sham-derived sera; mean values of cell viability (determined by flow cytometry with annexin V-PI) of HL-1 cells (a) and H9C2 cells (b). Prior to the analyses, cells were incubated with sera collected from rats 3 or 30 days after surgically induced MI or after
Results
Addition of Colchicine Decreases Viability in H9C2 Cells but Not in HL-1 Cells in the Presence of MI-Derived Sera Incubating HL-1 cells with MI-derived sera did not reduce cell viability in comparison to sham sera, as documented by flow cytometry: 47.3 ± 13% for 3-day MI, 41.9 ± 21% for 30-day MI and 32.9 ± 8.0% for sham controls (p = 0.16; fig. 1a). Adding colchicine to HL-1 cardiomyocytes exposed to MI or sham-derived sera did not significantly change their viability (fig. 1a). The relatively low percentage of annexin V-FITC-PI-negative cells, representing the viable HL-1 cells, may be attributed to the potential vulnerability of HL-1 cells to cell death due to their exposure to xenogenic sera, i.e. rat sera applied onto murine cells. In H9C2 cardiomyoblast cells, incubation with (3- or 30-day) MI-derived sera did not significantly change the percentage of viable cells in comparison to the sham controls, similar to data obtained for HL-1: 94.1 ± 0.5, 94.8 ± 0.5 versus 92.3 ± 0.5% for 3- and 30-day MI 30, and sham control, respectively (p > 0.1; fig. 1b). Interestingly, however, in H9C2, addition of colchicine to sera derived from post-MI animals (3 or 30 days) resulted in a significant (15%) increase in cell mortality compared to the same sera in the absence of colchicine: 94.1 ± 0.5 versus 80.3 ± 0.8% for 3-day MI without versus with colchicine, respectively (p < 0.05; fig. 1b); 94.8 ± 0.5 and 78.7 ± 0.9% for Effect of Colchicine on Cardiac Cells
*
60
0
b
*
MI 30 days
* **
**
90 80 70 60 50
c
Medium Medium +C
TNFį
TNFį +C
sham operation with or without subsequent addition of colchicine. c Bar graph represents cell viability of H9C2 grown in the presence of full DMEM with or without colchicine (1 or 10 μM) or in the presence of TNFα with or without colchicine. * p < 0.05, ** p < 0.01.
30-day MI without versus with colchicine, respectively (p < 0.05; fig. 1b). Likewise, addition of colchicine to H9C2 incubated in sham sera resulted in an increased mortality compared to the same sera in the absence of colchicine: 92.3 ± 0.5 versus 73.3 ± 3.1% without versus with colchicine, respectively (p < 0.05; fig. 1b). The detrimental nonspecific effect of colchicine on cell viability was confirmed in a third set of experiments in which colchicine was administered either at a concentration of 1 or 10 μM to H9C2 cells grown in regular DMEM. In both experiments, colchicine resulted in a significant reduction in the cell viability (p = 0.04) with no real difference between the two doses (reductions of 11 vs. 15.3% in the presence of 1 vs. 10 μM colchicine, respectively; p = 0.2). Equally important, adding colchicine to H9C2 cells which were kept for 24 h in the presence of TNFα (10 ng/ ml) resulted in an additive mortality rate (9.9% in regular DMEM, 15.9% in TNFα and 25.7% in TNFα + colchicine; p < 0.01; fig. 1c). Adding Colchicine to Oxygen-Deprived Cardiomyocytes Induces Apoptosis in H9C2 Cells but Not in HL-1 Cells Exposing both HL-1 cells and H9C2 cells to anoxic environment resulted in reduced viability of both cell types: 80.7 ± 11.3 versus 63.7 ± 15.4% for HL-1 cells in normoxic versus anoxic environment, respectively (p = 0.05; fig. 2a) and 81.5 ± 3.7 versus 68.3 ± 4.3% for H9C2 cells Cardiology 2016;134:57–64 DOI: 10.1159/000443369
59
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
60
*
80
100
Color version available online
Serum Serum + C
&HOO£YLDELOLW\£
&HOO£YLDELOLW\£
80
H9C2 cardiomyoblasts 100
+ &£FHOO£YLDELOLW\£
HL-1 cardiomyocytes 100
H9C2 cardiomyoblasts
*
100
£No C C
* 80 &HOO£YLDELOLW\£
60 40
0
10 5
Q3 103
Q4 10 4
Annexin FITC-A
Q1
103
Q3
102
10 5
52.5%
102
103
Q4 10 4
10 5
10 4
Q2
PI-A
103
A+C H9C2 11122014-A1 + C
10 5
10 4
Q2
PI-A
PI-A
Q1
102
Anoxia
0
H9C2 cardiomyoblasts N+C A H9C2 11122014-N1 + C H9C2 11122014-A1
10 5
c
b
Anoxia
N H9C2 11122014-N1
86.5%
Normoxia
40
Q1
103
Q3
102
10 5
Annexin FITC-A
10 4
Q2
72.7%
102
103
Q4 10 4
Annexin FITC-A
102
10 5
Q1
Q2
Q3
Q4
103
102
48.5%
103
10 4
10 5
FITC-A
Fig. 2. Colchicine (C) attenuates cell viability of cardiomyoblasts (H9C2) but not of cardiomyocytes (HL-1) exposed to anoxic environment; mean values of cell viability of HL-1 cells (a) and H9C2 cells (b) exposed to normoxic/anoxic environment with or without colchicine; * p < 0.05; ** p < 0.01. c Representative captures of
flow-cytometric data demonstrating the percent annexin V-FITCPI-negative H9C2 cells grown under normoxia (N) or anoxia (A) with or without colchicine. Cells which stained negative for both annexin-FITC and PI in flow cytometry (lower left quadrant) represent viable cells.
exposed to normoxic versus anoxic environment, respectively (p < 0.05; fig. 2b). Adding colchicine to HL-1 cells exposed to both normoxic or anoxic environment for 18 h did not significantly affect cell death: 68.8 ± 7.6 versus 80.7 ± 11.3% with versus without colchicine, respectively, in normoxic environment (p = 0.47; fig. 2a), and 61.2 ± 6.2 versus 63.7 ± 15.4% with versus without colchicine, respectively, in anoxic environment (p = 0.37; fig. 2a). On the other hand, adding colchicine to H9C2 cells kept in normoxia resulted in a significant (30%) decrease in viability: 50.8 ± 6.9 versus 81.5 ± 3.7% with versus without colchicine, respectively (p < 0.01; fig. 2b). Adding colchicine to H9C2 cells exposed to anoxic environment re-
sulted in an even greater increase in cardiac cell mortality: 42.3 ± 6.6 with versus 68.3 ± 4.3% without colchicine, respectively (p < 0.01; fig. 2b). Representative captures of flow-cytometric analyses of H9C2 cells grown under normoxia/anoxia with or without cholchicine are given in figure 2c.
60
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
Adherence to Tissue Culture Dishes after Incubation with Colchicine Is Reduced in H9C2 Cells but Not in HL-1 Cells Using light microscopy, we observed a marked reduction in the percentage of cells adherent to tissue culture plates in the presence of anoxia relative to normoxia in both H9C2 (fig. 3) and HL-1 (fig. 4) cells, in line with the Margolis/Hertzberg-Bigelman/Levy/ Ben-Shoshan/Keren/Entin-Meer
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
Normoxia
102
**
20
20
10 4
**
60
PI-A
&HOO£YLDELOLW\£
80
a
Color version available online
HL-1 cardiomyocytes 100
Color version available online
a
b Floating cells
c
d
e
flow-cytometric data given in figure 2. Nevertheless, regardless of the biological environment (normoxia vs. anoxia), treating H9C2 cells with colchicine resulted in detachment of approximately 50% of the cells from the tissue culture plates (21.7 ± 7.1% in normoxia vs. 51.1 ± 15.5% in normoxia + colchicine and 30.3 ± 13.7% in anoxia vs. 78.8 ± 7.3% in anoxia + colchicine; p < 0.05 for the colchicine effect and p < 0.05 for the lack of oxygen effect; fig. 3). This phenomenon did not apply to HL-1 cells (fig. 4). Induced Apoptosis in the Presence of Colchicine In order to assess whether the increased mortality of H9C2 cells in the presence of colchicine is mediated at least in part by apoptotic pathways, we subjected lysates of H9C2 cells grown under normoxia or anoxia with or without colchicine to Western blot analysis with two anEffect of Colchicine on Cardiac Cells
80
**
*
60 40 20 0
N
N+C
A
A+C
tibodies. The first one was anti-PARP-1, which recognizes both the full-length molecule (116 kDa) as well as a cleaved apoptotic fragment (89 kDa), which is produced upon activation of upstream caspase-3 in various cell types, including H9C2 [21, 22]. The second was an antibody directed to the cleaved form of caspase-3 (17 kDa). Representative data depicted in figure 5 demonstrated an increased expression of cleaved PARP-1 as well as cleaved caspase-3 in anoxia versus normoxia (p < 0.01). Incubation with colchicine induced further cleavage of PARP-1 under both normoxia and anoxia (p ≤ 0.05). The cleaved form of caspase-3 presented a trend towards a significant increase under normoxia (p = 0.07) and a significant upregulation under anoxia (p = 0.03). The data suggest that colchicine induced caspase-dependent apoptosis of H9C2 cells. Cardiology 2016;134:57–64 DOI: 10.1159/000443369
61
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
£ 3HUFHQW£IORDWLQJ£FHOOV
H9C2 cardiomyoblasts to culture dishes under normoxic/anoxic conditions (assessed by light microscopy); cellular morphology of H9C2 cells under normoxic (a) or anoxic conditions (b); H9C2 cells under normoxic conditions + colchicine (c) or anoxic conditions + colchicine (d). e The percentage of floating cells/total cells was calculated in four different fields in each experimental plate (two plates/treatment). A = Anoxia; C = colchicine; N = normoxia. * p < 0.05, ** p < 0.01.
*
100
Fig. 3. Colchicine reduces adherence of
Color version available online
a
b Floating cells
c
d
e
Discussion
In the current study, we have demonstrated differential effects of colchicine on cell viability of cardiac cells in an in vitro model simulating AMI versus normal physiological conditions. Interestingly, while colchicine attenuated cellular viability of H9C2 cardiomyoblasts, even when the cells were grown in full DMEM, the viability of HL-1 cardiomyocytes was not affected by this agent. Nevertheless, no salvage effect of colchicine on cell viability was observed. Moreover, surprisingly, colchicine per se exhibited a detrimental effect on the morphology of H9C2. In line with our data, a different study, which as62
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
25 20 15 10 5 0
N
N+C
A
A+C
sessed the cellular basis of the contractility abnormalities of feline myocytes in an experimental model of severe left ventricular hypertrophy (LVH), showed no significant difference in contractility properties between control and LVH myocytes after microtubule depolymerization by colchicine [23]. The authors suggested that the contractile abnormalities of feline LVH myocytes resulted from changes in cellular Ca2+ regulation and myofibrillar Ca2+ sensitivity, but not from changes in internal loading. These results were followed by reports from other authors [24], in which no effect on contraction dynamics by colchicine or Taxol, which induces microtubule polymerization, was observed. On the other hand, in recent studies, Margolis/Hertzberg-Bigelman/Levy/ Ben-Shoshan/Keren/Entin-Meer
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
£ 3HUFHQW£IORDWLQJ£FHOOV
duce adherence of HL-1 cardiomyocytes to culture dishes under normoxic/anoxic conditions (assessed by light microscopy); HL-1 culture morphology under normoxic (a) or anoxic conditions (b); HL-1 cells under normoxic conditions + colchicine (c) or anoxic conditions + colchicine (d). e The percentage of floating cells/total cells was calculated in four different fields in each experimental plate (two plates/treatment). A = Anoxia; C = colchicine; N = normoxia. * p < 0.05.
*
30
Fig. 4. Colchicine does not significantly re-
Color version available online
Full PARP-1 (116 kDa) Cleaved PARP-1 (89 kDa) GAPDH (37 kDa) Cleaved caspase-3 (17 kDa)
**
20 15
N+C
*
*
10 5 0
N
A A+C
N+C
A
A+C
c
N N+C
A A+C
**
30
*
25 20 15 10 5 0
N
N+C
A
A+C
Fig. 5. Colchicine induces caspase-dependent apoptosis in H9C2 cells. a Levels of cleaved relative to full-length PARP-1 as well as cleaved caspase-3 relative to the housekeeping protein GAPDH. b Bar graph for the mean percentage of cleaved PARP-1 out
of total PARP-1 (full length and cleaved). c Bar graph for the mean percentage of cleaved caspase-3 normalized to the levels of GAPDH. A = Anoxia; C = colchicine; N = normoxia. * p < 0.05, ** p < 0.01.
a cardioprotective role for microtubule depolymerization by colchicine was demonstrated in models of hearts exposed to chronic pressure overload [25, 26]. More specifically, a previous study has shown that reduction in microtubule hyperpolymerization by colchicine treatment reversed cardiomyocyte contraction abnormalities and suppressed apoptosis in HF induced by chronic pressure overload models [11]. Moreover, other authors have also reported cardioprotective effects against cell death in primary adult cardiomyocytes treated with proapoptotic agents such as TNFα [14]. These inconsistencies regarding the role of microtubules in HF could be explained by the observation that a substantial pressure overload is needed for the cytoskeletal change to happen [27]. Furthermore, one cannot rule out the possibility that primary cardiac cells may be affected differently by colchicine compared to H9C2 or HL-1 cells. Nevertheless, it was recently suggested that H9C2 could be successfully utilized as an in vitro model simulating cardiac ischemia/reperfusion injury [28].
In the current study, we observed enhanced cell death of cardiac cells incubated in sera collected from animals 30 days after MI and after treatment with colchicine. This observation may indicate that colchicine may not be effective to prevent late complications of AMI such as HF, in contrast to the benefits of its use in the pressure overload HF setting, as described before. Moreover, we assume that in the setting of AMI, the potential roles of microtubules are probably different from their role in chronic pressure overload. It has been shown that microtubules are reversibly disrupted in an ischemia/reperfusion model [29]. A recent study has demonstrated improved cardiomyocyte functional recovery and apoptosis rate in ischemia/reperfusion following exposure to Taxol – a microtubule stabilizer and polymerization promoter [30]. The differential effect of colchicine on H9C2 and HL-1 cells observed in our study could be explained by inherent metabolic differences between the two cell lines. It has been suggested that H9C2 cells are more sensitive to ischemia/reperfusion injury than HL-1 cells due to dif-
Effect of Colchicine on Cardiac Cells
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
63
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
b
N
Cleaved£FDVSDVH*$3'+£
Cleaved£3$53WRWDO£3$53£
a
ferent mitochondrial processes, explained in part by the presence of β-tubulin II in H9C2 cells and its absence in HL-1 cells [28]. Although we did not directly examine microtubular structural changes, these recent observations are consistent with our results, indicating that colchicine has no beneficial or even detrimental effect on cardiac cells in the AMI setting, which is in contrast to the cardioprotective effect shown before in pressure overload HF models by various investigators. We thus concluded that colchicine cannot inhibit myocardial apoptosis following AMI.
Study Limitation
In the experimental setting described in the current study, we were unable to test the potential immunosuppressive effects of colchicine on cardiac cell viability. This issue will be addressed in a future in vivo study, in which small animals subjected to AMI will be treated with colchicine and cardiac cell viability will be assessed by immunohistochemistry using apoptotic and necrotic markers.
References
64
11 Tagawa H, Wang N, Narishige T, Ingber DE, Zile MR, Cooper G 4th: Cytoskeletal mechanics in pressure-overload cardiac hypertrophy. Circ Res 1997;80:281–289. 12 Bhattacharyya B, Panda D, Gupta S, Banerjee M: Anti-mitotic activity of colchicine and the structural basis for its interaction with tubulin. Med Res Rev 2008;28:155–183. 13 Terkeltaub RA: Colchicine update: 2008. Semin Arthritis Rheum 2009;38:411–419. 14 Saji K, Fukumoto Y, Suzuki J, Fukui S, Nawata J, Shimokawa H: Colchicine, a microtubule depolymerizing agent, inhibits myocardial apoptosis in rats. Tohoku J Exp Med 2007; 213:139–148. 15 Deftereos S, Giannopoulos G, Angelidis C, et al: Anti-inflammatory treatment with colchicine in acute myocardial infarction: a pilot study. Circulation 2015;132:1395–1403. 16 Bonfoco E, Ceccatelli S, Manzo L, Nicotera P: Colchicine induces apoptosis in cerebellar granule cells. Exp Cell Res 1995;218:189–200. 17 Boyd AE 3rd, Bolton WE, Brinkley BR: Microtubules and beta cell function: effect of colchicine on microtubules and insulin secretion in vitro by mouse beta cells. J Cell Biol 1982; 92:425–434. 18 Ishibashi Y, Takahashi M, Isomatsu Y, et al: Role of microtubules versus myosin heavy chain isoforms in contractile dysfunction of hypertrophied murine cardiocytes. Am J Physiol Heart Circ Physiol 2003; 285:H1270– H1285. 19 Gomez AM, Kerfant BG, Vassort G: Microtubule disruption modulates Ca(2+) signaling in rat cardiac myocytes. Circ Res 2000;86:30– 36. 20 Kerfant BG, Vassort G, Gomez AM: Microtubule disruption by colchicine reversibly enhances calcium signaling in intact rat cardiac myocytes. Circ Res 2001;88:E59–E65. 21 D’Amours D, Sallmann FR, Dixit VM, Poirier GG: Gain-of-function of poly(ADP-ribose) polymerase-1 upon cleavage by apoptotic proteases: implications for apoptosis. J Cell Sci 2001;114:3771–3778.
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
22 Youn HJ, Kim HS, Jeon MH, et al: Induction of caspase-independent apoptosis in H9C2 cardiomyocytes by Adriamycin treatment. Mol Cell Biochem 2005;270:13–19. 23 Bailey BA, Dipla K, Li S, Houser SR: Cellular basis of contractile derangements of hypertrophied feline ventricular myocytes. J Mol Cell Cardiol 1997;29:1823–1835. 24 de Tombe PP: Altered contractile function in heart failure. Cardiovasc Res 1998; 37: 367– 380. 25 Cheng G, Kasiganesan H, Baicu CF, Wallenborn JG, Kuppuswamy D, Cooper G 4th: Cytoskeletal role in protection of the failing heart by beta-adrenergic blockade. Am J Physiol Heart Circ Physiol 2012; 302:H675– H687. 26 Zhang C, Chen B, Guo A, et al: Microtubulemediated defects in junctophilin-2 trafficking contribute to myocyte transverse-tubule remodeling and Ca2+ handling dysfunction in heart failure. Circulation 2014; 129: 1742– 1750. 27 Cooper G 4th: Proliferating cardiac microtubules. Am J Physiol Heart Circ Physiol 2009; 297:H510–H511. 28 Kuznetsov AV, Javadov S, Sickinger S, Frotschnig S, Grimm M: H9C2 and HL-1 cells demonstrate distinct features of energy metabolism, mitochondrial function and sensitivity to hypoxia-reoxygenation. Biochim Biophys Acta 2015;1853:276–284. 29 Sato H, Hori M, Kitakaze M, et al: Reperfusion after brief ischemia disrupts the microtubule network in canine hearts. Circ Res 1993; 72:361–375. 30 Xiao J, Liang D, Liu Y, et al: Taxol, a microtubule stabilizer, improves cardiac functional recovery during postischemic reperfusion in rat in vitro. Cardiovasc Ther 2012;30:12–30.
Margolis/Hertzberg-Bigelman/Levy/ Ben-Shoshan/Keren/Entin-Meer
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
1 Go AS, Mozaffarian D, Roger VL, et al: Heart disease and stroke statistics – 2013 update: a report from the American Heart Association. Circulation 2013;127:e6–e245. 2 Gonzalez A, Fortuno MA, Querejeta R, et al: Cardiomyocyte apoptosis in hypertensive cardiomyopathy. Cardiovasc Res 2003; 59: 549–562. 3 Kang PM, Haunstetter A, Aoki H, Usheva A, Izumo S: Morphological and molecular characterization of adult cardiomyocyte apoptosis during hypoxia and reoxygenation. Circ Res 2000;87:118–125. 4 Yang B, Ye D, Wang Y: Caspase-3 as a therapeutic target for heart failure. Expert Opin Ther Targets 2013;17:255–263. 5 Depre C, Taegtmeyer H: Metabolic aspects of programmed cell survival and cell death in the heart. Cardiovasc Res 2000;45:538–548. 6 Aikawa R, Komuro I, Yamazaki T, et al: Oxidative stress activates extracellular signal-regulated kinases through Src and Ras in cultured cardiac myocytes of neonatal rats. J Clin Invest 1997;100:1813–1821. 7 Krown KA, Page MT, Nguyen C, et al: Tumor necrosis factor alpha-induced apoptosis in cardiac myocytes. Involvement of the sphingolipid signaling cascade in cardiac cell death. J Clin Invest 1996;98:2854–2865. 8 Chen QM, Tu VC: Apoptosis and heart failure: mechanisms and therapeutic implications. Am J Cardiovasc Drugs 2002;2:43–57. 9 Hein S, Kostin S, Heling A, Maeno Y, Schaper J: The role of the cytoskeleton in heart failure. Cardiovasc Res 2000;45:273–278. 10 Tagawa H, Koide M, Sato H, Zile MR, Carabello BA, Cooper G 4th: Cytoskeletal role in the transition from compensated to decompensated hypertrophy during adult canine left ventricular pressure overloading. Circ Res 1998;82:751–761.
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
Differential Effects of Colchicine on Cardiac Cell Viability in an in vitro Model Simulating Myocardial Infarction Gilad Margolis a, c Einat Hertzberg-Bigelman b, c Ran Levy b Jeremy Ben-Shoshan b, c Gad Keren b, c Michal Entin-Meer b, c a
Department of Internal Medicine H and b Cardiovascular Research Laboratory, Department of Medicine, Tel Aviv Sourasky Medical Center, and c Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Key Words Apoptosis · Cardiomyocytes · Cardiomyoblasts · Cell viability · Colchicine · Heart failure · Myocardial infarction
ing AMI and reduced their adherence capability. HL-1 was not affected by colchicine; nevertheless, no salvage effect was observed. We thus conclude that colchicine may not inhibit myocardial apoptosis following AMI. © 2016 S. Karger AG, Basel
© 2016 S. Karger AG, Basel 0008–6312/16/1341–0057$39.50/0 E-Mail [email protected] www.karger.com/crd
Introduction
Congestive heart failure (HF) is a clinical presentation resulting from a common pathway of multiple pathophysiological processes, e.g. ischemia secondary to coronary artery disease [1]. Apoptosis, or planned cell death, has a significant role in the development of HF [2]. In particular, caspase-dependent apoptosis of cardiomyocytes has been demonstrated in animal models of cardiac injury as well as in patients with congestive HF or acute myocardial infarction (AMI) [3, 4]. Therefore, apoptosis has been proposed as an important process in cardiac remodeling and HF progression. Indeed, coronary ischemia exerts an apoptosis-promoting effect on myocardial cells [5]. During cardiac ischemia, several apoptosis-promoting factors are present in the circulation, among them are reactive oxygen species [6] and inflammatory cytokines such as tumor necrosis factor (TNF)α [7]. Michal Entin-Meer, PhD Cardiovascular Research Laboratory Tel Aviv Sourasky Medical Center, Weizmann 6 Street Tel Aviv 64239 (Israel) E-Mail michale @ tlvmc.gov.il
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
Abstract Objectives: We aimed to examine the effects of colchicine, currently in clinical trials for acute myocardial infarction (AMI), on the viability of cardiac cells using a cell line model of AMI. Methods: HL-1, a murine cardiomyocyte cell line, and H9C2, a rat cardiomyoblast cell line, were incubated with TNFα or sera derived from rats that underwent AMI or sham operation followed by addition of colchicine. In another experiment, HL-1/H9C2 cells were exposed to anoxia with or without subsequent addition of colchicine. Cell morphology and viability were assessed by light microscopy, flow cytometry and Western blot analyses for apoptotic markers. Results: Cellular viability was similar in both sera; however, exposing both cell lines to anoxia reduced their viability. Adding colchicine to anoxic H9C2, but not to anoxic HL-1, further increased their mortality, at least in part via enhanced apoptosis. Under any condition, colchicine induced detachment of H9C2 cells from their culture plates. This phenomenon did not apply to HL-1 cells. Conclusions: Colchicine enhanced cardiomyoblast mortality under in vitro conditions mimick-
TNFα has a significant role in the death receptor pathway. It activates the apoptosis cascade by binding to its receptor on the cardiomyocyte membrane. Then, through mediator proteins, caspase-8 is activated. One of the caspase-8 substrates is BID (BH3 interacting domain death agonist), a proapoptotic protein inducing the release of cytochrome c from the mitochondria, which eventually activates caspase-9. Both caspase-8 and caspase-9 activate other caspases, which in turn cause DNA fragmentation and apoptosis [8]. Microtubules are dynamic polymers which serve as key elements of the cytoskeleton; they are composed of the heterodimers α- and β-tubulin. A constant turnover of microtubules by polymerization and depolymerization occurs regularly. In cardiomyocytes, only 30% of total tubulin are present in the polymerized form as microtubules, whereas 70% present as nonpolymerized cytosolic protein. Previous studies have shown that in HF tubulin is hyperpolymerized, contributing to abnormalities in cellular contractility [9, 10]. Interestingly, it has been reported that exposure of cardiomyocytes from hypertrophied right ventricles to colchicine, a depolymerizing agent, reversed the contractility abnormalities fully back to normal [11]. Colchicine is an alkaloid approved by the Food and Drug Administration for treatment of inflammatory conditions such as familial Mediterranean fever and gout. Apart from its anti-inflammatory action, colchicine exerts a direct effect on the cell cycle by binding to microtubules. There are two main subtypes of microtubulebinding agents: polymerizing and depolymerizing agents. Given its high concentration, colchicine is considered a depolymerizing agent [12, 13]. A previous study has shown that exposure of rat-derived cardiomyocytes to paclitaxel, a microtubule polymerizing agent, resulted in apoptosis in vitro. In contrast, the same study reported that incubating cardiomyocytes with colchicine prevented apoptosis following exposure to proapoptotic agents such as TNFα [14]. Since colchicine is prescribed for inflammatory conditions and is currently in clinical trials for ST-elevation MI [15], we aimed to examine the potential effects of this drug on the viability of cardiac cells using a cell line model simulating AMI.
Materials and Methods
as the rat cardiomyoblast cell line H9C2 (ATCC® clone CRL1446TM; a generous gift from Gania Kessler-Icekson, Felsenstein Medical Research Center, Petach Tikva, Israel). Serum Effect/TNFα In the first set of experiments, HL-1 as well as H9C2 cells were incubated for 24 h with media containing 5% sera derived from rats that had surgically induced MI (lasting 3 or 30 days) or with 5% sera from sham-operated controls. This was done in order to simulate the inflammation-prone environment that occurs after MI. Another set of H9C2 dishes was treated with the proapoptotic agent TNFα for 24 h. Later, colchicine (10 μM) was added for 1 h to half of the wells, as described by others [16–18], followed by recovery with full media containing 10% FBS for 16 h (n = 4 for each experimental well). As a control, H9C2 cells were incubated with DMEM with or without colchicine (1 or 10 μM) [19, 20] for doseresponse analysis. Effect of Oxygen Deprivation In the second set of experiments, we exposed HL-1/H9C2 cells to anoxic conditions for 24 h to measure the effect of direct stress on cardiomyocytes. Later, colchicine (1 μM) was added for 1 h to half of the wells, and then recovery with full media containing 10% FBS was allowed for 16 h (n = 4 for each experimental well). Cell Morphology and Viability Analysis In both sets of experiments, cell morphology was assessed by light microscopy. Photos were taken using an Olympus DP73 camera and cellSense standard 1.6 software. The numbers of floating/ total cells were determined in two captures representing two plates/experimental group, each one divided into four fields. The cells were then stained with annexin V-propidium iodide (PI; Abcam, USA) and percent viable cells, i.e. cells which stained negative for both annexin V and PI, were determined by flow cytometry (BD Biosciences FACSCanto II). Western Blot Analyses Cells from experimental culture dishes were extracted using a commercial lysis buffer (Sigma). Equal protein amounts (50 μg) were loaded on a 4–20% acrylamide gel followed by electric transfer to nitrocellulose membranes (n = 4/arm). Following blocking with 5% low-fat milk diluted in TBS-Tween for 1 h, the membranes were incubated with the following antibodies: anti-PARP-1 (Santa Cruz, sc-7150), anti-cleaved caspase-3 (Sigma, C8487) or antiGAPDH (Abcam, clone 6C5) antibody used for validating equal loading. The primary antibodies were followed by blotting with HRP-conjugated secondary antibodies. After rapid incubation with an ECL substrate (Biological Industries, Israel), the membranes were exposed to an imaging film. Band intensity was determined using GelQuant.Net BiochemicalLabSolutions.com software. Statistical Analysis Groups were compared using one-way ANOVA. The Tukey post hoc correction was taken to account for multiple testing (IBM SPSS statistics 20). Results are expressed as means ± SE. p < 0.05 was accepted as statistically significant.
58
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
Margolis/Hertzberg-Bigelman/Levy/ Ben-Shoshan/Keren/Entin-Meer
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
The study was performed using the murine cardiomyocyte cell line HL-1 (a generous gift from the laboratory of Prof. W. Claycomb, LSU Health Sciences Center, New Orleans, La., USA) as well
40 20 0
a
Sham
MI 3 days MI 30 days
40 20 Sham
MI 3 days
Fig. 1. Colchicine (C) attenuates cellular viability of cardiomyoblasts (H9C2) but not of cardiomyocytes (HL-1) incubated with MI or sham-derived sera; mean values of cell viability (determined by flow cytometry with annexin V-PI) of HL-1 cells (a) and H9C2 cells (b). Prior to the analyses, cells were incubated with sera collected from rats 3 or 30 days after surgically induced MI or after
Results
Addition of Colchicine Decreases Viability in H9C2 Cells but Not in HL-1 Cells in the Presence of MI-Derived Sera Incubating HL-1 cells with MI-derived sera did not reduce cell viability in comparison to sham sera, as documented by flow cytometry: 47.3 ± 13% for 3-day MI, 41.9 ± 21% for 30-day MI and 32.9 ± 8.0% for sham controls (p = 0.16; fig. 1a). Adding colchicine to HL-1 cardiomyocytes exposed to MI or sham-derived sera did not significantly change their viability (fig. 1a). The relatively low percentage of annexin V-FITC-PI-negative cells, representing the viable HL-1 cells, may be attributed to the potential vulnerability of HL-1 cells to cell death due to their exposure to xenogenic sera, i.e. rat sera applied onto murine cells. In H9C2 cardiomyoblast cells, incubation with (3- or 30-day) MI-derived sera did not significantly change the percentage of viable cells in comparison to the sham controls, similar to data obtained for HL-1: 94.1 ± 0.5, 94.8 ± 0.5 versus 92.3 ± 0.5% for 3- and 30-day MI 30, and sham control, respectively (p > 0.1; fig. 1b). Interestingly, however, in H9C2, addition of colchicine to sera derived from post-MI animals (3 or 30 days) resulted in a significant (15%) increase in cell mortality compared to the same sera in the absence of colchicine: 94.1 ± 0.5 versus 80.3 ± 0.8% for 3-day MI without versus with colchicine, respectively (p < 0.05; fig. 1b); 94.8 ± 0.5 and 78.7 ± 0.9% for Effect of Colchicine on Cardiac Cells
*
60
0
b
*
MI 30 days
* **
**
90 80 70 60 50
c
Medium Medium +C
TNFį
TNFį +C
sham operation with or without subsequent addition of colchicine. c Bar graph represents cell viability of H9C2 grown in the presence of full DMEM with or without colchicine (1 or 10 μM) or in the presence of TNFα with or without colchicine. * p < 0.05, ** p < 0.01.
30-day MI without versus with colchicine, respectively (p < 0.05; fig. 1b). Likewise, addition of colchicine to H9C2 incubated in sham sera resulted in an increased mortality compared to the same sera in the absence of colchicine: 92.3 ± 0.5 versus 73.3 ± 3.1% without versus with colchicine, respectively (p < 0.05; fig. 1b). The detrimental nonspecific effect of colchicine on cell viability was confirmed in a third set of experiments in which colchicine was administered either at a concentration of 1 or 10 μM to H9C2 cells grown in regular DMEM. In both experiments, colchicine resulted in a significant reduction in the cell viability (p = 0.04) with no real difference between the two doses (reductions of 11 vs. 15.3% in the presence of 1 vs. 10 μM colchicine, respectively; p = 0.2). Equally important, adding colchicine to H9C2 cells which were kept for 24 h in the presence of TNFα (10 ng/ ml) resulted in an additive mortality rate (9.9% in regular DMEM, 15.9% in TNFα and 25.7% in TNFα + colchicine; p < 0.01; fig. 1c). Adding Colchicine to Oxygen-Deprived Cardiomyocytes Induces Apoptosis in H9C2 Cells but Not in HL-1 Cells Exposing both HL-1 cells and H9C2 cells to anoxic environment resulted in reduced viability of both cell types: 80.7 ± 11.3 versus 63.7 ± 15.4% for HL-1 cells in normoxic versus anoxic environment, respectively (p = 0.05; fig. 2a) and 81.5 ± 3.7 versus 68.3 ± 4.3% for H9C2 cells Cardiology 2016;134:57–64 DOI: 10.1159/000443369
59
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
60
*
80
100
Color version available online
Serum Serum + C
&HOO£YLDELOLW\£
&HOO£YLDELOLW\£
80
H9C2 cardiomyoblasts 100
+ &£FHOO£YLDELOLW\£
HL-1 cardiomyocytes 100
H9C2 cardiomyoblasts
*
100
£No C C
* 80 &HOO£YLDELOLW\£
60 40
0
10 5
Q3 103
Q4 10 4
Annexin FITC-A
Q1
103
Q3
102
10 5
52.5%
102
103
Q4 10 4
10 5
10 4
Q2
PI-A
103
A+C H9C2 11122014-A1 + C
10 5
10 4
Q2
PI-A
PI-A
Q1
102
Anoxia
0
H9C2 cardiomyoblasts N+C A H9C2 11122014-N1 + C H9C2 11122014-A1
10 5
c
b
Anoxia
N H9C2 11122014-N1
86.5%
Normoxia
40
Q1
103
Q3
102
10 5
Annexin FITC-A
10 4
Q2
72.7%
102
103
Q4 10 4
Annexin FITC-A
102
10 5
Q1
Q2
Q3
Q4
103
102
48.5%
103
10 4
10 5
FITC-A
Fig. 2. Colchicine (C) attenuates cell viability of cardiomyoblasts (H9C2) but not of cardiomyocytes (HL-1) exposed to anoxic environment; mean values of cell viability of HL-1 cells (a) and H9C2 cells (b) exposed to normoxic/anoxic environment with or without colchicine; * p < 0.05; ** p < 0.01. c Representative captures of
flow-cytometric data demonstrating the percent annexin V-FITCPI-negative H9C2 cells grown under normoxia (N) or anoxia (A) with or without colchicine. Cells which stained negative for both annexin-FITC and PI in flow cytometry (lower left quadrant) represent viable cells.
exposed to normoxic versus anoxic environment, respectively (p < 0.05; fig. 2b). Adding colchicine to HL-1 cells exposed to both normoxic or anoxic environment for 18 h did not significantly affect cell death: 68.8 ± 7.6 versus 80.7 ± 11.3% with versus without colchicine, respectively, in normoxic environment (p = 0.47; fig. 2a), and 61.2 ± 6.2 versus 63.7 ± 15.4% with versus without colchicine, respectively, in anoxic environment (p = 0.37; fig. 2a). On the other hand, adding colchicine to H9C2 cells kept in normoxia resulted in a significant (30%) decrease in viability: 50.8 ± 6.9 versus 81.5 ± 3.7% with versus without colchicine, respectively (p < 0.01; fig. 2b). Adding colchicine to H9C2 cells exposed to anoxic environment re-
sulted in an even greater increase in cardiac cell mortality: 42.3 ± 6.6 with versus 68.3 ± 4.3% without colchicine, respectively (p < 0.01; fig. 2b). Representative captures of flow-cytometric analyses of H9C2 cells grown under normoxia/anoxia with or without cholchicine are given in figure 2c.
60
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
Adherence to Tissue Culture Dishes after Incubation with Colchicine Is Reduced in H9C2 Cells but Not in HL-1 Cells Using light microscopy, we observed a marked reduction in the percentage of cells adherent to tissue culture plates in the presence of anoxia relative to normoxia in both H9C2 (fig. 3) and HL-1 (fig. 4) cells, in line with the Margolis/Hertzberg-Bigelman/Levy/ Ben-Shoshan/Keren/Entin-Meer
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
Normoxia
102
**
20
20
10 4
**
60
PI-A
&HOO£YLDELOLW\£
80
a
Color version available online
HL-1 cardiomyocytes 100
Color version available online
a
b Floating cells
c
d
e
flow-cytometric data given in figure 2. Nevertheless, regardless of the biological environment (normoxia vs. anoxia), treating H9C2 cells with colchicine resulted in detachment of approximately 50% of the cells from the tissue culture plates (21.7 ± 7.1% in normoxia vs. 51.1 ± 15.5% in normoxia + colchicine and 30.3 ± 13.7% in anoxia vs. 78.8 ± 7.3% in anoxia + colchicine; p < 0.05 for the colchicine effect and p < 0.05 for the lack of oxygen effect; fig. 3). This phenomenon did not apply to HL-1 cells (fig. 4). Induced Apoptosis in the Presence of Colchicine In order to assess whether the increased mortality of H9C2 cells in the presence of colchicine is mediated at least in part by apoptotic pathways, we subjected lysates of H9C2 cells grown under normoxia or anoxia with or without colchicine to Western blot analysis with two anEffect of Colchicine on Cardiac Cells
80
**
*
60 40 20 0
N
N+C
A
A+C
tibodies. The first one was anti-PARP-1, which recognizes both the full-length molecule (116 kDa) as well as a cleaved apoptotic fragment (89 kDa), which is produced upon activation of upstream caspase-3 in various cell types, including H9C2 [21, 22]. The second was an antibody directed to the cleaved form of caspase-3 (17 kDa). Representative data depicted in figure 5 demonstrated an increased expression of cleaved PARP-1 as well as cleaved caspase-3 in anoxia versus normoxia (p < 0.01). Incubation with colchicine induced further cleavage of PARP-1 under both normoxia and anoxia (p ≤ 0.05). The cleaved form of caspase-3 presented a trend towards a significant increase under normoxia (p = 0.07) and a significant upregulation under anoxia (p = 0.03). The data suggest that colchicine induced caspase-dependent apoptosis of H9C2 cells. Cardiology 2016;134:57–64 DOI: 10.1159/000443369
61
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
£ 3HUFHQW£IORDWLQJ£FHOOV
H9C2 cardiomyoblasts to culture dishes under normoxic/anoxic conditions (assessed by light microscopy); cellular morphology of H9C2 cells under normoxic (a) or anoxic conditions (b); H9C2 cells under normoxic conditions + colchicine (c) or anoxic conditions + colchicine (d). e The percentage of floating cells/total cells was calculated in four different fields in each experimental plate (two plates/treatment). A = Anoxia; C = colchicine; N = normoxia. * p < 0.05, ** p < 0.01.
*
100
Fig. 3. Colchicine reduces adherence of
Color version available online
a
b Floating cells
c
d
e
Discussion
In the current study, we have demonstrated differential effects of colchicine on cell viability of cardiac cells in an in vitro model simulating AMI versus normal physiological conditions. Interestingly, while colchicine attenuated cellular viability of H9C2 cardiomyoblasts, even when the cells were grown in full DMEM, the viability of HL-1 cardiomyocytes was not affected by this agent. Nevertheless, no salvage effect of colchicine on cell viability was observed. Moreover, surprisingly, colchicine per se exhibited a detrimental effect on the morphology of H9C2. In line with our data, a different study, which as62
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
25 20 15 10 5 0
N
N+C
A
A+C
sessed the cellular basis of the contractility abnormalities of feline myocytes in an experimental model of severe left ventricular hypertrophy (LVH), showed no significant difference in contractility properties between control and LVH myocytes after microtubule depolymerization by colchicine [23]. The authors suggested that the contractile abnormalities of feline LVH myocytes resulted from changes in cellular Ca2+ regulation and myofibrillar Ca2+ sensitivity, but not from changes in internal loading. These results were followed by reports from other authors [24], in which no effect on contraction dynamics by colchicine or Taxol, which induces microtubule polymerization, was observed. On the other hand, in recent studies, Margolis/Hertzberg-Bigelman/Levy/ Ben-Shoshan/Keren/Entin-Meer
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
£ 3HUFHQW£IORDWLQJ£FHOOV
duce adherence of HL-1 cardiomyocytes to culture dishes under normoxic/anoxic conditions (assessed by light microscopy); HL-1 culture morphology under normoxic (a) or anoxic conditions (b); HL-1 cells under normoxic conditions + colchicine (c) or anoxic conditions + colchicine (d). e The percentage of floating cells/total cells was calculated in four different fields in each experimental plate (two plates/treatment). A = Anoxia; C = colchicine; N = normoxia. * p < 0.05.
*
30
Fig. 4. Colchicine does not significantly re-
Color version available online
Full PARP-1 (116 kDa) Cleaved PARP-1 (89 kDa) GAPDH (37 kDa) Cleaved caspase-3 (17 kDa)
**
20 15
N+C
*
*
10 5 0
N
A A+C
N+C
A
A+C
c
N N+C
A A+C
**
30
*
25 20 15 10 5 0
N
N+C
A
A+C
Fig. 5. Colchicine induces caspase-dependent apoptosis in H9C2 cells. a Levels of cleaved relative to full-length PARP-1 as well as cleaved caspase-3 relative to the housekeeping protein GAPDH. b Bar graph for the mean percentage of cleaved PARP-1 out
of total PARP-1 (full length and cleaved). c Bar graph for the mean percentage of cleaved caspase-3 normalized to the levels of GAPDH. A = Anoxia; C = colchicine; N = normoxia. * p < 0.05, ** p < 0.01.
a cardioprotective role for microtubule depolymerization by colchicine was demonstrated in models of hearts exposed to chronic pressure overload [25, 26]. More specifically, a previous study has shown that reduction in microtubule hyperpolymerization by colchicine treatment reversed cardiomyocyte contraction abnormalities and suppressed apoptosis in HF induced by chronic pressure overload models [11]. Moreover, other authors have also reported cardioprotective effects against cell death in primary adult cardiomyocytes treated with proapoptotic agents such as TNFα [14]. These inconsistencies regarding the role of microtubules in HF could be explained by the observation that a substantial pressure overload is needed for the cytoskeletal change to happen [27]. Furthermore, one cannot rule out the possibility that primary cardiac cells may be affected differently by colchicine compared to H9C2 or HL-1 cells. Nevertheless, it was recently suggested that H9C2 could be successfully utilized as an in vitro model simulating cardiac ischemia/reperfusion injury [28].
In the current study, we observed enhanced cell death of cardiac cells incubated in sera collected from animals 30 days after MI and after treatment with colchicine. This observation may indicate that colchicine may not be effective to prevent late complications of AMI such as HF, in contrast to the benefits of its use in the pressure overload HF setting, as described before. Moreover, we assume that in the setting of AMI, the potential roles of microtubules are probably different from their role in chronic pressure overload. It has been shown that microtubules are reversibly disrupted in an ischemia/reperfusion model [29]. A recent study has demonstrated improved cardiomyocyte functional recovery and apoptosis rate in ischemia/reperfusion following exposure to Taxol – a microtubule stabilizer and polymerization promoter [30]. The differential effect of colchicine on H9C2 and HL-1 cells observed in our study could be explained by inherent metabolic differences between the two cell lines. It has been suggested that H9C2 cells are more sensitive to ischemia/reperfusion injury than HL-1 cells due to dif-
Effect of Colchicine on Cardiac Cells
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
63
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
b
N
Cleaved£FDVSDVH*$3'+£
Cleaved£3$53WRWDO£3$53£
a
ferent mitochondrial processes, explained in part by the presence of β-tubulin II in H9C2 cells and its absence in HL-1 cells [28]. Although we did not directly examine microtubular structural changes, these recent observations are consistent with our results, indicating that colchicine has no beneficial or even detrimental effect on cardiac cells in the AMI setting, which is in contrast to the cardioprotective effect shown before in pressure overload HF models by various investigators. We thus concluded that colchicine cannot inhibit myocardial apoptosis following AMI.
Study Limitation
In the experimental setting described in the current study, we were unable to test the potential immunosuppressive effects of colchicine on cardiac cell viability. This issue will be addressed in a future in vivo study, in which small animals subjected to AMI will be treated with colchicine and cardiac cell viability will be assessed by immunohistochemistry using apoptotic and necrotic markers.
References
64
11 Tagawa H, Wang N, Narishige T, Ingber DE, Zile MR, Cooper G 4th: Cytoskeletal mechanics in pressure-overload cardiac hypertrophy. Circ Res 1997;80:281–289. 12 Bhattacharyya B, Panda D, Gupta S, Banerjee M: Anti-mitotic activity of colchicine and the structural basis for its interaction with tubulin. Med Res Rev 2008;28:155–183. 13 Terkeltaub RA: Colchicine update: 2008. Semin Arthritis Rheum 2009;38:411–419. 14 Saji K, Fukumoto Y, Suzuki J, Fukui S, Nawata J, Shimokawa H: Colchicine, a microtubule depolymerizing agent, inhibits myocardial apoptosis in rats. Tohoku J Exp Med 2007; 213:139–148. 15 Deftereos S, Giannopoulos G, Angelidis C, et al: Anti-inflammatory treatment with colchicine in acute myocardial infarction: a pilot study. Circulation 2015;132:1395–1403. 16 Bonfoco E, Ceccatelli S, Manzo L, Nicotera P: Colchicine induces apoptosis in cerebellar granule cells. Exp Cell Res 1995;218:189–200. 17 Boyd AE 3rd, Bolton WE, Brinkley BR: Microtubules and beta cell function: effect of colchicine on microtubules and insulin secretion in vitro by mouse beta cells. J Cell Biol 1982; 92:425–434. 18 Ishibashi Y, Takahashi M, Isomatsu Y, et al: Role of microtubules versus myosin heavy chain isoforms in contractile dysfunction of hypertrophied murine cardiocytes. Am J Physiol Heart Circ Physiol 2003; 285:H1270– H1285. 19 Gomez AM, Kerfant BG, Vassort G: Microtubule disruption modulates Ca(2+) signaling in rat cardiac myocytes. Circ Res 2000;86:30– 36. 20 Kerfant BG, Vassort G, Gomez AM: Microtubule disruption by colchicine reversibly enhances calcium signaling in intact rat cardiac myocytes. Circ Res 2001;88:E59–E65. 21 D’Amours D, Sallmann FR, Dixit VM, Poirier GG: Gain-of-function of poly(ADP-ribose) polymerase-1 upon cleavage by apoptotic proteases: implications for apoptosis. J Cell Sci 2001;114:3771–3778.
Cardiology 2016;134:57–64 DOI: 10.1159/000443369
22 Youn HJ, Kim HS, Jeon MH, et al: Induction of caspase-independent apoptosis in H9C2 cardiomyocytes by Adriamycin treatment. Mol Cell Biochem 2005;270:13–19. 23 Bailey BA, Dipla K, Li S, Houser SR: Cellular basis of contractile derangements of hypertrophied feline ventricular myocytes. J Mol Cell Cardiol 1997;29:1823–1835. 24 de Tombe PP: Altered contractile function in heart failure. Cardiovasc Res 1998; 37: 367– 380. 25 Cheng G, Kasiganesan H, Baicu CF, Wallenborn JG, Kuppuswamy D, Cooper G 4th: Cytoskeletal role in protection of the failing heart by beta-adrenergic blockade. Am J Physiol Heart Circ Physiol 2012; 302:H675– H687. 26 Zhang C, Chen B, Guo A, et al: Microtubulemediated defects in junctophilin-2 trafficking contribute to myocyte transverse-tubule remodeling and Ca2+ handling dysfunction in heart failure. Circulation 2014; 129: 1742– 1750. 27 Cooper G 4th: Proliferating cardiac microtubules. Am J Physiol Heart Circ Physiol 2009; 297:H510–H511. 28 Kuznetsov AV, Javadov S, Sickinger S, Frotschnig S, Grimm M: H9C2 and HL-1 cells demonstrate distinct features of energy metabolism, mitochondrial function and sensitivity to hypoxia-reoxygenation. Biochim Biophys Acta 2015;1853:276–284. 29 Sato H, Hori M, Kitakaze M, et al: Reperfusion after brief ischemia disrupts the microtubule network in canine hearts. Circ Res 1993; 72:361–375. 30 Xiao J, Liang D, Liu Y, et al: Taxol, a microtubule stabilizer, improves cardiac functional recovery during postischemic reperfusion in rat in vitro. Cardiovasc Ther 2012;30:12–30.
Margolis/Hertzberg-Bigelman/Levy/ Ben-Shoshan/Keren/Entin-Meer
Downloaded by: University of Illinois at Chicago 198.143.32.33 - 2/24/2016 2:39:02 PM
1 Go AS, Mozaffarian D, Roger VL, et al: Heart disease and stroke statistics – 2013 update: a report from the American Heart Association. Circulation 2013;127:e6–e245. 2 Gonzalez A, Fortuno MA, Querejeta R, et al: Cardiomyocyte apoptosis in hypertensive cardiomyopathy. Cardiovasc Res 2003; 59: 549–562. 3 Kang PM, Haunstetter A, Aoki H, Usheva A, Izumo S: Morphological and molecular characterization of adult cardiomyocyte apoptosis during hypoxia and reoxygenation. Circ Res 2000;87:118–125. 4 Yang B, Ye D, Wang Y: Caspase-3 as a therapeutic target for heart failure. Expert Opin Ther Targets 2013;17:255–263. 5 Depre C, Taegtmeyer H: Metabolic aspects of programmed cell survival and cell death in the heart. Cardiovasc Res 2000;45:538–548. 6 Aikawa R, Komuro I, Yamazaki T, et al: Oxidative stress activates extracellular signal-regulated kinases through Src and Ras in cultured cardiac myocytes of neonatal rats. J Clin Invest 1997;100:1813–1821. 7 Krown KA, Page MT, Nguyen C, et al: Tumor necrosis factor alpha-induced apoptosis in cardiac myocytes. Involvement of the sphingolipid signaling cascade in cardiac cell death. J Clin Invest 1996;98:2854–2865. 8 Chen QM, Tu VC: Apoptosis and heart failure: mechanisms and therapeutic implications. Am J Cardiovasc Drugs 2002;2:43–57. 9 Hein S, Kostin S, Heling A, Maeno Y, Schaper J: The role of the cytoskeleton in heart failure. Cardiovasc Res 2000;45:273–278. 10 Tagawa H, Koide M, Sato H, Zile MR, Carabello BA, Cooper G 4th: Cytoskeletal role in the transition from compensated to decompensated hypertrophy during adult canine left ventricular pressure overloading. Circ Res 1998;82:751–761.
