- Email: [email protected]
Ischemic injury to the conducting system of the heart
Ischemic injury to the conducting system of the heart Involvement of myocardial lysosomes The conducting system was studied in an in situ perfused swine heart preparation with reduced coronary flow (ischemia) using perfusate containing high and low levels of glucose (26.6 versus 8.6mM) with and without insulin. Coronary flow was maintained at normal levels for 60 minutes in control hearts. In ischemic hearts flow was reduced to about 50 percent of control levels for 30 minutes. Ultrastructural studies documented only subtle modifications of Purkinje fibers in ischemic hearts. Glycogen depletion and disruption of cell junctions were observed in some fibers. One consistent finding was the activation of the lysosomal system. The outer membranes of primary lysosomes appeared herniated and in some cases disrupted. and small vesicles containing hydrolytic enzymes were seen in association with the Golgi apparatus and larger primary lysosomes . Specimens prepared for the demonstration of acid phosphatase indicated a redistribution of hydrolytic enzymes in Purkinje fibers with a deposition of acid hydrolases in smaller lysosomal vesicles, the transverse and side-to-side junctions between cells, and occasionally in the sarcoplasmic reticulum. Enriched perfusate containing high levels of glucose with insulin appeared to have no therapeutic effects in terms of the structure of the Purkinje fibers. The results suggest that alterations in the lysosomal system may be one of the earliest structural changes which occur in oxygen-deficient hearts.
Lawrence P. McCallister, Ph.D., Arthur J. Liedtke, M.D., and Howard C. Hughes, V.M.D., M.S., Hershey, Pa.
Alterations in lysosomal enzymes and acid phosphatase activity have been studied extensively in the working myocardial cells of ischemic hearts.I " Morphologic studies indicate a redistribution of hydrolytic enzymes and a loss of acid phosphatase activity associated with cell damage and the destruction of subcellular organelles. 3-5 Biochemical studies also indicate an intimate involvement of cardiac Iysosomes in ischemic injury to working myocardial cells.P"" Riciutti" found that Iysosomes isolated from the site of uniform ischFrom the Departments of Anatomy and Medicine and Comparative Medicine, The Milton S. Hershey Medical Center, The Pennsylvania State University, Hershey, Pa. Supported in part by Contract NHLI- 71- 2499 and Grant HL-1863l from the National Institutes of Health. Dr, McCallister is the recipient of a Research Career Development Award. Received for publication Aug. 28, 1978. Accepted for publication Oct. 2, 1978. Address for reprints: L. P. McCallister, Ph.D" Department of Anatomy, The Milton S. Hershey Medical Center, Hershey, Pa. 17033. 0022-5223/79/050647+ 15$01.50/0
©
emic injury in the left posterior papillary muscle of the dog heart had much less enzymatic activity or were disrupted. He!" also noted that lysosome stability was significantly reduced I hour after the onset of ischemia, and continued to decrease as the duration of ischemia increased. Further, it has been proposed that the increase in free activity of hydrolytic enzymes at the expense of particle-bound acid hydrolases in ischemic myocardial cells may represent the autolytic phase of cell death and destruction. 7 Much less is known about the effects of oxygen deprivation on the conducting tissue of the heart. Baba and associates'" found that ischemia of the isolated perfused rat heart resulted in decreased glycogen, mitochondrial swelling, myofibrillar supercontraction, and sarcoplasmic vesiculation of sinoatrial (SA) nodal cells apparent as early as 5 minutes after the onset of injury.f" Morphologic alterations of atrioventricular (A V) nodal cells were not identified until 10 to 15 minutes after oxygen deprivation.P However, ultrastructural and biochemical studies have not docu-
1979 The C. V. Mosby Co.
647
The Journal of
648
McCallister et at.
mented changes in the distribution of hydrolytic enzymes in the conducting tissue of the heart as a result of ischemia. The present experiments were designed to study the involvement of cardiac lysosomes in ischemic injury to the Purkinje fibers of an in situ working swine heart preparation with reduced coronary flow. Studies were carried out in animals perfused with media containing normal levels of glucose without insulin and animals which received supplements with high levels of glucose and insulin to determine if such an intervention might improve the stability of the myocardial cells. Attention was paid to alterations in the transverse and side-to-side junction of the Purkinje fibers, since electrophysiological studies indicate that Purkinje fibers surviving in an infarct have action potentials diminished in amplitude and rapidity and this impairment may be one of the major causes of myocardial arrhythmias. Materials and methods Twenty-five swine of either sex, weighing an average of 42.1 ± 1.1 kilograms, were studied following anesthesia with pentobarbital (35 mg. per kilogram) and the establishment of controlled positive-pressure ventilation with 100 percent oxygen. The animal's arterial pH, Po 2 , and carbon dioxide combining power were determined frequently throughout each study to assure adequacy of ventilation and acid-base balance. The technique of Liedtke and associates 12 was used to induce whole-heart ischemia. Following bilateral thoracotomy with transsternotomy, right heart bypass from the venae cavae to the pulmonary artery was constructed, with antegrade control of systemic cardiac output maintained by a high-flow Sarns roller pump (No. 6003). The circuit was primed with I to 2 L. of heparinized, fresh whole blood from a donor pig. Total coronary perfusion was established with a recirculation closed loop from the right ventricular drainage sump to cannulated left and right coronary arteries at flow rates controlled by a low-flow Sarns perfusion pump (No. 6050). Coronary perfusate was reoxygenated at 37.5° C. by a Bentley blood oxygenator (Model Q-130) with a 97: 3 percent gas mixture of oxygen and carbon dioxide. The central coronary loop volume in each study averaged 1,500 ml. Coronary perfusate consisted of donor pig erythrocytes, which were washed twice with Krebs-Henseleit bicarbonate buffer within 24 hours of the experiment and combined on the day of use with buffer to a final average hemoglobin concentration of 8.0 ± 0.1 Gm. per 100 ml. Palmitic acid was com-
Thoracic and Cardiovascular Surgery
bined with albumin, dialyzed overnight against KrebsHenseleit buffer, and added to the washed red blood cells on the day of study to give a final albumin concentration of 4 Gm. per 100 ml. and free fatty acid levels of 1.26 ± 0.04mM. The final concentration of salts in the buffered (pH 7.40) medium, in millimoles, was as follows: NaC!, 118; KCl, 4.7; CaCI2 , 5; MgS0 4 , 1.2; KH 2P04 , 1.2; Ca-EDTA, 0.5; and NaHC0 3,25. In 14 animals, the arterial glucose concentration in the coronary loop averaged 8.6 ± 0.8mM (154 mg. per 100 ml.); in these hearts no insulin was added to the perfusate. In II swine, coronary perfusate was enriched approximately threefold with excess glucose (loop concentration 26.6 ± 1.3mM, 478 mg. per 100 ml.), in addition to regular crystalline zinc insulin (0.025 units per milliliter). Experimental design. The experiment was designed to restrict coronary flow in an in situ working swine heart preparation and to observe the effects of oxygen deprivation on the structure and lysosomal system of the conducting tissue of the heart in animals perfused with media containing normal levels of glucose without insulin and in animals which received supplements with high levels of glucose and insulin. Experimental animals were compared to control animals perfused with identical media and for similar time periods. Control coronary flows, determined initially by measuring right ventricular drainage, were maintained for 60 minutes in II hearts (seven perfused with little glucose and no insulin and four with excess glucose and insulin added). In 14 hearts (seven perfused with little glucose and no insulin and seven with excess glucose and insulin added), normal coronary perfusion was maintained for 20 minutes and then critically restricted over a 10 minute period (45 and 53 percent reductions in total coronary flow, respectively, for the remaining 30 minutes of perfusion) to determine the early effects of ischemia on the heart. The initial 20 minute period of control perfusions in these hearts permitted adequate mixing of the various ingredients in the central coronary loop volume. Gradual rather than precipitous reductions in coronary flows were necessitated, since in this preparation sudden drops in flow of large magnitude resulted almost invariably in ventricular fibrillation. Microscopy. Good samples were obtained of the Purkinje fibers forming the left limb of the bundle of His where it descends in the form of a flattened band beneath the endocardium of the left ventricle by taking tissues from the left side of the interventricular septum
Volume 77 Number 5 May, 1979
Lysosomes in ischemic hearts
Fig. I. Electron micrograph showing the appearance of Purkinje fibers (P) in the descending left limb of the His bundle in control hearts perfused for 60 minutes with normal levels of glucose without insulin. Note that the cells possess a glycogen -rich sarcoplasm and often are joined together by transverse and side-to-side junctions (arr o ws) forming groups of two to four cells . As shown in the micrograph, the myofibrils often occupied a peripheral position in the sarcoplasm of the cells. (x4,190.)
649
The Journal of
650 McCallister et al.
near its origin from the atrioventricular valve. Hearts were prepared for examination under light and electron microscopes by perfusion fixation. Before fixation, red cells were removed by retrograde perfusion of the coronary arteries through the aorta with physiological saline (0.9 percent sodium chloride) at 37° C. After the saline rinse, the hearts were fixed for 3 to 5 minutes at 37° C. by perfusion with the Karnovsky formaldehyde-glutaraldehyde fixative. The fixative, which was diluted I: 7 with O.IM cacodylate buffer and adjusted to pH 7.2, contained 4 percent dextran to prevent swelling of the extracellular space. For the light microscope, a portion of the left ventricular septum was immediately placed in 10 percent neutral buffered formalin and processed for paraffin sectioning. Paraffin sections were cut at 6 foL and stained with hematoxylin and eosin and periodic acid-Schiff (PAS), with and without diastase, for the demonstration of glycogen. For the demonstration of acid phosphatase, a portion of the left ventricular septum was quench frozen in isopentane cooled to liquid nitrogen temperatures; it was stored frozen for subsequent cryostat sectioning. Sections cut on a cryostat at 10 to 15 foL were processed for the demonstration of acid phosphatase by the Gomori technique. For the electron microscope, small blocks of tissue (cut in cubic millimeters) were fixed for an additional 2 hours in full-strength Karnovsky fixative and then post fixed in 1.33 percent s-collidine buffered osmium tetroxide. Next, they were dehydrated rapidly in a graded series of ethanol baths of increasing concentration and were embedded in araldite (Dukupan, Fluka). Thin sections were cut with diamond knives, stained with uranyl acetate and lead hydroxide, and examined in an RCA EMU IV or a Philips 300 electron microscope. For the ultrastructural demonstration of acid phosphatase, small blocks of tissue were postfixed in fullstrength Karnovsky fixative for 30 minutes and were stored in the refrigerator in O.IM cacodylate buffer containing 7.5 percent sucrose. The blocks of tissue were sectioned at 40 foL with a Sorvall MT-l tissue chopper. The thick sections were then incubated for 2 hours in the reaction media for acid phosphatase and were rinsed, postfixed in osmium, dehydrated, and embedded as before. Results Morphology of control hearts perfused with normal levels of glucose without insulin. In the present study the cells of the His bundle were observed to contain typical Purkinje fibers similar in structure to those previously described for the conducting system of large
Thoracic and Cardiovascular Surgery
hearts such as the goat, sheep, and dog.!" Nodal or transitional cells!" 15 were not encountered. The Purkinje fibers could be distinguished easily from the underlying working myocardial cells by the low density of myofibrils and abundance of glycogen-rich sarcoplasm. In addition, the fact that these cells stained intensely PAS positive aided in the identification of the Purkinje fibers with the light microscope. With the electron microscope, the cells were seen to contain one centrally placed nucleus, although cells containing more than one nuclei were occasionally encountered. The myofibrils usually were located near the periphery of the cell (Fig. I); however, aggregates of contractile filaments were also seen in the central portion of the sarcoplasm, the position resembling that of myofibrils found in avian species 16. 17 and in the rat heart. 18 The sarcoplasm also contained numerous mitochondria in all shapes and configurations (Fig. 2). The cristae of the mitochondria were tightly packed and similar in appearance to those of working myocardial cells. Mitochondria with a single, central crista as reported in the conducting tissue of other hearts, were not observed. 13. 19 In addition to the contractile and mitochondrial compartments, the Purkinje fibers contained a highly developed sarcoplasmic reticulum (Fig. 2). The sarcoplasmic reticulum consisted of a ramifying system of tubules and cisternae which were found throughout the sarcoplasm. The tubules formed intimate associations with the mitochondria, myofibrils, and the cell membrane. Where the sarcoplasmic reticulum came in contact with the cell membrane, subsarcolemmal cisternae often were formed (Fig. 2). That lumen of the cisternae appeared to be more electron dense suggests the ability of these organelles to actively sequester calcium ions. Pinocytotic vesicles also were found in association with the sarcolemma and appeared as invaginations of the cell membrane lined with basal lamina. Juxtaposed to the system of tubules forming the sarcoplasmic reticulum were dense-cored vesicles 150 to 600 mfoL in diameter and bounded by a single limiting membrane (Figs. 2 and 3). The vesicles were particularly numerous near the Golgi apparatus, which consisted of several separate stacks of Golgi cisternae located in the nuclear pole zone of the conducting cells. In this region of the Purkinje fibers were also found deposits of lipofuscin pigment. Because of their morphologic appearance and the presence of acid hydrolase activity, the dense-cored vesicles were interpreted as being primary lysosomes. Transverse and side-to-side junctions of the Purkinje fibers were not in the form of conventional intercalated discs, but consisted of meandering stretches of macula
Volume n Number5
Lysosomes in ischemic hearts
May, 1979
Fig. 2. Portion of a Purkinje fiber from a control heart perfused with normal levels of glucose without insulin. Note that the sarcoplasmic reticulum (SR) consists of a ramifying system of tubules which form intimate associations with the myofibrils and mitochondria (M). Where the sarcoplasmic reticulum comes in contact with the surface membrane, subsarcolemmal cisternae (inset, arrow), are formed, which appear more electron dense than pinocytotic vesicles and tubules of the sarcoplasmic reticulum. Note the numerous dense-cored vesicles 150 to 600 mf.L in diameter and resembling primary Iysosomes (PL). The vesicles were particularly numerous in the nuclear pole zone of the Purkinje cells, and often formed intimate associations with the tubules of the sarcoplasmic reticulum and cisternae of the Golgi apparatus (G). (X 16,370; inset, x 37,660.)
651
652 McCallister et al .
The Journal of Thoracic and Cardiovascular Surgery
Fig. 3. Electron micrograph of the nuclear pole zone of a Purkinje fiber from a control heart perfu sed with normal levels of glucose without insulin . Note the numerous primary Iysosomes (PL) found in this portion of the cell. G. Golgi apparatus . M. Mitochondria. ( x 27.750.) Fig. 4. Junction between two Purkinje fibers from a control heart perfused with normal levels of glucose without insulin . Note that the complex consists of macula adheren s, fascia adherens, undifferent iated regions (double arrows) and gap-junctional membrane s (arrows). ( x21 ,620.)
Volume rr Number 5 May, 1979
Lysosomes in ischemic hearts
Fig. 5. Localization of acid phosphatase in the Purkinje fibers of control hearts perfused with normal levels of glucose without insulin . Electron-den se precipitation indicative of acid phosphatase activity is seen in association with primary lysosome s (PL) and the Golgi apparatus (G). N , Nucleus. ( x 16,150 .)
653
The Journal of Thoracic and Cardiovascular Surgery
654 McCallister et al,
I
Fig. 6. A. Control heart perfused with nonnallevels of glucose without insulin . The periodic acid-Schiff (PAS) technique was used. Large deposits of glycogen in the Purkinje fibers can be seen forming the left limb of the bundle of His (arrows). (PAS X425) . B. Ischemic heart perfused with glucose without insulin for 30 minutes . Note that the PAS reaction is much lighter in the Purkinje cells of this heart , and some cells appear to have lost almost all of their glycogen (arrows) . Enriched perfusate containing high levels of glucose with insulin appeared to have no effect in these hearts. (PAS X425.)
Volume n Number5
Lysosomes in ischemic hearts
655
May, 1979
adherens, fascia adherens, undifferentiated regions, and gap-junctional membranes (Fig. 4). These intimate contacts between the conducting cells often formed groups of two to four fibers (Fig. 1). However, typical Purkinje strands as observed in the trabeculae carinae of the species of large mammals 13. 20 were not observed in the descending left limb of the His bundle in the pig heart. Acid phosphatase activity in control hearts perfused with normal levels of glucose without insulin. To determine the involvement of cardiac Iysosomes in ischemic injury to the conducting tissue, we studied pig hearts by means of the Gomori reaction for the demonstration of acid phosphatase at the electron microscopic level. Fig. 5 is an electron micrograph of a control heart perfused for I hour with media containing glucose without insulin. The micrograph shows the deposition of acid hydrolases in the nuclear pole zone of the Purkinje cells. Electron-dense precipitate indicative of acid phosphatase activity is seen in association with stacks of Golgi cisternae and large primary Iysosomes. The remainder of the sarcoplasm, including the nucleus, appears to be relatively free of precipitate. Finally, although the lysosomal compartment appeared normal in control hearts perfused for I hour with media containing glucose without insulin, perfusion of control hearts for 3 hours with buffer containing glucose without insulin may result in the disruption of the lysosomal system and an increase in the number of autophagic vacuoles (BL Munger, personal communication). Morphology and acid phosphatase activity of control hearts perfused with supplements containing high levels of glucose plus insulin. The structure of the conducting tissue in hearts perfused with media containing high levels of glucose plus insulin was not markedly different from that of hearts perfused with normal levels of glucose without insulin. The cell boundaries again consisted of macula adherens alternating with undifferentiated regions, fascia adherens, and gap junctional membranes. The nuclei were centrally placed, and the myofibrils usually occupied a peripheral position in the sarcoplasm just beneath the cell membrane. Around the nucleus were seen areas with electron-dense bodies resembling primary Iysosomes. These dense-cored vesicles stained intensely positive with the Gomori reaction and were found in association with larger deposits of lipofuscin pigment, which also showed a positive reaction for acid hydrolase activity. Morphology and acid phosphatase activity of ischemic hearts perfused with normal levels of glucose without insulin and supplements with high
levels of glucose and insulin. The conducting tissue of swine hearts perfused under ischemic conditions with media containing glucose without insulin showed subtle ultrastructural modifications. Although many areas of the His bundle contained Purkinje fibers which appeared normal, other areas contained cells which showed alterations in glycogen content, modifications of the junctions between cells, and changes in the lysosomal compartment. Cells showing large expanses of glycogen-free sarcoplasm, similar to the loss of glycogen which occurs in ischemic working myocardial cells;" were occasionally observed. Glycogen depletion was still evident in hearts perfused with supplements containing high levels of glucose and insulin (Fig. 6). In addition, large separations between undifferentiated and presumably gap-junctional regions of the boundaries between Purkinje fibers were observed in some cells and were not prevented by perfusion with media containing high levels of glucose and insulin (Figs. 7 and 8). One of the most consistent findings was the disruption of the lysosomal system with occasional redistribution of acid hydrolases. Figs. 9 and 10 show the disruption of primary Iysosomes in hearts perfused under ischemic conditions for 30 minutes, and the spreading of acid hydrolases within the sarcoplasm, particularly to the sarcoplasmic reticulum (Fig. 10). Under these circumstances, primary lysosomal vesicles appeared to be herniated or to lose their outer limiting membranes, which was coincident with a "budding" of larger Iysosomes to form smaller vesicles (Fig. 11). "Budding" of larger primary Iysosomes to form smaller vesicles was also demonstrable by the Gomori technique in ischemic hearts perfused with supplements with high levels of glucose and insulin (Fig. 12). Further, acid hydrolase activity was also detectable between undifferentiated regions and gap-junctional membranes of the intercalated disc and was still present in hearts perfused by metabolic solutions such as supplements containing high levels of glucose and insulin (Fig. 13). Discussion In the early 1960's treatment with polarizing solutions containing glucose, insulin, and potassium (GIK) was described for the relief of some of the symptoms associated with acute myocardial infarction.F Shortly after the introduction of GIK therapy, several studies reported a decreased incidence of arrhythmias, decreased mortality rate, and a return to a normal electrocardiogram in patients receiving GIK treatment. 23-25
The Journal of
656 McCallister et al.
Thoracic and Cardiovascular Surgery
Fig. 7. Ischemic heart perfused with supplements containing high levels of glucose and insulin. Many of these cells still showed large separations of the membranes making up the transverse and side-to-side junctions of the Purkinje fibers (arrows). (x8,060.) Fig. 8. High-power electron micrograph of an ischemic heart perfused with high levels of glucose and insulin. Note the large separations of undifferentiated and presumably gap-junctional regions between cells (arrows). (x20,960.)
Volume n Number 5 May, 1979
Lysosomes in ischemic hearts
Fig. 9. Acid phosphatase in ischemic hearts perfused with normal levels of glucose without insulin. Large lysosomal vesicles (Ly) showing a positive reaction for acid hydrolases appeared to lose their outer limiting membranes and to form smaller vesicles by a budding process (arrows). (x31,080.) Fig. 10. Ischemic heart showing that in some cases acid hydrolases occasionally spread to the sarcoplasmic reticulum, which gave a more intense reaction product for acid phosphatase (arrows) using the Gomori reaction. M, Mitochondria. (x2? ,980.)
657
The Journal of
65 8 McCallister et al.
Since ligation of the coronary arteries previously had been shown to cause a loss of intracellular potassium from the myocardial cells of the dog heart.i" it was proposed that the beneficial effects of GIK therapy were due to potassium entering into damaged and hypopolarized cardiac muscle cells and restoring them to their normal state of polarization. 23 Further, Maroko and associates'? also postulated that the major benefits of GIK treatment might arise primarily from glucose rather than potassium, through an increase in energy production from enhanced glycolysis of the myocardial cells in infarcted areas of the heart. Unfortunately, these early claims regarding GIK treatment have been difficult to confirm, and further studies generally have been disappointingP' ''? Malach'" reported that 65 percent of patients receiving polarizing therapy manifested life-threatening arrhythmias. Fletcher and associates" noted no significant difference in electrocardiographic changes, arrhythmias, and mortality rate of patients who did not receive treatment and those who received infusions of I L. of solution containing either glucose, glucose and insulin, glucose and potassium, or glucose, insulin, and potassium daily for 3 days. Further, Lesch and colleagues'" reported that the tolerance to ischemia was not extended with GIK therapy, and that patients receiving such treatment might have been adversely affected. The present experiments are in accord with those that report little or no beneficial effects of glucose and insulin in the oxygen-deprived heart. The structure of the conducting system did not appear to be significantly different in hearts perfused with low coronary flows for 30 minutes, whether or not the perfusate was enriched with high levels of glucose and insulin. Although some cells of the conducting system appeared the same as in control hearts, other Purkinje fibers showed ultrastructural modifications. Of particular interest was the decrease in subcellular glycogen stores observed in many of the fibers of the descending left limb of the His bundle. This observation is in direct agreement with metabolic studies on the effects of glucose and insulin on carbohydrate utilization previously reported for ischemic hearts by Liedtke;" Rovetto.:" and their colleagues, and clinical findings which show no beneficial effects of GIK therapy. 28-30 The present experiments also form a good basis for understanding the involvement of myocardial Iysosomes in ischemic injury to the heart. Numerous primary lysosomal vesicles, 150 to 600 mf-t in diameter, and bounded by a single limiting membrane, were found associated with stacks of Golgi cisternae and the tubules of the sarcoplasmic reticulum in the nuclear
Thoracic and Cardiovascular Surgery
pole zone of Purkinje fibers; these findings were observed both in hearts perfused at control levels of coronary flow with media containing normal levels of glucose without insulin and in hearts perfused with supplements containing high levels of glucose and insulin. With the Gomori reaction used for the demonstration of acid phosphatase, electron-dense precipitate indicative of acid hydrolase activity was seen in association with these vesicles, the Golgi apparatus, and lipofuscin pigment. In hearts perfused at approximately 50 percent of normal coronary flow for 30 minutes, there appeared to be a disruption of the lysosomal compartment and in some cases a spreading of acid hydrolases throughout the sarcoplasm. Large primary Iysosomes appeared to have herniated or disrupted outer membranes and to have formed smaller lysosomal vesicles by a budding process. The morphologic data favor a budding process rather than a coalescing of Iysosomes, since the vesicles observed were below the lower limit of primary Iysosomes measured in control hearts, i.e., less than 150 mu. The breakdown of primary Iysosomes was coincident with the deposition of acid hydrolases at the junctions between cells, and in some instances reaction product was also observed in the tubules of the sarcoplasmic reticulum. At the same time there appeared to be a dissolution of a small number of undifferentiated and gap-junctional regions of the intercalated disc in some cells. Thus the lysosomal alterations observed in the present study may be among the earliest changes which occur in ischemic hearts and could be intimately involved in the destruction of subcellular organelles. Jennings and associates'" demonstrated that irreversibly damaged myocardial cells have structural defects in the plasma membrane which are reflected by an increased insulin space, increased H 20 and Na" content of the myocardial cells, and failure of the tissue to maintain high levels of K+ and Mg" ". In the present study both morphologic and histochemical evidence indicates that the sarcolemma can be altered during ischemic injury. The lesions of the plasma membrane observed during the early phase of oxygen deprivation were primarily localized to the gap-junctional membranes and undifferentiated regions of the junctions between cells, whereas the extrajunctional sarcolemma appeared relatively normal. An increase in the spacing of membranes making up gap junctions has previously been noted by this laboratory after I hour of normothermic ischemic arrest in the dog heart" and in the isolated perfused rat heart" made ischemic according to the technique of Neely and associates. 35 In the latter model the dissolution of gap junctions occurred as a function of time, was correlated with the breakdown of
Volume n Number 5
Lysosomes in ischemic hearts
May, 1979
Fig. II. Nuclear pole zone of a Purkinje fiber from a heart made ischemic and perfused with supplements containing high levels of glucose and insulin. Note that the lysosomal compartment was disrupted in these cells. and large lysosomal vesicles (Ly) were seen to lose their outer limiting membranes and to form smaller structures by a budding process (arrows). (x27 ,990.) Fig. 12. Micrograph similar to Fig. II, but showing the localization of acid phosphatase in ischemic hearts perfused with supplements containing high levels of glucose and insulin. Acid hydrolase activity is seen in association with lysosomal vesicles (Ly) which appear to be in a budding process (arrows). (X27,350.) Fig. 13. Localization of acid phosphatase in ischemic hearts perfused with supplements containing high levels of glucose and insulin. Note the electron dense precipitate in association with the undifferentiated and gap-junctional regions of the intercalated disc (arrows). (xn,080.)
659
660 McCallister et al.
primary lysosomal vesicles, and was coincident with the onset of irreversible tissue damage. After normothermic ischemic arrest the separation of gap junctions was associated with a widening of the QRS complex of the electrocardiogram or T-wave inversion and ST-segment depression in 80 percent of the animals
studied.":
36
These results are particularly interesting in view of the electrophysiological changes which have been reported to occur in Purkinje fibers surviving in an infarct.!': 37. 38 Lazzara and associates!\' 38 using bipolar electrograms to assess the activity of Purkinje fibers I day after occlusion of the anterior descending coronary artery in the dog heart, reported Purkinje potentials which were diminished in amplitude and rapidity. In addition to diminished resting and action potentials, reduced upstroke velocity, enhanced automaticity, and phase 4 depolarization also were reported. These investigators concluded that the arrhythmias which occur as a result of ligation of the coronary arteries may originate in the altered Purkinje cells of the infarcted zone." Thus the present results, which indicate dissolution of the undifferentiated and gap-junctional membranes of the transverse and side-to-side junctions between cells and the deposition of acid hydrolases in these regions of hearts perfused at approximately 50 percent of their normal coronary flows, may represent a morphologic basis for the physiological phenomena associated with myocardial infarction. We would like to acknowledge the excellent technical assistance of Earle W. Christman II.
2
3
4
5
6
REFERENCES Brachfeld N: Maintenance of cell viability. Circulation 50:202-215, 1969 Brachfeld N, Gemba T: Mechanisms of myocardial cell death. Release of lysosomal hydrolases after ischemia. Clin Res 13:524, 1965 Hoffstein S, Gennaro DE, Weissmann G, Hirsch J, Streuli F, Fox AC: Cytochemical localization of lysosomal enzyme activity in normal and ischemic dog myocardium. Am J Pathol 79:193-206, 1975 McCallister LP, Munger BL, Neely JR: Electron microscopic observations and acid phosphatase activity in the ischemic rat heart. J Mol Cell Cardiol 9:353-364, 1977 McCallister LP, Munger BL, Tyers GFO, Hughes HC: The effect of different methods of protecting the myocardium on lysosomal activation and acid phosphatase activity in the dog heart after one hour of cardiopulmonary bypass. J THoRAc CARDIOVASC SURG 69:644-663, 1975 Riciutti MA: Myocardial lysosome stability in the early stages of acute ischemic injury. Am J Cardiol 30:492497, 1972
The Journal of Thoracic and Cardiovascular Surgery
6a Baba N, Leighton RF, Weissler AN: Experimental cardiac ischemia. Observation of the sinoatrial and atrioventricular nodes. Lab Invest 23: 168-178, 1970 7 Ravens KG, Gudbjamson S: Changes in the activities of lysosomal enzymes in infarcted canine heart muscle. Circ Res 24:851-856. 1969 8 Smith AL, Bird JWC: Distribution and particle properties of the vacuolar apparatus of cardiac muscle tissue. I. Biochemical characterization of cardiac muscle Iysosomes and the isolation and characterization of acid, neutral and alkaline proteases. J Mol Cell Cardiol 7:39-61, 1975 9 Tallan HH, Jones ME, Fruton JS: On the proteolytic enzymes of animal tissues. X. Beef spleen cathepain C. J Bioi Chern 194:793-805, 1952 10 Riciutti MA: Lysosomes and myocardial cellular injury. Am J Cardiol 30:498-502, 1972 II Lazzara R, EI-Sherif N, Sherlag J: Early and late effects of coronary artery occlusion on canine Purkinje fibers. Circ Res 35:4391-399, 1974 12 Liedtke AJ, Hughes HC, Neely JR: An experimental model for studying myocardial ischemia. Correlations of hemodynamic performance and metabolism in the working swine heart. J THoRAc CARDIOVASC SURG 69:203211, 1975 13 Sommer JR, Johnson EA: Cardiac muscle. A comparative study of Purkinje fibers and ventricular fibers. J Cell Bioi 36:497-526, 1968 14 James TN, Sherf L: Ultrastructure of the human atrioventricular node. Circulation 37: 1049-1070, 1968 15 Mochet M, Maravec J, Guillemot H, Hatt PY: The ultrastructure of the rat conductive tissue. An electron microscopic study of the atrioventricular node and the bundle of His. J Mol Cell Cardiol 7:879-889, 1975 16 Bogusch G: Investigations on the fine structure of Purkinje fibers in the atrium of the avian heart. Cell Tissue Res 150:43-56, 1974 17 Gossrau R: Uber das Reizleitungssystem der Vogel. Histochemische une elektronen microskopische Untersuchungen. Histochemie 13: 111-159, 1968 18 Lindner E: Submikroskopische Ringbinden-funktionelle Strukturen der "Weichteilmuskulatur." Z Zelloforsch 88:370-386, 1968 19 Arluk OJ, Rhodin JAG: The ultrastructure of calf heart conducting fibers with special reference to nexuses and their distribution. J Ultrastruct Res 49: 11-23, 1974 20 Johnson EA, Sommer JR: A strand of cardiac muscle. Its ultrastructure and the electrophysio1ogica1 implications of its geometry. J Cell Bioi 33:103-129, 1967 21 Jennings RB, Baum JH, Herdson PB: Fine structural changes in myocardial ischemic injury. Arch Pathol 79:135-143, 1965 22 Sodi-Pallares D: Possibility of a therapy of cellular ion integration in cardiovascular disease. Arch Inst Cardiol Mex 31:557-574, 1961 23 Mittra B: Potassium, glucose, and insulin treatment of myocardial infarction. Lancet 2:607-609, 1965 24 Sodi-Pallares D, Ponce de Leon J, Bisteni A, Medrano
Volume
n
Number 5 May, 1979
25
26
27
28
29
30 31
GA: Potassium, glucose, and insulin in myocardial infarction. Lancet I: 1315-1316, 1969 Sodi-Pallares 0, Testille NR, Fishleder BL, Bisteni A, Medrano GA, Friedland C, DeMicheli A: Effects of an intravenous infusion of potassium-glucose-insulin solution on the electrocardiographic signs of myocardial infarction. A preliminary report. Am J CardioI9:166-181, 1962 Jennings RB, Grout JR, Smelters GW: Studies on distribution and localization of potassium in early myocardial ischemic injury. Arch Pathol 63:586-592, 1957 Maroko PR, Libby P, Sobel BE, Bloor CM, Sybers HD, Shell WE, Covell JW, Braunwald E: Effect of glucoseinsulin-potassium infusion on myocardial infarction following experimental coronary artery occlusion. Circulation 45: 1160-1175, 1972 Fletcher GF, Hurst JW, Schlant RC: "Polarizing" solutions in patients with acute myocardial infarctions. A double-blind study with negative results. Am Heart J 75:319-324, 1968 Lesch M, Teichholz LE, Soeldner JS, Gorlin R: Ineffectiveness of glucose, potassium, and insulin infusion during pacing stress in chronic ischemic heart disease. Circulation 49: 1028-1037, 1974 Malach M: Polarizing solution in acute myocardial infarction (abstr). Am J Cardiol 19: 141, 1967 Rovetto MJ, Whitmer JT, Neely JR: Comparison of the effects of anoxia and whole heart ischemia on carbohydrate utilization in isolated working rat hearts. Circ Res 32:699-711, 1973
Lysosomes in ischemic hearts
66 I
32 Liedtke AJ, Hughes HC, Neely JR: Effects of excess glucose and insulin on' glycolytic metabolism during experimental myocardial ischemia. Am J Cardiol 38: 17-27, 1976 33 Herdson PB, Sommer HB, Jennings RB: A comparative study of the fine structure of normal and ischemic dog myocardium with special reference to early changes following temporary occlusion of a coronary artery. Am J Pathol 46:367-386, 1965 34 Jennings RB, Ganote CE, Reimer KA: Ischemic tissue injury. Am J Pathol 81:179-195, 1975 35 Neely JR, Rovetto MJ, Whitmer JT, Morgan HE: Effects of ischemia on function and metabolism of the isolated working rat heart. Am J Physiol 255:651-658, 1973 36 Tyers GFO, Hughes HC, Todd GJ, Williams DR, Andrews EJ, Prophet GA, Waldhausen JA: Protection from ischemic cardiac arrest by coronary perfusion with cold Ringer's lactate solution. J THORAC CARDIOVASC SURG 67:411-418, 1974 37 Friedman PL, Fenoglio 11, Wit AL: Time course for reversal of electrophysiological and ultrastructural abnormalities in subendocardial Purkinje fibers surviving extensive myocardial infarction in dogs. Circ Res 36:127-144, 1975 38 Lazzara R, El-Sherif N, Sherlag J: Electrophysiological properties of canine Purkinje cells in one-day-old myocardial infarction. Circ Res 33:722-734, 1973
Lawrence P. McCallister, Ph.D., Arthur J. Liedtke, M.D., and Howard C. Hughes, V.M.D., M.S., Hershey, Pa.
Alterations in lysosomal enzymes and acid phosphatase activity have been studied extensively in the working myocardial cells of ischemic hearts.I " Morphologic studies indicate a redistribution of hydrolytic enzymes and a loss of acid phosphatase activity associated with cell damage and the destruction of subcellular organelles. 3-5 Biochemical studies also indicate an intimate involvement of cardiac Iysosomes in ischemic injury to working myocardial cells.P"" Riciutti" found that Iysosomes isolated from the site of uniform ischFrom the Departments of Anatomy and Medicine and Comparative Medicine, The Milton S. Hershey Medical Center, The Pennsylvania State University, Hershey, Pa. Supported in part by Contract NHLI- 71- 2499 and Grant HL-1863l from the National Institutes of Health. Dr, McCallister is the recipient of a Research Career Development Award. Received for publication Aug. 28, 1978. Accepted for publication Oct. 2, 1978. Address for reprints: L. P. McCallister, Ph.D" Department of Anatomy, The Milton S. Hershey Medical Center, Hershey, Pa. 17033. 0022-5223/79/050647+ 15$01.50/0
©
emic injury in the left posterior papillary muscle of the dog heart had much less enzymatic activity or were disrupted. He!" also noted that lysosome stability was significantly reduced I hour after the onset of ischemia, and continued to decrease as the duration of ischemia increased. Further, it has been proposed that the increase in free activity of hydrolytic enzymes at the expense of particle-bound acid hydrolases in ischemic myocardial cells may represent the autolytic phase of cell death and destruction. 7 Much less is known about the effects of oxygen deprivation on the conducting tissue of the heart. Baba and associates'" found that ischemia of the isolated perfused rat heart resulted in decreased glycogen, mitochondrial swelling, myofibrillar supercontraction, and sarcoplasmic vesiculation of sinoatrial (SA) nodal cells apparent as early as 5 minutes after the onset of injury.f" Morphologic alterations of atrioventricular (A V) nodal cells were not identified until 10 to 15 minutes after oxygen deprivation.P However, ultrastructural and biochemical studies have not docu-
1979 The C. V. Mosby Co.
647
The Journal of
648
McCallister et at.
mented changes in the distribution of hydrolytic enzymes in the conducting tissue of the heart as a result of ischemia. The present experiments were designed to study the involvement of cardiac lysosomes in ischemic injury to the Purkinje fibers of an in situ working swine heart preparation with reduced coronary flow. Studies were carried out in animals perfused with media containing normal levels of glucose without insulin and animals which received supplements with high levels of glucose and insulin to determine if such an intervention might improve the stability of the myocardial cells. Attention was paid to alterations in the transverse and side-to-side junction of the Purkinje fibers, since electrophysiological studies indicate that Purkinje fibers surviving in an infarct have action potentials diminished in amplitude and rapidity and this impairment may be one of the major causes of myocardial arrhythmias. Materials and methods Twenty-five swine of either sex, weighing an average of 42.1 ± 1.1 kilograms, were studied following anesthesia with pentobarbital (35 mg. per kilogram) and the establishment of controlled positive-pressure ventilation with 100 percent oxygen. The animal's arterial pH, Po 2 , and carbon dioxide combining power were determined frequently throughout each study to assure adequacy of ventilation and acid-base balance. The technique of Liedtke and associates 12 was used to induce whole-heart ischemia. Following bilateral thoracotomy with transsternotomy, right heart bypass from the venae cavae to the pulmonary artery was constructed, with antegrade control of systemic cardiac output maintained by a high-flow Sarns roller pump (No. 6003). The circuit was primed with I to 2 L. of heparinized, fresh whole blood from a donor pig. Total coronary perfusion was established with a recirculation closed loop from the right ventricular drainage sump to cannulated left and right coronary arteries at flow rates controlled by a low-flow Sarns perfusion pump (No. 6050). Coronary perfusate was reoxygenated at 37.5° C. by a Bentley blood oxygenator (Model Q-130) with a 97: 3 percent gas mixture of oxygen and carbon dioxide. The central coronary loop volume in each study averaged 1,500 ml. Coronary perfusate consisted of donor pig erythrocytes, which were washed twice with Krebs-Henseleit bicarbonate buffer within 24 hours of the experiment and combined on the day of use with buffer to a final average hemoglobin concentration of 8.0 ± 0.1 Gm. per 100 ml. Palmitic acid was com-
Thoracic and Cardiovascular Surgery
bined with albumin, dialyzed overnight against KrebsHenseleit buffer, and added to the washed red blood cells on the day of study to give a final albumin concentration of 4 Gm. per 100 ml. and free fatty acid levels of 1.26 ± 0.04mM. The final concentration of salts in the buffered (pH 7.40) medium, in millimoles, was as follows: NaC!, 118; KCl, 4.7; CaCI2 , 5; MgS0 4 , 1.2; KH 2P04 , 1.2; Ca-EDTA, 0.5; and NaHC0 3,25. In 14 animals, the arterial glucose concentration in the coronary loop averaged 8.6 ± 0.8mM (154 mg. per 100 ml.); in these hearts no insulin was added to the perfusate. In II swine, coronary perfusate was enriched approximately threefold with excess glucose (loop concentration 26.6 ± 1.3mM, 478 mg. per 100 ml.), in addition to regular crystalline zinc insulin (0.025 units per milliliter). Experimental design. The experiment was designed to restrict coronary flow in an in situ working swine heart preparation and to observe the effects of oxygen deprivation on the structure and lysosomal system of the conducting tissue of the heart in animals perfused with media containing normal levels of glucose without insulin and in animals which received supplements with high levels of glucose and insulin. Experimental animals were compared to control animals perfused with identical media and for similar time periods. Control coronary flows, determined initially by measuring right ventricular drainage, were maintained for 60 minutes in II hearts (seven perfused with little glucose and no insulin and four with excess glucose and insulin added). In 14 hearts (seven perfused with little glucose and no insulin and seven with excess glucose and insulin added), normal coronary perfusion was maintained for 20 minutes and then critically restricted over a 10 minute period (45 and 53 percent reductions in total coronary flow, respectively, for the remaining 30 minutes of perfusion) to determine the early effects of ischemia on the heart. The initial 20 minute period of control perfusions in these hearts permitted adequate mixing of the various ingredients in the central coronary loop volume. Gradual rather than precipitous reductions in coronary flows were necessitated, since in this preparation sudden drops in flow of large magnitude resulted almost invariably in ventricular fibrillation. Microscopy. Good samples were obtained of the Purkinje fibers forming the left limb of the bundle of His where it descends in the form of a flattened band beneath the endocardium of the left ventricle by taking tissues from the left side of the interventricular septum
Volume 77 Number 5 May, 1979
Lysosomes in ischemic hearts
Fig. I. Electron micrograph showing the appearance of Purkinje fibers (P) in the descending left limb of the His bundle in control hearts perfused for 60 minutes with normal levels of glucose without insulin. Note that the cells possess a glycogen -rich sarcoplasm and often are joined together by transverse and side-to-side junctions (arr o ws) forming groups of two to four cells . As shown in the micrograph, the myofibrils often occupied a peripheral position in the sarcoplasm of the cells. (x4,190.)
649
The Journal of
650 McCallister et al.
near its origin from the atrioventricular valve. Hearts were prepared for examination under light and electron microscopes by perfusion fixation. Before fixation, red cells were removed by retrograde perfusion of the coronary arteries through the aorta with physiological saline (0.9 percent sodium chloride) at 37° C. After the saline rinse, the hearts were fixed for 3 to 5 minutes at 37° C. by perfusion with the Karnovsky formaldehyde-glutaraldehyde fixative. The fixative, which was diluted I: 7 with O.IM cacodylate buffer and adjusted to pH 7.2, contained 4 percent dextran to prevent swelling of the extracellular space. For the light microscope, a portion of the left ventricular septum was immediately placed in 10 percent neutral buffered formalin and processed for paraffin sectioning. Paraffin sections were cut at 6 foL and stained with hematoxylin and eosin and periodic acid-Schiff (PAS), with and without diastase, for the demonstration of glycogen. For the demonstration of acid phosphatase, a portion of the left ventricular septum was quench frozen in isopentane cooled to liquid nitrogen temperatures; it was stored frozen for subsequent cryostat sectioning. Sections cut on a cryostat at 10 to 15 foL were processed for the demonstration of acid phosphatase by the Gomori technique. For the electron microscope, small blocks of tissue (cut in cubic millimeters) were fixed for an additional 2 hours in full-strength Karnovsky fixative and then post fixed in 1.33 percent s-collidine buffered osmium tetroxide. Next, they were dehydrated rapidly in a graded series of ethanol baths of increasing concentration and were embedded in araldite (Dukupan, Fluka). Thin sections were cut with diamond knives, stained with uranyl acetate and lead hydroxide, and examined in an RCA EMU IV or a Philips 300 electron microscope. For the ultrastructural demonstration of acid phosphatase, small blocks of tissue were postfixed in fullstrength Karnovsky fixative for 30 minutes and were stored in the refrigerator in O.IM cacodylate buffer containing 7.5 percent sucrose. The blocks of tissue were sectioned at 40 foL with a Sorvall MT-l tissue chopper. The thick sections were then incubated for 2 hours in the reaction media for acid phosphatase and were rinsed, postfixed in osmium, dehydrated, and embedded as before. Results Morphology of control hearts perfused with normal levels of glucose without insulin. In the present study the cells of the His bundle were observed to contain typical Purkinje fibers similar in structure to those previously described for the conducting system of large
Thoracic and Cardiovascular Surgery
hearts such as the goat, sheep, and dog.!" Nodal or transitional cells!" 15 were not encountered. The Purkinje fibers could be distinguished easily from the underlying working myocardial cells by the low density of myofibrils and abundance of glycogen-rich sarcoplasm. In addition, the fact that these cells stained intensely PAS positive aided in the identification of the Purkinje fibers with the light microscope. With the electron microscope, the cells were seen to contain one centrally placed nucleus, although cells containing more than one nuclei were occasionally encountered. The myofibrils usually were located near the periphery of the cell (Fig. I); however, aggregates of contractile filaments were also seen in the central portion of the sarcoplasm, the position resembling that of myofibrils found in avian species 16. 17 and in the rat heart. 18 The sarcoplasm also contained numerous mitochondria in all shapes and configurations (Fig. 2). The cristae of the mitochondria were tightly packed and similar in appearance to those of working myocardial cells. Mitochondria with a single, central crista as reported in the conducting tissue of other hearts, were not observed. 13. 19 In addition to the contractile and mitochondrial compartments, the Purkinje fibers contained a highly developed sarcoplasmic reticulum (Fig. 2). The sarcoplasmic reticulum consisted of a ramifying system of tubules and cisternae which were found throughout the sarcoplasm. The tubules formed intimate associations with the mitochondria, myofibrils, and the cell membrane. Where the sarcoplasmic reticulum came in contact with the cell membrane, subsarcolemmal cisternae often were formed (Fig. 2). That lumen of the cisternae appeared to be more electron dense suggests the ability of these organelles to actively sequester calcium ions. Pinocytotic vesicles also were found in association with the sarcolemma and appeared as invaginations of the cell membrane lined with basal lamina. Juxtaposed to the system of tubules forming the sarcoplasmic reticulum were dense-cored vesicles 150 to 600 mfoL in diameter and bounded by a single limiting membrane (Figs. 2 and 3). The vesicles were particularly numerous near the Golgi apparatus, which consisted of several separate stacks of Golgi cisternae located in the nuclear pole zone of the conducting cells. In this region of the Purkinje fibers were also found deposits of lipofuscin pigment. Because of their morphologic appearance and the presence of acid hydrolase activity, the dense-cored vesicles were interpreted as being primary lysosomes. Transverse and side-to-side junctions of the Purkinje fibers were not in the form of conventional intercalated discs, but consisted of meandering stretches of macula
Volume n Number5
Lysosomes in ischemic hearts
May, 1979
Fig. 2. Portion of a Purkinje fiber from a control heart perfused with normal levels of glucose without insulin. Note that the sarcoplasmic reticulum (SR) consists of a ramifying system of tubules which form intimate associations with the myofibrils and mitochondria (M). Where the sarcoplasmic reticulum comes in contact with the surface membrane, subsarcolemmal cisternae (inset, arrow), are formed, which appear more electron dense than pinocytotic vesicles and tubules of the sarcoplasmic reticulum. Note the numerous dense-cored vesicles 150 to 600 mf.L in diameter and resembling primary Iysosomes (PL). The vesicles were particularly numerous in the nuclear pole zone of the Purkinje cells, and often formed intimate associations with the tubules of the sarcoplasmic reticulum and cisternae of the Golgi apparatus (G). (X 16,370; inset, x 37,660.)
651
652 McCallister et al .
The Journal of Thoracic and Cardiovascular Surgery
Fig. 3. Electron micrograph of the nuclear pole zone of a Purkinje fiber from a control heart perfu sed with normal levels of glucose without insulin . Note the numerous primary Iysosomes (PL) found in this portion of the cell. G. Golgi apparatus . M. Mitochondria. ( x 27.750.) Fig. 4. Junction between two Purkinje fibers from a control heart perfused with normal levels of glucose without insulin . Note that the complex consists of macula adheren s, fascia adherens, undifferent iated regions (double arrows) and gap-junctional membrane s (arrows). ( x21 ,620.)
Volume rr Number 5 May, 1979
Lysosomes in ischemic hearts
Fig. 5. Localization of acid phosphatase in the Purkinje fibers of control hearts perfused with normal levels of glucose without insulin . Electron-den se precipitation indicative of acid phosphatase activity is seen in association with primary lysosome s (PL) and the Golgi apparatus (G). N , Nucleus. ( x 16,150 .)
653
The Journal of Thoracic and Cardiovascular Surgery
654 McCallister et al,
I
Fig. 6. A. Control heart perfused with nonnallevels of glucose without insulin . The periodic acid-Schiff (PAS) technique was used. Large deposits of glycogen in the Purkinje fibers can be seen forming the left limb of the bundle of His (arrows). (PAS X425) . B. Ischemic heart perfused with glucose without insulin for 30 minutes . Note that the PAS reaction is much lighter in the Purkinje cells of this heart , and some cells appear to have lost almost all of their glycogen (arrows) . Enriched perfusate containing high levels of glucose with insulin appeared to have no effect in these hearts. (PAS X425.)
Volume n Number5
Lysosomes in ischemic hearts
655
May, 1979
adherens, fascia adherens, undifferentiated regions, and gap-junctional membranes (Fig. 4). These intimate contacts between the conducting cells often formed groups of two to four fibers (Fig. 1). However, typical Purkinje strands as observed in the trabeculae carinae of the species of large mammals 13. 20 were not observed in the descending left limb of the His bundle in the pig heart. Acid phosphatase activity in control hearts perfused with normal levels of glucose without insulin. To determine the involvement of cardiac Iysosomes in ischemic injury to the conducting tissue, we studied pig hearts by means of the Gomori reaction for the demonstration of acid phosphatase at the electron microscopic level. Fig. 5 is an electron micrograph of a control heart perfused for I hour with media containing glucose without insulin. The micrograph shows the deposition of acid hydrolases in the nuclear pole zone of the Purkinje cells. Electron-dense precipitate indicative of acid phosphatase activity is seen in association with stacks of Golgi cisternae and large primary Iysosomes. The remainder of the sarcoplasm, including the nucleus, appears to be relatively free of precipitate. Finally, although the lysosomal compartment appeared normal in control hearts perfused for I hour with media containing glucose without insulin, perfusion of control hearts for 3 hours with buffer containing glucose without insulin may result in the disruption of the lysosomal system and an increase in the number of autophagic vacuoles (BL Munger, personal communication). Morphology and acid phosphatase activity of control hearts perfused with supplements containing high levels of glucose plus insulin. The structure of the conducting tissue in hearts perfused with media containing high levels of glucose plus insulin was not markedly different from that of hearts perfused with normal levels of glucose without insulin. The cell boundaries again consisted of macula adherens alternating with undifferentiated regions, fascia adherens, and gap junctional membranes. The nuclei were centrally placed, and the myofibrils usually occupied a peripheral position in the sarcoplasm just beneath the cell membrane. Around the nucleus were seen areas with electron-dense bodies resembling primary Iysosomes. These dense-cored vesicles stained intensely positive with the Gomori reaction and were found in association with larger deposits of lipofuscin pigment, which also showed a positive reaction for acid hydrolase activity. Morphology and acid phosphatase activity of ischemic hearts perfused with normal levels of glucose without insulin and supplements with high
levels of glucose and insulin. The conducting tissue of swine hearts perfused under ischemic conditions with media containing glucose without insulin showed subtle ultrastructural modifications. Although many areas of the His bundle contained Purkinje fibers which appeared normal, other areas contained cells which showed alterations in glycogen content, modifications of the junctions between cells, and changes in the lysosomal compartment. Cells showing large expanses of glycogen-free sarcoplasm, similar to the loss of glycogen which occurs in ischemic working myocardial cells;" were occasionally observed. Glycogen depletion was still evident in hearts perfused with supplements containing high levels of glucose and insulin (Fig. 6). In addition, large separations between undifferentiated and presumably gap-junctional regions of the boundaries between Purkinje fibers were observed in some cells and were not prevented by perfusion with media containing high levels of glucose and insulin (Figs. 7 and 8). One of the most consistent findings was the disruption of the lysosomal system with occasional redistribution of acid hydrolases. Figs. 9 and 10 show the disruption of primary Iysosomes in hearts perfused under ischemic conditions for 30 minutes, and the spreading of acid hydrolases within the sarcoplasm, particularly to the sarcoplasmic reticulum (Fig. 10). Under these circumstances, primary lysosomal vesicles appeared to be herniated or to lose their outer limiting membranes, which was coincident with a "budding" of larger Iysosomes to form smaller vesicles (Fig. 11). "Budding" of larger primary Iysosomes to form smaller vesicles was also demonstrable by the Gomori technique in ischemic hearts perfused with supplements with high levels of glucose and insulin (Fig. 12). Further, acid hydrolase activity was also detectable between undifferentiated regions and gap-junctional membranes of the intercalated disc and was still present in hearts perfused by metabolic solutions such as supplements containing high levels of glucose and insulin (Fig. 13). Discussion In the early 1960's treatment with polarizing solutions containing glucose, insulin, and potassium (GIK) was described for the relief of some of the symptoms associated with acute myocardial infarction.F Shortly after the introduction of GIK therapy, several studies reported a decreased incidence of arrhythmias, decreased mortality rate, and a return to a normal electrocardiogram in patients receiving GIK treatment. 23-25
The Journal of
656 McCallister et al.
Thoracic and Cardiovascular Surgery
Fig. 7. Ischemic heart perfused with supplements containing high levels of glucose and insulin. Many of these cells still showed large separations of the membranes making up the transverse and side-to-side junctions of the Purkinje fibers (arrows). (x8,060.) Fig. 8. High-power electron micrograph of an ischemic heart perfused with high levels of glucose and insulin. Note the large separations of undifferentiated and presumably gap-junctional regions between cells (arrows). (x20,960.)
Volume n Number 5 May, 1979
Lysosomes in ischemic hearts
Fig. 9. Acid phosphatase in ischemic hearts perfused with normal levels of glucose without insulin. Large lysosomal vesicles (Ly) showing a positive reaction for acid hydrolases appeared to lose their outer limiting membranes and to form smaller vesicles by a budding process (arrows). (x31,080.) Fig. 10. Ischemic heart showing that in some cases acid hydrolases occasionally spread to the sarcoplasmic reticulum, which gave a more intense reaction product for acid phosphatase (arrows) using the Gomori reaction. M, Mitochondria. (x2? ,980.)
657
The Journal of
65 8 McCallister et al.
Since ligation of the coronary arteries previously had been shown to cause a loss of intracellular potassium from the myocardial cells of the dog heart.i" it was proposed that the beneficial effects of GIK therapy were due to potassium entering into damaged and hypopolarized cardiac muscle cells and restoring them to their normal state of polarization. 23 Further, Maroko and associates'? also postulated that the major benefits of GIK treatment might arise primarily from glucose rather than potassium, through an increase in energy production from enhanced glycolysis of the myocardial cells in infarcted areas of the heart. Unfortunately, these early claims regarding GIK treatment have been difficult to confirm, and further studies generally have been disappointingP' ''? Malach'" reported that 65 percent of patients receiving polarizing therapy manifested life-threatening arrhythmias. Fletcher and associates" noted no significant difference in electrocardiographic changes, arrhythmias, and mortality rate of patients who did not receive treatment and those who received infusions of I L. of solution containing either glucose, glucose and insulin, glucose and potassium, or glucose, insulin, and potassium daily for 3 days. Further, Lesch and colleagues'" reported that the tolerance to ischemia was not extended with GIK therapy, and that patients receiving such treatment might have been adversely affected. The present experiments are in accord with those that report little or no beneficial effects of glucose and insulin in the oxygen-deprived heart. The structure of the conducting system did not appear to be significantly different in hearts perfused with low coronary flows for 30 minutes, whether or not the perfusate was enriched with high levels of glucose and insulin. Although some cells of the conducting system appeared the same as in control hearts, other Purkinje fibers showed ultrastructural modifications. Of particular interest was the decrease in subcellular glycogen stores observed in many of the fibers of the descending left limb of the His bundle. This observation is in direct agreement with metabolic studies on the effects of glucose and insulin on carbohydrate utilization previously reported for ischemic hearts by Liedtke;" Rovetto.:" and their colleagues, and clinical findings which show no beneficial effects of GIK therapy. 28-30 The present experiments also form a good basis for understanding the involvement of myocardial Iysosomes in ischemic injury to the heart. Numerous primary lysosomal vesicles, 150 to 600 mf-t in diameter, and bounded by a single limiting membrane, were found associated with stacks of Golgi cisternae and the tubules of the sarcoplasmic reticulum in the nuclear
Thoracic and Cardiovascular Surgery
pole zone of Purkinje fibers; these findings were observed both in hearts perfused at control levels of coronary flow with media containing normal levels of glucose without insulin and in hearts perfused with supplements containing high levels of glucose and insulin. With the Gomori reaction used for the demonstration of acid phosphatase, electron-dense precipitate indicative of acid hydrolase activity was seen in association with these vesicles, the Golgi apparatus, and lipofuscin pigment. In hearts perfused at approximately 50 percent of normal coronary flow for 30 minutes, there appeared to be a disruption of the lysosomal compartment and in some cases a spreading of acid hydrolases throughout the sarcoplasm. Large primary Iysosomes appeared to have herniated or disrupted outer membranes and to have formed smaller lysosomal vesicles by a budding process. The morphologic data favor a budding process rather than a coalescing of Iysosomes, since the vesicles observed were below the lower limit of primary Iysosomes measured in control hearts, i.e., less than 150 mu. The breakdown of primary Iysosomes was coincident with the deposition of acid hydrolases at the junctions between cells, and in some instances reaction product was also observed in the tubules of the sarcoplasmic reticulum. At the same time there appeared to be a dissolution of a small number of undifferentiated and gap-junctional regions of the intercalated disc in some cells. Thus the lysosomal alterations observed in the present study may be among the earliest changes which occur in ischemic hearts and could be intimately involved in the destruction of subcellular organelles. Jennings and associates'" demonstrated that irreversibly damaged myocardial cells have structural defects in the plasma membrane which are reflected by an increased insulin space, increased H 20 and Na" content of the myocardial cells, and failure of the tissue to maintain high levels of K+ and Mg" ". In the present study both morphologic and histochemical evidence indicates that the sarcolemma can be altered during ischemic injury. The lesions of the plasma membrane observed during the early phase of oxygen deprivation were primarily localized to the gap-junctional membranes and undifferentiated regions of the junctions between cells, whereas the extrajunctional sarcolemma appeared relatively normal. An increase in the spacing of membranes making up gap junctions has previously been noted by this laboratory after I hour of normothermic ischemic arrest in the dog heart" and in the isolated perfused rat heart" made ischemic according to the technique of Neely and associates. 35 In the latter model the dissolution of gap junctions occurred as a function of time, was correlated with the breakdown of
Volume n Number 5
Lysosomes in ischemic hearts
May, 1979
Fig. II. Nuclear pole zone of a Purkinje fiber from a heart made ischemic and perfused with supplements containing high levels of glucose and insulin. Note that the lysosomal compartment was disrupted in these cells. and large lysosomal vesicles (Ly) were seen to lose their outer limiting membranes and to form smaller structures by a budding process (arrows). (x27 ,990.) Fig. 12. Micrograph similar to Fig. II, but showing the localization of acid phosphatase in ischemic hearts perfused with supplements containing high levels of glucose and insulin. Acid hydrolase activity is seen in association with lysosomal vesicles (Ly) which appear to be in a budding process (arrows). (X27,350.) Fig. 13. Localization of acid phosphatase in ischemic hearts perfused with supplements containing high levels of glucose and insulin. Note the electron dense precipitate in association with the undifferentiated and gap-junctional regions of the intercalated disc (arrows). (xn,080.)
659
660 McCallister et al.
primary lysosomal vesicles, and was coincident with the onset of irreversible tissue damage. After normothermic ischemic arrest the separation of gap junctions was associated with a widening of the QRS complex of the electrocardiogram or T-wave inversion and ST-segment depression in 80 percent of the animals
studied.":
36
These results are particularly interesting in view of the electrophysiological changes which have been reported to occur in Purkinje fibers surviving in an infarct.!': 37. 38 Lazzara and associates!\' 38 using bipolar electrograms to assess the activity of Purkinje fibers I day after occlusion of the anterior descending coronary artery in the dog heart, reported Purkinje potentials which were diminished in amplitude and rapidity. In addition to diminished resting and action potentials, reduced upstroke velocity, enhanced automaticity, and phase 4 depolarization also were reported. These investigators concluded that the arrhythmias which occur as a result of ligation of the coronary arteries may originate in the altered Purkinje cells of the infarcted zone." Thus the present results, which indicate dissolution of the undifferentiated and gap-junctional membranes of the transverse and side-to-side junctions between cells and the deposition of acid hydrolases in these regions of hearts perfused at approximately 50 percent of their normal coronary flows, may represent a morphologic basis for the physiological phenomena associated with myocardial infarction. We would like to acknowledge the excellent technical assistance of Earle W. Christman II.
2
3
4
5
6
REFERENCES Brachfeld N: Maintenance of cell viability. Circulation 50:202-215, 1969 Brachfeld N, Gemba T: Mechanisms of myocardial cell death. Release of lysosomal hydrolases after ischemia. Clin Res 13:524, 1965 Hoffstein S, Gennaro DE, Weissmann G, Hirsch J, Streuli F, Fox AC: Cytochemical localization of lysosomal enzyme activity in normal and ischemic dog myocardium. Am J Pathol 79:193-206, 1975 McCallister LP, Munger BL, Neely JR: Electron microscopic observations and acid phosphatase activity in the ischemic rat heart. J Mol Cell Cardiol 9:353-364, 1977 McCallister LP, Munger BL, Tyers GFO, Hughes HC: The effect of different methods of protecting the myocardium on lysosomal activation and acid phosphatase activity in the dog heart after one hour of cardiopulmonary bypass. J THoRAc CARDIOVASC SURG 69:644-663, 1975 Riciutti MA: Myocardial lysosome stability in the early stages of acute ischemic injury. Am J Cardiol 30:492497, 1972
The Journal of Thoracic and Cardiovascular Surgery
6a Baba N, Leighton RF, Weissler AN: Experimental cardiac ischemia. Observation of the sinoatrial and atrioventricular nodes. Lab Invest 23: 168-178, 1970 7 Ravens KG, Gudbjamson S: Changes in the activities of lysosomal enzymes in infarcted canine heart muscle. Circ Res 24:851-856. 1969 8 Smith AL, Bird JWC: Distribution and particle properties of the vacuolar apparatus of cardiac muscle tissue. I. Biochemical characterization of cardiac muscle Iysosomes and the isolation and characterization of acid, neutral and alkaline proteases. J Mol Cell Cardiol 7:39-61, 1975 9 Tallan HH, Jones ME, Fruton JS: On the proteolytic enzymes of animal tissues. X. Beef spleen cathepain C. J Bioi Chern 194:793-805, 1952 10 Riciutti MA: Lysosomes and myocardial cellular injury. Am J Cardiol 30:498-502, 1972 II Lazzara R, EI-Sherif N, Sherlag J: Early and late effects of coronary artery occlusion on canine Purkinje fibers. Circ Res 35:4391-399, 1974 12 Liedtke AJ, Hughes HC, Neely JR: An experimental model for studying myocardial ischemia. Correlations of hemodynamic performance and metabolism in the working swine heart. J THoRAc CARDIOVASC SURG 69:203211, 1975 13 Sommer JR, Johnson EA: Cardiac muscle. A comparative study of Purkinje fibers and ventricular fibers. J Cell Bioi 36:497-526, 1968 14 James TN, Sherf L: Ultrastructure of the human atrioventricular node. Circulation 37: 1049-1070, 1968 15 Mochet M, Maravec J, Guillemot H, Hatt PY: The ultrastructure of the rat conductive tissue. An electron microscopic study of the atrioventricular node and the bundle of His. J Mol Cell Cardiol 7:879-889, 1975 16 Bogusch G: Investigations on the fine structure of Purkinje fibers in the atrium of the avian heart. Cell Tissue Res 150:43-56, 1974 17 Gossrau R: Uber das Reizleitungssystem der Vogel. Histochemische une elektronen microskopische Untersuchungen. Histochemie 13: 111-159, 1968 18 Lindner E: Submikroskopische Ringbinden-funktionelle Strukturen der "Weichteilmuskulatur." Z Zelloforsch 88:370-386, 1968 19 Arluk OJ, Rhodin JAG: The ultrastructure of calf heart conducting fibers with special reference to nexuses and their distribution. J Ultrastruct Res 49: 11-23, 1974 20 Johnson EA, Sommer JR: A strand of cardiac muscle. Its ultrastructure and the electrophysio1ogica1 implications of its geometry. J Cell Bioi 33:103-129, 1967 21 Jennings RB, Baum JH, Herdson PB: Fine structural changes in myocardial ischemic injury. Arch Pathol 79:135-143, 1965 22 Sodi-Pallares D: Possibility of a therapy of cellular ion integration in cardiovascular disease. Arch Inst Cardiol Mex 31:557-574, 1961 23 Mittra B: Potassium, glucose, and insulin treatment of myocardial infarction. Lancet 2:607-609, 1965 24 Sodi-Pallares D, Ponce de Leon J, Bisteni A, Medrano
Volume
n
Number 5 May, 1979
25
26
27
28
29
30 31
GA: Potassium, glucose, and insulin in myocardial infarction. Lancet I: 1315-1316, 1969 Sodi-Pallares 0, Testille NR, Fishleder BL, Bisteni A, Medrano GA, Friedland C, DeMicheli A: Effects of an intravenous infusion of potassium-glucose-insulin solution on the electrocardiographic signs of myocardial infarction. A preliminary report. Am J CardioI9:166-181, 1962 Jennings RB, Grout JR, Smelters GW: Studies on distribution and localization of potassium in early myocardial ischemic injury. Arch Pathol 63:586-592, 1957 Maroko PR, Libby P, Sobel BE, Bloor CM, Sybers HD, Shell WE, Covell JW, Braunwald E: Effect of glucoseinsulin-potassium infusion on myocardial infarction following experimental coronary artery occlusion. Circulation 45: 1160-1175, 1972 Fletcher GF, Hurst JW, Schlant RC: "Polarizing" solutions in patients with acute myocardial infarctions. A double-blind study with negative results. Am Heart J 75:319-324, 1968 Lesch M, Teichholz LE, Soeldner JS, Gorlin R: Ineffectiveness of glucose, potassium, and insulin infusion during pacing stress in chronic ischemic heart disease. Circulation 49: 1028-1037, 1974 Malach M: Polarizing solution in acute myocardial infarction (abstr). Am J Cardiol 19: 141, 1967 Rovetto MJ, Whitmer JT, Neely JR: Comparison of the effects of anoxia and whole heart ischemia on carbohydrate utilization in isolated working rat hearts. Circ Res 32:699-711, 1973
Lysosomes in ischemic hearts
66 I
32 Liedtke AJ, Hughes HC, Neely JR: Effects of excess glucose and insulin on' glycolytic metabolism during experimental myocardial ischemia. Am J Cardiol 38: 17-27, 1976 33 Herdson PB, Sommer HB, Jennings RB: A comparative study of the fine structure of normal and ischemic dog myocardium with special reference to early changes following temporary occlusion of a coronary artery. Am J Pathol 46:367-386, 1965 34 Jennings RB, Ganote CE, Reimer KA: Ischemic tissue injury. Am J Pathol 81:179-195, 1975 35 Neely JR, Rovetto MJ, Whitmer JT, Morgan HE: Effects of ischemia on function and metabolism of the isolated working rat heart. Am J Physiol 255:651-658, 1973 36 Tyers GFO, Hughes HC, Todd GJ, Williams DR, Andrews EJ, Prophet GA, Waldhausen JA: Protection from ischemic cardiac arrest by coronary perfusion with cold Ringer's lactate solution. J THORAC CARDIOVASC SURG 67:411-418, 1974 37 Friedman PL, Fenoglio 11, Wit AL: Time course for reversal of electrophysiological and ultrastructural abnormalities in subendocardial Purkinje fibers surviving extensive myocardial infarction in dogs. Circ Res 36:127-144, 1975 38 Lazzara R, El-Sherif N, Sherlag J: Electrophysiological properties of canine Purkinje cells in one-day-old myocardial infarction. Circ Res 33:722-734, 1973