Citation
Pijnappels, D. A. (2009, June 18). Electrophysiological deterioration and resurrection in the scarred heart. Retrieved from
https://hdl.handle.net/1887/13851
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/13851
Note: To cite this publication please use the final published version (if applicable).
V
Chapter
P rogressive increase in conduction velocity across human mesenchymal stem cells is mediated by enhanced electrical coupling
Daniël A. Pijnappels,1 Martin J. Schalij,1 John van Tuyn,1,2 Dirk L. Ypey,1 Antoine A. F.
de Vries,2 Ernst E. van der Wall,1 Arnoud van der Laarse,1 Douwe E. Atsma1
Departments of Cardiology,1 and Molecular Cell Biology,2 Leiden University Medical Center, Leiden, the Netherlands
Cardiovasc Res. 2006;72:282-91
V
Chapter
P rogressive increase in conduction velocity across human mesenchymal stem cells is mediated by enhanced electrical coupling
Daniël A. Pijnappels,1 Martin J. Schalij,1 John van Tuyn,1,2 Dirk L. Ypey,1 Antoine A. F.
de Vries,2 Ernst E. van der Wall,1 Arnoud van der Laarse,1 Douwe E. Atsma1
Departments of Cardiology,1 and Molecular Cell Biology,2 Leiden University Medical Center, Leiden, the Netherlands
Cardiovasc Res. 2006;72:282-91
Abstract
Objective: Purpose of the study was to investigate the development of electrical trans- mission across human adult bone marrow-derived mesenchymal stem cells (hMSCs) during long-term co-incubation with cardiomyocytes (CMCs).
Methods: Neonatal rat CMCs were cultured in multi-electrode array (MEA) dishes.
A conduction block was induced by creating a central a-cellular channel, yielding two asynchronously beating CMC fields. Enhanced Green Fluorescent Protein (eGFP)- labeled hMSCs from ischemic heart disease patients (n=8), eGFP-labeled hMSCs having RNA interference-mediated connexin43 (Cx43) knock down (n=6), 1,1’-dioc- tadecyl-3,3,3’,3’-tetramethylindocarbocyanine (Dil)-labeled CMCs (n=6), or no cells (n=9) were seeded in the channel. Assessment of conduction velocity (CV), Cx ex- pression and localization, gap junctional coupling, and intracellular electrical recor- dings were performed for up to 14 days.
Results: Resynchronization of the two CMC fields occurred within 24 h after see- ding of hMSCs. CV across hMSCs increased from 1.4±0.4 cm/s at day 7 to 3.5±0.1 cm/s (p<0.05) at day 14. CV across seeded CMCs was 16.8±0.2 cm/s throughout this period. No resynchronization occurred in the absence of seeded cells. Knock down of Cx43 in hMSCs abolished conduction across the channel completely. Time-de- pendent increase of CV across hMSCs was associated with increased Cx43 mRNA and protein expression resulting in increased gap junctional coupling. Intracellular recordings in coupled hMSCs showed increased conducted action potential (AP) am- plitude, lower resting membrane potential, and decreased duration of conducted AP (p<0.05, day 14 versus day 1).
Conclusions: CV across hMSCs increases progressively after 7 days of co-incubation with CMCs, most likely via improved electrotonic interaction. This is associated with increased Cx43 expression, increased functional gap junctional coupling and enhan- ced intercellular electrical coupling.
Introduction
C
ell therapy is a novel treatment option for ischemic heart disease such as acute myocardial infarction1 and chronic ischemia.2 Stem cells3 or progenitor cells4 are able to enhance function of ischemic myocardium through improvement of myocar- dial perfusion and/or contractile performance. However, structural and functional integration of injected cells into host myocardium is crucial to achieve a beneficial therapeutic outcome. Recently, injection of skeletal myoblasts into damaged myo- cardium of heart failure patients was found to be associated with the occurrence of life-threatening ventricular arrhythmias.5 Lack of electrical coupling between injec- ted skeletal myoblasts and resident cardiomyocytes (CMCs) most likely produced a pro-arrhythmogenic substrate.6,7 The failure of injected skeletal myoblasts to couple to host CMCs is considered to be caused by an absent expression of connexin 43 in skeletal myoblasts, the major connexin isoform in gap junctions of working myocar- dium. The essential role of intact gap junctional coupling between CMCs in the pre- vention of arrhythmias is further demonstrated by the occurrence of life-threatening arrhythmias during pathophysiological states (such as myocardial ischemia), resul- ting in impaired gap junctional function.8Gap junctions form low-resistance intercellular pathways allowing intercellular transport of low molecular traffic (up to 1 kD) and conduction of electrical impulses thereby ensuring coordinated propagation of action potentials across the myocardi- um.9 Gap junctions contain connexin40 (Cx40), connexin43 (Cx43), and connexin45 (Cx45), predominantly localized in the atria, ventricles, and conduction system, res- pectively.
Human mesenchymal stem cells (hMSCs) might differentiate into CMCs in vitro and in vivo and have been used in a number of clinical stem cell trials.10 In addition, isola- ted autologous MSCs have already been used in a small scale clinical trial in patients with ischemic heart disease, with favorable initial results.11 However, detailed long- term electrophysiological characterization of these hMSCs in the context of persi- stent co-incubation with host CMCs is not yet available.
Recently, we reported that adult hMSCs are able to conduct action potentials bet- ween two fields of CMCs divided by an experimental conduction block, thereby syn- chronizing these two fields for at least 48 h.12
In the present report, we studied the development of electrical transmission across hMSCs during 14 days of co-incubation with CMCs in a model of experimental con- duction block.
Materials and Methods
Preparation of Primary Neonatal Cardiomyocytes Cultures
All animal experiments were approved by the Animal Experiments Committee of the Leiden University Medical Center and conform to the Guide for the Care and Use of Laboratory Animals according to the US National Institutes of Health.
Cultures of neonatal rat ventricular CMCs were prepared as described previously.13 Briefly, hearts were dissected aseptically from anesthetized, 2-day-old male Wistar rats, from which the ventricles were minced and dissociated using collagenase and DNase. The cells were suspended in Ham’s F10 medium (ICN Biomedicals, Irvine, CA, USA) with 10% horse serum (HS, Invitrogen, Carlsbad, CA, USA) and 10% fe- tal bovine serum (FBS, Invitrogen), and pre-plated to allow preferential attachment of non-CMCs. After 1 h, the non-adherent cells (predominantly CMCs) were col- lected and plated (1.5x106 cells/well) on multi-electrode arrays (MEA, Multi Chan- nel Systems, Reutlingen, Germany) or onto collagen-coated glass coverslips in 6-well culture dishes (2x106 cells/well) and incubated in a humidified incubator at 37°C and 5% CO2. Prior to plating of CMCs on the MEAs and glass coverslips, the hydrop- hilic character of the surface was increased by glow-discharge treatment (K950X, Emitech, Ashford, UK), which improved long-term cell attachment. In addition, the MEAs were pre-coated for 24 h with DMEM/Ham’s F10 (1:1) containing 10% FBS and 10% HS. Furthermore, CMCs were collected in culture flasks and labeled with the viable fluorescent dye CM-DiI (CellTracker®, Molecular Probes, Eugene, OR, USA).
Overgrowth of residual cardiac fibroblasts was prevented by incubation with 100 μM 5-bromo-2-deoxyuridine (BrdU, Sigma, Saint Louis, MO, USA) for 24 h, in a 1:1 mix- ture of DMEM (Invitrogen) and Ham’s F10 supplemented with 5% HS, penicillin (100 U/ml) and streptomycin (100 µg/ml). After 3 days, a confluent spontaneously beating monolayer of CMCs was present.
Harvesting and Preparation of Bone Marrow-Derived Human Adult Mesenchy- mal Stem Cells
Bone marrow samples were obtained from four adult ischemic heart disease patients (no further conventional treatment options, aged 60-75 years) scheduled for cardiac stem cell therapy. hMSCs were purified and characterized as described previously.14 The institutional ethical committee approved this therapy, and the patients had given informed consent. The present study is conform the Declaration of Helsinki for use of human tissue or subjects.
hMSCs were expanded by serial passage, and used from passages 3 to 8 in culture conditions. To ensure identification, hMSCs were transfected with a first generation
adenoviral vector encoding enhanced green fluorescent protein (eGFP; hAd5/F50.
CMV.eGFP). For immunostaining, hMSCs were labeled with CM-DiI.
Induction and Restoration of Experimental Conduction Block
Multi-electrode high density mapping of cultured CMCs was performed using a multi-electrode array (MEA) data acquisition system. Standard planar MEAs con- taining 60 titanium nitride electrodes (inter-electrode distance: 200 μm; electrode diameter: 30 μm) allowed simultaneous recording of extracellular electrograms from all electrodes (one electrode served as reference electrode) at a sample rate of 5 kHz.
Electrograms were analyzed off-line using MC-Rack software (version 3.2.2.0, Multi Channel Systems).
Local activation time was measured at the maximal negative intrinsic deflection (-dV/
dtmax). The local activation time values were used to construct two-dimensional co- lor-coded activation maps using appropriate software (S-Plus, version 6.2, Insightful Corp., Seattle, WA, USA).
Activation maps were generated after 2 days of culture to confirm the presence of a synchronously beating monolayer. After assessing impulse propagation, a conduction block was generated as described previously.19 Briefly, by manual abrasion a 250-350 μm wide a-cellular channel was created across the center of the 60 electrodes, exten- ding over the entire length of the MEA culture dish and perpendicular to the excita- tion wave front, thereby dividing the cell culture into two fields. After ensuring that no cells nor cell debris were present in the channel and after ensuring the presence of conduction block between the two CMC fields, either 30x103 eGFP-labeled hMSCs or 30x103 CM-Dil-labeled CMCs were applied in a channel-crossing pattern. In a sepa- rate series of experiments, channels in MEA culture dishes were left empty.
During the following 14 days, impulse propagation across seeded cells was assessed daily. The two separated and asynchronously beating culture fields were considered electrically coupled if the timing of the electrograms of the two fields correlated con- sistently with each other for 30 consecutive local activation times recorded at elec- trodes in the upper fields plotted against local activation times recorded at electrodes in the lower fields.
Reverse Transcription-Polymerase Chain Reaction Analysis
Co-cultures of CMCs and hMSCs were prepared in a configuration similar to that in MEA culture dishes. Total RNA was extracted from the cells after 1, 7, and 14 days of co-culture using RNAeasy (Qiagen, Valencia, CA, USA). cDNA was synthesized in 50 µl volumes using 2 µg of RNA, 0.25 µg of random hexanucleotides, 25 nmol of dNTPs, and 500 U of Superscript II RNase H-reverse transcriptase (Invitrogen).
To amplify human mRNAs, 1 µl of cDNA was subjected to PCR using 2.5 U of Su- perTaq (Fermentas, Hanover, MD, USA) and 10 pmol of human-specific primer pairs listed in Table 1.
The amplification scheme consisted of a 2 min incubation at 94°C, followed by 25 to 35 cycles of 15 s at 94°C (melting), 30 s at 60 to 64°C (annealing), and 30 s at 72°C (extension). For standardization purposes, RT-PCR of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was performed in parallel. RNA samples from human atrial and human ventricular myocardium were also subjected to RT-PCR to provide po- sitive controls and to determine the specificity of the various marker genes assayed.
All primers were chosen to hybridize to human cDNA only. PCRs carried out with water instead of cDNA served as negative controls. The PCR products were separated in 1.5% agarose gels containing 1 µg/ml ethidium bromide and visualized with a Gel Doc 2000 digital imaging system (BioRad, Hercules, CA, USA) and Scion Image Beta (Version 4.02, Scion Corporation, Frederick, MD, USA).
Immunofluorescence Microscopy
Co-cultures of CMCs and CM-DiI-labeled hMSCs, cultured on collagen-coated glass coverslips in a configuration similar to that in the MEA culture dishes, were subjected to immunostaining at day 1, 7, and 14 after application of hMSCs. The cells were fixated and permeabilized at 4⁰C in PBS-1% formalin (30 min) (Merck, Darmstadt, Germany) and PBS-0.1% Triton X-100 (30 min) (BDH Laboratories, Poole, England), respectively. Next, the cells were labeled with one of the following antibodies, goat anti-Cx40, rabbit anti-Cx43, goat anti-Cx45, or goat anti-SCN5A (Santa Cruz Biotechnology, Santa Cruz, CA, USA) at a dilution of 1:200 in PBS-1%
FBS at 4⁰C for 24 h. Excess primary antibody was removed by a triple wash in PBS-1%
FBS, and the cells were stained with either FITC-conjugated anti-goat IgG (Sigma),
Table 1. Human specific PCR primers used to examine mRNA expression of Cx40, 43, 45, GATA4, SCN5A and GAPDH. (30 cycles, annealing temperature 60°C).
Alexa-conjugated anti-rabbit or Alexa-conjugated anti-goat IgG (Molecular Probes), at dilutions of 1:100 in PBS-1% FBS at 4⁰C for 1 h. After three washes with PBS-1%
FBS, the cells were incubated with Hoechst 33342 (Molecular Probes) at 1 µg/ml in PBS-1% FBS for 5 min. After washing three times with PBS-1% FBS, the coverslips were mounted onto glass slides in Vectashield (Vector Laboratories, Burlingame, CA, USA). Examination of the slides was performed using a fluorescence microscope equi- pped with a digital camera (Eclipse, Nikon Europe, Badhoevedorp, The Netherlands).
The signal intensities recorded from the samples were quantified using Image-Pro Plus (Version 4.1.0.0, Media Cybernetics, Silver Spring, MD, USA). All co-cultures of CMCs and CM-DiI-labeled hMSCs were stained on the same day, using the same solutions and exposure times. Quantifications were performed by an independent investigator.
Cx43 knockdown in hMSCs by RNA interference
hMSCs were plated in a 6-well culture dish at densities of 1x105 per well and trans- fected with self-inactivating third-generation lentiviral vectors coding for enhanced green fluorescent protein (eGFP) and for recombinant microRNAs directed against specific mRNAs. These vectors were named LV.SM2C.Cx43.hPGK.eGFP, targeting Cx43, and LV.SM2C.pLuc.hPGK.eGFP, targeting pLuc (firefly luciferase, control). The lentiviral vectors contained commercially available short hairpins (Open Biosystems, Huntsville, AL, USA). Cx43: TGCTGTTGACAGTG-AGCGAAGGTGCATGTTGG- TATTTAAATAGTGAAGCCACAGATGTATTTAAATACCAACATGCACCTCT- GCCTACTGCCTCGA, and pLuc: CTTACGCTGAGTACTTCGA. These vectors were produced using the pSHAG-MAGIC2 Vector (Open Biosystems), pEGFP (Clontech, Mountain View, CA, USA), pGL3.hPGK, and pLV.MCS. Transfections were performed at multiplicity of infection (MOI) 0, 4, 16, and 32, and western blot analysis was performed using rabbit anti-Cx43 antibodies (Santa Cruz Biotechnolo- gy) to test the efficiency of LV.SM2C.Cx43.hPGK.eGFP. The control vector LV.SM2C.
pLuc.hPGK.eGFP was tested for selectivity and efficiency using a luminescence assay (Steady-Glo® Luciferase Assay System, Promega, Madison, WI, USA). hMSCs were transfected for 4 h, and after 4 days they were trypsinized, counted, and applied in the channel between the two CMC fields in the MEA culture dish. Electrical current transmission across the channel in the MEA culture dish was recorded at 24 and 48 h after hMSC application, and from the local activation times, activation maps were drawn.
Gap Junctional Uncoupling between hMSCs by Carbenoxolone
Carbenoxolone (Sigma), a reversible gap junction uncoupler, was dissolved in PBS (10 mM) and final concentrations were made using culture medium (5 μM-1600 μM).
After restoration of conduction was confirmed, carbenoxolone was applied to the MEA cultures for 15 min. As control experiment, only culture medium was added.
The effect of a certain concentration of carbenoxolone on conduction velocity (CV) across the channel seeded with hMSCs was evaluated using the MEA data acquisi- tion system. Next, the cultures were washed twice, cells were incubated with culture medium for 25 min, and a higher dose of carbenoxolone was applied. This series of steps was repeated until a dose of carbenoxolone completely blocked transmission across the channel. Experiments were repeated at day 14 using the same MEA culture dishes.
Patch-Clamp Technique
CMCs and CM-DiI-labeled hMSCs were co-cultured on collagen-coated glass co- verslips in a conformation similar to that in the MEA culture dishes. After identifi- cation of hMSCs using fluorescence microscopy, action potentials from CMCs and conducted action potentials from hMSCs were recorded by glass patch-electrodes.15 Whole-cell recordings were performed at day 1, 7, and 14 after application of hM- SCs, at 25°C using a L/M-PC patch-clamp amplifier (3 kHz filtering) (List-Medical, Darmstadt, Germany). The pipette solution contained (in mM) 10 Na2ATP, 115 KCl, 1 MgCl2, 5 EGTA, 10 HEPES/KOH (pH 7.4). Tip resistance was 2.0-2.5 MΩ, and seal resistance >1 GΩ. The bath solution contained (in mM) 137 NaCl, 4 KCl, 1.8 CaCl2, 1 MgCl2, 10 HEPES (pH 7.4). For data acquisition and analysis pClamp/Clampex8 software (Axon Instruments, Molecular Devices, Sunnyvale, CA, USA) was used. We analyzed (transmitted) action potential characteristics in CMCs and hMSCs at day 1, 7, and 14.
Statistics
Statistical analysis was performed using SPSS 11.0 for Windows (SPSS Inc., Chicago, IL, USA). Data were compared with Student’s t-test, ANOVA test or the two-sided chi-square test when appropriate, and expressed as mean±SD. P-values <0.05 were considered statistically significant.
Results
Restoration of Conduction Block by hMSCs
After two days of culture, a spontaneously and synchronously beating monolayer of CMCs was present in the MEA culture dishes (Figure 1B-1).
Figure 1. (A) Electrograms recorded in a 60-electrode grid using the multi electrode array data acquisi- tion system. (B) Time needed for the electrical current to travel over a distance of 4 electrodes (600 μm) was calculated at baseline, after creation of conduction block, 1 and 14 days after application of hMSCs.
(C) Typical activation maps from cell culture at baseline, after creation of conduction block and 1 and 14 days after hMSC application. The activation map at day 14 shows more color segments in the two CMC- fields, than at day 1, because of the decreased traveling time across hMSCs in the channel. (D) Correlation between local activation time above and local activation time below the channel. The tight correlation in first, third and fourth panel indicates synchronized beating. Second panel shows asynchrony between beating in the upper and lower culture halves.
The presence of a conduction block (280±25 μm wide) between the two CMC fields resulted in asynchronized beating (Figure 1B-2). Asynchronously beating CMC fields were resynchronized after applying hMSCs (n=8) in MEA culture dishes within 24 h (Figure 1B-3). Restoration of conduction was determined by tight correlation of local activation times at both CMC fields separated by the channel seeded with hMSCs, from which activation maps were created (Figure 1C, D). Cultures were used in long- term experiments if spontaneously and synchronously beating monolayers of CMCs were firmly attached to the MEA culture dish, as judged by gross appearance and by quality of the recorded electrograms. In general, 1 out of 3 MEA culture dishes could be used. After 14 days, the cultures spontaneously detached from the culture dish precluding further analysis.
Twenty four h after seeding of hMSCs in the channel, the CV across the hMSCs was 1.2±0.2 cm/s. CV was relatively stable during the first 6 days after which CV incre- ased progressively, reaching 3.5±0.1 cm/s at day 14 (Figure 2).
-
Immunofluorescence Microscopy
Connexin 43 (Cx43) staining was present in-between adjacent CMCs, adjacent hMSCs, and CMCs adjacent to hMSCs. Expression of Cx40 and Cx45 was very low (2-3% compared to Cx43 staining) or undetectable in-between adjacent hMSCs and CMCs adjacent to hMSCs. However, hMSCs showed a cytoplasmic distribution pat-
Figure 2. Temporal development of electrical conduc- tion velocity (CV) across hMSCs connecting two fields of CMCs (lower panel). CV across CMCs was stable for 14 days (upper panel), whereas CV across hMSCs increased gradually from day 7 onward (*p<0.05 vs day 7).
No migration of CMCs was ob- served in this period in any of the MEA culture dishes, as de- termined by the absence of any other cell type than GFP-labeled hMSCs. In addition, MEAs with- out cells seeded in the channel (n=9) all remained asynchroni- zed throughout the follow up.
After seeding CMCs in the chan- nel (control, n=6), the cultures were resynchronized at 24 h. CV across CMC was 16.8±0.2 cm/s at day 1 and did not change sig- nificantly throughout the follo- wing days (Figure 2).
tern of Cx40 and Cx45. Cx43 expression did not change between day 1 and 7, but was significantly increased at day 14 (Student’s t-test, Figure 3). This increased Cx43 staining was present between adjacent CMCs, adjacent hMSCs, and between CMCs adjacent to hMSCs. The punctuated pattern of Cx43 expression in adjacent hMSCs did not change over time, but remained relatively uniformly distributed. For quan- titative analysis of immunohistochemistry, 24 random samples were taken from 8 adjacent cell pairs (CMCs, CMC-hMSC, hMSCs), from 3 coverslips at each time- point. Typical examples of adjacent cell pairs are shown in Figure 3. Cx43 staining was considered to be positive as Cx43 signal intensity was exceeding the threshold value of 50 on the 0-255 gray intensity scale, and values were averaged for each type of cell pair and time-point. During the first 14 days SCN5A staining was observed in CMCs, but not in hMSCs.
Reverse Transcription-Polymerase Chain Reaction Analysis
To assess gene expression of Cx40, Cx43, Cx45, SCN5A and GATA4 in hMSCs, RNA was extracted from cells at day 1, 7, and 14 of co-incubation.
Figure 3. Upper panel shows immunofluorescence images of hMSCs stained with DiI (red). Expression of Cx43 (green) is shown at day 1, 7, and 14 of co-incubation. Significant increases in Cx43 expression were seen between adjacent CMCs, MSCs, and CMCs to MSCs at day 14 as compared to day 1 and 7. No time-dependent changes in local distribution of Cx43 in hMSCs were found. Lower panel presents the fluorescent signal, expressed as the percentage fluorescent signal compared to day 1 (100%). As compared to day 1, Cx43 expression was not increased at day 7, but did increase at day 14 (*p<0.05 vs day 7).
Using human-specific primers, analysis of mRNA encoding Cx43 by RT-PCR showed a 30% increase at day 14 compared to day 1 and 7. RT-PCR products specific for Cx40 or Cx45 were hardly detectable, and no time-dependent effect was noticed. No ex- pression of SCN5A or GATA4 was detected in hMSCs at day 1 nor at day 14.
Effect of Cx43 Knockdown in hMSCs on Conduction Block Restoration
Cultured hMSCs were subjected to western blot analysis 4 days after transfection with LV.SM2C.Cx43.hPGK.eGFP or the control vector LV.SM2C.pLuc.hPGK.eGFP.
Cx43 protein expression in hMSCs transfected by LV.SM2C.Cx43.hPGK.eGFP sho- wed a vector dose-dependent decrease from MOI 0 to 32 (Figure 4).
In contrast, no knockdown of Cx43 protein expression took place after transfection by LV.SM2C.pLuc.hPGK.eGFP. Next, hMSCs transfected with LV.SM2C.Cx43.hPGK.
eGFP at MOIs ranging from 0 to 32 were applied in the channel in the MEA culture dish dividing the two CMC fields. Resynchronization of the two CMC fields occurred 24 h after application of hMSCs transfected at MOI 0 (control, n=4) and MOI 4 (n=4,
~10% knockdown), and was still present after 48 h. However, no resynchronization occurred after application of hMSCs transfected at MOI 16 (n=4, ~70% knockdown) or 32 (n=4, ~95% knockdown) after 24 h nor 48 h. All CMC fields were resynchroni- zed after application of hMSCs transfected by LV.SM2C.pLuc.hPGK.eGFP (MOI 0, 4, 16, and 32, n=4 in all cases), with no significant differences in conduction velocity across transfected hMSCs at 24 or 48 h after application.
Figure 4. Western blot analysis of connexin43 (Cx43) expression after transfection of hMSCs with LV.SM2C.Cx43.hPGK.eGFP, or LV.SMC.pLuc.hPGK.eGFP. Cx43 protein level is dose dependently decre- ased in hMSCs transfected by the lentiviral vector LV.SM2C.Cx43.hPGK.eGFP, encoding for recombinant microRNAs directed against Cx43. Cx43 protein expression decreased by 8, 70, and 95% at multiplicity of infection (MOI) 4, 16 and 32, respectively, compared to control (MOI 0). Cx43 protein levels did not change in hMSCs transfected by the control vector LV.SMC.pLuc.hPGK.eGFP, that targeted pLuc.
Effect of Gap Junctional Uncoupling between hMSCs on Conduction Block Res- toration
Carbenoxolone was added to MEA culture dishes at day 1 and day 14 after application of hMSCs. At day 1, hMSCs were still electrically coupled after 15 min of incubation with carbenoxolone at concentrations from 5 till 100 μM. However, electrical trans- mission across hMSCs was abolished after incubation with carbenoxolone at con- centration 200 μM or higher (IC50=118±57 μM), leading to asynchronously beating CMC fields (n=6) (Figure 5).
At day 14, electrical transmission across hMSCs was still present after incubation with 200 μM carbenoxolone. Only a higher concentration of carbenoxolone (250 μM) did abolish electrical transmission across hMSCs (n=4). CV) across hMSCs and CMCs returned to baseline after a washout of 25 min, associated with resynchronization of the two CMC fields. CV across CMCs also decreased dose dependently, although electrical conduction was still present at 1.6 mM of carbenoxolone (IC50=187±29 μM). Assessment of the extracellular electrograms from hMSCs and CMCs at day 1 and day 14 showed a decrease in total amplitude and maximal upstroke velocity (dV/
dTmax) at increasing concentrations of carbenoxolone (Table 2). Incubation with cul- ture medium for 15 min did not affect CV across hMSCs or CMCs.
Figure 5. Effect of gap junctional uncoupling on conduction velocity (CV) across hMSCs and CMCs 1 day after application of hMSCs. Carbenoxolone dose dependently decreased CV across hMSCs and CMCs.
Electrical transmission across hMSCs was abolished at 200 µM, accompanied by asynchronously beating CMC fields as the conduction block was reestablished. After a washout step CV returned to baseline values, resulting in resynchronized beating again. *p<0.05 vs 100 mM.
Intracellular Electrical Recordings
Intracellular electrical recordings were performed in co-cultures of hMSCs and CMCs in a configuration similar to that in the MEA culture dishes at day 1-2 (hMSCs, n=6;
CMCs, n=4), day 6-7 (n=6, n=4), and day 13-14 (n=6, n=4).
Adjacent hMSCs and CMCs remained electrically coupled for at least 14 days, as- sociated with transmission of action potentials from the CMCs across the hMSCs (Figure 6). No significant differences were found in the characteristics of conducted action potentials recorded in hMSCs between day 1 and day 7. At day 14, however, hMSCs had a more negative resting potential (-40±10 mV), a larger action potential amplitude (54±11 mV), and shorter action potential duration until 50% repolarization (330±40 ms) than at day 7 (-23±5 mV, 31±6 mV and 500±40 ms, respectively).
Figure 6. Typical tracings from patch clamp experiments showing conducted action potentials across hMSCs at day 1, 7, and 14 after application of the hMSCs in the channel between two CMC fields. Con- ducted action potential amplitude recorded in hMSCs was increased at day 14 as compared to day 1 and 7, but still lower as compared to CMCs. Furthermore, resting membrane potential of hMSCs decreased at day 14 as compared to day 1 and 7, but was higher than the resting membrane potential in CMCs.
Table 2. Effect of carbenoxolone on extracellular electrograms from hMSCs and CMCs at day 1 and 14 after co-culture. Total Amp= total amplitude, dV/dTmax= maximal upstroke velocity of electrogram.
*p<0.05 lower vs higher carbenoxolone concentration.
Action potential characteristics of CMCs remained constant during the first 14 days (Table 3). From voltage-step responses of CMC-coupled hMSCs we derived the input resistance, as an over-estimate of the coupling resistance of 70±20 MΩ (n=4, day 2 of co-culture), which remained in that range throughout the follow up.
Discussion
Key findings of this study are: 1) Bone marrow-derived MSCs from patients with ischemic heart disease conduct an electrical signal over a considerable distance for at least 14 days, thereby resynchronizing two separated fields of CMCs; 2) Conduction velocity across hMSCs increased 3-fold during follow-up, which was accompanied by an increase in Cx43 expression (mRNA and protein) between hMSCs and CMCs, as well as between coupled hMSCs. Knock-down of Cx43 in hMSCs with RNAi abolis- hed transmission across hMSCs; and 3) The time-dependent increase of Cx43 expres- sion was related to an increase in functional gap junctional coupling as a higher dose of carbenoxolone, a reversible gap junction uncoupler, was needed to block electrical transmission across hMSCs at day 14 as compared to day 1. In addition, intracellular recordings in coupled hMSCs showed increased action potential amplitude, lower resting membrane potential, and decreased action potential duration until 50% re- polarization of conducted action potentials at day 14 compared to day 1 or day 7.
This indicates good coupling between the hMSCs and CMCs. However we found no evidence for excitability of the hMSCs at day 1 nor at day 14 of co-culture, compatible with absence of SCN5A staining of hMSCs.
These results highlight the ability of hMSCs to adapt to an electrically active environ- ment, thereby increasing their electrical compatibility with host CMCs. The impor- tance of electrical coupling between donor and host cells was demonstrated by studies showing that electrical isolation between certain types of transplanted cells, such as
Table 3. Electrophysiological parameters from hMSCs and CMCs at day 1, 7, and 14 after co-culture.
*P<0.05, day 1 vs day 14. †P<0.05, day 7 vs day 14.
skeletal myoblasts, and host CMCs created a potential arrhythmogenic substrate.7-9 The absence of Cx43 is considered to be one of the most important factors causing this electrical incompatibility. Other studies have shown the presence of Cx43 bet- ween donor and host CMCs, indicating the presence of gap junctions16 and inter- calated discs.17 However, so far direct assessment of functional electrical cell-to-cell coupling in vivo has been technically impossible.
Time-Dependent Changes in Electrophysiological Parameters
Electrical transmission across the hMSCs in the channel occurred within 24 h after seeding, and improved during the second week. Detachment of the cultures after 14 days precluded a longer observation period. It is unlikely that hMSCs were acti- vely involved in action potential propagation, as no SCN5A expression was found in hMSCs at day 1 nor at day 14, suggesting that the hMSCs were non-excitable.
CV across CMCs is in agreement with previously reported data,18 although no long- term data concerning CV across CMCs cultured for 14 days are available. Further- more, other CMC characteristics as action potential amplitude, resting membrane potential and amplitude duration till 50% were in agreement with previously publis- hed data.19 Apart from our previous study,14 no other studies focused on CV across hMSCs connected to CMCs. However, electrical coupling was demonstrated hours after formation of pairs of hMSCs and adult canine CMCs, associated with immature gap junctional coupling.20
Role of Gap Junctions and Ion Channels in Conduction Velocity
In cardiac tissue, CV is determined by cell-to-cell coupling, tissue architecture, and excitability. Our data showed that cell-to-cell coupling between CMCs and hMSCs, and between adjacent hMSCs improves progressively, as demonstrated by changes in membrane potentials, increase in Cx43 expression, and increased functional gap junctional coupling. Intracellular measurements from hMSCs pro- vided a coupling resistance of the order of 70 MΩ, indicating good coupling of hMSCs with CMCs.
In this study, we found Cx43, the major gap junction protein present in working myo- cardium, to be essential for electrical coupling by hMSCs to CMCs. The increase in CV observed during the first 14 days was closely accompanied by higher expres- sion of Cx43 at protein as well as mRNA level at day 14, as compared to correspon- ding measurements at day 1 or 7. RNAi-mediated knock-down of Cx43 (MOI 32) in hMSCs was associated with a ~95% reduction in protein levels. This reduction resulted in conduction block across hMSCs, again highlighting the crucial role of Cx43 in establishing and maintaining electrical transmission. Studies using hetero-
zygous (Cx43+/-) and homozygous (Cx43-/-) Cx43 knock-out mice showed 30-40%21 and 42-55%22 reduction in ventricular CV, accompanied by 50% and 95% reduction in expression of Cx43 mRNA and protein, respectively. In contrast, CV across Cx43-/- CMC strands decreased 96% compared to Cx43+/+ CMC strands resulting in propa- gation at 2.1 cm/s vs 52 cm/s.23 In these studies excitable cells were subjected to Cx43 knockdown, however, we used non-excitable cells in our study. Cx43 knockdown in non-excitable hMSCs may influence CV more effectively than in excitable CMCs, as in non-excitable cells electrical conduction depends only on passive electronic cur- rent flow across gap junctions.
We found that carbenoxolone dose-dependently decreased CV across hMSCs and CMCs, but that the influence of carbenoxolone on CV across CMCs was less than that on CV across hMSCs (as was shown by the IC50 values). In a previous study, ventricular myocytes were still electrically coupled after incubation with 50 µM car- benoxolone, although coupling was diminished.24 In our study, CV across CMCs de- creased significantly after incubation with 200 µM of carbenoxolone, but conduction remained intact up to concentrations as high as 1.6 mM. However, how these hi- gher concentrations act on gap junctional coupling and excitability is not fully un- derstood.25 We performed additional whole cell patch clamp experiments on hMSCs (n=4) coupled to CMCs with or without carbenoxolone at day 1 of co-culture. Upon addition of 200 μM carbenoxolone, no action potentials could be recorded anymore and resting membrane potential became less negative. During these experiments the CMCs were still beating. This confirms uncoupling of hMSCs with maintained exci- tability of CMCs.
In our model, changes in CV due to altered tissue architecture can be considered as less likely because of standardized culture characteristics and the application of the antimitoticum BrdU to prevent cardiac fibroblast overgrowth. In addition, no migra- tion of CMCs or cardiac fibroblasts into the channel was observed. Furthermore, the two CMC fields in the multi-electrode array culture dishes without cells seeded in the channel did not resynchronize throughout the follow up.
As no sodium channels (SCN5A), the major channel proteins involved in excitation of cardiac cells, were found to be expressed in hMSCs at day 1, nor at day 14, is it most likely that electrical current is conducted across hMSCs by electrotonic con- duction. Besides gap junctions, ion channels play an important role in maintaining electrical propagation across excitable tissues. Recent studies revealed at least three types of outward currents (large-conductance Ca2+-activated K+ current (IKCa), Ito, IKDR or heag1), and two types of inward currents (L-type Ca2+ current (ICa.L) and INa,TTX) in mesenchymal stem cells.26,27 However, the percentage of hMSCs expres- sing these ion-channels was rather low, ranging from 8 till 30%. A TTX-sensitive Na+
channel was detected, however this was not associated with SCN5A gene expression, but SCN9A gene expression.27 The role of these and other ion channels in the elec- trophysiological, proliferation, and differentiation properties of hMSCs remains to be established.
Limitations
Ideally, adult human CMCs should have been used. However, there are considerable logistical, technical and ethical impediments to their use. Therefore, we used neona- tal rat CMCs, as they are spontaneously beating, and easily available. The coupling resistance measurements in this study could be refined further by using double patch clamp techniques. However, the scale of such experiments exceeds the goal of the present research.
Conclusions
hMSCs from patients with ischemic heart disease are functionally and electrically coupled to CMCs for at least 14 days. Time-dependent increase in electrical conduc- tion velocity is associated with increased Cx43 expression, improved functional gap junctional coupling and improved intercellular electrical coupling between hMSCs and CMCs.
Acknowledgments
We thank H. K. Koerten (Dept. of Molecular Cell Biology, LUMC) for the use of the glow-discharge equipment and M. van de Watering (Dept. of Molecular Cell Biology, LUMC) for constructing lentiviral vectors coding for recombinant microRNAs.
References
1. Wollert KC, Meyer GP, Lotz J, Ringes-Lichtenberg S, Lippolt P, Breidenbach C, et al. Intracoronary autologous bone-marrow cell transfer after myocardial infarction: the BOOST randomised control led clinical trial. Lancet. 2004;364:141-48.
2. Beeres SL, Bax JJ, Kaandorp TA, Zeppenfeld K, Lamb HJ, Dibbets-Schneider P, et al. Usefulness of intramyocardial injection of autologous bone marrow-derived mononuclear cells in patients with severe angina pectoris and stress-induced myocardial ischemia. Am J Cardiol. 2006;97:1326-31.
3. Perin EC, Dohmann HF, Borojevic R, Silva SA, Sousa AL, Mesquita CT, et al. Transendocardial, autologous bone marrow cell transplantation for severe, chronic ischemic heart failure. Circula- tion. 2003;107:2294-2302.
4. Taylor DA, Atkins BZ, Hungspreugs P, Jones TR, Reedy MC, Hutcheson KA, et al. Regenerating functional myocardium: improved performance after skeletal myoblast transplantation. Nat Med.
1998;4:929-33.
5. Menasche P, Hagege AA, Vilquin JT, Desnos M, Abergel E, Pouzet B, et al. Autologous
skeletal myoblast transplantation for severe postinfarction left ventricular dysfunction. J Am Coll Cardiol. 2003;41:1078-83.
6. Leobon B, Garcin I, Menasche P, Vilquin JT, Audinat E, Charpak S. Myoblasts transplanted into ratinfarcted myocardium are functionally isolated from their host. Proc Natl Acad Sci U S A.
2003;100:7808-11.
7. Abraham MR, Henrikson CA, Tung L, Chang MG, Aon M, Xue T, et al. Antiarrhythmic engineering of skeletal myoblasts for cardiac transplantation. Circ Res 2005;97:159-167.
8. Peters NS, Wit AL. Myocardial architecture and ventricular arrhythmogenesis. Circulation.
1998;97:1746-54.
9. Rohr S. Role of gap junctions in the propagation of the cardiac action potential. Cardiovasc Res.
2004;62:309-22.
10. Pittenger MF, Martin BJ. Mesenchymal stem cells and their potential as cardiac therapeutics. Circ Res. 2004;95:9-20.
11. Chen SL, Fang WW, Ye F, Liu YH, Qian J, Shan SJ, et al. Effect on left ventricular function of in tracoronary transplantation of autologous bone marrow mesenchymal stem cell in patients with acute myocardial infarction. Am J Cardiol. 2004;94:92-95.
12. Beeres SL, Atsma DE, van der Laarse A, Pijnappels DA, van Tuyn J, Fibbe WE, et al. Human adult bone marrow mesenchymal stem cells repair experimental conduction block in rat cardiomyocyte cultures. J Am Coll Cardiol. 2005;46:1943-52.
13. Ruwhof C, van Wamel AE, Egas JM, van der Laarse A. Cyclic stretch induces the release of growth promoting factors from cultured neonatal cardiomyocytes and cardiac fibroblasts. Mol Cell Bio chem. 2000;208:89-98.
14. van Tuyn J, Knaan-Shanzer S, van de Watering MJ, de Graaf M, van der Laarse A, Schalij MJ, et al.
Activation of cardiac and smooth muscle-specific genes in primary human cells after forced expres sion of human myocardin. Cardiovasc Res. 2005;67:245-55.
15. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflϋgers Arch.
1981;391:85-100.
16. Rubart M, Pasumarthi KB, Nakajima H, Soonpaa MH, Nakajima HO, Field LJ. Physiological coupling of donor and host cardiomyocytes after cellular transplantation. Circ Res. 2003;92:
1217-24.
17. Soonpaa MH, Koh GY, Klug MG, Field LJ. Formation of nascent intercalated disks between grafted fetal cardiomyocytes and host myocardium. Science. 1994;264:98-101.
18. Meiry G, Reisner Y, Feld Y, Goldberg S, Rosen M, Ziv N, et al. Evolution of action potential pro pagation and repolarization in cultured neonatal rat ventricular myocytes. J Cardiovasc Electrophy siol. 2001;12:1269-77.
19. Guo W, Kamiya K, Cheng J, Toyama J. Changes in action potentials and ion currents in long-term cultured neonatal rat ventricular cells. Am J Physiol. 1996;271:C93-102.
20. Valiunas V, Doronin S, Valiuniene L, Potapova I, Zuckerman J, Walcott B, et al. Human mesenchy mal stem cells make cardiac connexins and form functional gap junctions. J Physiol. 2004;555:
617-26.
21. Thomas SA, Schuessler RB, Berul CI, Beardslee MA, Beyer EC, Mendelsohn ME, et al. Disparate effects of deficient expression of connexin43 on atrial and ventricular conduction: evidence for
chamber-specific molecular determinants of conduction. Circulation. 1998;97:686-91.
22. Gutstein DE, Morley GE, Tamaddon H, Vaidya D, Schneider MD, Chen J, et al. Conduction slo wing and sudden arrhythmic death in mice with cardiac-restricted inactivation of connexin43. Circ Res. 2001;88:333-39.
23. Beauchamp P, Choby C, Desplantez T, de Peyer K, Green K, Yamada KA, et al. Electrical propagation in synthetic ventricular myocyte strands from germline connexin43 knockout mice.
Circ Res. 2004;95:170-78.
24. de Groot JR, Veenstra T, Verkerk AO, Wilders R, Smits JP, Wilms-Schopman FJ, et al. Conduction slowing by the gap junctional uncoupler carbenoxolone. Cardiovasc Res. 2003;60:288-97.
25. Harris AL. Emerging issues of connexin channels: biophysics fills the gap. Q Res Biophys. 2001;34:
325-472.
26. Heubach JF, Graf EM, Leutheuser J, Bock M, Balana B, Zahanich I, et al. Electrophysiological properties of human mesenchymal stem cells. J Physiol. 2004;554:659-72.
27. Li GR, Sun H, Deng X, Lau CP. Characterization of ionic currents in human mesenchymal stem cells from bone marrow. Stem Cells. 2005;23:371-82.
