the response of the atrium to direct mechanical wounding in the adult heart of the newt,...

Cell Tissue Res (1984) 235:583-592 Cell and Tissue Research �9 Springer-Verlag 1984

The response of the atrium to direct mechanical wounding in the adult heart of the newt, Notophthalmus viridescens An electron-microscopic and autoradiographic study

Timothy J. McDonnell and John O. Oberpriller Department of Anatomy, University of North Dakota, Grand Forks, North Dakota, USA

Summary. Wound repair and proliferation were examined in the injured newt atrium with light- and electron-microscopic techniques including autoradiography. Hearts were injured by removing a piece approximately 0.5 mm 2 of the atrial wall. The five-day wound was an endothelial and mesothelial-lined blood clot bordered by a 150-gm necrotic zone. Repair progressed from the periphery inward with areas of macrophage activity replaced by fibroblasts and connective tissue. The wound at 25 days consisted of a scar with few myocytes. There was no difference in the proliferative behavior between the right and left atria. Pro- liferative cells were localized to a 500-gm reactive zone surrounding the wound. The maximum mesothelial cell thymidine-labeling index of 20.5% and mitotic index of 1.4% were seen 5 days after injury. The peak connective tissue cell thymidine-labeling index of 10.2% and mitotic index of 0.4% were seen 10 days after wounding. The peak thymidine-labeling index of 9.8% for myocardial cells was re- corded 10 days after injury with a mitotic index of 0.2%. Proliferation returned to control levels by 25 days post- injury. Electron microscopy demonstrated that myocytes engaged in DNA synthesis were indistinguishable from control myocytes. Z-band material was not observed in mitotic myocytes, but myofilaments and junctions were present.

Key words: Heart - Wound healing - Cell division - No- tophthalmus viridescens (Urodela)

Most investigations concerned with the wound repair process in the injured adult heart have examined events subsequent to experimental infarction in the mammalian ventricular myocardium (Bing 1971 ; Dusek et al. 1971 ; Rumyant- sev 1966; Rumyantsev and Mirakyan 1968a). A common finding of these studies was that the infarcted area healed by the formation of a connective tissue scar and that perinecrotic ventricular myocytes exhibited little, if any, potential for cell division in the mammalian heart. It also has been demonstrated that perinecrotic ventricular myocytes engaged in both DNA synthesis and mitosis following mechanical wounding in the adult amphibian heart (Ober- priller and Oberpriller 1971, 1974; Oberpriller et al. 1979, 1981; Rumyantsev 1966, 1973, 1977, 1979, 1981). Despite

Send offprint requests to: Dr. Timothy J. McDonnell, Department of Anatomy, Medical Science Building, University of North Da- kota, Grand Forks, ND 58202, USA

this proliferation the wounded amphibian ventricle also heals by connective tissue overgrowth.

In addition, it is known that atrial myocytes are capable of significant proliferative activity subsequent to ventricular infarction caused by left coronary ligation in the adult rat heart (Oberpriller et al. 1983; Rumyantsev 1974; Rumyant- sev and Kassem 1976; Rumyantsev and Mirakyan 1968a, b) and partial ventricular amputation in the adult newt heart (McDonnell and Oberpriller 1983 a).

The response of the amphibian atrium to direct injury has not previously been studied. Because of the importance of understanding wound healing in the heart, particularly in systems (such as the amphibian) which exhibit a potential for generating new myocytes, an investigation was under- taken to examine this process in the adult newt atrium. Light- and electron-microscopic techniques including autoradiography were used to define the histological and fine structural events involved in wound repair and the atrial proliferative activity associated with removal of a portion of the atrial wall.

Materials and methods

Adult red-spotted newts, Notophthalmus viridescens, were used in this study. Iridectomy scissors were used to expose the heart by a ventral midline incision through the body wall and pericardium of newts anesthetized with MS222 (Ayerst). Injury was achieved by removing a piece (approximately 0.5 mm 2) of the anterior atrial wall located in the area of the interatrial septum. In this manner it was possible to wound, simultaneously, the left and right atria. The body wall was then closed with surgical silk sutures. Sham-operated and unoperated animals served as controls. Animals were killed at intervals of 5, 10, 15, 20, and 25 days after wounding. One hour prior to fixation each newt received a single intraperitoneal injection of 3H-thymidine (2.5 gCi for each animal processed for light microscopy and 20.0 gCi for each animal of the EM series). Atrial tissue was processed for either light- or electron-microscopic autoradiography as described by McDonnell and Oberpriller (1983a).

Mitotic and thymidine-labeling indices were based on counts from four animals for each experimental time period of the light microscopic autoradiographic series. Indices were calculated for the epicardial mesothelium, epicardial connective tissue, and myocardium of both the right and left atrium. Because the majority of proliferative cells were confined to a reactive zone adjacent to the original wound,

584

cells were included in the counting procedure only if they were present within a 500 lam distance from the wound site. Statistical significance between indices of different experimental time periods was established with Students t-test.

Resul t s

It was apparent from preliminary observations o f light-microscopic autoradiographs that the great majority of atrial proliferative cells were localized to a reactive zone which extended approximately 500 ~tm from the wound site. Be- cause the proliferative response was localized, only cells within the 500 p.m reactive zone were included in the counting procedure. Because no consistent differences in the proliferative response were exhibited between the two atria, the data for the left and right atria were combined and presented together in Tables 1, 2, and 3 for the epicardial mesothelium, epicardial connective tissue, and myocardium, respectively.

The ultrastructure of the atrial wall of sham-operated animals was similar to that of unoperated control or normal atria. However, an increase in the extent of intercellular spaces and number o f cytoplasmic projections was observed in the mesothelial cells of the five day sham-operated atrium.

Five day experimental group. The mechanically wounded atrial wall consisted of three zones or regions. The wound site proper was initially composed of a blood clot. The blood clot was seen, at the electron-microscopic level of resolution, to consist of an abundance of necrotic cells, presumably of hematogenous origin, and cellular debris em- bedded within an extensive fibrin matrix. The only cells which appeared viable within the blood clot of the five day experimental group were macrophages. Immediately surrounding the blood clot was a zone of degeneration, or necrosis. This zone (Fig. 1) was characterized by the reduction or absence of striated myofibrillae and frequent nuclear pyknosis and could extend a distance of 150 p,m from the wound site. Degenerative changes were most severe in the area immediately apposed to the wound site and became progressively less severe with increasing distance from the blood clot as the zone of necrosis became indistinguishable from the third zone, consisting of the mor- phologically unaltered atrial wall.

The endothelial cells formed a complete lining on the inner surface of the wound and were more attenuated than those of the uninjured atrial wall. They possessed slender, elongated filopodia which overlapped with those of adjacent endothelial cells. Intercellular junctions, such as the punctum adherens, which are normally present between endothelial cells were not observed. Mesothelial cells over, and adjacent to, the wound site were frequently more than one cell layer thick. These cells showed more extensive regions o f overlap, perhaps resulting from a relaxation of cellular interdigitation. Also, they exhibited increased intercellular spaces and a reduced number of intercellular junctions, loosening from the subjacent basal lamina, and an increased number of cytoplasmic projections. Many of these cells possessed spherical lysosome-like granules. Mesothe- lial cells which had migrated over the blood clot rested directly on the fibrin matrix, in contrast to the mesothelial cells over the zone of necrosis which possessed a well- formed basal lamina.

Table l. Mitotic and thymidine-labeling indices for the epicardial mesothelium *

Days after injury Thymidine-labeling index Mitotic index

5 20.5• 1.4• 10 9.0• 0.5• 15 7.6• 0.2• 20 5.5• 0.0• 25 3.0• 0.1• Unoperatedcontrol 0.1• 0.0• 10Daysham 1.8• 1.1•

* In tables 1, 2, and 3, indices are presented as the mean• standard deviation and are expressed as percentages. Cells were counted only within a 500 [am distance adjacent to the wound site. Four animals were counted for each experimental time period and two animals for both the unoperated and sham-operated controls

Table 2. Mitotic and thymidine-labeling indices for the epicardial connective tissue


5 7.9&6.4 0.0• 10 10.2• 0.4• 15 5.0• 0.6• 20 5.0• 0.0• 25 2.7• 0.0• Unoperatedcontrol 0.7&0.9 0.0• 10Daysham 1.3• 0.2•

Table 3. Mitotic and thymidine-labeling indices for the myocardium


5 1.5• 0.0• 10 9.8• 0.2• 15 3.0• 0.1• 20 1.3• 0.3• 25 1.5• 0.3• Unoperatedcontrol 0.0• 0.0• 10Daysham 0.0• 0.0•

The necrotic zone bordering the wound was characterized by accumulations of cellular debris, marked swelling of tissue with corresponding increases in the size of intercellular spaces, numerous macrophages which appeared to be engaged in phagocytosis, and the presence of necrotic myocytes. The necrotic rnyocytes possessed pyknotic nuclei, ap- parently in the process o f karyolysis. They also possessed swollen mitochondria with increased intercristal spaces, increased amounts of lipid, loss of myofibrillar and sarcomeric organization, decreased amounts of glycogen, and discontinuities of the sarcolemma (Fig. 2). The laminar coat surrounding the necrotic myocytes remained intact. Myo- cytes located further from the wound site exhibited less extensive alterations. Mitochondria were slightly swollen and myofibrils were usually in a state of partial contraction. The nuclei of these cells did not show margination of chromatin and possessed prominent nucleoli. Other myocytes had completely normal ultrastructure. Occasional myocytes in this region were partially withdrawn from the surround-

585

Fig. 1. Light micrograph of the five day atrial wound site. The zone of necrosis (ZN) is seen adjacent to the blood clot (BC) of the wound site and is characterized by the presence of pyknotic nuclei and the reduction or absence of striated myofibrillae. Mesothelium (M). x 220

Fig. 2. Electron-microscopic detail of a degenerative myocyte from the necrotic zone bordering the five day atrial wound site. Note the loss of myofibrillar (MF) sarcomeric organization, swollen mitochondria (M), lipid droplets (L), apparent membrane discontinuities (arrow), and the intact laminar coat (arrowhead). x 17330

ing laminar coat, making contact with it through thin cytoplasmic project ions which resembled f i lopodia (Fig. 3).

Ten day experimental group. A t ten days the b lood clot of the wound site differed from that seen in the five day experimental group in that cellular elements did not appear as densely packed, and consequently the extracellular space of the clot was more apparen t (Fig. 4). Mesothel ial cells ranged up to several layers in thickness at the wound site, and fibrin was less abundan t than in the five day group.

Nuclear pyknosis was less evident in cells of the wound site than in the five day group.

Mos t of the cells in the wound site were identified at the ul t ras t ructural level as macrophages , but f ibroblasts and myofibroblas ts were present also. Myof ibroblas t s differed from fibroblasts in that they formed desmosomal at tach- ments with each other and contained microf i lament bundles. A t this time collagen fibrils first appeared in the matr ix immediately surrounding the fibroblasts. The mesothelial and endothel ial linings of the wound site were ul t ras t ructur-

586

ally similar to that seen in the five day experimental group. However, mesothelial cells which contained ingested cellular debris in the form of phagocytic vacuoles were observed infrequently.

The zone of necrosis extended a distance of up to 100-200 gm from the wound site and consisted of widely scattered debris and cellular elements. On the basis of limited nuclear pyknosis, most of the cells which remained in this region were believed to be viable. Although degener- ating myocytes were identified within the zone of necrosis, they were less common than in the five day experimental

Fig. 3. Electron micrograph of an atrial myocyte within the reactive zone of the five day atrial wound site. The myocyte (M) exhibits normal ultrastructure but is partially withdrawn from the surrounding laminar coat (arrow), making contact with it via filopodial-like (F) projections, x 14590

Fig. 4. Light micrograph of the ten day atrial wound site (W). Nuclear pyknosis and fibrin (F) accumulations are less common than in the five day wound site. The mesothelium (M) covering the wound site is several layers in thickness. Endothelium (E). x 280

Fig. 5. Electron-microscopic detail of the myocardium in the zone of necrosis adjacent to the ten day wound site. A myocyte (M) is seen by an extracellular space (ES) which is surrounded by an intact laminar coat (arrow). Such spaces remain after the complete degeneration of necrotic myocytes, x 14670

Fig. 6. Electron micrograph of an atrial myocyte present in the zone of necrosis adjacent to the ten day atrial wound site. The myofibrillae (MF) are in a state of disarray, the cytoplasm is vesiculated and possesses reduced amounts of glycogen and slightly swollen mitochondria. The nucleus (N) contains a prominent nucleolus (NL). x 4000

group. Well-formed basal laminae surrounded the spaces which remained after complete degeneration of the necrotic myocytes (Fig. 5). Myocytes in this region commonly possessed a vesiculated cytoplasm with slightly swollen mitochondria, reduced amounts of glycogen, and myofibrillae in a state of disarray (Fig. 6). The nuclei of these cells were, with the exception of a thin rim of heterochromatin, euchro- matic and contained prominent nucleoli. These myocytes were considered to be viable and in the process of recover- ing from the initial wound trauma. Other myocytes in this region were identical to those of control atria.

587

Fig. 7. Light-microscopic autoradiograph of the reactive zone adjacent to the ten day atrial wound site. A mitotic figure (M) is present within the myocardium. Epicardial mesothelium (EM). • 760

Fig. 8. Electron-microscopic autoradiograph of an atrial myocyte in the reactive zone adjacent to the ten day wound site. The myocyte nucleus (N) is labeled with 3H- thymidine. Obliquely sectioned myofibrillae (ME) are present within the cytoplasm, x 10000

The thymidine-labeling index of the epicardial rnesothe- lium value was significantly higher (p<0.01) than that of the ten day sham-operated animals indicating that there are higher levels of proliferation within the 500 ~tm reactive zone as a consequence of atrial wounding. The DNA synthetic activity of the epicardial connective tissue was significantly elevated (p<0.05) over that observed for the ten day sham-operated controls.

The DNA synthetic activity within the myocardium was significantly higher (p < 0.02) than the ten day sham-operated animals. As many as 14.4% of myocardial cells within the 500 tim reactive zone could be labeled with a single injection of 3H-thymidine. These cells at the light-microscopic level were in close proximity to striated myofibrillae and appeared to be myocytes. Mitotic figures were seen within the myocardium at this time (Fig. 7). Electron-microscopic autoradiography of ten day experimental animals confirmed that atrial myocytes were engaging in DNA synthesis subsequent to atrial wounding (Fig. 8). Myocytes labeled with 3H-thymidine exhibited normal ultrastructure including striated myofibrillae and intact intercellular junctions. The degenerative changes described above were not observed in labeled myocytes.

Fifteen day experimental group. The perimeter of the wound site, the area subjacent to the mesothelium and endothelium, and the original necrotic zone were less cellular and accumulations of connective tissue were more prominent than at previous time periods. By this time melanophores had migrated into the would area. Macrophages were the most common cell type encountered in the central portion of the wound. The amount of cellular debris, necrotic cells and fibrin was reduced from that found in the ten day experimental group. Connective tissue was still scant in the central wound area. Fibroblasts were located primarily in the would periphery. The fibroblasts contained extensive rough endoplasmic reticulum and were usually encom-

passed by a finely flocculent extracellular material in which collagen fibrils were seen.

Mesothelial basal laminae were not yet apparent, but instead of resting on a matrix of fibrin and cellular dibris as in previous time periods, the mesothelial cells lining the wound were usually in contact with a finely flocculent extracellular matrix. The endothelial lining of the wound resembled that seen during earlier time periods. However, many of the endothelial cells now possessed an extensive network of Golgi and rough endoplasmic reticulum.

The myocytes in the region of the initial zone of necrosis did not exhibit the vesiculated appearance of those found in the ten day group. The myofibrillae were more numerous and more regularly oriented than in the ten day animals. Incompletely formed sarcomeric units suggested that myofi- brillogenesis was in progress during this time period. Mito- chondria were numerous and showed no detectable signs of swelling, and glycogen was more abundant.

Several mitotic myocytes were observed at the ultrastructural level in the metaphase or early anaphase stage. Mitotic myocytes displayed an apparent absence of Z-band material and maintenance of intercellular junctions (Fig. 9). Myofibrillar architecture in the adjacent interphase myocytes was unaltered.

Twenty day experimentalgroup. The connective tissue depo- sition was no longer apparent only at the wound periphery but was apparent also in the central regions. Concomitant with the decline in the number of macrophages present within the wound site was a decrease in the amount of fibrin and cellular debris. The number of fibroblasts and amount of connective tissue within the wound increased from that previously observed. Melanophores also were more numerous and occasionally contained phagocytic vacuoles. In addition, there were indications that atrial myocytes may have migrated into the wound site. These cells possessed striated myofibrillae and were located well within

588

Fig. 9. Electron micrograph of a metaphase atrial myocyte in the reactive zone adjacent to the 15 day atrial wound site. Sister chromatids (C) are seen to be paired. Striated myofibrillae (MF) are present in the cytoplasm of the adjacent interphase myocyte (N) but Z-band material is not seen in the mitotic myocyte. However, bundles of myofilaments (arrowheads) are present in the mitotic myocyte cytoplasm. Intercellular junctions (arrow) remain intact, x 7420

the connective tissue substance of the wound (Fig. 10). It was not possible to determine with certainty whether these myocytes actually migrated into the wound or simply survived the initial t r auma within the necrotic zone, gradual ly becoming encompassed by connective tissue. Proliferative ceils were more common within the wound site proper than observed during earlier experimental t ime periods and consisted o f macrophages, melanophores , and fibroblasts. However, the proliferat ive indices for cells within the reactive zone cont inued to decline.

A faint basal lamina could now be observed in scattered locations beneath the mesothelial lining of the wound. However, the mesothelium still rested predominant ly on the flocculent connective tissue ground substance. Endothe- lial cells appeared much like those lining the wound of the 15 day experimental atr ium but the Golgi complex and rough endoplasmic reticulum were less extensive.

There was no apparent cellular degenerat ion within the original zone of necrosis. The basal laminae-l ined spaces which remained after the per iod of myocyte degenerat ion

589

Fig. 10. Light-microscopic autoradiograph of the 20 day atrial wound site. The wound (W) consists of widely scattered cells in an abundant connective tissue matrix. A myocyte (M) possessing striated myofibrillae is present within the connective tissue of the wound. Atrial lumen (L). x 480

Fig. 11. Electron micrograph of the myocardium adjacent to the 25 day atrial wound site. Myocytes (M) in this region form an irregular anastomotic network. Myofibrillae (MF) are well-formed and oriented. Connective tissue (CT) spaces have replaced myocytes which have undergone degeneration. • 4000

were still evident, but the spaces now contained connective tissue.

Twenty-five day experimental group. The appearance of the 25 day wound site was not remarkably different from that observed in the 20 day experimental animals. The wound site p roper healed pr imari ly by the format ion of a connective tissue scar. Small amounts of fibrin and cellular debris still could be found in the region o f the original wound site. F ibroblas t s and macrophages were interspersed

throughout the abundan t connective tissue matrix. Myo- cytes were widely scattered within the scar tissue.

The mesothelial covering of the wound still possessed a number of cytoplasmic project ions and widened intercellular spaces and was commonly more than one cell layer in thickness. The mesothelial basal lamina, a l though more delicate than that found in other areas of the atr ium, was now present as a more-or-less cont inuous layer with only occasional discontinuities. Therefore, the wound mesothelium could be dist inguished from that which bordered the

590

wound. The endothelial lining of the wound site was essen- tially the same as that found in control atria.

The myocardium of the original necrotic zone was dis- tinctly spongy in appearance (Fig. 11). This is believed to be the consequence of myocyte degeneration during the initial phases of wound repair. The surviving myocytes that remained formed an irregular anastomotic network in this region and were interspersed with connective tissue spaces formerly occupied by other myocytes.

The thymidine-labeling indices for cells of the 25 day experimental animals could not be statistically distinguished from those of unoperated control animals (p > 0.19). Simi- larly, the mitotic indices of 25 day experimental and unoperated control animals could not be differentiated (p > 0.42). These data indicate that the proliferative events within the 500 rtm reactive zone involved in the processes of atrial wound repair have subsided 25 days after injury.

Discussion

The electron-microscopic and autoradiographic evidence presented demonstrate that the amphibian atrial myocyte is capable of significant proliferative activity in response to mechanical wounding of the atrium. The proliferation observed in this experimental system is expressed as increased levels of both DNA synthesis and mitosis.

Myocytes which are actively participating in DNA synthesis, as indicated by the incorporation of 3H-thymidine, are indistinguishable from either control or unlabeled myocytes at the ultrastructural level. Myocytes exhibiting degenerative changes are never observed to be labeled with 3H- thymidine.

Several mitotic atrial myocytes were examined at the ultrastructural level following atrial wounding. Mitotic myocytes maintained junctional continuity with adjacent interphase cells but displayed dramatically altered myofibrillar structure. Z-band material failed to be demonstrated in these cells. The myofibrillar structure of adjacent, non- proliferative, myocytes remained unaltered and exhibited the normal sarcomeric banding pattern. The myofibrillar alterations noted in this study have been previously reported in other systems including the embryonic rat (Ru- myantsev 1972; Rumyantsev and Snigerevskaya 1968) and chicken heart (Hay and Low 1972; Manasek 1968; Wein- stein and Hay 1970). Similarly, Z-band material is not observed or only rarely observed in mitotic myocytes of the wounded amphibian ventricle (Oberpriller and Oberpriller 1971; Rumyantsev 1973), adult rat atrium following left ventricular infarction (Rumyantsev 1974), and adult amphibian atrium following partial ventricular amputation (McDonnell and Oberpriller 1983a). This alteration is believed to be an ultrastructural modification which facilitates the completion of cytokinesis while enabling the myocyte to maintain a differentiated phenotype.

The proliferative atrial myocytes observed in this investigation are believed to represent fully differentiated myocytes and not to have arisen from an undifferentiated stem cell. Myogenic stem cells have not been demonstrated in the mammalian heart (Rumyantsev 1977) or normal newt atrium (McDonnell and Oberpriller 1983b) and there was no evidence of their presence in this study.

It was found that a maximum of 14.4% of the atrial myocardial cells present within a 500 ~tm reactive zone adjacent to the site of injury could be labeled with a single

injection of 3H-thymidine 10 days after wounding. Limited levels of mitosis were also observed within the myocardium of the reactive zone beginning five days after wounding and continued throughout the observation period. Cells which incorporated 3H-thymidine or engaged in mitosis were not apparent within the myocardium of either unoperated or ten day sham-operated animals. Therefore, the proliferative response within the reactive zone is concluded to be a consequence of the atrial injury.

It is believed that the majority of the proliferative atrial myocardial cells observed at the light-microscopic level, following direct atrial injury, are myocytes. The reactive cells were located within the myocardium and frequently appeared to have myofibrillae. Furthermore, it was demonstrated by McDonnell and Oberpriller (1983b) that fibroblasts, or other interstitial cells, were rarely identified within the atrial myocardium at the ultrastructural level. With EM autoradiography it was observed that interstitial cells engaging in DNA synthetic or mitotic activity were virtually absent from the myocardium in experimental animals. From these observations it may be concluded that the thymidine-labeling and mitotic indices, calculated from data obtained at the light microscopic level for the myocardium, represent mostly myocytes.

No consistent difference was detected in the proliferative response between the two atria subsequent to injury. How- ever, the response was localized with respect to distance from the wound site. Most of the cells were localized to a "reactive zone" which extended approximately 500 lam from the site of injury. It is the localization of proliferative activity to a reactive zone adjacent to the wound site which differentiates the response of the atrium to a direct injury from that resulting from partial ventricular amputation. Results from a previous study (McDonnell and Oberpriller 1983a) indicated that atrial myocardial cells of adult newts exhibit a generalized response to ventricular injury. Similar findings for the atria of adult rats suggest that the proliferative myocytes are diffusely distributed throughout the atria following left ventricular infarction (Rumyantsev 1974; Ru- myantsev and Mirakyan 1968 a) and abdominal aortic stenosis (Rumyantsev 1970). Rumyantsev and Kassem (1976) propose that the removal, or inactivation through infarction, of a portion of the ventricular myocardium functions as a stimulus for atrial hypertrophy resulting in the DNA synthetic and mitotic activity of atrial myocytes.

The distribution of proliferative myocytes within the atrium, therefore, seems to be a function of the type of wound stimulus inflicted on the heart and may be indicative of two discrete proliferative mechanisms. The localization of proliferative atrial myocardial cells to a reactive zone adjacent to the site of a direct injury indicates that the response is related to the processes of wound repair. A similar response has been observed in the wounded amphibian ventricle. Rumyantsev (1973) reported that 13-15% of the myocytes within a 300-500 ~tm perinecrotic zone could be labeled with a single injection of 3H-thymidine one week following a ventricular crush injury in adult frogs. Ober- priller and Oberpriller (1974) reported a comparable response for the injured adult newt ventricle with a maximum of 9-10% of the myocardial cells labeled within a 250-300 ~tm perinecrotic zone 20 days after wounding.

However, despite the pronounced proliferative activity reported by Rumyantsev (1973) and Oberpriller and Ober- priller (1971, 1974) in the injured amphibian ventricle, and

591

by the present investigation for the amphibian atrium, the wound heals primarily by the formation of a connective tissue scar. Possible explanations for the inability of the amphibian heart to generate functional contractile tissue in response to injury, despite high levels of myocyte proliferative activity, may be obtained from an understanding of the wound repair process in the heart.

Degenerative changes were most pronounced in the myocardium adjacent to the wound site in the five day experimental group. These changes were expressed by the loss of sarcomeric organization and myofibrillar orienta- tion, mitochondrial swelling, loss of glycogen and increased amounts of lipid, margination and clumping of chromatin, karyolysis, and plasmalemmal discontinuities. These changes are characteristic of mammalian ventricular myocytes following infarction or ischemic injury (Dusek et al. 1971 ; Jennings and Ganote 1974; Kloner and Ingwall 1980; Kloner et al. 1979; Lautsch 1979; Sybers et al. 1979; To- manek et al. 1981) and in amphibian ventricular myocytes following mechanical wounding (Oberpriller et al. 1981; Rumyantsev 1973).

Several factors are believed to be responsible for the degenerative changes observed in the myocardium adjacent to the wound site. Direct mechanical damage to the myocytes would, obviously, result from the removal of tissue during wound formation. In addition, it may be presumed that oxygen tension would be decreased in this region due to the blood loss from the initial wound trauma and the high metabolic rate of the numerous macrophages present. It might be expected that the pH in the periphery of the wound site would decline due to the release of lytic enzymes from the neutrophils and lactate production by macrophages (Hunt and VanWinkle 1976). It may be concluded that the environment of the wound site during the early phases of the reparative process is not conducive to myocyte viability. The fibroblast, in contrast, is stimulated to proliferate and produce collagen in a slightly acidic environment with reduced oxygen tension (Hunt and VanWinkle 1976).

Within the connective tissue scar 25 days after wounding cells were seen at the light-microscopic level of resolution that possessed striated myofibrillae. Corresponding electron-microscopic observations revealed that differentiated myocytes were present, in apparent isolation from the adjacent myocardium, in the connective tissue of the wound site. It appears from these observations that myocytes might be capable of limited migration into the wound. Rumyant- sev (1981) states that the maintenance of intercellular junctions by perinecrotic myocytes may limit their outgrowth into the wound site. However, it has been demonstrated that migrating epithelial cells maintain a limited number of junctional attachments during limb regeneration in adult newts (Repesh and Oberpriller 1980). In this investigation desmosomal attachments were observed between mesothelial cells which were migrating over the wound site. There- fore, the presence of intercellular junctions does not seem to be a limitation to epithelial migration, and thus, it might not be a limitation to myocyte migration. It is more reason- able to speculate that the laminar coat which surrounds small bundles of myocytes would be more likely to function as a barrier to migration. Indeed, structures resembling filopodia, which are characteristic of migrating epithelial cells, were demonstrated in perinecrotic myocytes in the five day experimental group. I f the laminar coat does serve to im- pede myocyte migration, then it may be presumed that myo-

cytes present within the wound site originated from the myocardium directly adjacent to the site of injury because the laminar coat at this point would be disrupted. Other myocytes in the region, which might also be stimulated to migrate, would be unable to do so due to the presence of an intact laminar coat. The appearance of myocytes within the wound site, thus, would be severely limited.

The possibility that these myocytes found within the connective tissue scar simply survived the original wound trauma and subsequently became surrounded by connective tissue also must be considered. Environmental conditions in the region bordering the wound are more favorable for survival of fibroblasts over myocytes. In fact, decreases in 02 concentration and pH, which could occur in this region, may stimulate fibroblast proliferation and collagen production. The same conditions would be detrimental to myocyte survival. This is reflected in the relatively more rapid onset of connective tissue cell proliferation within the reactive zone following injury. Also, the thymidine-labeling index for connective tissue cells always exceeded that for the myocardium during the experimental time periods examined. It is believed that atrial myocytes proliferate about two times more slowly than connective tissue cells. Rumyantsev (1981) reports that the length of the S phase following coronary artery ligation for left atrial connective tissue cells in the rat is seven hours, while that for left atrial myocytes is 13 h. It appears that several factors are present in the healing of the direct atrial injury, which would favor the production of a connective tissue scar rather than functional myocardium.

In summary, it may be stated that amphibian atrial myocytes are capable of significant D N A synthetic and mitotic activity following either ventricular (McDonnell and Oberpriller 1983 a) or direct atrial injury. The diffuse distribution of proliferative atrial myocytes following ventricular injury indicates that they may be responding to a hyper- trophic stimulus or an increase in functional demand. With direct atrial injury, the localization of proliferative atrial myocytes to a 500 gm reactive zone adjacent to the site of injury suggests that they are involved in the process of wound repair. Despite the pronounced proliferative activity, the atrial wound heals by the formation of a connective tissue scar with only a few, widely scattered, myocytes present. It is believed that environmental conditions of the atrial wound site favor connective tissue overgrowth and that the laminar coat surrounding groups of myocytes may serve as an effective barrier in preventing significant myocyte migration.

Acknowledgments. This research was supported in part by a Hamre Fellowship awarded to Timothy McDonnell and by BRSG 507-RR05407-18 awarded by the Biomedical Research Support Program, Division of Research Resources NIH to John Ober- priller.

References

Bing RJ (1971) Reparative processes in heart muscle following myocardial infarction. Cardiology 56: 314-324

Dusek J, Rona G, Kahn DS (1971) Healing process in the marginal zone of an experimental myocardial infarct. Am J Pathol 62 : 321-338

Hay DA, Low FN (1972) The fine structure of progressive stages of myocardial mitosis in chick embryos. Am J Anat 134:175-202

592

Hunt TK, VanWinkle W (1976) Fundamentals of wound manage- ment in surgery. Wound healing: Normal repair. Chirwigecom, Inc, South Plainfield, New Jersey, p 68

Jennings RB, Ganote CE (1974) Structural changes in myocardium during acute ischemia. Supplement III to Circulation Research, Vols 34 and 35, pp III-156-III-172

Kloner RA, Ingwall JS (1980) The cultured fetal mouse heart as a model for studying myocardial ischemic necrosis. Exp Mol Pathol 32 : 317-335

Kloner RA, Fishbein MC, Hare CM, Maroko PR (1979) Early ischemic ultrastructural and histochemical alterations in the myocardium of the rat following coronary artery occlusion. Exp Mol Pathol 30:129-143

Lautsch EV (1979) Morphological factors of clinical significance in myocardial infarction - A review. Texas Reports on Biology and Medicine 39 : 371-386

Manasek FJ (1968) Mitosis in developing cardiac muscle. J Cell Biol 37:191-196

McDonnell TJ, Oberpriller JO (1983a) The atrial proliferative response following partial ventricular amputation in the heart of the adult newt, Notophthalmus viridescens. A light and electron microscopic autoradiographic study. Tissue&Cell 15:351 364

McDonnell T J, Oberpriller JO (1983b) The ultrastructure of the atrium in the adult newt, Notophthalmus viridescens (Amphibia : Salimandridae). J Morphol 175:235-251

Oberpriller JO, Oberpriller JC (1971) Mitosis in adult newt ventricle. J Cell Biol 49:560-563

Oberpriller JO, Oberpriller JC (1974) Response of the adult newt ventricle to injury. J Exp Zool 187: 249-260

Oberpriller JO, Bader DM, Oberpriller JC (1979) The regenerative potential of cardiac muscle in the newt, Notophthalmus viridescens. In: Mauro A (ed) Muscle regeneration. Raven Press, New York, pp 323-333

Oberpriller JO, Oberpriller JC, Bader DM, McDonnell TJ (1981) Cardiac muscle and its potential for regeneration in the adult newt heart. In : Becker RO (ed) Mechanisms of growth control. Charles C Thomas. Springfield, Ill, pp 343 372

Oberpriller JO, Ferrans VJ, Carroll RJ (1983) Changes in DNA content, number of nuclei and cellular dimensions of young rat atrial myocytes in response to left coronary artery ligation. J Mol Cell Cardiol 15:31-42

Repesh LA, Oberpriller JC (1980) Ultrastructural studies on migrating epidermal cells during the wound healing stage of regeneration in the adult newt, Notophthalmus viridescens. Am J Anat 159 : 187-208

Rumyantsev PP (1966) Autoradiographic study on the synthesis of DNA, RNA and proteins in normal cardiac muscle cells and those changed by experimental injury. Folia Histochem Cytochem 4:397-424

Rumyantsev PP (1970) DNA synthesis and mitosis in atrial myocytes of rats with aortal stenosis. Experientia 26:773-774

Rumyantsev PP (1972) Electron microscope study of the myofibril partial disintegration and recovery in the mitotically dividing cardiac muscle cells. Z Zellforsch 129:471-499

Rumyantsev PP (1973) Post-injury DNA synthesis, mitosis and ultrastructural reorganization of adult frog cardiac myocytes. Z Zellforsch 139 : 431-450

Rumyantsev PP (1974) Ultrastructural reorganization, DNA synthesis and mitotic division of myocytes of atria of rats with left ventricle infarction. Virchows Arch [Cell Pathol] 15 : 357-378

Rumyantsev PP (1977) Interrelations of the proliferation and dif- ferentiation processes during cardiac myogenesis and regeneration. Int Rev Cytol 51:182273

Rumyantsev PP (1979) Some comparative aspects of myocardial regeneration. In: Muscle regeneration. Mauro A (ed) Raven Press, New York, pp 335355

Rumyantsev PP (1981) New comparative aspects of myocardial regeneration with special reference to cardiomyocyte proliferative behavior. In: Becker RO (ed) Mechanisms of growth control. Charles C Thomas, Springfield, Ill, pp 311-342

Rumyantsev PP, Kassem AM (1976) Cumulative indices of DNA synthesizing myocytes in different compartments of the working myocardium and conductive system of the rat's heart muscle following extensive left ventricle infarction. Virchows Arch [Cell Pathol] 20: 329-342

Rumyantsev PP, Mirakyan VO (1968 a) Reactive synthesis of DNA and mitotic division in atrial heart muscle cells following ventricle infarction. Experientia 24: 1234-1235

Rumyantsev PP, Mirakyan VO (1968 b) Increased activity of DNA synthesis and mitoses in rat atrial muscle cells under ventricular myocardial infarction and local injury of auricles. Tsitologiia |0:1276-1287

Rumyantsev PP, Snigirevskaya ES (1968) The ultrastructure of differentiating cells of the heart muscle in the state of mitotic division. Acta Morphol Acad Sci Hung 16:271-283

Sybers MD, Ingwall J, Deluca M (1979) Autophagic response to sublethal injury in cardiac myocytes. J Mol Cell Cardiol 11 : 331-338

Tomanek R J, Grimes JC, Diana JN (1981) Relationship between the magnitude of myocardial ischemia and ultrastructural alterations. Exp Mol Pathol 35:65 83

Weinstein RB, Hay ED (1970) Deoxyribonucleic acid synthesis and mitosis in differentiated cardiac muscle cells of chick embryos. J Cell Biol 47:310-316

Accepted September 13, 1983

the response of the atrium to direct mechanical wounding in the adult heart of the newt,...

Documents