Fas (CD95/Apo-1)–Mediated Damage to Ventricular Myocytes Induced by Cytotoxic T Lymphocytes From Perforin-Deficient Mice

A Major Role for Inositol 1,4,5-Trisphosphate

From the Rappaport Family Institute for Research in the Medical Sciences (B.F., M.S., H.L., G.M., R.C., O.B.), Bruce Rappaport Faculty of Medicine, The Bernard Katz Center for Cell Biophysics, Technion-Israel Institute of Technology, Haifa, Israel; the Department of Heart Surgery (I.S.), Carmel Medical Center, Haifa, Israel; the Department of Pharmacology (R.B.R.), College of Physicians & Surgeons of Columbia University, New York, NY; and the Department of Immunology (G.B.), Weizmann Institute of Science, Rehovot, Israel.

Correspondence to Ofer Binah, DSc, Rappaport Institute, POB 9697, Haifa 31096, Israel. E-mail binah{at}tx.technion.ac.il

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract—Cytotoxic T lymphocytes (CTLs) that infiltrate the heart are important immune effectors implicated in heart transplant rejection, myocarditis, and other cardiomyopathies. To investigate the mechanism(s) underlying CTL damage to the myocardium through activation of the Fas receptor (Fas/CD95/Apo-1) by the Fas ligand, we explored the interaction between peritoneal exudate CTLs (PELs), derived from perforin gene–knockout (P-/-) mice, and murine ventricular myocytes. Fas expression on isolated ventricular myocytes was demonstrated immunohistochemically. Action potentials, [Ca²⁺]_i transients, and contractions of myocytes conjugated to P-/- PELs or treated with the apoptosis-inducing anti-Fas monoclonal antibody Jo2 were recorded. Action potential characteristics of nonconjugated myocytes and myocytes conjugated with P-/- PELs were, respectively, as follows: V_m, -73.2±1.5 and -53.6±6.4 mV (mean±SEM); action potential amplitude, 117.9±3.9 and 74.3±21.2 mV; and action potential duration at 80% repolarization, 17±6 and 42±13 milliseconds (all P<.05). P-/- PELs also induced early and delayed afterdepolarizations as well as arrhythmogenic activity. Diastolic [Ca²⁺]_i increased during the cytocidal interaction with P-/- PELs, from a fluorescence ratio of 0.82±0.05 (n=7) to 1.98±0.09 (n=13) (P<.05). All of the effects caused by P-/- PELs were reproduced by incubating the myocytes with Jo2. Heparin (50 µg/mL), an antagonist of inositol trisphosphate (IP₃)–operated sarcoplasmic reticulum Ca²⁺ channels, or U-73122 (2 µmol/L), a phospholipase C inhibitor, but not the inactive agonist U-73343, prevented Fas-mediated myocyte dysfunction. Additionally, intracellular application (through the patch pipette) of the active IP₃ analogue, inositol 1,4,5-trisphosphate, but not the inactive analogue, inositol 1,3,4-trisphosphate, caused electrophysiological changes resembling those resulting from P-/- PELs and Jo2, suggesting that CTL-induced Fas-based myocyte dysfunction is mediated by IP₃. We conclude that a Fas-based perforin-independent mechanism of CTL action can account for the immunopathology seen in the allotransplanted heart, myocarditis, and dilated cardiomyopathy.

Key Words: cytotoxic T lymphocyte • Fas • inositol trisphosphate • ventricular myocyte • perforin gene–knockout mouse

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Immunological mechanisms are involved in the pathophysiology of major heart diseases, such as transplant rejection, myocarditis and the resulting DCM, and Chagas' heart disease. Of the multiple immunological components involved in heart diseases, the present study is focused on the contribution(s) of CTLs. The involvement of both CD4⁺ and CD8⁺ T lymphocytes in heart transplant rejection is well documented,^{1 2} and the degree of lymphocyte infiltration is the basis for the commonly used Billingham's criteria for grading acute cardiac rejection.^{3 4} Additionally, studies show good correlation between the level of lymphocyte infiltration and the decline in cardiac performance. In myocarditis and in DCM, there is wide agreement that T lymphocytes contribute to ongoing cardiac damage, often leading to terminal heart failure.^{5 6} The critical role of T lymphocytes in myocarditis has been conclusively demonstrated by immunodepletion and adoptive transfer experiments.^{7 8} Autoimmunity is also a major effector mechanism responsible for Chagas' heart disease.⁹

Cytocidal lymphocytes (ie, CTLs and natural killer cells) possess at least two molecularly distinct fast-acting lytic mechanisms.¹⁰ In the first (secretory) pathway, granzymes cosecreted with perforin are believed to penetrate the target cell through polyperforin-induced transmembrane pores, thus bringing about the demise of the cell by activating the interleukin-1ß–converting enzyme death pathway. In the second (nonsecretory) pathway, FasL expressed on the surface of the killer lymphocytes binds to the Fas receptor (Fas/CD95/Apo-1) expressed by the target cell. This encounter triggers a cascade of intracellular protein-protein interactions and proteolytic activities culminating in apoptosis. Importantly, myocardial cells constitutively express Fas^{11 12} and are therefore likely to be affected by CTLs promptly and adversely.

Apoptosis, a major hallmark of Fas-mediated damage, appears to participate in the genesis and pathophysiology of paroxysmal arrhythmias, conduction disturbances, and arrhythmogenic right ventricular dysplasia.¹³ Myocyte destruction due to apoptosis was reported in myocarditis, in end-stage cardiomyopathy,¹⁴ and in arrhythmogenic right ventricular dysplasia.¹⁵ To investigate the mechanism(s) of Fas-induced myocyte dysfunction, we investigated the interaction of CTLs from perforin gene–knockout (P-/-) mice and ventricular myocytes, thus excluding the lytic action of the pore-forming protein perforin. Action potentials, [Ca²⁺]_i transients, and myocyte contraction were recorded from BALB/c (H-2^d) ventricular myocytes conjugated to and interacting with H-2^b anti–H-2^d PELs obtained from P-/- mice (P-/- PELs).

Because our previous studies on the interaction of murine CTLs with guinea pig myocytes demonstrated that cytocidal damage was prevented by inhibiting the IP₃ pathway,¹⁶ we have now tested the hypothesis that IP₃ is the intracellular agent mediating Fas-based damage to ventricular myocytes. The important finding that this damage was prevented by blocking IP₃ production or IP₃-operated SR channels suggests that the IP₃ pathway may be targeted to attenuate the injury inflicted on the heart by killer lymphocytes, thus restoring heart function.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Preparation of Ventricular Myocytes and Electrophysiological Measurements
Ventricular myocytes from adult BALB/c mice (H-2^d) were obtained by an enzymatic dissociation procedure.¹⁷ Ten minutes after intraperitoneal administration of 500 U heparin, mice were anesthetized with pentobarbital (60 mg/kg). The chest was opened, and the heart was rapidly removed, washed with ice-cold dissociation solution containing (mmol/L) NaCl 126, KCl 4.4, MgCl₂ 1, NaHCO₃ 18, glucose 11, HEPES 4, and butanedione monoxime 30, along with 0.13 U/mL insulin (pH 7.4), and equilibrated with a gas mixture of 5% CO₂/95% O₂. Subsequently, the heart was attached to an aortic cannula and perfused at a rate of 2 mL/min with gassed dissociation solution at 37°C. The heart was perfused for 5 minutes with the dissociation solution, followed by 25 minutes of perfusion with the same solution containing 0.4 mg/mL collagenase (type II, Worthington). The heart was detached from the cannula, and the ventricle was cut off and minced with fine scissors. The mixture was gently pipetted for 2 to 3 minutes and then filtered through a nylon mesh (200 µm) into Tyrode's solution containing 50 µmol/L Ca²⁺ and 2% bovine albumin (Sigma). After a 15-minute incubation at 37°C, myocytes were centrifuged at 300 rpm for 3 minutes and resuspended in the same solution containing 200 µmol/L Ca²⁺ and 2% bovine albumin. After an additional 15-minute incubation at 37°C, myocytes were transferred to Tyrode's solution and kept at room temperature (24°C to 25°C) until studied (on the same day). For the electrophysiological experiments, myocytes were transferred to the recording bath mounted on the stage of an inverted microscope (Zeiss IM). The bath was perfused (unless otherwise indicated) with Tyrode's solution at a rate of 1 to 2 mL/min at 31°C to 32°C. The Tyrode's solution contained (mmol/L) NaCl 140, KCl 5.4, glucose 10, MgCl₂ 1, sodium pyruvate 2, CaCl₂ 1, and HEPES 10 (pH 7.4). Action potentials were recorded from ventricular myocytes stimulated at 0.5 Hz by means of an Axon 200A patch-clamp amplifier (Axon Instruments, Inc) as previously described.¹⁸ Patch electrodes were prepared from glass micropipettes and had tip resistances of 2 to 4 M when filled with the pipette solution containing (mmol/L) potassium aspartate 120, KCl 20, MgCl₂ 3.5, KH₂PO₄ 20, Na₂ATP 3, glucose 10, and EGTA 1 (pH 7.4).

Cytotoxic T Lymphocytes and Conjugate Formation
In vivo–primed H-2^b anti–H-2^d PELs were obtained from P-/- or C57BL/6 (P+/+) mice 4 to 5 days after the secondary intraperitoneal immunization with 25x10⁶ leukemia L1210 cells.^{19 20} The P-/- mice used to establish the colony at the Weizmann Institute were a gift from W.R. Clark (University of California at Los Angeles). Perforin deficiency of P-/- PELs was confirmed by polymerase chain reaction analysis. On the day of the experiment, mice were killed by cervical dislocation, and PELs were obtained by rinsing the peritoneal cavities with PBS containing 10% heat-inactivated FCS (PBS-FCS), as previously described.^{21 22} All procedures for handling animals were in accordance with institutional guidelines. Allogeneic CTL–target cell conjugates were formed between H-2^b anti–H-2^d PELs (from P-/- or P+/+ mice) and BALB/c (H-2^d) ventricular myocytes. Since centrifugation, commonly used to encourage conjugation between killer and target cells, was deleterious to myocytes, the myocytes were pretreated with the plant lectin Con A before exposure to PEL, enabling conjugate formation in the recording bath. After obtaining control measurements from a nonconjugated myocyte, flow to the bath was stopped, and Con A was added to a final concentration of 10 µg/mL. Ten minutes later, 10 to 20 µL of PEL suspension was added to the bath, resulting in myocytes to which one to several PELs became attached. In control experiments (not shown), we found that in Con A–treated nonconjugated myocytes, action potential characteristics were unchanged during 60 minutes of superfusion with Tyrode's solution.

LF⁺ and LF^- Leukemia Cells
LF⁺ cells are L1210 leukemia cells of DBA/2 mice stably transfected with the Fas expression vector, kindly provided by P. Golstein (Marseilles-Luminy, France). LF^- cells are L1210 leukemia cells stably transfected with the Fas antisense expression vector, kindly provided by W.R. Clark (University of California at Los Angeles). To determine Fas expression in in vitro–cultured L1210 Fas⁺ (LF⁺) and L1210 Fas^- (LF^-) leukemia, cells (0.25x10⁶) were washed in staining medium (0.5% to 1% BSA in PBS+0.02% azide) and pelleted. Thirty microliters (0.25 µg) anti-Fas Armenian hamster mAb (Jo2, PharMingen) was added, and cells were incubated on ice (30 minutes with occasional shaking). After washing and pelleting the cells, 30 µL (1:100 dilution) of FITC goat anti–Ham F(ab)₂ (Jackson Immune Research) was added, and cells were incubated as described above. After they were washed, the cells were resuspended in PBS+0.02% azide and analyzed by FACS.

Incubation of Myocytes With Anti-Fas mAb Jo2
To activate the Fas receptor directly, myocytes were incubated for various periods, up to 180 minutes, at 37°C in the presence of 10 µg/mL of the anti-Fas mAb Jo2. The incubation medium consisted of a 1:1 mixture of the dissociation solution containing 200 µmol/L Ca²⁺ and 2% albumin, and Tyrode's solution was used for the electrophysiological experiments (see composition above). Because of the long incubation time required for Jo2 action, in these experiments we studied three groups of myocytes: nonincubated myocytes (control) and myocytes incubated for various intervals in the absence or presence of Jo2 (or another mAb).

Measurement of [Ca²⁺]_i Transients and Myocyte Contraction
Ventricular myocytes were loaded for 25 minutes at room temperature (24°C to 25°C) with fura 2-AM (Molecular Probes), at a final concentration of 5 µmol/L, in a 1:1 mixture of Tyrode's solution and dissociation solution containing 2% bovine albumin. Excess fura 2 was removed by rinsing twice with Tyrode's solution. Myocytes were then transferred to a nonfluorescent chamber mounted on the stage of an inverted microscope (Diaphot 300, Nikon) and visualized with a x40 oil immersion Neofluor objective.²³ The chamber was perfused (unless otherwise indicated) with Tyrode's solution at a rate of 1 mL/min. Experiments were performed at 31°C to 32°C. Fura 2 fluorescence was measured using a dual-wavelength system (Delta- Scan, Photon Technology Intl). Briefly, light emitted from a xenon arc lamp was fed in parallel into two independent monochromators to obtain quasimonochromatic light beams of two different wavelengths exciting the cell at 340 and 380 nm. Either a 340- or 380-nm wavelength was switched by a rotating chopper disk at a frequency enabling ratio measurements at a rate of 150 per second. The two separate monochromator outputs were collected by the ends of a bifurcated quartz fiberoptic bundle. The emitted fluorescence (510 nm) was collected by the microscope optics, passed through an interference filter, and detected by a photomultiplier tube (710 PMT Photon Counting Detection System, Photon Technology Intl). Raw data were stored for off-line analysis by Felix software (Photon Technology Intl) as 340- and 380-nm counts and as the following ratio: R=F₃₄₀/F₃₈₀, where F₃₄₀ and F₃₈₀ indicate fluorescence at 340 and 380 nm, respectively. For scaling the fluorescence ratio, cell-derived autofluorescence and noncell fluorescence were subtracted from the measured fluorescence. In these experiments, myocytes were stimulated at 0.5 Hz using platinum wires embedded in the walls of the perfusion chamber.²³

To monitor myocyte contraction (represented by cell shortening) while measuring [Ca²⁺]_i transients, myocytes were simultaneously illuminated with red light, and a dichroic mirror (630-nm cutoff) placed in the emission path deflected the cell image to a video optical system (Crescent Electronic). The cursors of the optical system tracked motion of the cell edge along a raster line segment of the image during electrically stimulated contractions. The analog voltage output from the motion detector was calibrated into micrometers of motion. The motion signal obtained at 60 Hz was digitized and stored along with the fluorescence data.

Immunohistochemical Determination of Fas Expression in Ventricular Myocytes
Ventricular myocytes were spread on a precleaned polylysine-coated slides. The slides were dried at room temperature, fixed for 45 minutes with 5% paraformaldehyde, and washed three times with PBS. Endogenous peroxidase activity was neutralized by 20 minutes of incubation with 3% H₂O₂ in methanol, followed by a wash with PBS. Slides were then incubated at room temperature for 90 minutes with Jo2 (diluted 1:25), followed by incubations with biotinylated anti–Armenian hamster IgG secondary antibody (Jackson Immune Research, diluted 1:750) with streptavidin-peroxidase conjugate, and 3-amino-g-ethylcarbazol in N,N-dimethylformamide dissolved in acetate buffer, pH 5.2 as a substrate (Histostain-SP kit, Zymed Laboratory Inc). Incubation with nonimmune hamster serum served as a control. Counterstaining was performed with hematoxylin.

Cytotoxicity Assay
Effector cells were incubated together with ⁵¹ Cr-labeled target cells at various ratios. Cells were centrifuged (1000 rpm, 5 minutes, room temperature) to facilitate conjugation and incubated at 37°C in 5% CO₂ atmosphere for different times. After incubation, the plates were centrifuged (2000 rpm, 10 minutes), and 100 µL of supernatant from each well was harvested and counted in a gamma counter. The percentage of lysis was calculated as follows: ([experimental cpm-spontaneous cpm)/(total releasable cpm-spontaneous cpm])x100.

Statistical Analysis
Results were expressed as mean±SEM. To compare means of two populations, Student's t test for paired or unpaired observations was used. To compare the effects of PEL or Jo2 versus control, two-way ANOVA was performed. A value of P<.05 was considered significant. Figures were plotted with ORIGIN software (Microcal).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Interaction of P-/- and P+/+ PELs With Ventricular Myocytes: Action Potentials, [Ca²⁺]_i Transients, and Myocyte Contraction
As a first step in investigating the mechanisms of Fas-induced myocyte dysfunction, we thought it necessary to confirm the results of earlier studies demonstrating expression of the Fas receptor in myocardial tissue.^{11 12} In the present study, we found that freshly dissociated ventricular myocytes also express Fas (Fig 1A). Incubation of myocytes with the anti-Fas mAb Jo2 (see "Materials and Methods" for details) resulted in specific red staining, representing positive Fas expression. In contrast, incubation of myocytes under the same conditions, in the presence of nonimmune hamster serum (instead of Jo2), did not result in red staining (Fig 1B).

View larger version (90K): [in this window] [in a new window]   Figure 1. Fas expression in freshly dissociated ventricular myocytes. Fas was determined by the "indirect" immunohistochemistry method using anti-Fas mAb Jo2 (A) or nonimmune hamster serum (B). See "Materials and Methods" for further details.

To explore the contribution of the FasL/Fas pathway to myocardial immune pathology (associated with heart-infiltrating diseases), we developed an allogeneic model of murine ventricular myocytes interacting with P-/- PELs. After the addition of P-/- PELs to the recording bath, a variable number of lymphocytes promptly adhere to myocytes, forming stable conjugates (Fig 2). It appears that soon after conjugate formation, "reshaping" of lymphocytes occurs, a process possibly associated with lymphocyte activation and delivery of the lethal hit. We then investigated electrophysiological properties of affected myocytes by recording action potentials initially from an unbound myocyte and then during conjugate formation with P-/- (or P+/+) PELs and throughout the cytotoxic interaction. The main electrophysiological alterations (Fig 3) were reduction of V_m and action potential amplitude and lengthening of APD. In addition to these alterations, in 5 of 10 P-/- PEL–conjugated myocytes, delayed or early afterdepolarizations developed, as depicted by a representative experiment (Fig 3C). The effects of P-/- PELs on action potential characteristics in 10 conjugated myocytes (Fig 4) illustrate progressive changes with the advancement of the cytotoxic interaction. In control experiments (Fig 4) in which action potentials from nonconjugated myocytes were recorded, V_m and action potential amplitude were stable during the 60-minute experiment, whereas APD₈₀ gradually shortened, possibly because of the rundown of ion currents contributing to repolarization. Because similar APD shortening occurred in all three groups during the first 40 minutes, it is unlikely that this commonly seen phenomenon affected the outcome of the interaction with the lymphocytes. Next, we compared the effect of perforin-deficient CTLs (P-/- PELs) with that of perforin-containing CTLs (P+/+ PELs) by recording action potentials from myocytes conjugated with P+/+ PELs (Figs 3 and 4). We found that the electrophysiological effects of P-/- and P+/+ PELs were comparable (Fig 4). Additionally, the interaction of P-/- and P+/+ PELs with myocytes was associated with typical shortening (10%) of myocyte diastolic length, initially observed 50 to 60 minutes after conjugation.

View larger version (111K): [in this window] [in a new window]   Figure 2. Interaction of P-/- PELs and a ventricular myocyte. A, A ventricular myocyte conjugated with several lymphocytes 30 minutes after lymphocytes were added. Arrows indicate lymphocytes attached to a myocyte (1-µm epoxy resin section stained with alkaline toluidine blue). B, Transmission electron micrograph showing the contact area between a killer lymphocyte (CTL) and a ventricular myocyte (VM).

View larger version (13K): [in this window] [in a new window]   Figure 3. Action potentials from ventricular myocytes conjugated with P+/+ and P-/- PELs. Representative action potentials from a nonconjugated myocyte (A), a myocyte conjugated with P+/+ PELs (B), and a myocyte conjugated with P-/- PELs (C) are shown. In panel C, the lower trace illustrates arrhythmogenic activity in a myocyte conjugated with P-/- PELs (60 minutes after conjugation). The arrows mark two stimulated action potentials. In panel A, action potentials are shown in a nonconjugated myocyte at the onset of the experiment (t=0) and 60 minutes later. In panels B and C, action potentials are shown before (t=0) and 60 minutes after conjugate formation. In panel C, bottom trace, the arrows are at 2-second intervals.

View larger version (12K): [in this window] [in a new window]   Figure 4. Action potentials in myocytes conjugated with P-/- PELs and P+/+ PELs as a function of the interaction duration. A, Resting potential. B, Action potential amplitude. C, APD80.

Effect of Anti-Fas mAb on Ventricular Myocytes
In the previous section, we demonstrated how P-/- PELs, acting via the FasL/Fas pathway, affected conjugated myocytes. To directly test the involvement of Fas activation in the deleterious action of P-/- PELs, we explored the effect of the apoptosis-inducing anti-Fas mAb Jo2 on ventricular myocytes (Fig 5). In these experiments, action potentials were recorded from myocytes incubated for as much as 180 minutes with 10 µg/mL Jo2 at 37°C. Whereas incubation per se (180 minutes at 37°C) did not appear to affect action potential configuration (Fig 5A, two different myocytes), incubation with Jo2 induced prominent action potential alterations (Fig 5B), similar to the alterations caused by P-/- PELs: reduction in V_m and action potential amplitude and increase in APD (the results are summarized in Fig 6). Frequently, in myocytes treated with Jo2, APD prolongation was associated with large early afterdepolarizations (Fig 5B). In control experiments (n=5 myocytes, not shown), we found that 10 µg/mL hamster IgG (Jackson Immune Research Laboratory) did not alter action potential characteristics.

View larger version (10K): [in this window] [in a new window]   Figure 5. Effect of anti-Fas mAb Jo2 on action potentials of ventricular myocytes. Myocytes were incubated for 180 minutes at 37°C in the absence (A) or presence (B) of 10 µg/mL Jo2. Representative action potentials (recorded from five different myocytes) are shown in control nonincubated myocytes (t=0) and after 180 minutes of incubation. In panel B, at 180 minutes of incubation with Jo2 (two traces on right), the lower recording depicts an action potential in which an early afterdepolarization has developed.

View larger version (12K): [in this window] [in a new window]   Figure 6. Action potential changes in myocytes incubated with anti-Fas mAb Jo2. A, Resting potential. B, Action potential amplitude. C, APD80. Myocytes were incubated in the absence (control) or presence of 10 µg/mL Jo2 at 37°C for up to 180 minutes.

Effect of P-/- PELs and Jo2 on [Ca²⁺]_i Transients and Myocyte Contraction
To further investigate Fas-mediated myocyte dysfunction, we monitored [Ca²⁺]_i transients and contraction (represented by cell shortening) in myocytes conjugated with P-/- PELs. In these experiments, fura 2 signals (fluorescence counts at 340 and 380 nm, yielding the ratio R=F₃₄₀/F₃₈₀) and cell shortening were recorded simultaneously, immediately after addition of PELs to the bath (t=0, providing control values), and then during 60 minutes of interaction. During the course of the experiment, fura 2 signals were obtained for a period of 30 to 40 seconds at various time intervals. Before testing the effect of P-/- PELs, we performed control experiments to assess the stability of [Ca²⁺]_i transients and myocyte contraction during a 60-minute experiment in nonconjugated myocytes. As shown in Fig 7A, [Ca²⁺]_i transients and cell shortening were reasonably stable throughout the experiment (60 minutes). In seven nonconjugated myocytes (Table 1), the diastolic [Ca²⁺]_i level (represented by R=F₃₄₀/F₃₈₀) did not change significantly during the 60-minute experiment. In contrast, marked changes in [Ca²⁺]_i transients and cell shortening were observed in a conjugated myocyte (Fig 7B). The cytocidal interaction with P-/- PELs (at 60 minutes; compare with 0 minutes immediately after conjugation) resulted in two important (and likely related) occurrences: (1) an elevation in diastolic [Ca²⁺]_i, from 0.8 to 1.25, and (2) arrhythmogenic activity (indicated by the thin arrows) represented by "aftercontractions" and nonstimulated contractions, probably resulting from early or delayed afterdepolarizations. A summary of PEL-induced changes in [Ca²⁺]_i in 13 conjugated myocytes is shown in Table 1, illustrating that after 60 minutes of interaction with P-/- PELs, the fluorescence ratio was doubled. A smaller rise in [Ca²⁺]_i (R=1.25) seen in Fig 7B is due to the presence of a burst of arrhythmic activity, which tends to transiently lower [Ca²⁺]_i, probably as a result of enhanced extrusion mechanisms.

View larger version (29K): [in this window] [in a new window]   Figure 7. Effect of P-/- PELs on [Ca2+]i transients and contraction of conjugated myocytes. [Ca2+]i levels are represented by the ratio of fura 2 fluorescence at 340 and 380 nm, R=F340/F380. Myocyte contraction represented by cell length is expressed in micrometers. A, A nonconjugated myocyte at the onset of the experiment (t=0) and 60 minutes later. B, A P-/- PEL–conjugated myocyte immediately (t=0) and 60 minutes after conjugate formation. In panel B, the stimulated response is depicted by the thick arrow. The spontaneous responses and the "aftercontraction" are depicted by thin arrows. The horizontal dotted line depicts the diastolic [Ca2+]i level, which was increased during the cytocidal interaction.

Next, we examined the effects of Jo2 on [Ca²⁺]_i transients and myocyte contraction (Fig 8). The prominent effects of Jo2, compared with control, were elevation of diastolic [Ca²⁺]_i (data from 12 Jo2-treated myocytes is summarized in Table 1), induction of "aftercontractions" (Fig 8), and arrhythmogenic activity (nonstimulated contractions) (Fig 8), closely resembling the effects of P-/- PELs. The Fas-induced elevation in diastolic [Ca²⁺]_i resulted in myocyte shortening, expressed as percent decrease of the diastolic length. In control (n=18) and Fas-treated (n=21, 180 minutes) myocytes, diastolic length was, respectively, 114.8±3.2 and 97.7±5.3 µm (P<.05).

View larger version (23K): [in this window] [in a new window]   Figure 8. Effect of anti-Fas mAb Jo2 on [Ca2+]i transients and contraction of ventricular myocytes (four different cells). [Ca2+]i levels are represented by the ratio of fura 2 fluorescence at 340 and 380 nm, R=F340/F380. Myocyte contraction represented by cell length is expressed in micrometers. Myocytes were incubated for 180 minutes at 37°C in the absence (A) or presence (B and C) of 10 µg/mL Jo2. In panels A and B, representative traces are shown in nonincubated myocytes (t=0) and in myocytes incubated for 180 minutes (a different cell). In panel C, the trace was obtained from a myocyte incubated for 180 minutes with Jo2. In panel B, the double-headed arrow depicts a Jo2-induced aftercontraction. In panel C, thick arrows depict stimulated contractions, and thin arrows depict spontaneous contractions. Note that the diastolic [Ca2+]i level (shown by the horizontal dashed line) was increased by Jo2.

Involvement of IP₃ in Fas-Mediated Myocyte Dysfunction
Previous sections demonstrated that P-/- PELs and Jo2 caused a marked elevation in diastolic [Ca²⁺]_i in affected myocytes. An intracellular signaling molecule likely to be involved in the Fas-mediated rise in [Ca²⁺]_i is IP₃. By binding to IP₃-operated Ca²⁺-release channels in the SR, IP₃ elevates [Ca²⁺]_i, which may then trigger transsarcolemmal Ca²⁺ influx,²⁴ contributing to myocyte dysfunction. The IP₃ hypothesis was tested by investigating whether the changes in myocyte functional properties induced by P-/- PELs or Jo2 are modified by the PLC antagonist U-73122 (Research Biochemicals Intl)^{25 26} or by heparin, a blocker of IP₃-operated SR Ca²⁺-release channels.^{27 28} In support of the IP₃ hypothesis, heparin or U-73122 included in the recording pipette solution (to exclude a possible effect of these drugs on the killer lymphocytes) prevented PEL-induced changes in action potential characteristics (Fig 9). Both drugs also prevented the occurrence of arrhythmogenic activity as well as myocyte shortening. Next, we tested whether U-73343 (Research Biochemicals Intl), an inactive isomer of U-73122, can interfere with P-/- PEL action. Unlike U-73122, treating myocytes with U-73343 (2 µmol/L) before conjugate formation did not provide protection against killer lymphocytes. Before (t=0) and 60 minutes after conjugate formation in the presence of U-73343 (n=3, t=60 minutes), action potential characteristics were, respectively, as follows: V_m, -74.4±0.5 and -60.7±0.9 mV; action potential amplitude, 110.7±4.5 and 54.6±9.2 mV; and APD₈₀, 27±7 and 22±2 milliseconds (all P<.05). For an unknown reason, APD was not increased in the presence of U-73343. Additionally, U-73343 did not prevent P-/- PEL–induced arrhythmogenic activity and myocyte shortening. The IP₃ hypothesis was further supported by the finding that U-73122 prevented Jo2-induced alterations in action potential characteristics. In these experiments, U-73122 was added 30 minutes before Jo2. In Jo2-treated myocytes (10 µmol/L, 180-minute incubation) in the absence (n=7) or presence of 2 µmol/L U-73122 (n=6), action potential characteristics were, respectively, as follows: V_m, -66.5±1.3 and -77.5±0.7 mV; action potential amplitude, 94.1±4.8 and 137.3±7.3 mV; and APD₈₀, 165±35 and 26±5 milliseconds (all P<.01). Action potential characteristics of myocytes exposed to Jo2 in the presence of U-73122 were similar to control action potential characteristics.

View larger version (15K): [in this window] [in a new window]   Figure 9. Summary of the effects of heparin (50 µg/mL) and U-73122 (2 µmol/L) on action potential characteristics of P-/- PEL–conjugated myocytes. A, Resting potential. B, Action potential amplitude. C, APD80. In control experiments, we found that heparin and U-73122 included in the pipette solution did not affect action potential characteristics of nonconjugated myocytes (not shown).

Because [Ca²⁺]_i elevation was a prominent effect of direct Fas activation, we tested the ability of U-73122 (the active PLC inhibitor) and U-73343 (the inactive form) to modify the Jo2-induced rise in [Ca²⁺]_i. In further support of the IP₃ hypothesis, U-73122 completely prevented the Jo2-induced rise in diastolic [Ca²⁺]_i levels, whereas U-73343 did not (Table 1). Accordingly, aftercontractions and arrhythmogenic activity occurred in the presence of U-73343 but not in the presence of U-73122.

To determine whether the rise in [Ca²⁺]_i is a necessary component of Fas-mediated cell injury, myocytes were pretreated with 10 µmol/L ryanodine (Alomone Labs) for 30 minutes before and during the entire exposure (180 minutes) to Jo2. Importantly, ryanodine prevented Jo2-induced myocyte electrophysiological and arrhythmogenic effects (ie, generation of early afterdepolarizations) as well as myocyte shortening. In the presence of ryanodine, Jo2 did not alter action potential characteristics, which were (n=6) as follows: V_m, -74.5±1.6 mV; action potential amplitude, 118.3±4.8 mV; and APD₈₀, 22.2±3.5 milliseconds.

Direct Intracellular Effect of IP₃ on the Action Potential
To test directly the involvement of IP₃ in Fas-induced myocyte dysfunction, 1,4,5-IP₃ (2 µmol/L), the Ca²⁺-releasing compound, was applied intracellularly by including it in the patch pipette solution (Fig 10). As seen by the representative action potential traces, intracellular application of 1,4,5-IP₃ (traces b and c) induced oscillations in the membrane potential and arrhythmogenic activity, which was observed during the first 10 minutes after establishing of the whole-cell configuration. In contrast, intracellular application of the nonfunctional derivative 1,3,4-IP₃ was ineffective (Fig 10d). The effects of 1,4,5-IP₃ and 1,3,4-IP₃ on action potential characteristics are summarized in Fig 11. Whereas 1,3,4-IP₃ was ineffective, 1,4,5-IP₃ significantly decreased V_m and action potential amplitude and prolonged APD, resembling the effects of P-/- PELs and Jo2.

View larger version (15K): [in this window] [in a new window]   Figure 10. Electrophysiological effects of intracellular application IP3. Representative traces of action potentials from control ventricular myocytes (a) and myocytes treated with 2 µmol/L 1,4,5-IP3 (b and c) and with 2 µmol/L 1,3,4-IP3 (d). IP3 compounds were included in the pipette solution.

View larger version (17K): [in this window] [in a new window]   Figure 11. Summary of the effects of 1,4,5-IP3 (n=12, 2 µmol/L) and 1,3,4-IP3 (n=6, 2 µmol/L) on action potential characteristics of ventricular myocytes. A, Resting potential. B, Action potential amplitude. C, APD80. In the 1,4,5-IP3 experiments, measurements were obtained at the time of peak effect, occurring 5 to 10 minutes after establishing the whole-cell configuration. Control and 1,3,4-IP3 measurements were obtained at comparable time intervals. In panels A through C, significance (P<.05) was determined for 1,4,5-IP3 compared with control and 1,3,4-IP3.

P-/- PELs Damage Ventricular Myocytes by Activating the Fas Receptor
To ascertain initiation of Fas-based cytotoxicity by P-/- PELs, we have tested their cytocidal activity against Fas-expressing and -nonexpressing target cells (LF⁺ and LF^-, respectively). The LF⁺ and LF^- cell lines were derived from leukemia L1210 of DBA/2 mice, transfected with Fas overexpression and anti-sense constructs, respectively. By means of routine FACS analysis, we have found that 77% of LF⁺ cells were positively stained for Fas, versus only 2.2% of the LF^- cells. To test whether retarded Fas expression abrogated the susceptibility of LF^- to Fas-based apoptosis, we examined the effect of Jo2 on LF⁺ and LF^- cells. Indeed, the results (Table 2) reveal refractoriness to Fas-induced apoptosis of the low-level Fas–expressing LF^- cells compared with the LF⁺ cells. Although these results clearly show refractoriness to Jo2, we further established comparable differential refractoriness to Fas-based CTL-mediated apoptosis. To ascertain clear-cut Fas-based CTL action, P-/- PELs were used. The results (Table 3) clearly show that whereas LF⁺ cells were effectively lysed by P-/- PELs, LF^- cells were not. Hence, LF^- cells are refractory to Fas-based apoptosis induced either by Jo2 or by CTLs.

Finally, we tested the effect of prior occupation of Fas by Jo2 (10 µg/mL, 10 minutes, 24.0°C to 25.0°C) on P-/- PEL–myocyte interaction. In P-/- PEL–conjugated myocytes (60 minutes after conjugation) untreated or treated with Jo2 (before conjugation), action potential characteristics were, respectively, as follows: V_m, -52.6±5.5 and –74.4±1.2 mV; action potential amplitude, 84.6±12.2 and 116.8±3.9 mV; and APD₈₀, 42.9±17.2 and 10.8±1.2 milliseconds (all P<.05). Thus, prior occupation of the Fas receptor completely prevented PEL-induced myocyte dysfunction, providing a clear-cut proof that P-/- PELs damage myocytes by activating the Fas receptor.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
In the present study, we have investigated the cellular mechanisms responsible for Fas-based functional derangements in ventricular myocytes induced either by CTLs or antibody-mediated activation of the Fas pathway. Specifically, we have tested the hypothesis that Fas-based myocyte dysfunction is mediated by increased IP₃ production. To this end, we have developed an in vitro model system of murine allogeneic interaction of ventricular myocytes with peritoneal exudate CTLs (PELs), obtained from perforin gene–knockout (P-/-) mice, thus excluding perforin-mediated lytic action. The model has two advantages over the xenogeneic model of murine CTL–guinea pig myocytes that we previously used.^{16 21 22} First, it is a closer representation of the in vivo setting. Second, it offers an opportunity to investigate the contribution of distinct mutations in the FasL/Fas system (eg, gld, lpr, and lpr^cg) to heart-directed lymphocytotoxicity.

The novel observation made was that myocyte dysfunction caused by P-/- PELs or by the apoptosis-inducing anti-Fas mAb Jo2 was prevented by blocking the IP₃ pathway. Importantly, it was demonstrated that freshly dissociated ventricular myocytes express Fas (Fig 1). We have found that in P-/- PEL–conjugated or Jo2-treated myocytes, V_m and action potential amplitude were reduced, APD was markedly prolonged, and delayed or early afterdepolarizations (frequently culminating into triggered arrhythmias) were generated. Comparable electrophysiological effects were induced by PELs from perforin-containing (P+/+) mice interacting with guinea pig ventricular myocytes,¹⁶ suggesting that at least in this in vitro setting, the absence of the pore-forming protein perforin did not diminish the capacity of P-/- PELs. Additionally, diastolic [Ca²⁺]_i levels were elevated in myocytes interacting with P-/- PELs or treated with Jo2. We have previously reported¹⁶ that in conjugated myocytes P+/+ PELs induced higher (R=4.0) [Ca²⁺]_i elevation than in the present study (R=2.0). This difference can result from the following: (1) a contribution of Ca²⁺ influx through perforin channels (in P+/+ PELs) to the [Ca²⁺]_i accumulation, (2) an IP₃-induced Ca²⁺ release stronger in guinea pig than in murine ventricular myocytes, and (3) the use of 1.8 mmol/L [Ca²⁺]_o in the previous study compared with 1.0 mmol/L [Ca²⁺]_o in the present study.

In support of the IP₃ hypothesis, heparin, an antagonist of IP₃-operated Ca²⁺ SR channels, and U-73122, a specific PLC inhibitor, prevented P-/- PEL–induced and Jo2-induced action potential changes and arrhythmogenic activity, as well as myocyte shortening. Accordingly, U-73122 prevented Jo2-induced elevation in diastolic [Ca²⁺]_i levels. Clear-cut support for the involvement of IP₃ in Fas-based myocyte dysfunction is found in the novel observation that intracellular application of the active IP₃ analogue, 1,4,5-IP₃, but not the inactive analogue, 1,3,4-IP₃, caused electrophysiological changes resembling those brought about by P-/- PELs and Jo2. These results are in accord with recent work¹⁶ on the cytocidal interaction of mouse P+/+ PELs with guinea pig ventricular myocytes. The major electrophysiological difference was that the APD of conjugated guinea pig ventricular myocytes was decreased¹⁶ compared with the marked increase seen in the present study. This difference may reflect the experimental models used.

Involvement of IP₃ in Fas-Mediated Myocyte Dysfunction
An important question regarding the finding that Fas-based damage of myocytes was inhibited by blocking the IP₃ pathway is how IP₃ accumulation can cause myocyte dysfunction. It is well established that the second messenger IP₃ plays an important role in [Ca²⁺]_i mobilization by opening IP₃-operated SR (or ER) Ca²⁺ channels.^{29 30 31} Recently, several laboratories have suggested that IP₃ also has a role in intracellular Ca²⁺ homeostasis in cardiac preparations.^{32 33 34} For example, Borgatta et al³² demonstrated the occurrence of a low-conductance Ca²⁺-release channel sensitive to IP₃ in SR vesicles from the canine ventricular septum. It appears that at least under normal conditions, the heart differs from other tissues. Whereas in T lymphocytes³⁵ IP₃ triggers a rapid and transient release of Ca²⁺ followed by transmembrane Ca²⁺ influx, in both atria and ventricular muscle IP₃ causes a slow leakage of Ca²⁺ from intracellular stores.^{36 37 38 39} Several studies support the association of IP₃ accumulation and functional (mostly electrophysiological) derangements in a number of disease states.^{23 40 41} For example, IP₃-enhanced spontaneous Ca²⁺ oscillations (characteristics of Ca²⁺-overloaded SR) occur in saponin-skinned rat papillary muscle.³⁹ Ca²⁺ oscillations, commonly seen in postischemic reperfusion, contribute to ventricular arrhythmogenesis and are perhaps compatible with Fas-induced early and delayed afterdepolarizations seen in the present study. Interestingly, adding PLC to guinea pig ventricular myocytes caused functional alterations resembling those seen in PEL-bound myocytes, including a reduction of V_m, induction of delayed afterdepolarizations, increased [Ca²⁺]_i, and cell destruction.⁴² Strong support for the IP₃ hypothesis is drawn from recent studies investigating the association between IP₃ and arrhythmogenic activity resulting from reperfusion arrhythmias in rat hearts.^{40 43 44} Jacobsen et al⁴⁴ have found that reperfusion following acute ischemia causes a rapid transient increase in IP₃ levels, which is dependent on local release of norepinephrine. They have also shown that the PLC inhibitor U-73122 (but not its inactive isomer U-73343) inhibits IP₃ generation and thrombin-induced arrhythmias, but not those initiated by epinephrine, thus establishing the requirement for IP₃ production in arrhythmogenesis under these conditions. It should be noted that U-73343 differs chemically from U-73122 only in one double bond and that the only functional difference lies in their efficacy for inhibiting IP₃-specific PLC.⁴⁵ Thus, the effectiveness of U-73122 in preventing Fas-mediated myocyte dysfunction in the present study indicates the involvement of PLC.

A probable candidate mediating IP₃-based damage is elevated [Ca²⁺]_i, known to be cytotoxic to a variety of cell types, including cardiac myocytes.⁴⁶ To the best of our knowledge, no other adverse action of IP₃ has been reported. Electrophysiologically, increased [Ca²⁺]_i has at least three important consequences: (1) stimulation of a transient inward current evoking delayed afterdepolarizations, so that triggered activity can develop in otherwise quiescent ventricular muscle, (2) generation of Ca²⁺-dependent slow responses in depolarized fibers, so that conditions for reentry are favored, and (3) intracellular uncoupling with slowing of conduction.⁴⁷ Although [Ca²⁺]_i elevation in myocytes cannot account for all of the Fas-induced (IP₃-mediated) adverse effects, such as attenuated V_m and action potential amplitude, it can trigger the arrhythmogenic activity seen in myocytes conjugated with P-/- PELs or exposed to Jo2. Additionally, [Ca²⁺]_i elevation can cause morphological changes in myocytes. Nevertheless, although Fas activation indeed caused prominent [Ca²⁺]_i elevation, it is questionable whether it is sufficient to trigger cell destruction (and apoptosis), suggesting that other elements of the Fas cascade may contribute to myocyte dysfunction. In addition to direct effects, elevated [Ca²⁺]_i can affect myocyte function indirectly, by activating a variety of intracellular components, such as Ca²⁺-dependent phosphatases and endonucleases.^{46 48}

That [Ca²⁺]_i elevation may be associated with Fas-based action (eg, apoptosis) was suggested by several groups. Oshimi and Miyazaki⁴⁹ have found that in the human B cell line FMO, apoptosis induced by anti-Fas mAb is associated with [Ca²⁺]_i elevation, which is proposed by the authors (on the basis of chelation of [Ca²⁺]_i) to be a prerequisite for DNA and chromatin fragmentation. The SR-ER Ca²⁺-ATPase blocker thapsigargin, which initiates [Ca²⁺]_i rise by depleting [Ca²⁺]_i stores and thus generating Ca²⁺ influx through specific plasma membrane Ca²⁺ channels (I_CRAC), commits to apoptosis cells belonging to different lineages, such as human hepatoma cells⁵⁰ and mouse lymphoma cells.⁵¹ Furthermore, Rovere et al⁵² have suggested that in normal human V9/vd2⁺ T-cell clones, engagement of the Fas receptor causes [Ca²⁺]_i mobilization, resulting from activation of I_CRAC, which is required for the subsequent apoptosis. Accordingly, the bcl-2 gene product, which represses apoptosis, has been suggested to interfere with the [Ca²⁺]_i mobilization that is associated with apoptosis induced by growth factor withdrawal and, in particular, to protect from apoptosis via inhibition of Ca²⁺ release and subsequent I_CRAC activation.⁵¹

An important question, yet to be answered, is how Fas activation causes IP₃ accumulation. An alternative explanation of the IP₃ hypothesis (although unlikely) is that both heparin and U-73122 prevent Fas-mediated damage by interfering with one or more downstream mediators of the Fas signaling pathway. At least theoretically, none of the cardiac receptors (eg, ₁-adrenergic, muscarinic, endothelin, angiotensin II, and thrombin) that are coupled to IP₃ turnover^{23 53 54 55} are likely to be directly triggered by Fas activation; therefore, determining how the IP₃ pathway is activated by FasL requires further investigation. Finally, support for the involvement of the IP₃R in myocyte damage has been provided by Jayaraman and Marks.⁵⁶ Based on the findings that IP₃R1-deficient T cells are resistant to apoptosis induced by dexamethasone, TCR stimulation, ionizing radiation and Fas, they suggest that intracellular calcium release via the IP₃R1 is a critical mediator of apoptosis.

Fas Activation Is Directly Responsible for Myocyte Dysfunction Induced by P-/- PELs or by Jo2
Of the two (granzyme and perforin–based and FasL/Fas-mediated) mechanisms of lymphocytotoxicity, discussed in the present study, we investigated the latter. To this end, CTLs from gene-knockout (P-/-) mice have been used (P-/- PELs). That P-/- PELs affected target cells by activating Fas was demonstrated by the following findings: (1) only L1210 leukemia cells transfected with Fas overexpression (LF⁺; 77% Fas-positive, determined by FACS analysis) were lysed by P-/- PELs or Jo2, but not L1210 cells transfected with the Fas antisense construct (LF^-; 2% Fas-positive), and (2) ventricular myocytes in which Fas was initially occupied by Jo2 were subsequently refractory to the damaging action of P-/- PELs.

Although activation of the FasL/Fas pathway may eventually lead to cell death, it does not preclude the possibility that Fas-based myocardial dysfunction can also result from contribution of FasL-affected diseased myocytes to the global decline in cardiac function. That Fas-induced myocyte damage is potentially reversible is supported by the frequent recovery of cardiac function in many cases of clinical myocarditis, a T-cell–dependent disease, and by the observations that systolic dysfunction occurring during heart transplant rejection may be reversed by treating the rejection process.

In summary, the finding that IP₃ is involved in Fas-induced damage to myocytes contributes to the understanding of mechanisms of lymphocytotoxicity as they relate to myocardial pathologies in which heart-infiltrating lymphocytes play a key role, such as heart transplant rejection and DCM. Hence, the intracellular messenger IP₃ mediating myocardial damage may be a target for pharmaceuticals aimed at attenuating the injury inflicted to the affected heart by killer lymphocytes.

	Selected Abbreviations and Acronyms

 

APD = action potential duration APD80 = APD at 80% repolarization Con A = concanavalin A CTL = cytotoxic T lymphocyte DCM = dilated cardiomyopathy ER, SR = endoplasmic and sarcoplasmic reticulum FACS = fluorescence-automated cell sorting FasL = Fas ligand ICRAC = Ca2+ release–activated Ca2+ channels IP3 = inositol trisphosphate 1,3,4-IP3 = inositol 1,3,4-trisphosphate 1,4,5-IP3 = inositol 1,4,5-trisphosphate IP3R = IP3 receptor mAb = monoclonal antibody P-/- = perforin-deficient P+/+ = perforin-containing PEL = peritoneal exudate CTL PLC = phospholipase C R=F340/F380 = fluorescence ratio Vm = resting potential

	Acknowledgments

 
This study was supported by grants to Drs Binah and Berke from the US-Israel Binational Science Foundation, the DKFZ, and the Israel Academy of Sciences. The research was also funded by the Minerva Foundation through the Bernard Katz Center for Cell Biophysics and by the Rappaport Family Institute for Research in the Medical Sciences. The research was also supported by National Heart, Lung, and Blood Institute grant HL-28958 (Dr Robinson). The authors wish to thank Drs Michael Rosen and Elizabeth Woodcock for their critical review of the manuscript.

	Footnotes

 
¹ Both authors contributed equally to this study.

Received July 28, 1997; accepted December 16, 1997.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Barry WH. Mechanisms of immune-mediated myocyte injury. Circulation. 1994;89:2421–2432.[Abstract/Free Full Text]
2. Binah O. Immune effector mechanisms in heart transplant rejection. Cardiol Res. 1994;28:1748–1757.
3. Caves PC, Stinson EB, Billingham ME, Rider AK, Shumway NE. Diagnosis of human. J Heart Transplant. 1988;7:292–297.[Medline] [Order article via Infotrieve]
4. Chomette G, Auriol M, Cabrol C. Chronic rejection in human heart transplantation: autoimmune myocarditis. Clin Exp Immunol. 1994;96:470–475.[Medline] [Order article via Infotrieve]
5. Hanawa H, Kodama M, Inomata T, Izumi T, Shibata A, Tsuchida M, Matsumoto Y, Abo T. Anti-ßT receptor antibody prevents the progression of experimental cardiac allograft rejection by serial cardiac biopsy. J Thorac Cardiovasc Surg.. 1973;66:461–466.[Medline] [Order article via Infotrieve]
6. Leslie K, Blay R, Haisch C, Lodge A, Weller A, Huber S. Clinical and experimental aspects of viral myocarditis. Clin Microbiol Rev. 1989;2:191–203.[Abstract/Free Full Text]
7. Neu N, Ploier B, Ofner C. Cardiac myosin-induced myocarditis: heart autoantibodies are not involved in the induction of the disease. J Immunol. 1990;145:4094–4100.[Abstract]
8. Smith SC, Allen PM. The role of T cells in myosin-induced autoimmune myocarditis. Clin Immunol Immunopathol. 1993;68:100–106.[Medline] [Order article via Infotrieve]
9. Cabeza-Mackert P, Hontebeyrie-Joskowicz M, Chambo J, Levin M, Laguens RP. Tripanosoma cruzi: aberrant expression of class II major histocompatibility complex molecules in skeletal and heart muscle cells of chronically infected mice. J Mol Cell Cardiol. 1991;72:8–14.
10. Berke G. The Fas-based mechanism of lymphocytotoxicity. Hum Immunol.. 1997;54:1–7.[Medline] [Order article via Infotrieve]
11. Liu Y, Cigola E, Cheng W, Kajstura J, Olivetti G, Hintze TH, Anversa P. Myocyte nuclear mitotic division and programmed myocyte cell death characterize the cardiac myopathy induced by rapid ventricular pacing in dogs. Lab Invest. 1995;73:771–787.[Medline] [Order article via Infotrieve]
12. Watanabe-Fukunaga R, Brannan CI, Itoh N, Yonehara S, Copeland NG, Jemkins NA, Nagat S. The cDNA structure, expression, and chromosomal assignment of the mouse Fas antigen. J Immunol. 1992;148:1274–1279.[Abstract]
13. James TN. Normal and abnormal consequences of apoptosis in the human heart: from postnatal morphogenesis to paroxysmal arrhythmias. Circulation. 1994;90:556–573.[Abstract/Free Full Text]
14. Narula J, Haider N, Virmani R, DiSalvo TG, Kolodgie FD, Hajjar RJ, Schmidt U, Semigran MJ, Dec W, Khaw BN. Apoptosis in myocytes in end-stage heart failure. N Engl J Med. 1996;335:1182–1189.[Abstract/Free Full Text]
15. Mallat Z, Tedgui A, Fontaliran F, Frank R, Durigon M, Fontaine G. Evidence of apoptosis in arrhythmogenic right ventricular dysplasia. N Engl J Med. 1996;335:1190–1196.[Abstract/Free Full Text]
16. Felzen B, Berke G, Gardner P, Binah O. Involvement of IP₃ cascade in the damage to guinea pig ventricular myocytes induced by cytotoxic T lymphocytes. Pflugers Arch. 1997;433:721–726.[Medline] [Order article via Infotrieve]
17. Wagoner LE, Zhao L, Bishop K, Chan S, Xu S, Barry WH. Lysis of adult ventricular myocytes by cells infiltrating rejecting murine cardiac allografts. Circulation. 1996;93:111–119.[Abstract/Free Full Text]
18. Binah O, Marom S, Rubinstein I, Robinson RB, Berke G, Hoffman B. Immunological rejection of heart transplant: how lytic cytotoxic T lymphocytes damage guinea-pig ventricular myocytes. Pflugers Arch. 1992;420:172–179.[Medline] [Order article via Infotrieve]
19. Berke G, Sullivan KA, Amos DB. Rejection of ascites tumor allografts, l: isolation, characterization and in vitro reactivity of peritoneal lymphoid effector cells from BALB/c mice immune to EL4 leukosis. J Exp Med. 1972;135:1334–1350.[Abstract]
20. Walsh CM, Matloubian M, Liu CC, Ueda R, Kurahara CG, Christensen JL, Huang MY, Young JD, Ahmed R, Clark WR. Immune function in mice lacking the perforin gene. Proc Natl Acad Sci U S A. 1994;91:10854–10858.[Abstract/Free Full Text]
21. Felzen B, Berke G, Rosen D, Binah O. Mechanisms whereby cytotoxic T lymphocytes damage guinea pig ventricular myocytes. Pflugers Arch. 1995;427:422–431.
22. Felzen B, Lavy R, Garcia M, Berke G, Gardner P, Binah O. Interaction of cytotoxic T lymphocytes and guinea pig ventricular myocytes: pharmacological modulation by blocking K⁺ currents in cytotoxic T lymphocytes. Circ Res.. 1996;78:253–261.[Abstract/Free Full Text]
23. Steinberg SF, Robinson RB, Lieberman HB, Stern DM, Rosen MR. Thrombin modulates phosphoinositide metabolism, cytosolic calcium, and impulse initiation in the heart. Circ Res. 1991;68:1216–1229.[Abstract/Free Full Text]
24. Berrige MJ. Inositol trisphosphate and calcium signaling. Nature. 1993;135:1334–1350.
25. Berven LA, Barritt GJ. Evidence obtained using single hepatocytes for inhibition by the phospholipase C inhibitor U73122 of store-operated Ca²⁺ inflow. Biochem Pharmacol. 1995;49:1373–1379.[Medline] [Order article via Infotrieve]
26. Smallridge RC, Kiang JG, Gost ID, Fein HG, Galloway RJ. U-73122, an aminosteroid phospholipase-C antagonist, noncompetitively inhibits thyrotropin releasing hormone effects in GH3 rat. Endocrinology. 1992;131:1883–1888.[Abstract]
27. Kobayashi S, Kitazawa T, Somlyo AV, Somlyo AP. Cytosolic heparin inhibits muscarinic and -adrenergic Ca²⁺ release in smooth muscle: physiological role of inositol 1,4,5-trisphosphate in pharmacological coupling. J Biol Chem. 1989;264:17997–18004.[Abstract/Free Full Text]
28. Rao GH, Fareed J, White JG. Influence of heparins on inositol 1,4,5-trisphosphate induced calcium mobilization in permeabilized human platelets. Biochem Med Metab Biol. 1991;45:171–180.[Medline] [Order article via Infotrieve]
29. Huisamen B, Mouton R, Opie LH, Lochner A. Demonstration of a specific [³H]INS(1,4,5)P₃ binding site in rat heart sarcoplasmic reticulum. J Mol Cell Cardiol. 1994;26:341–349.[Medline] [Order article via Infotrieve]
30. Irvine RF. Inositol phosphates and Ca²⁺ entry: toward a proliferation or simplification? FASEB J. 1992;6:3085–3090.[Abstract]
31. Supattapone S, Worley PF, Barban JM, Snyder SH. Solubilization, purification, and characterization of an inositol trisphosphate receptor. J Biol Chem. 1988;63:1530–1534.
32. Borgatta L, Watras J, Katz AM, Ehrlich BE. Regional differences in calcium-release channels from heart. Proc Natl Acad Sci U S A. 1991;88:2486–2489.[Abstract/Free Full Text]
33. Fitzgerald M, Anderson KE, Woodcock EA. Inositol-1,4,5-trisphosphate and ins(1,4,5)P3 receptor concentrations in heart tissues. Clin Exp Pharmacol Physiol. 1994;21:257–260.[Medline] [Order article via Infotrieve]
34. Moschella MC, Marks AR. Inositol 1,4,5-trisphosphate receptor expression in cardiac myocytes. J Cell Biol. 1993;120:1137–1146.[Abstract/Free Full Text]
35. Premack BA, Gardner P. Signal transduction by T-cell receptors: mobilization of Ca and regulation of Ca-dependent effector molecules. Am J Physiol. 1992;263:C1119–C1140.[Abstract/Free Full Text]
36. Eckel J, Gerlach-Eskuchen E, Reinauer H. Alpha-adrenoreceptor-mediated increase in cytosolic free calcium in isolated cardiac myocytes. J Mol Cell Cardiol. 1991;23:617–625.[Medline] [Order article via Infotrieve]
37. Fabiato A. Inositol 1,4,5-trisphosphate-induced release of Ca²⁺ from the SR of skinned cardiac cells. Biophys J. 1986;49:190a. Abstract.
38. Kentish JC, Barsotti RJ, Lea TJ, Mulligan IP, Patel JR, Ferenczi MA. Calcium release from cardiac sarcoplasmic reticulum induced by photorelease of calcium of Ins(1,4,5)P₃. Am J Physiol. 1990;58:H610–H615.
39. Zhu Y, Nosek TM. Inositol trisphosphate enhances Ca²⁺ oscillations but not Ca²⁺ induced Ca²⁺ release from cardiac sarcoplasmic reticulum. Pflugers Arch. 1991;418:1–6.[Medline]</