This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Clarke, M. S. F. Articles by McNeil, P. L. Search for Related Content PubMed PubMed Citation Articles by Clarke, M. S. F. Articles by McNeil, P. L. Pubmed/NCBI databases Compound via MeSH Substance via MeSH

(Circulation Research. 1995;76:927-934.)
© 1995 American Heart Association, Inc.

Articles

Contraction-Induced Cell Wounding and Release of Fibroblast Growth Factor in Heart

Mark S. F. Clarke, Robert W. Caldwell, Hsi Chiao, Katsuya Miyake, Paul L. McNeil

From the Departments of Anatomy and Cellular Biology (M.S.F.C., K.M., P.L.M.) and Pharmacology and Toxicology (R.W.C., H.C.), The Medical College of Georgia, Augusta.

Correspondence to Dr Paul L. McNeil, Department of Anatomy and Cellular Biology, The Medical College of Georgia, Augusta, GA 30912-2000.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract The heart hypertrophies in response to certain forms of increased mechanical load, but it is not understood how, at the molecular level, the mechanical stimulus of increased load is transduced into a cell growth response. One possibility is that mechanical stress provokes the release of myocyte-derived autocrine growth factors. Two such candidate growth factors, acidic and basic fibroblast growth factor (aFGF and bFGF, respectively), are released via mechanically induced disruptions of the cell plasma membrane. In the present study, we demonstrate that transient, survivable disruption (wounding) of the cardiac myocyte plasma membrane is a constitutive event in vivo. Frozen sections of normal rat heart were immunostained to reveal the distribution of the wound event marker, serum albumin. Quantitative image analysis of these sections indicated that an average of 25% of the myocytes contained cytosolic serum albumin; ie, this proportion had suffered a plasma membrane wound. Wounding frequency increased approximately threefold after ß-adrenergic stimulation of heart rate and force of contraction. Heparin-Sepharose chromatography, enzyme-linked immunosorbent assay, growth assay coupled with antibody neutralization, and two-dimensional SDS-PAGE followed by immunoblotting were used to demonstrate that both aFGF and bFGF were released from an ex vivo beating rat heart. Importantly, ß-adrenergic stimulation of heart rate and force of contraction increased FGF release. Cell wounding is a fundamental but previously unrecognized aspect of the biology of the cardiac myocyte. We propose that contraction-induced cardiac myocyte wounding releases aFGF and bFGF, which then may act as autocrine growth-promoting stimuli.

Key Words: fibroblast growth factor • myocardium • cell injury • plasma membrane • heart hypertrophy

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Myocardial hypertrophy is an adaptive process initiated by the imposition of additional mechanical load on the heart.¹ Cardiac myocytes within the hypertrophying adult heart enlarge and display a correspondingly elevated rate of contractile protein synthesis² but, according to most studies, do not proliferate.³ In vitro, stretching of cardiac myocytes initiates a "hypertrophic" response characterized by the activation of an array of second messenger systems,⁴ the expression of a number of immediate-early response genes,⁵ and an increase in contractile protein synthesis.³ This mechanically initiated cascade of events is similar to that initiated on stimulation of cultured myocytes with several polypeptides, including angiotensin II⁶ and fibroblast growth factor (FGF).⁷

We have recently shown that during normal contractile activity in vivo skeletal myofibers suffer transient disruptions or "wounds" of their plasma membranes and that the frequency of such wounding is greatly increased by eccentric exercise-induced contractions.⁸ We found that such wounded myofibers contain less basic FGF (bFGF) in their cytoplasm than do uninjured myofibers,⁹ suggesting that contraction-induced fiber wounding results in bFGF release. On the basis of these and numerous other findings (summarized in Reference 10¹⁰ ), we postulate that mechanically induced transient disruptions of the plasma membrane provide a route for FGF release. This mechanism, termed the "wound hormone hypothesis," can explain how, in vivo, proteins such as acidic FGF (aFGF) and bFGF are exported despite the lack of a classic signal peptide sequence,¹¹ which is thought to be required for secretion via the exocytotic pathway.

The heart, like skeletal muscle, is a mechanically active tissue composed of FGF-responsive and FGF-producing cells.^{12 13 14} However, to date, there have been no studies linking the placement of mechanical stress on cardiac myocytes or myocardium with the release of FGF or studies that provide a mechanistic basis for FGF release. In the present study, we have asked the following questions that are raised when one applies the wound hormone hypothesis to the heart. First, do cardiac myocytes suffer plasma membrane wounds in the normal heart in vivo? Second, if so, does the frequency of wounding increase following the imposition on the heart of additional mechanical load? Third, is FGF released from the mechanically active (eg, beating) heart? Fourth, do the rate and force of contraction influence the amount of FGF released from the heart?

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cells and Reagents
Bovine retinal microvascular endothelial (BRME) cells (passage 7), the kind gift of Dr Ruth Caldwell (Medical College of Georgia, Augusta), were maintained in DMEM containing 10% fetal calf serum (FCS) and standard antibiotics. The rabbit anti-bFGF polyclonal antibodies, CR1 (catalog No. 40095, aFGF- and bFGF-neutralizing activity) and CR2 (catalog No. 40013, recognizing bFGF), were obtained from Collaborative Biomedical Products. A rabbit anti-aFGF polyclonal antibody, RD 1 (catalog No. AB-32-NA, recognizing aFGF), and purified human recombinant aFGF and bFGF were purchased from R&D Systems.

Detection of Cardiac Myocyte Wounding in the Rat Heart In Vivo
Sprague-Dawley rats (male, 200 to 250 g) were deeply anesthetized (50 mg/kg IP pentobarbital) and monitored from lead II of the ECG. Heart rate was recorded before and after administration of 0.5 µg/kg isoproterenol or vehicle alone via the femoral vein. After a period of 20 minutes, the dorsal aorta was cannulated, the vena cava was cut to allow efficient flow of perfusate, and then the heart was back-perfused through the coronary vasculature with 150 mL warm Dulbecco's phosphate-buffered saline (DPBS, pH 7.2) containing 0.1% procaine at a perfusion rate of 10 mL/min. This was followed by perfusion with 60 mL DPBS containing 8% freshly generated paraformaldehyde. Surrounded with fixative-soaked cotton swabs to prevent drying, the heart was left in place for 60 minutes. The heart was then carefully excised from the animal and placed in fresh fixative for a further 24 hours. Heart tissue was processed for frozen sectioning and serum albumin immunostaining as previously described.⁹ All procedures involving animal experimentation were in accordance with institutional guidelines.

Quantification of Cardiac Myocyte Wounding In Vivo
Control and isoproterenol-stimulated hearts were mounted side by side on a sectioning stub so that the apex of each heart was at the same level. The tissue block was trimmed to a depth of 8000 µm from the apex of the hearts, and 5-µm frozen sections were collected at random to a depth of 12 000 µm from the apex by using a Zeiss HM 500 cryostat microtome. Sections were then stained for the presence of cytoplasmic serum albumin as previously described.⁹ Quantification of cardiac myocyte immunostaining intensity was carried out by using an Image 1 (Universal Imaging Corp) image analysis system. The methods for random acquisition of digitized transmitted light images of muscle sections, for delimiting a cytoplasmic portion of the myocyte for analysis, and for quantitatively analyzing the data thus generated were essentially as previously described.⁹ A total of 7300 individual cardiac myocytes, obtained from three separate experiments, were analyzed in this manner. The number of wounded myocytes present in the left ventricular wall of control and isoproterenol-stimulated hearts was expressed as a percentage of the total myocyte population per condition analyzed.

Immunoelectron Microscopy
Left ventricle tissue from perfusion-fixed hearts was cut into 5-mm³ pieces and dehydrated in ethanol. Tissue was then incubated overnight in LR-White acrylic resin (EM Services), followed by polymerization at 60°C for 2 hours. Sections (3 µm) were cut for light microscopy by using an Ultracut E ultramicrotome (Reichert Scientific Instruments) and stained for the presence of serum albumin as above, except that primary antibody binding was detected with a biotinylated rabbit anti-goat secondary antibody that was disclosed by using streptavidin (10 nm)–gold (Auroprobe EM kit, Amersham) followed by silver enhancement (Intense Silver Enhancement kit, Amersham). For electron microscopy, ultrathin sections (70 nm) were cut and mounted on a nickel grid coated with a Formvar (Monsanto Co) membrane. Sections were stained by using the immunogold protocol described above and incubated overnight in sodium cacodylate buffer (pH 7.2) containing 1% osmium tetroxide. Sections were viewed with a Zeiss EM902 electron microscope after staining with uranyl acetate and lead citrate.

Detection of FGF Release From the Rat Langendorff Preparation
Sprague-Dawley rats (male, 200 to 250 g) were decapitated, and the heart was immediately removed from the animal and placed in cold Chenoweth-Koelle buffer. Excess tissue was trimmed from the preparation, and the heart was back-perfused through the aorta with warm Chenoweth-Koelle buffer (37°C) as previously described¹⁵ at a flow rate of 8 mL/min. This procedure was performed within 4 minutes of removing the heart from the animal. After a 20-minute equilibration period, the isolated heart had achieved a steady heart rate (260 beats per minute), and the experiment could begin. Perfusate was collected for a period of 40 minutes in a reservoir maintained at 4°C in the presence of 5 µg/mL aprotinin, 5 µg/mL leupeptin, 1 µg/mL pepstatin, and 1 mmol/L phenylmethylsulfonyl fluoride. The perfusate was collected and was pumped through a 1-mL Hi-Trap heparin-Sepharose column (Pharmacia), and the column was then washed with 1000 column bed volumes of 50 mmol/L Tris buffer (pH 7.2) at 4°C. Heparin binding proteins were eluted from the column in a stepwise fashion by using 10 mL of 0.5 mol/L NaCl, 1.2 mol/L NaCl, and 2.0 mol/L NaCl made up in 50 mmol/L Tris buffer (pH 7.2). The column was washed with 10 mL of Tris buffer (pH 7.2) between each salt wash. The column fractions were concentrated and desalted by using ultrafiltration (10 000–molecular weight cutoff filter, Amicon) to a final volume of 200 µL and stored at -70°C until FGF analysis. Fractions destined for assessment of growth-promoting activity were dialyzed by using ultrafiltration against serum-free DMEM to a final volume of 200 µL.

Detection of FGF by Enzyme-Linked Immunosorbent Assay
Aliquots (50 µL) of concentrated column fractions were dispensed into Immulon 96-well microtiter plates (Dynatech Laboratories) and incubated overnight at 37°C. The plates were washed with three changes of calcium- and magnesium-free PBS (pH 7.2) containing 0.05% Tween 20 (wash buffer) over a 20-minute period. Each well was blocked with wash buffer containing 3% bovine serum albumin (BSA) for a period of 30 minutes at 37°C. Blocking solution was removed and replaced with primary antibody solution (CR2 or RD 1, both used at 2.5 µg IgG per milliliter of blocking solution) and incubated for a further 2 hours at 37°C. The plates were washed with three changes of wash buffer over a 20-minute period, and primary antibody was detected by using a peroxidase-conjugated anti-rabbit biotinylated antibody kit (Vector Laboratories) and phenylenediamine dihydrochloride as a substrate. The reaction was halted by the addition of 2 mol/L sulfuric acid, and the absorbance was measured at 490 nm with a plate reader (Cambridge Technology, Inc). The amount of FGF detected in each column fraction was determined by comparison with a standard curve constructed on the same plate by using purified human recombinant aFGF and bFGF.

Growth Assay
BRME cells were plated in 2% FCS-DMEM containing standard antibiotics at a cell density of 2500 cells per well in 24-well plates. Cells were allowed to attach for 16 hours, and the medium was replaced with 0.5 mL of fresh 2% FCS-DMEM alone or 2% FCS-DMEM containing one of the following: recombinant aFGF (25 ng/mL), recombinant bFGF (5 ng/mL), 50 µL of concentrated 1.2 mol/L NaCl column fraction previously dialyzed against serum-free DMEM, and the above mitogens plus 10 µg IgG per milliliter of CR1. Cells were grown for a further 48 hours, after which time cell number was assayed by using a previously described method.¹⁶

Two-Dimensional SDS-PAGE and Immunoblotting
A concentrated desalted 1.2 mol/L NaCl column fraction (100 µL) was mixed with 20 ng bFGF, lyophilized, dissolved in sample buffer (0.1% Nonidet P-40 [vol/vol], 1% [wt/vol] dithiothreitol [DTT], and 10% [vol/vol] glycerol in distilled water), and boiled for 5 minutes. bFGF alone (20 ng) was treated in a similar manner. Samples were then separated on a pH 3 to 10 isoelectric focusing (IEF) gel (NOVEX) for a period of 2 hours at 100 V, followed by focusing for a further 3 hours at 200 V. The IEF gel was cut into vertical strips and equilibrated in 0.015 mol/L Tris buffer (pH 6.8) containing 3% (wt/vol) SDS and 10% (vol/vol) glycerol for 2 minutes with gentle agitation. Focused proteins were then separated on 15% SDS-PAGE gels and transferred onto 0.2-µm supported nitrocellulose membranes by using a method previously described.¹⁷ Immunoblots were blocked with Tris-buffered saline (TBS, pH 7.2) containing 4% BSA and 0.05% Tween 20 for 2 hours and probed for the presence of aFGF or bFGF by using a mixture of the anti-aFGF (RD 1) and anti-bFGF (CR2) primary antibodies described above (both used at 2.5 µg IgG per milliliter in TBS containing 2% BSA and 0.05% Tween 20, pH 7.4) for 2 hours at room temperature. Primary antibody binding was detected with an alkaline phosphatase–conjugated secondary antibody ABC kit as previously described.⁹ The pI and molecular weight coordinates for any proteins disclosed by immunoblotting were calculated by comparison with IEF standards and molecular weight standards (BioRad) run in parallel with the samples on the relevant gels.

Measurement of Isoproterenol-Stimulated FGF Release
Langendorff preparations were prepared as detailed above and allowed to equilibrate over a period of 20 minutes. After this time, a 50-mL aliquot of perfusate was collected and mixed with a heparin-Sepharose slurry in the presence of protease inhibitors at 4°C. The heart was then stimulated with 10^-7 mol/L isoproterenol for a period of 5 minutes to induce an increase in both heart rate and force of contraction. A second 50-mL aliquot of perfusate from the stimulated heart was then collected and mixed with the heparin-Sepharose slurry. After a period of 4 hours, the heparin-Sepharose was collected by centrifugation and washed with 100 mL of 0.5 mol/L NaCl, followed by 100 mL of 50 mmol/L Tris buffer (pH 7.2), to remove the non–FGF-inhibitable activity detected in the 0.5 mol/L NaCl wash by heparin-Sepharose affinity chromatography. The heparin-Sepharose was then washed with 2 mL of 1.2 mol/L NaCl to elute any bound FGF, and this fraction was then concentrated and desalted by using ultrafiltration to a final volume of 200 µL as described above. The amount of FGF detected in equal amounts of concentrated perfusate, collected before and after stimulation with isoproterenol, was determined by the enzyme-linked immunosorbent assay (ELISA) protocol described above.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cardiac Myocyte Wounding In Vivo
Wounded cardiac myocytes, cytoplasmically labeled with a horseradish peroxidase–conjugated antibody to rat serum albumin,⁸ were observed in both the normal (Fig 1A) and the isoproterenol-stimulated (Fig 1B) rat heart. Quantification of the proportion of wounded cardiac myocytes in the normal heart revealed that an average of 25% had suffered sarcolemma wounds (Fig 2, left). Administration of a bolus injection of 0.5 µg/kg IV isoproterenol, a dose associated with increased heart rate (40%) and force of contraction (50%) but not arrhythmia or disruption of electrical conductance,¹⁸ caused a highly significant (P<.001) approximately threefold increase in the proportion of wounded cardiac myocytes (Fig 2). The high level of myocyte wounding detected in normal, as well as stimulated, heart muscle raised the possibility that we could not resolve cytosolic serum albumin from that potentially present in the T-tubule system. Light microscopic examination of thin plastic sections of isoproterenol-stimulated heart tissue stained for serum albumin by use of an immunogold procedure did not reveal any T-tubule staining (Fig 3A). More important, electron microscopic examination of ultrathin sections confirmed that the serum albumin wound marker was present only in the myocyte cytosol (Fig 3B). The electron micrographs of Fig 3 also rule out pinocytotic uptake of the rat serum albumin: none of the gold label was seen in a vesicular compartment. The present work is consistent with numerous previous studies validating the use of albumin as a marker of a membrane disruption event (reviewed in Reference 10¹⁰ ). Therefore, the serum albumin staining we have observed is cytosolic and is not associated with the T-tubule or any other vesicular system; ie, it is a valid indicator of a plasma membrane wound. Our data indicate that mechanically induced transient membrane damage is a common event in the normal heart and that an increase in heart rate and force of contraction increases the frequency of such cardiac myocyte wounding.

View larger version (0K): [in this window] [in a new window]   Figure 1. Immunoperoxidase staining of serum albumin in the left ventricular wall of a normal (A) and an isoproterenol-stimulated (B) rat heart. Wounded cardiac myocytes, identified by the presence of cytosolic serum albumin (horseradish peroxidase staining), are seen in both cross sections but appear to be more numerous in the isoproterenol-stimulated heart. Primary antibody specificity was demonstrated by preincubation of the antibody with a 40-mol/L excess of rat serum albumin for 2 hours at 37°C, which resulted in complete abolition of staining.

View larger version (12K): [in this window] [in a new window]   Figure 2. Image analysis of frozen sections of rat myocardium immunostained for serum albumin. Left, Histogram showing the average staining intensities of 3510 individual cardiomyocytes in the left ventricular wall of three control hearts (a value of 1 on this gray scale represents the highest possible staining intensity). Middle, Histogram showing the average staining intensities of 3790 individual cardiomyocytes in the left ventricular wall of three stimulated hearts. Arrows in left and middle panels represent the staining intensity threshold (150), below which a myocyte was deemed to have been wounded. Right, Bar graph showing the number of wounded cardiac myocytes expressed as a percentage of the total population analyzed. The percentage of wounded myocytes detected in isoproterenol-stimulated hearts was approximately threefold greater (*P<.005, Student's t test; error bar, 1 SD) than in normal hearts.

View larger version (137K): [in this window] [in a new window]   Figure 3. Silver-enhanced immunogold staining of serum albumin in the left ventricular wall of an isoproterenol-stimulated rat heart viewed in a thin plastic section with a light microscope (A) and an ultrathin plastic section with an electron microscope (B). A diffuse cytosolic staining pattern was observed in both thin and ultrathin plastic sections with no staining associated with the T-tubule system (arrowhead, B). As expected, extracellular matrix serum albumin staining was also observed. Incubation of sections with secondary antibody alone followed by silver enhancement resulted in no staining of either the cytosol or extracellular matrix (not shown).

Detection of FGF in the Perfusate of an Ex Vivo Beating Rat Heart
To determine whether FGF is released from the beating heart, perfusate from the Langendorff preparation was collected and passed over a heparin-Sepharose column as illustrated in Fig 4A. Heparin-binding proteins were eluted with buffer containing 0.5, 1.2, and 2.0 mol/L of NaCl, desalted, concentrated, and assayed by ELISA for the presence of FGF. Fig 4B illustrates the results obtained from a representative experiment using the CR2 antibody, which recognizes human recombinant bFGF. Primary antibody binding was detected in both the 0.5 and 1.2 mol/L NaCl column fractions. However, unlike the activity detected in the 1.2 mol/L NaCl fraction, the immunoreactivity detected in the 0.5 mol/L NaCl fraction could not be inhibited with 40-mol/L excess of bFGF, indicating that the activity detected in this but not the 1.2 mol/L NaCl fraction was due to nonspecific primary antibody binding. Similar results were obtained when using the RD 1 antibody, which recognizes human recombinant aFGF (data not shown). These data indicate that heparin-binding molecules, which could be eluted with 1.2 mol/L NaCl and had the immunoreactivity of both aFGF and bFGF, are released from the beating heart.

View larger version (18K): [in this window] [in a new window]   Figure 4. Detection of fibroblast growth factor (FGF) in the perfusate of the rat Langendorff preparation. A, Diagrammatic representation of the procedure used to collect, concentrate, and quantify the amount of FGF released into the perfusate of the beating heart. ELISA indicates enzyme-linked immunosorbent assay. B, Bar graph showing results of a representative experiment illustrating the detection of basic FGF (bFGF) in the perfusate of the Langendorff preparation after heparin-Sepharose column affinity chromatography separation followed by ELISA. Ab indicates antibody. Primary Ab (CR2) specificity was confirmed by coincubation of the sample with 20-fold excess of human recombinant (hr) bFGF in the presence of the primary Ab, which resulted in the abolition of color development. Incubation with secondary Ab alone resulted in no color development. bFGF concentration was determined by comparison to a standard curve constructed using known amounts of human recombinant bFGF. Similar results were obtained using a primary Ab (RD 1) directed against acidic FGF (data not shown).

Biological Activity of the 1.2 mol/L NaCl Fraction Obtained From the Heart Perfusate
The FGF-like activity detected in the 1.2 mol/L NaCl heparin-Sepharose column fraction was tested for its ability to induce BRME proliferation.¹⁹ A significant increase over control values in absorbance (the measure in this assay of cell number) was observed when BRME cells were grown in medium containing aFGF (25 ng/mL) and bFGF (5 ng/mL), respectively (Fig 5). This increase was significantly reduced when cells were grown in the presence of 10 µg/mL of the FGF-neutralizing antibody CR1 (Fig 5). The FGF-like activity present in the 1.2 mol/L NaCl column fraction also significantly increased absorbance from control values. This 1.2 mol/L fraction growth-promoting activity was abolished by the presence of 10 µg/mL of the FGF-neutralizing antibody CR1 (Fig 5). These data indicate that the 1.2 mol/L NaCl heparin-Sepharose column fraction contains a molecule or molecules that have a characteristic biological activity of aFGF or bFGF,¹⁸ namely, growth-promoting activity for endothelial cells.

View larger version (21K): [in this window] [in a new window]   Figure 5. Bar graph showing assessment of growth-promoting activity in the perfusate of the rat Langendorff preparation eluted from a heparin-Sepharose column using 1.2 mol/L NaCl. bFGF indicates basic fibroblast growth factor; aFGF, acidic fibroblast growth factor. An anti-bFGF antibody (CR1) significantly inhibited the growth-promoting activity for bovine retinal endothelial cells of both recombinant human aFGF and bFGF as well as the growth-promoting activity detected in the 1.2 mol/L NaCl fraction (*P<.01 and **P<.005, Student's t-test; error bar, 1 SD). Substitution of the neutralizing antibody, CR1, with a nonspecific rabbit IgG did not significantly alter the growth-promoting activity of aFGF, bFGF, or the 1.2 mol/L NaCl fraction (data not shown).

Identification of Heparin-Binding Molecules in the Heart Perfusate
Human recombinant bFGF was separated by using two-dimensional SDS-PAGE and immunoblotted onto a nitrocellulose membrane. The membrane was probed with a mixture of CR2 and RD 1 primary antibodies to detect both aFGF and bFGF. As can be seen in Fig 6A, human recombinant bFGF contained two major isoforms with coordinates of 7.7, 17 kD (pI, M_r) and 8.5, 17 kD (pI, M_r). When a mixture of the recombinant bFGF and the lyophilized 1.2 mol/L NaCl fraction of the heart perfusate was separated in a similar manner, two new proteins were detected on the blot (Fig 6B, arrows). These proteins had coordinates of 8.5, 15 kD (pI, M_r) and 5.5, 15 kD (pI, M_r), pI values very close to those previously reported for bFGF and aFGF, respectively.²⁰ Thus, our data indicate that both bFGF and aFGF are released from the beating heart.

View larger version (22K): [in this window] [in a new window]   Figure 6. Two-dimensional SDS-PAGE immunoblot of 20 ng human recombinant basic fibroblast growth factor (bFGF) (A) and the 1.2 mol/L NaCl fraction plus 20 ng human recombinant bFGF (B). Each blot was probed with a mixture of two primary antibodies directed against acidic fibroblast growth factor (aFGF, RD 1 antibody) and bFGF (CR2 antibody). Two distinct proteins with coordinates of 8.5, 15 kD (pI, Mr) and 5.5, 15 kD (pI, Mr) (arrows) were detected in the 1.2 mol/L NaCl fraction/bFGF mixture but not in the recombinant bFGF sample. This suggests that both aFGF and bFGF are released into the perfusate of the rat Langendorff preparation and can be eluted from the heparin-Sepharose column using 1.2 mol/L NaCl.

Effect of Stimulated Contraction on FGF Release
We next asked whether, as predicted by the wound hormone hypothesis, increased mechanical load placed on the heart results in increased FGF release. The amount of FGF present in the heart perfusate was measured before and after stimulation with the ß-adrenergic agent isoproterenol. Initial experiments indicated that placement of electrodes in the cardiac wall or a balloon catheter in the left ventricle to measure changes in heart rate and force of contraction, respectively, caused artifactual cardiac myocyte wounding (data not shown). Therefore, we measured FGF release before and after stimulation with 10^-7 mol/L isoproterenol in unmonitored hearts to avoid such artifactual wounding. The effect of 10^-7 mol/L isoproterenol on the heart rate and force of contraction in a monitored heart preparation is illustrated in Fig 7, left. Isoproterenol stimulated heart rate by 42% and force of contraction (left ventricular dP/dt) by 46%. The data presented in Fig 7, right, indicate that isoproterenol stimulation caused a significant increase (P<.01, paired t test) in the amount of FGF released from the Langendorff preparation.

View larger version (33K): [in this window] [in a new window]   Figure 7. The effect of ß-adrenergic stimulation on heart rate, force of contraction, and fibroblast growth factor (FGF) release. Left, Representative recording showing the effect of isoproterenol added for 5 minutes to the perfusate (final concentration, 10-7 mol/L) of a rat Langendorff preparation. Left ventricular (LV) dP/dtmax is the maximum rate of LV pressure developed per second within a fluid-filled balloon catheter inserted into the LV cavity. Heart rate is derived from the pressure signal. Right, Bar graph showing the amount of FGF released into the perfusate of seven unmonitored rat Langendorff preparations in the 5 minutes before and after stimulation with 10-7 mol/L isoproterenol. The amount of FGF released after stimulation was significantly increased (P<.01, paired t test) over that detected in an equal volume of perfusate before stimulation.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Membrane Wounding in the Normal and Stimulated Beating Heart
Transient survivable disruptions of cell plasma membrane integrity, termed "cell wounds," are now known to occur in a variety of mechanically active tissues under physiological conditions.¹⁰ These tissues include skin, gut, aorta, and, perhaps most relevant to the heart, skeletal muscle. In skeletal muscle subject to the eccentric contractions produced during downhill running, the percentage of wounded myocytes increases approximately sevenfold, from a control nonrunning level of 4% to a postrunning level of >28%.⁸

In the present study, we provide direct evidence that cell wounding also occurs in the heart, that the cell type affected is the cardiac myocyte, and that myocyte wounding is a frequently occurring event. Approximately 25% of the myocytes examined in the normal rat heart had suffered constitutive nonlethal plasma membrane disruptions (Fig 1) before organ perfusion. We are unable to temporally correlate this basal level of cardiac myocyte wounding with any particular mechanical stimuli, because the clearance of serum albumin from the cytoplasm of a wounded muscle cell takes 24 to 48 hours.⁸ However, an acute increase in rate and force of contraction in vivo, caused by administration of the ß-adrenergic agonist isoproterenol, caused a highly significant increase (3-fold over control levels) in the frequency of cardiac myocyte wounding that occurred within the 20-minute period between isoproterenol injection and heart perfusion (Fig 1).

Several studies previous to ours had indicated that myocyte wounding might occur in the normal heart. For example, the cardiac-specific protein troponin I is readily detected in the blood of normal individuals.²¹ Additionally, cardiac-specific myosin heavy chain fragments, also detectable in the blood of normal individuals, increase in concentration twofold after exercise.^{22 23} Other cardiac-specific markers, eg, cardiac troponin T, are not detectable in normal blood²⁴ : possibly troponin T is released from cells only after lethal injury, or possibly the assay used is insufficiently sensitive.²⁵

How do cardiac myocytes survive and continue to function despite frequent disruptions of sarcolemma integrity? Crucial to any cell's ability to survive a plasma membrane disruption is the capacity for resealing. Most eukaryotic cells possess this capacity, as is demonstrated by their ability to survive microinjection and other mechanically based cell-loading techniques.²⁶ Nevertheless, since serum albumin transiently enters the cytosolic compartment of the wounded cardiac myocyte, so too must ions such as Ca²⁺, Na⁺, and K⁺. The isoproterenol-stimulated heart continues to beat rhythmically and with increased force, suggesting that cardiac myocytes can withstand large, but transient, fluxes in cytoplasmic ion concentrations without permanent functional or electrical compromise. Highly efficient sarcolemma Ca²⁺- ATPase pumps²⁷ and sarcolemma Na⁺-Ca²⁺-exchangers activated by elevated intracellular Na⁺²⁸ may explain such a capacity. In addition, Ca²⁺-activated closure of gap junctions²⁹ at the intercalated disk of the cardiac myocyte may constitute an evolutionary adaptation of this and other cell types to life in a mechanically injurious environment, since it would prevent Ca²⁺ poisoning of adjacent cardiac myocytes until membrane resealing had occurred. Finally, eukaryotic cells use a Ca²⁺-activated kinesin-based vesicular shuttle in resealing membrane disruptions.³⁰ This mechanism delivers intracellular membrane to the site of plasma membrane disruption, where it is then added via exocytosis to the plasma membrane (K. Miyake and P.L. McNeil, unpublished data, 1994).

We chose an isoproterenol dose (0.5 µg/kg) sufficient for maximally increasing heart rate and force of contraction. Much higher (50 to 100 mg/kg) doses of this and other catecholamines cause gross damage to the myocardium.³¹ Proposed mechanisms explaining this cardiotoxicity include permeabilization of the cardiac myocyte plasma membrane.³² Indeed, such permeabilization was established in previous studies that used horseradish peroxidase as a wound marker, much as we have used serum albumin in the present study. Interestingly, these previous studies failed to reveal wounded myocytes in the normal heart.³² This discrepancy between our present work and this previous work probably reflects the greater sensitivity of our technique, which uses an endogenous marker, serum albumin, that is present at far higher concentration than can be achieved by using injected horseradish peroxidase.

FGF Release Into the Perfusate of the Contracting Heart
In the present study, we provide evidence from heparin-Sepharose chromatography, ELISA, an endothelial growth assay combined with antibody neutralization of growth-promoting activity, and two-dimensional gel electrophoresis followed by Western blotting that both aFGF and bFGF are present in heart perfusates. In addition (data not shown), we detected immunoreactivity in the 1.2 mol/L heparin-Sepharose eluate in Western blots and ELISA assays using a total of five additional antibodies to FGF. Moreover, we show that stimulation of heart rate and force of contraction with the ß-adrenergic agonist isoproterenol results in an increase in the amount of FGF released into the perfusate of the rat Langendorff preparation. In a separate study, we have found that cultured cardiac myocytes suffer plasma membrane disruptions and release bFGF at a greater rate after the initiation of electrical pacing of contraction (D. Kaye, D. Pimental, S. Prasad, T. Möki, H.J. Berger, P.L. McNeil, R.A. Kelly, and T.W. Smith, unpublished data, 1994). When cultures were subject to electrical pacing in the presence of the contraction-inhibiting drug verapamil, cell wounding and FGF release returned to levels characteristic of unpaced cells. All of these recent results suggest that in the myocardium, as in skeletal muscle,⁹ FGF release is stimulated by increased contractile activity.

Injury-induced increases in the expression of bFGF mRNA and protein have been observed in several systems, including brain³³ and cultured endothelial cells³⁴ and, most interestingly, in the isoproterenol-injured³⁵ and the transplanted^{36 37 38} heart. Moreover, exercise increases expression of bFGF in skeletal muscle.³⁹ Thus, in addition to the possible growth-stimulatory role bFGF may play after its acute release upon injury, FGF may also play a longer term role in heart recovery from injury, facilitated by an increase in its expression. Wounding of cultured endothelial cells induces the expression of bFGF message and protein,³⁴ suggesting that disruption of plasma membrane integrity may result both in enhanced release and production of bFGF.

The most likely source of the aFGF and bFGF released into the heart perfusate is the cardiac myocyte, since it is the cell type most strongly immunostained.^{10 11 13 40} Another potential source of FGF is extracellular matrix.⁴¹ However, immunostaining of heart tissue does not support this notion, although aFGF has been reported to be present in the extracellular matrix remaining after detergent lysis of cultures of cardiac myocytes.⁴² Finally, other cells of the myocardium, such as endothelial cells and smooth muscle cells, may produce these growth factors in heart and therefore cannot be ruled out as potential sources of FGF.

The Wound Hormone Hypothesis and Heart Hypertrophy
The signaling mechanism that transduces a mechanical stimulus into a cellular growth response is not yet fully understood in any tissue. One mechanism recently advanced for heart suggests that stretch-induced activation of second messenger systems,^{4 43} by a mechanism independent of stretch sensitive calcium channels,⁵ leads to secretion of angiotensin II.⁴⁴

The data in the present study suggest that an additional mechanism may apply to the mechanically challenged heart. We found that the heart fulfills several salient predictions of the wound hormone hypothesis for bFGF release. First, cardiac myocytes are frequently wounded under normal conditions of cardiac mechanical activity. Second, an increased level of mechanical activity increases the frequency of myocyte wounding. Third, the growth factors, aFGF and bFGF, are released from the mechanically active heart. And, fourth, such release increases with added mechanical load. We propose that FGF is released on contraction-induced wounding of the cardiac myocyte and then acts as an autocrine growth-promoting stimulus for this cell.

	Acknowledgments

 
This study was supported by grants from the National Institutes of Health (GM 48091 [Dr McNeil] and HL-32562 [Dr Caldwell]) and an American Heart Association Established Investigator Award (Dr McNeil). We wish to thank D. Ulrich and R. Khakee for providing technical assistance.

Received January 10, 1995; accepted March 13, 1995.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Cooper GI. Cardiocyte adaptation to chronically altered load. Ann Rev Physiol. 1987;49:501-518. [Medline] [Order article via Infotrieve]
2. Black FM, Packer SE, Parker TG, Michael LH, Roberts R, Schwartz RJ, Schneider MD. The vascular smooth muscle alpha-actin gene is reactivated during cardiac hypertrophy provoked by load. J Clin Invest. 1991;88:1581-1588.
3. Parker TG. Molecular biology of cardiac growth and hypertrophy. Herz. 1993;18:245-255. [Medline] [Order article via Infotrieve]
4. Sadoshima J, Izumo S. Mechanical stretch rapidly activates multiple signal transduction pathways in cardiac myocytes: potential involvement of an autocrine/paracrine mechanism. EMBO J. 1993;12:1681-1692. [Medline] [Order article via Infotrieve]
5. Sadoshima J-I, Takahashi T, Jahn L, Izumo S. Roles of mechano-sensitive ion channels, cyoskeleton, and contractile activity in stretch-induced immediate-early gene expression and hypertrophy of cardiac myocytes. Proc Natl Acad Sci U S A. 1992;89:9905-9909. [Abstract/Free Full Text]
6. Baker KM, Aceto JF. Angiotensin II stimulation of protein synthesis and cell growth in chick heart cells. Am J Physiol. 1990;259:H610-H618. [Abstract/Free Full Text]
7. Schneider MD, McLellan WR, Black FM, Parker TG. Growth factors, growth factor response elements, and the cardiac phenotype. Basic Res Cardiol. 1992;2:33-48.
8. McNeil PL, Khakee R. Disruptions of muscle fiber plasma membranes: role in exercise-induced damage. Am J Pathol. 1992;140:1097-1109. [Abstract]
9. Clarke MSF, Khakee R, McNeil PL. Loss of cytoplasmic basic fibroblast growth factor from physiologically wounded myofibers of normal and dystrophic muscle. J Cell Sci. 1993;106:121-133. [Abstract]
10. McNeil PL. Cellular and molecular adaptations to injurious mechanical force. Trends Cell Biol. 1993;3:302-307. [Medline] [Order article via Infotrieve]
11. Abraham JA, Mergia A, Whang JL, Tumolo A, Friedman J, Hjerrild KA, Gospodarowicz D, Fiddes JC. Nucleotide sequence of a bovine clone encoding the angiogenic protein, basic fibroblast growth factor. Science. 1986;233:545-548. [Abstract/Free Full Text]
12. Casscells W, Speir E, Sasse J, Klagsbrun M, Allen P, Lee M, Calvo B, Ciba M, Haggroth L, Folkman J, Epstein SP. Isolation, characterization, and localization of heparin-binding growth factors in the heart. J Clin Invest. 1990;85:433-441.
13. Speir E, Yi-Fu Z, Lee M, Shrivastav S, Casscells W. Fibroblast growth factors are present in adult cardiac myocytes, in vivo. Biochem Biophys Res Commun. 1988;157:1336-1341.
14. Speir E, Tanner V, Gonzalez AM, Farris J, Baird A, Casscells W. Acidic and basic fibroblast growth factors in adult rat heart myocytes: localization, regulation in culture, and effects of DNA synthesis. Circ Res. 1992;71:251-259. [Abstract/Free Full Text]
15. Broadley KJ. The Langendorff heart preparation: reappraisal of its role as a research tool and teaching model for coronary vasoactive drugs. J Pharmacol Methods. 1979;2:143-156.
16. Hansen MB, Nielsen SE, Berg K. Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J Immunol Methods. 1989;119:203-210. [Medline] [Order article via Infotrieve]
17. Khyse-Anderson J. Electroblotting of multiple gels: a simple apparatus without buffer tank for rapid transfer of proteins from polyacrylamide to nitrocellulose. J Biochem Biophys Methods. 1984;10:203-209. [Medline] [Order article via Infotrieve]
18. Mohamed GA, Caldwell RW, Carrier GO, Elmazar MM, Tuttle RR. Interactions of a novel catecholamine, GP-2-128, with adrenergic receptors. J Cardiovasc Pharmacol. 1994;23:492-500. [Medline] [Order article via Infotrieve]
19. Gospodarowicz D, Moran J, Braun D, Birdwell C. Clonal growth of bovine vascular endothelial cells: fibroblast growth factor as a survival agent. Proc Natl Acad Sci U S A. 1976;73:4120-4124. [Abstract/Free Full Text]
20. Klagsbrun M, Shing Y. Heparin affinity of anionic and cationic capillary endothelial growth factors: analysis of hypothalamus-derived growth factors and fibroblast growth factors. Proc Natl Acad Sci U S A. 1985;82:805-811. [Abstract/Free Full Text]
21. Cummins B, Auckland ML, Cummins P. Cardiac-specific troponin-I radioimmunoassay in the diagnosis of acute myocardial infarction. Am Heart J. 1987;113:1333-1344.[Medline] [Order article via Infotrieve]
22. Léger JO, Larue C, Ming T, Calzolavi C, Gautier P, Mouton C, Grolleau R, Louisot P, Puech P, Peperstraete B, Starokine M, Telerman M, Pau B, Léger JJ. Assay of serum cardiac myosin heavy chain fragments in patients with acute myocardial infarction: determination of infarct size and long-term follow-up. Am Heart J. 1990;120:781-790. [Medline] [Order article via Infotrieve]
23. Larue C, Calzolari C, Leger J, Leger J, Pau B. Immunoradiometric assay of myosin heavy chain fragments in plasma for investigation of myocardial infarction. Clin Chem. 1991;37:78-82. [Abstract/Free Full Text]
24. Gerhardt W, Katus H, Ravkilde J, Hamm C, Jorgensen PJ, Peheim E, Ljungdahl L, Lofdahl P. S-troponin T in suspected ischemic myocardial injury compared with mass and catalytic concentrations of S-creatine kinase isoenzyme MB. Clin Chem. 1991;37:1405-1411. [Abstract/Free Full Text]
25. Katus HA, Remppis A, Looser S, Hallermeier K, Scheffold T, Kubler W. Enzyme linked immuno assay of cardiac troponin T for the detection of acute myocardial infarction in patients. J Mol Cell Cardiol. 1989;21:1349-1353. [Medline] [Order article via Infotrieve]
26. McNeil PL. Incorporation of labeled macromolecules into living cells. Methods Cell Biol. 1989;29:153-173. [Medline] [Order article via Infotrieve]
27. Barry WH, Bridge JHB. Intracellular calcium homeostasis in cardiac myocytes. Circulation. 1993;87:1806-1815. [Abstract/Free Full Text]
28. Weir WG. Sodium-calcium exchange in intact cardiac cells. Ann N Y Acad Sci. 1991;639:366-374. [Medline] [Order article via Infotrieve]
29. Bennett MVL, Verselis VK. Biophysics of gap junctions. Semin Cell Biol. 1992;3:29-47. [Medline] [Order article via Infotrieve]
30. Steinhardt RA, Bi G, Alderton DJM. Cell membrane resealing by a vesicular mechanism similar to neurotransmitter release. Science. 1994;263:390-393. [Abstract/Free Full Text]
31. Rona G. Catecholamine cardiotoxicity. J Mol Cell Cardiol. 1985;17:291-306. [Medline] [Order article via Infotrieve]
32. Boutet M, Attner IH, Rona G. Permeability alteration of sarcolemmal membrane in catecholamine-induced cardiac muscle injury. Lab Invest. 1976;34:482-488. [Medline] [Order article via Infotrieve]
33. Finkelstein SP, Apostolides PJ, Cadoy CG, Prosser, Philips MF, Klagsbrun M. Increased basic fibroblast growth factor (bFGF) immunoreactivity at the site of focal brain wounds. Brain Res. 1988;460:253-259. [Medline] [Order article via Infotrieve]
34. Ku PT, D'Amore PA. Regulation of basic fibroblast growth factor (bFGF) gene and protein expression in endothelial cells following injury-induced bFGF release. Mol Biol Cell. 1993;4:H39.
35. Padua RR, Kardami E. Increased basic fibroblast growth factor (bFGF) accumulation and distinct patterns of localization in isoproterenol-induced cardiomyocyte injury. Growth Factors. 1993;8:291-306. [Medline] [Order article via Infotrieve]
36. Zhao XM, Yeoh TK, Frist WH, Porterfield DL, Miller GG. Induction of acidic fibroblast growth factor and full-length platelet-derived growth factor expression in human cardiac allografts: analysis by PCR, in situ hybridization, and immunohistochemistry. Circulation. 1994;90:677-685. [Abstract/Free Full Text]
37. Zhao XM, Yeoh TK, Hiebert M, Frist WH, Miller GG. The expression of acidic fibroblast growth factor (heparin-binding growth factor-1) and cytokine genes in human cardiac allografts and T cells. Transplantation. 1993;56:1177-1182. [Medline] [Order article via Infotrieve]
38. Antionu A, Carter N. Ventricular expression of basic fibroblast growth factor gene after orthotopic cardiac transplantation. Transplantation. 1994;57:1364-1366. [Medline] [Order article via Infotrieve]
39. Morrow NG, Kraus WE, Moore JW, Sanders-Williams R, Swain JL. Increased expression of fibroblast growth factors in a rabbit skeletal muscle model of exercise conditioning. J Clin Invest. 1990;85:1816-1820.
40. Kardami E, Fandrich RR. Basic fibroblast growth factor in atria and ventricles of the vertebrate heart. J Cell Biol. 1989;109:1865-1875. [Abstract/Free Full Text]
41. Klagsbrun M. The affinity of fibroblast growth factors (FGFS) for heparin: FGF-heparin sulfate interactions in cells and extracellular matrix. Curr Opin Cell Biol. 1990;2:857-863. [Medline] [Order article via Infotrieve]
42. Weiner HL, Swain JL. Acidic fibroblast growth factor mRNA is expressed by cardiac myocytes in culture and the protein is localized to the extracellular matrix. Proc Natl Acad Sci U S A. 1989;86:2683-2687. [Abstract/Free Full Text]
43. Sadoshima J, Izumo S. Signal transduction pathways of angiotensin II–induced c-fos gene expression in cardiac myocytes in vitro: roles of phospholipid-derived second messengers. Circ Res. 1993;73:424-438. [Abstract/Free Full Text]
44. Sadoshima J, Xu Y, Slayter HS, Izumo S. Autocrine release of angiotensin II mediates stretch-induced hypertrophy of cardiac myocytes in vitro. Cell. 1993;75:977-984.[Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				C. Etard, U. Roostalu, and U. Strahle Shuttling of the chaperones Unc45b and Hsp90a between the A band and the Z line of the myofibril J. Cell Biol., March 24, 2008; 180(6): 1163 - 1175. [Abstract] [Full Text] [PDF]

				R. C. Squires, S. M. Muehlbauer, and J. Brojatsch Proteasomes Control Caspase-1 Activation in Anthrax Lethal Toxin-mediated Cell Killing J. Biol. Chem., November 23, 2007; 282(47): 34260 - 34267. [Abstract] [Full Text] [PDF]

				K. Amano, K. Miyake, J.L. Borke, and P.L. McNeil Breaking Biological Barriers with a Toothbrush J. Dent. Res., August 1, 2007; 86(8): 769 - 774. [Abstract] [Full Text] [PDF]

				D. E. Ingber Cellular mechanotransduction: putting all the pieces together again FASEB J, May 1, 2006; 20(7): 811 - 827. [Abstract] [Full Text] [PDF]

				O. Farkas, J. Lifshitz, and J. T. Povlishock Mechanoporation induced by diffuse traumatic brain injury: an irreversible or reversible response to injury? J. Neurosci., March 22, 2006; 26(12): 3130 - 3140. [Abstract] [Full Text] [PDF]

				A. Zolkiewska Ecto-ADP-ribose Transferases: Cell-Surface Response to Local Tissue Injury Physiology, December 1, 2005; 20(6): 374 - 381. [Abstract] [Full Text] [PDF]

				A. McNeil and P. L. McNeil Yolk granule tethering: a role in cell resealing and identification of several protein components J. Cell Sci., October 15, 2005; 118(20): 4701 - 4708. [Abstract] [Full Text] [PDF]

				S. F. Vatner FGF Induces Hypertrophy and Angiogenesis in Hibernating Myocardium Circ. Res., April 15, 2005; 96(7): 705 - 707. [Full Text] [PDF]

				P. Camelliti, T. K. Borg, and P. Kohl Structural and functional characterisation of cardiac fibroblasts Cardiovasc Res, January 1, 2005; 65(1): 40 - 51. [Abstract] [Full Text] [PDF]

				M. Palmen, M. J.A.P. Daemen, L. J. De Windt, J. Willems, W. R.M. Dassen, S. Heeneman, R. Zimmermann, M. Van Bilsen, and P. A. Doevendans Fibroblast growth factor-1 improves cardiac functional recovery and enhances cell survival after ischemia and reperfusion: A fibroblast growth factor receptor, protein kinase c, and tyrosine kinase-dependent mechanism J. Am. Coll. Cardiol., September 1, 2004; 44(5): 1113 - 1123. [Abstract] [Full Text] [PDF]

				E. Kardami, Z.-S. Jiang, S. K Jimenez, C. J Hirst, F. Sheikh, P. Zahradka, and P. A Cattini Fibroblast growth factor 2 isoforms and cardiac hypertrophy Cardiovasc Res, August 15, 2004; 63(3): 458 - 466. [Abstract] [Full Text] [PDF]

				S. K Jimenez, F. Sheikh, Y. Jin, K. A Detillieux, J. Dhaliwal, E. Kardami, and P. A Cattini Transcriptional regulation of FGF-2 gene expression in cardiac myocytes Cardiovasc Res, June 1, 2004; 62(3): 548 - 557. [Abstract] [Full Text] [PDF]

				R. H. Singleton and J. T. Povlishock Identification and Characterization of Heterogeneous Neuronal Injury and Death in Regions of Diffuse Brain Injury: Evidence for Multiple Independent Injury Phenotypes J. Neurosci., April 7, 2004; 24(14): 3543 - 3553. [Abstract] [Full Text] [PDF]

				S. L. House, C. Bolte, M. Zhou, T. Doetschman, R. Klevitsky, G. Newman, and J. E. J. Schultz Cardiac-Specific Overexpression of Fibroblast Growth Factor-2 Protects Against Myocardial Dysfunction and Infarction in a Murine Model of Low-Flow Ischemia Circulation, December 23, 2003; 108(25): 3140 - 3148. [Abstract] [Full Text] [PDF]