Autocrine Regulation of Myocyte Cx43 Expression by VEGF
Cardiac myocytes can rapidly adjust their expression of gap junction channel proteins in response to changes in load. Previously, we showed that after only 1 hour of linear pulsatile stretch (110% of resting cell length; 3 Hz), expression of connexin43 (Cx43) by cultured neonatal rat ventricular myocytes is increased by ≈2-fold and impulse propagation is significantly more rapid. In the present study, we tested the hypothesis that vascular endothelial growth factor (VEGF), acting downstream of transforming growth factor-β (TGF-β), mediates stretch-induced upregulation of Cx43 expression by cardiac myocytes. Incubation of nonstretched cells with exogenous VEGF (100 ng/mL) or TGF-β (10 ng/mL) for 1 hour increased Cx43 expression by ≈1.8-fold, comparable to that observed in cells subjected to pulsatile stretch for 1 hour. Stretch-induced upregulation of Cx43 expression was blocked by either anti-VEGF antibody or anti-TGF-β antibody. Stretch-induced enhancement of conduction was also blocked by anti-VEGF antibody. ELISA assay showed that VEGF was secreted into the culture medium during stretch. Furthermore, stretch-conditioned medium stimulated Cx43 expression in nonstretched cells. This effect was also blocked by anti-VEGF antibody. Upregulation of Cx43 expression stimulated by exogenous TGF-β was blocked by anti-VEGF antibody, but VEGF-stimulation of Cx43 expression was not blocked by anti-TGF-β antibody. Thus, stretch-induced upregulation of Cx43 expression is mediated, at least in part, by VEGF, which acts downstream of TGF-β. Because the cultures contained only ≈5% nonmyocytic cells, these results indicate that myocyte-derived VEGF, secreted in response to stretch, acts in an autocrine fashion to enhance intercellular coupling.
- gap junctions
- vascular endothelial growth factor
- transforming growth factor-&bgr
- pulsatile stretch
Electrical activation of the heart depends on cell-to-cell transfer of current at gap junctions, the fundamental organelles of intercellular communication.1 To elucidate signaling pathways regulating intercellular coupling at gap junctions, we have developed an in vitro model of the early hypertrophic response in which monolayers of well-coupled neonatal rat ventricular myocytes are subjected to brief intervals of linear pulsatile stretch.2 After only 1 hour of stretch, expression of the major ventricular gap junction protein, connexin43 (Cx43), is enhanced by ≈2-fold, resulting in an increase in the number of gap junctions connecting neighboring cells and a significant gain in function manifested as an increase in conduction velocity.2 Thus, cardiac myocytes can rapidly adjust their expression of gap junction channel proteins and alter the number of gap junction channels in response to changes in load.2
The present study was designed to identify mechanisms mediating stretch-induced upregulation of Cx43 expression. We tested the hypothesis that vascular endothelial growth factor (VEGF), which is known to be synthesized and secreted by cardiac myocytes,3,4⇓ is released in response to pulsatile stretch and stimulates Cx43 expression in cardiac myocytes.3 We also tested the hypothesis that VEGF acts downstream of transforming growth factor-β (TGF-β), which has also been implicated in VEGF signaling in cardiac myocytes subjected to pulsatile stretch.3 We first showed that addition of exogenous TGF-β or VEGF to neonatal rat ventricular myocytes rapidly increased Cx43 expression. Antibodies against either TGF-β or VEGF blocked stretch-induced upregulation of Cx43. VEGF levels were increased in culture medium of cells subjected to pulsatile stretch, and Cx43 expression was upregulated in nonstretched cells incubated in stretch-conditioned medium. Upregulation of Cx43 expression induced by exogenous TGF-β was blocked by anti-VEGF antibody but VEGF-stimulation of Cx43 expression was not blocked by anti-TGF-β antibody. Because the cultures are nearly devoid of nonmyocytic cells, these results indicate that myocyte-derived VEGF, secreted in response to stretch, acts downstream of TGF-β in an autocrine fashion to enhance intercellular coupling in cardiac myocytes.
Materials and Methods
Primary Culture of Neonatal Rat Ventricular Myocytes
Primary cardiac myocyte cultures were prepared from ventricles of 1-day-old Wistar Kyoto rat pups (Charles River, Indianapolis, Ind) as described previously.2,5⇓ Cells were seeded on collagen-coated silicone membranes at a density of 2.4×105 cells/cm2 and grown at 35°C in an atmosphere of room air supplemented with 1% CO2 for 4 days before experimentation.2 All protocols were approved by the Washington University Animal Studies Committee.
Stretch and Exogenous TGF-β and VEGF Protocols
Monolayers of ventricular myocytes were subjected to linear pulsatile stretch using a custom-designed apparatus as described in detail in a previous study.2 Cells were stretched to 110% of resting length at a frequency of 3 Hz for 1 hour in serum-free medium (1:1, Ham’s F-12/DMEM, Fisher Life Sciences). In selected experiments, medium from cultures previously subjected to stretch for 1 hour was collected and applied to other cultures that had not been stretched. In other experiments, myocyte cultures in serum-free medium were incubated with recombinant human TGF-β1 (Sigma) or recombinant human VEGF165 (the 165 amino acid isoform of human VEGF; R&D Systems), or stretched for 1 hour in the presence of a chicken (IgY) anti-TGF-β antibody (R&D Systems), a mouse monoclonal anti-VEGF antibody (directed against recombinant human VEGF165) (R&D Systems), or a goat polyclonal anti-VEGF receptor [VEGF R2 (Flk-1)] antibody (R&D Systems). Nonimmune control antibodies included normal chicken IgY, mouse IgG2B, and normal goat IgG (all from R&D Systems). Control cultures (not stretched or exposed to exogenous TGF-β or VEGF) were also incubated in serum-free medium for 1 hour. At the conclusion of each experimental protocol, cells were either scraped from membranes and lysed for immunoblot analysis or fixed for confocal immunofluorescence analysis of Cx43 expression.
Immunoblot Assay of Cx43 Expression
Cell cultures were washed once in cold PBS and then scraped from the silicone membranes in a low ionic strength buffer containing protease inhibitors (NaHCO3 1 mmol/L; EDTA 5 mmol/L; EGTA 1 mmol/L; leupeptin 1 μmol/L; pepstatin 1 μmol/L; aprotinin 0.1 μmol/L; benzamidine 1 mmol/L; iodoacetamide 1 mmol/L; phenylmethylsulfonylfluoride 1 mmol/L). After centrifugation at 3000g for 6 minutes at 4°C, the pellet was resuspended in the same buffer and stored at −70°C. Aliquots containing 7 μg of total protein were analyzed by SDS polyacrylamide gel electrophoresis, and Cx43 signal was quantified as described in detail in previous studies.6 In all experiments, Cx43 signal in experimental preparations was compared with signal from control cultures (cells incubated in serum-free medium for 1 hour but not subjected to stretch or exposed to VEGF or antibodies), which was normalized to a value of 1.0.
Quantitative Confocal Immunofluorescence Analysis of Cx43 Expression
Cultures were rinsed in serum-free medium and then fixed in 4% paraformaldehyde. Cells were immunostained with a polyclonal rabbit anti-Cx43 antibody (Zymed) and specific immunoreactive signal was quantified by laser scanning confocal microscopy as described previously in detail.7 The amount of Cx43 signal present in discrete loci at intercellular junctions was expressed as a percentage of total myocyte area. In each culture, 3 to 5 fields were randomly selected for quantitative analysis, and a mean value was determined as previously described.7
Immunohistochemical Quantification of Myocytes and Nonmyocytic Cells
To assess the purity of neonatal rat ventricular myocyte cultures and measure the proportion of nonmyocytic cells, cultures were stained with a specific antibody against cardiac myosin as described previously.8 Multiple fields were examined by both fluorescence and bright-field microscopy, and the number of viable cells that failed to show a distinctive pattern of fluorescent cross striations was determined.
Optical Electrophysiological Mapping
Propagation velocity of the electrical impulse and the maximum upstroke velocity of the transmembrane action potential (dV/dtmax) were calculated from the initial portions of transmembrane action potentials measured in dense cultures at 64 sites (8×8 diode array, measurement area of each diode=15×15 μm2) subjected to stretch for 1 hour in the presence or absence of anti-VEGF antibody. The method to optically map transmembrane potential at multiple sites from the change in fluorescence of a voltage-sensitive dye has been previously described in detail.2 Data were analyzed with software written in LabView (National Instruments) and MATLAB (MathWorks).
Quantification of VEGF in Culture Medium by ELISA
Aliquots of culture medium collected from control cells and cells subjected to pulsatile stretch were applied to ELISA assay plates (R&D Systems) along with standards containing known amounts of recombinant VEGF. The ELISA assay was performed according to the manufacturer’s instructions and analyzed using a microplate reader (Softmax 2.5). Absorbance was measured spectrophotometrically at 450 nm and corrected by subtracting absorbance measured at 540 nm.
All values are expressed as mean±standard deviation. Data were analyzed using ANOVA (StatView). Multiple comparisons between groups were made using Fisher’s least significant differences method.
Exogenous VEGF and TGF-β Upregulate Cx43 Expression
To determine whether addition of exogenous VEGF can upregulate Cx43 expression in cardiac myocytes, cultures were incubated for 1 hour with selected concentrations of recombinant human VEGF (1 to 500 ng/mL), and Cx43 expression was measured by immunoblotting. As shown in a representative blot (Figure 1), exogenous VEGF increased the total amount of Cx43 in cultured neonatal ventricular myocytes in a dose-dependent and saturable manner with a peak response occurring at a concentration of 100 to 200 ng/mL. In subsequent experiments, cells were incubated for 1 hour with 100 ng/mL VEGF. Under these conditions, total Cx43 content as measured by immunoblotting increased by 1.78±0.80-fold (P=0.036) (Figure 2). Upregulation of Cx43 expression induced by 100 ng/mL VEGF was equivalent to the 1.89±1.31-fold increase (P=0.019) observed in cells subjected to pulsatile stretch (110% of resting length at 3 Hz) for 1 hour (Figure 2).
To determine whether VEGF and TGF-β increased the amount of Cx43 in gap junctions, cultures were incubated for 1 hour with 100 ng/mL VEGF or 10 ng/mL TGF-β1, and the proportion of total cell area occupied by specific Cx43 immunoreactive signal at sites of intercellular junctions was measured by quantitative confocal microscopy. Cx43 signal increased by 1.79-fold (from 0.97±0.29% to 1.73±0.20% of cell area; P<0.0001) in cells incubated with VEGF and by 1.86-fold (from 0.97±0.29% to 1.80±0.39% of cell area; P<0.0001) in cells incubated with TGF-β (Figure 3). A comparable change in Cx43 expression (1.84-fold increase from 0.97±0.29% to 1.78±0.29% of cell area; P<0.0001) occurred in cells stretched for 1 hour (Figure 3). These results indicate that addition of exogenous VEGF upregulates the total amount of Cx43 in cardiac myocytes and addition of either exogenous VEGF or TGF-β increases the amount of Cx43 immunoreactive signal at intercellular junctions to the same extent and within the same time frame as that observed in cells subjected to pulsatile stretch.
Stretch-Induced Upregulation of Cx43 Is Blocked by Anti-VEGF and Anti-TGF-β Antibodies
To determine whether the stretch-induced upregulation of Cx43 is mediated by VEGF and/or TGF-β, cells were stretched for 1 hour in the presence or absence of a specific antibody against VEGF (50 ng/mL) or TGF-β (100 ng/mL), and Cx43 expression was measured by confocal microscopy. As shown in Figure 4A, no increase in Cx43 signal occurred in cells that were stretched in the presence of either anti-VEGF or anti-TGF-β antibodies. Cx43 expression was upregulated, however, when cells were stretched in the presence of a nonspecific mouse IgG2B isotype antiserum (50 ng/mL) or a nonspecific chicken IgY antibody (100 ng/mL), which served as controls to demonstrate the specificity of the anti-VEGF and anti-TGF-β antibodies, respectively (Figure 4B). Incubation of control (nonstretched) cells with the anti-VEGF antibody for 1 hour had no effect on Cx43 expression levels (data not shown), but addition of the anti-VEGF antibody did block the effects of exogenous VEGF on Cx43 expression (Figure 4B). These results directly implicate signaling mediated by both VEGF and TGF-β in stretch-induced upregulation of Cx43 expression.
Stretch-Induced Increase in Conduction Velocity Is Blocked by Anti-VEGF Antibody
Optical mapping revealed that after 1 hour of pulsatile stretch, conduction velocity increased by 39% from 28.7±5.2 to 39.9±4.8 cm/s (n=4 for each, P=0.0081). Action potential upstroke velocity did not change significantly (98±13 versus 106±6 V/s in control and stretched cells, respectively; n=4 for each, P=0.39). These results confirm previous findings.2 In contrast, no change in conduction velocity (29.7±4.0 cm/s; n=4, P=0.78) or action potential upstroke velocity (104±5 V/s; n=4, P=0.50) was observed in cells stretched for 1 hour in the presence of anti-VEGF antibody (Figure 5). Thus, the stretch-induced increase in conduction velocity was blocked by anti-VEGF antibody.
Stretch-Conditioned Medium Contains VEGF and Can Stimulate Cx43 Expression in Nonstretched Cells
To substantiate the hypothesis that stretch enhances VEGF secretion, we used ELISA to measure the amount of VEGF in culture media obtained from cells subjected to pulsatile stretch for 1 or 6 hours. After 1 hour of stretch, the VEGF content of the culture medium increased by 50% (from 4.6±1.5 to 6.9±2.0 ng/mL; n=11, P=0.046). After 6 hours of stretch, the VEGF content increased by ≈4-fold (to 21.7±3.2 ng/mL; n=4, P<0.0001).
To further prove that VEGF and/or other mediators secreted into culture medium in response to stretch are responsible for upregulating Cx43 expression, experiments were performed with stretch-conditioned medium. As shown in Figure 6, Cx43 expression as measured by confocal microscopy was enhanced in cells that were incubated for 1 hour in culture medium collected from cells that had previously been stretched for 1 hour. This effect was blocked by addition of anti-VEGF antibody to the conditioned medium, but addition of the nonspecific IgG had no effect. Medium collected from control (nonstretched) cells also had no effect on Cx43 expression. These results further strengthen the conclusion that stretch increases secretion of VEGF and/or other mediators into the culture medium, which in turn, act in a VEGF-dependent manner to upregulate Cx43 expression.
VEGF Regulates Cx43 Expression by Interacting With Receptors on Cardiac Myocytes
To provide independent evidence that exogenous VEGF upregulates Cx43 expression by interacting with specific receptors, cells were incubated with a monoclonal anti-VEGF receptor (Flk-1) antibody (1 μg/mL) for 30 minutes before the addition of exogenous VEGF (100 ng/mL). After 1 hour, cells were fixed and Cx43 expression levels were quantified by confocal microscopy. As shown in Figure 7, incubation of cells with exogenous VEGF led to the expected increase in Cx43 immunoreactive signal. However, these effects were blocked in cells that had first been incubated with anti-VEGF receptor antibody. Addition of a control nonspecific goat IgG (1 μg/mL) had no effect on stimulation of Cx43 expression by VEGF.
VEGF Acts Downstream of TGF-β to Upregulate Cx43 Expression
To define the sequence of signaling events mediated by VEGF and TGF-β, cells were incubated with one mediator in the presence of an antibody against the other. As shown in Figure 8, both VEGF and TGF-β caused significant upregulation of Cx43 expression, as seen in an earlier experiment (see Figure 3). Upregulation of Cx43 expression stimulated by TGF-β was blocked by anti-VEGF antibody, but VEGF-stimulation of Cx43 expression was not blocked by anti-TGF-β antibody (Figure 8). Thus, VEGF acts downstream of TGF-β to upregulate Cx43 expression.
Potential Role of Nonmyocytic Cells in Stretch-Induced Upregulation of Cx43 Expression
To address the possibility that the effects of exogenous and/or secreted VEGF were mediated by interactions with receptors on nonmyocytic cells, which in turn, activated myocyte expression of Cx43, cells grown for 4 days in 2 representative cultures were stained with an antibody against cardiac myosin. Randomly selected high-power fields were examined by fluorescence and bright-field microscopy and the proportion of cells exhibiting fluorescent cross striations was determined. A typical field contained 146±8 myocytes and 8±2 nonmyocytic cells (n=10 fields) present either singly or in clusters of 2 to 5 cells. Thus, only ≈5% of the total cells in culture were not cardiac myocytes. These results strongly suggest that VEGF acts in an autocrine fashion to upregulate myocyte Cx43 expression.
In this study, we showed that exogenously supplied VEGF and TGF-β rapidly increased Cx43 expression in cultured neonatal rat ventricular myocytes in a manner similar to the effects induced by brief intervals of linear pulsatile stretch. We also showed that specific antibodies against either VEGF or TGF-β blocked stretch-induced upregulation of Cx43 expression. Stretch-conditioned medium contained increased levels of VEGF and was able to stimulate Cx43 expression in cells not subjected to stretch. Anti-VEGF antibody blocked the effects of exogenous TGF-β on Cx43 expression but anti-TGF-β antibody did not block the VEGF effect. Thus, we confirmed original observations by Seko et al3 that pulsatile stretch stimulates secretion of VEGF, mediated at least in part by TGF-β, and we now provide evidence that directly implicates these pathways in stretch-mediated upregulation of Cx43.
It is likely that pulsatile stretch causes secretion of multiple mediators from cardiac myocytes and nonmyocytic cells that can affect expression or function of myocyte proteins. The present studies were focused on TGF-β and VEGF, but other factors probably play a role in regulation of Cx43 expression. Shyu et al9 have recently reported that the angiotensin II receptor blocker losartan attenuates increased Cx43 expression in neonatal rat cardiac myocytes subjected to pulsatile stretch (120% of resting length at 1 Hz) for 24 hours. They also reported that angiotensin II levels in culture media were increased after 1 hour of pulsatile stretch. Although increased angiotensin II levels in stretch-conditioned medium have not been reported by all authors,10 previous work has demonstrated that angiotensin II upregulates Cx43 expression in cultured neonatal rat myocytes.11 Thus, a role for angiotensin II as a mediator of Cx43 upregulation in response to pulsatile stretch is likely, especially under the more prolonged stretch conditions (24 hours) used by Shyu et al.9
Stretch-induced enhancement of conduction velocity was blocked by anti-VEGF antibody, but no change in dV/dtmax of the action potential upstroke occurred in cells subjected to stretch. These results suggest that no major changes in ion channels involved in propagation (INa and possibly IL,Ca)12 occurred in response to stretch. In a modeling study, Rudy and Quan13 showed a clear dependence of average dV/dtmax of the action potential upstroke on coupling resistance. This dependence was steep in partially uncoupled cells but flat in well-coupled tissue. Thus, a significant change in dV/dtmax would not be expected in cells that are extensively interconnected by gap junctions. Spach et al14 have shown in a simulation study that the dependence of dV/dtmax on passive electrical properties is related not only to the absolute level of cell-to-cell coupling, but also to the ratio between the cytoplasmic resistance and the gap junction resistance (in addition to active membrane properties). This dependence is much less apparent in a continuous system in which gap junction resistance is low relative to myoplasm resistance than in a discontinuous system in which gap junction resistance predominates. The theoretical simulations of Rudy and Quan13 and Spach et al14 fit experimental data in anisotropic rat cell cultures.15 For these reasons, a significant decrease in dV/dtmax would not likely be seen in well-coupled myocytes subjected to pulsatile stretch or stimulated to increase Cx43 expression by addition of exogenous mediators.
The fact that cultured cardiac myocytes respond to exogenous VEGF and pulsatile stretch by nearly doubling their Cx43 content within 1 hour is dramatic evidence of the capacity for dynamic regulation of intercellular communication by the heart. It is becoming increasingly clear that myocytes regulate their level of coupling by changing the amount of connexin (and, presumably, the number of functional channels) in gap junctions. For example, under conditions such as acute ischemia that leads to uncoupling, there is rapid translocation of Cx43 from gap junctions to intracellular pools, associated with changes in Cx43 phosphorylation.16 In contrast, under conditions in which myocytes are subjected to an external load to stimulate a hypertrophic response2 or incubated with chemical mediators of hypertrophy such as cAMP17 and angiotensin II,11 Cx43 expression is enhanced, resulting in an increase in the number of gap junctions between adjacent cells and increased conduction velocity. It remains to be determined whether the marked increase in Cx43 expression observed in the present study was due to increased connexin synthesis, decreased connexin degradation, or a combination of these mechanisms. Based on previous studies demonstrating that Cx43 turns over rapidly (t1/2≈1.5 hours) in cardiac myocytes18,19⇓ and that turnover of the contractile protein pool is suppressed in neonatal rat cardiac myocytes stretched by 1% to 5%,20 we predict that at least part of the increase in Cx43 content observed in the present study was due to a diminished rate of degradation of Cx43.
Compelling evidence for a direct role of cardiac myocyte-derived VEGF in the development of the coronary vasculature and maintenance of contractile function was reported recently by Giordano et al,4 who used Cre-lox methods to selectively delete the gene encoding VEGF-A in cardiac myocytes. Hearts of affected animals contained fewer coronary microvessels and exhibited thin ventricular walls and depressed contractile function. These results revealed a critical role of the cardiac myocyte as secretory cell and paracrine effector,4 but little consideration was given to the possibility that myocyte-derived VEGF also acts in an autocrine fashion to regulate myocyte gene expression and function.
Stretch has been applied to cultured cardiac myocytes and to isolated perfused hearts as an experimental model of load-induced cardiac hypertrophy. Izumo and coworkers21,22⇓ showed that angiotensin II played a role in modulating the hypertrophic response to static stretch, which included activation of immediate early response genes (c-jun, c-erg, c-myc), induction of “fetal” gene expression (skeletal α-actin, ANF, and β-myosin heavy chain),23 and activation of mitogen-activated protein kinase (MAPK) and S6 kinase pathways.24 They also showed that the hypertrophic response to static stretch could be induced by exogenous angiotensin II25 and inhibited by AT1A receptor blockade.22,25⇓ Yazaki and coworkers26 have elucidated signaling pathways and changes in patterns of gene expression activated by pulsatile stretch of cultured neonatal rat cardiac myocytes. Multiple MAPK family members, including p44/p42 MAPKs, stress-activated protein kinase (p38 MAPK), and focal adhesion kinase [p125(FAK)], are all activated by pulsatile stretch.26 In a related study, it was shown that exogenous VEGF activates Raf-1, MAPKs, and S6 kinase in neonatal rat cardiac myocytes.27
Both endothelial cells and fibroblasts secrete VEGF, which regulates important biological processes in these cells, including chemotaxis, proliferation, and angiogenesis.28 Although we cannot completely exclude the possibility that VEGF produced and secreted by fibroblasts and/or endothelial cells played a role in the stretch-induced upregulation of Cx43 in cardiac myocytes, the low level of contamination by these cells in the myocyte cultures makes this possibility unlikely.
Stretch-induced VEGF expression is not limited to cultured neonatal rat myocyte preparations. Li et al29 observed a nearly 6-fold increase in VEGF mRNA expression in isolated perfused rat hearts subjected to increased left ventricular end-diastolic load of 35 mm Hg for 30 minutes. Increased VEGF expression has also been reported in experimental models of myocardial ischemia and infarction30–33⇓⇓⇓ and in patients with atrial fibrillation.34 These results are consistent with the hypothesis that VEGF expression and secretion by cardiac myocytes is enhanced in response to mechanical and metabolic stress and that VEGF exerts multiple paracrine and autocrine actions to promote angiogenesis and compensatory hypertrophy of cardiac myocytes. Our results indicate that an important component of the autocrine myocyte response to VEGF secretion is upregulation of Cx43, which contributes to enhanced intercellular coupling.
This work was supported by a Fellowship Grant from the American Heart Association, National Institutes of Health Grants HL-50598 and HL-66350, and grants from the Swiss National Science Foundation and the Swiss Heart Foundation. The authors thank Karen Green, Lilly Bircher-Lehmann, William Kraft, and Anita Iannone for outstanding technical assistance.
Original received November 11, 2001; revision received February 14, 2002; accepted February 25, 2002.
- ↵Saffitz JE, Yeager M. Intracardiac cell communication and gap junctions.In: Spooner PM, Rosen M, eds. Foundations of Cardiac Arrhythmias. New York, NY: Marcel Dekker; 2000: 171–204.
- ↵Zhuang J, Yamada KA, Saffitz JE, Kléber AG. Pulsatile stretch remodels cell-to-cell communication in cultured myocytes. Circ Res. 2000; 87: 316–322.
- ↵Giordano FJ, Gerber H-P, Williams S-P, VanBruggen N, Bunting S, Ruiz-Lozano P, Gu Y, Nath AK, Huang Y, Hickey R, Dalton N, Peterson KL, Ross JJr, Chien KR, Ferrara N. A cardiac myocyte vascular endothelial growth factor paracrine pathway is required to maintain cardiac function. Proc Natl Acad Sci U S A. 2001; 98: 5780–5785.
- ↵Fast VG, Darrow BJ, Saffitz JE, Kleber AG. Anisotropic activation spread in heart cell monolayers assessed by high-resolution optical mapping: role of tissue discontinuities. Circ Res. 1996; 79: 115–27.
- ↵Saffitz JE, Green KG, Kraft WJ, Schechtman KB, Yamada KA. Effects of diminished expression of connexin43 on gap junction number and size in ventricular myocardium. Am J Physiol Heart Circ Physiol. 2000; 278: H1662–H1670.
- ↵Thomas SP, Bircher-Lehmann L, Thomas SA, Zhuang J, Saffitz JE, Kléber AG. Synthetic strands of neonatal mouse cardiac myocytes: structural and electrophysiological properties. Circ Res. 2000; 87: 467–473.
- ↵Rudy Y, Quan WL. A model study of the effects of the discrete cellular structure on electrical propagation in cardiac tissue. Circ Res. 1987; 61: 815–823.
- ↵Spach MS, Heidlage JF, Dolber PC, Barr RC. Electrophysiological effects of remodeling cardiac gap junctions and cell size: experimental and model studies of normal cardiac growth. Circ Res. 2000; 86: 302–311.
- ↵Fast VG, Kleber AG. Anisotropic conduction in monolayers of neonatal rat heart cells cultured on collagen substrate. Circ Res. 1994; 75: 591–595.
- ↵Beardslee MA, Lerner DL, Tadros PN, Laing JG, Beyer EC, Yamada KA, Kléber AG, Schuessler RB, Saffitz JE. Dephosphorylation and intracellular redistribution of ventricular Cx43 during electrical uncoupling induced by ischemia. Circ Res. 2000; 87: 656–662.
- ↵Darrow BJ, Fast VG, Kléber AG, Beyer EC, Saffitz JE. Functional and structural assessment of intercellular communication: increased conduction velocity and enhanced connexin expression in dibutyryl cAMP-treated cultured cardiac myocytes. Circ Res. 1996; 79: 174–183.
- ↵Laird DW, Puranam KL, Revel JP. Turnover and phosphorylation dynamics of connexin43 gap junction protein in cultured cardiac myocytes. Biochem J. 1991; 273: 67–72.
- ↵Darrow BJ, Laing JG, Lampe PD, Saffitz JE, Beyer EC. Expression of multiple connexins in cultured neonatal rat ventricular myocytes. Circ Res. 1995; 76: 381–387.
- ↵Malhotra R, Sadoshima J, Brosius F, Izumo S. Mechanical stretch and angiotensin II differentially upregulate the renin-angiotensin system in cardiac myocytes in vitro. Circ Res. 1999; 85: 137–146.
- ↵Sadoshima J, Jahn L, Takahashi T, Kulik T, Izumo S. Molecular characterization of the stretch-induced adaptation of cultured cardiac cells. J Biol Chem. 1992; 267: 10551–10560.
- ↵Sadoshima J, Izumo S. Molecular characterization of angiotensin II–induced hypertrophy of cardiac myocytes and hyperplasia of cardiac fibroblasts: critical role of the AT1 receptor subtype. Circ Res. 1993; 73: 413–423.
- ↵Seko Y, Imai Y, Suzuki S, Kamijukkoku S, Hayasaki K, Sakomura Y, Tobe K, Kadowaki T, Maekawa H, Takahashi N, Yazaki Y. Serum levels of vascular endothelial growth factor in patients with acute myocardial infarction undergoing reperfusion therapy. Clin Sci. 1997; 92: 453–454.