THE ANATOMICAL RECORD 254:521–533 (1999) Long-term Regeneration of Fast and Slow Murine Skeletal Muscles after Induced Injury by ACL Myotoxin Isolated from Agkistrodon contortrix laticinctus (Broad-banded Copperhead) Venom TANIA DE FÁTIMA SALVINI,1* CLÁUDIO CÉSAR MORINI,1 HELOISA SOBREIRO SELISTRE DE ARAÚJO,2 AND CHARLOTTE LEDBETTER OWNBY3 1Laboratório de Neurociências, Departamento de Fisioterapia, Universidade Federal de São Carlos, 13565-905 São Carlos, São Paulo, Brazil 2Departamento de Ciências Fisiológicas, Universidade Federal de São Carlos, 13565-905 São Carlos, São Paulo, Brazil 3Department of Anatomy, Pathology, and Pharmacology, Oklahoma State University, Stillwater, Oklahoma 74078-2007 ABSTRACT The aim of the present work was to analyze the regenerated muscle types I and II fibers of the soleus and gastrocnemius muscles of mice, 8 months after damage induced by ACL myotoxin (ACLMT). Animals received 5 mg/kg of ACLMT into the subcutaneous lateral region of the right hind limb, near the Achilles tendon; contralateral muscles received saline. Longitudinal and cross sections (10 µm) of frozen muscle tissue were evaluated. Eight months after ACLMT injection, both muscle types I and II fibers of soleus and gastrocnemius muscles still showed centralized nuclei and small regenerated fibers. Compared with the left muscle, the incidence of type I fibers increased in the right muscle (21% ⫾ 03% versus 12% ⫾ 06%, P ⫽ 0.009), whereas type II fibers decreased (78% ⫾ 02% versus 88% ⫾ 06%, P ⫽ 0.01). The incidence of type IIC fibers was normal. These results confirm that ACLMT induced muscle type fiber transformation from type II to type I, through type IIC. The area analysis of types I and II fibers of the gastrocnemius revealed that injured right muscles have a higher percentage of small fibers in both types I and II fibers (0–1,500 µm2) than left muscles, which have larger normal type I and II fibers (1,500–3,500 µm2). These results indicate that ACLMT can be used as an excellent model to study the rearrangement of motor units and the transformation of muscle fiber types during regeneration. Anat Rec 254:521–533, 1999. r 1999 Wiley-Liss, Inc. Key words: muscle injury; muscle regeneration; gastrocnemius muscle; soleus muscle; ACL myotoxin Skeletal muscle fibers are injured after exposure to several kinds of snake venom (Ownby, 1990). Damage in skeletal muscles frequently leads to degeneration and subsequent regeneration of muscle fibers (Schmalbruch, 1976; Mauro, 1979), usually by repair of surviving fiber r 1999 WILEY-LISS, INC. *Correspondence to: Tania F. Salvini, Departamento de Fisioterapia, Universidade Federal de São Carlos, CEP: 13565-905, São Carlos, SP, Brazil. Fax: 0055-16/261-2081. E-mail: email@example.com Received 14 July 1998; Accepted 4 December 1998 522 SALVINI ET AL. Fig. 1. Cross section of the right gastrocnemius muscle 8 months after ACLMT injection (toluidine blue staining). Note the difference in the diameter of the muscle fibers and the abundant presence of muscle fibers with a centralized nucleus (arrowheads) and small regenerated split fibers (asterisks). Scale bar ⫽ 40 µm. fragments and by the development of new fibers. Mononucleate satellite cells, which are located between the plasma membrane and basal lamina, are the source of new muscle fibers (Snow, 1977, 1978). New myonuclei can come only from mononuclear myoblasts, which are descendants of satellite cells in mature muscles (Mauro et al., 1970; Holtzer et al., 1975). Injured skeletal muscle regenerates rapidly, forming myotubules by the end of 3 days, functionally reinnervated muscle fibers by days 4–5, and fully repaired fibers after 21–28 days (Schmalbruch, 1976; Grubb et al., 1991; Wernig et al., 1991a,b; Morini et al., 1998). Although the process of muscle fiber regeneration has been thoroughly studied, questions remain to be answered, especially concerning the regeneration process of different types of muscle fibers months after induced injury. Previous reports from our laboratory showed that ACL myotoxin (ACLMT) purified from the venom of the broadbanded copperhead (Agkistrodon contortrix laticinctus) is an excellent model to induce a homogeneous site of muscle injury in both soleus and gastrocnemius muscles of mice (Morini et al., 1998). ACLMT was first isolated by Johnson and Ownby (1993), and it was determined to be a Lys49 type II phospholipase A2 (PLA2) (Selistre de Araujo et al., 1996). Members of this class of PLA2 have no or very low phospholipase activity despite high myotoxic activity. The myotoxic effect of ACLMT was evaluated 3 hr and 3 and 21 days after subcutaneous injection of the toxin in the soleus and gastrocnemius muscles of mice (Morini et al., 1998). ACLMT injured both muscle type I and type II fibers in the soleus and gastrocnemius muscles. Twenty-one days after ACLMT injection, both muscles were completely regenerated, and there were many muscle fibers with centralized nuclei, split fibers, and clusters of newly regenerated muscle fibers. Muscle fiber type transformation was also observed, with a significant increase in the incidence of type IIC and a decreased incidence of type II fibers. Although ACLMT is known for its myotoxic activity, these results indicate that it can also be used as a model to induce rearrangement of the motor units and a change in muscle fiber types. These data stimulated new questions about the morphologic characteristics of regenerated types I and II fibers several months after damage induced by ACLMT. Could there be significant differences between types I and II Fig. 2. Longitudinal sections of the right gastrocnemius muscle 8 months after ACLMT injection (toluidine blue staining). (A) Muscle fibers with a small diameter (asterisk) and a centralized nucleus (arrowhead). (B) Presence of branched fibers (arrow) and a centralized nuclei (arrowheads). (C) A row of centralized nucleus along the muscle fiber (arrowheads). Scale bar ⫽ 20 µm. REGENERATION OF FAST AND SLOW MURINE MUSCLES Figure 2. 523 524 SALVINI ET AL. Fig. 3. Serial cross sections in the deep (A–C) and the superficial regions (D–F) of the right gastrocnemius muscle. Note that both regions of the muscle show split fibers (asterisks) and a centralized nucleus (arrowheads). In the deep region of the muscle, there are split fibers and centralized nuclei in type I and type II fibers. Some of the type I fibers are indicated by an asterisk. (B: ac-mATPase, pH 4.3; C: alc-mATPase, pH 10.3). In the superficial region of the muscle, where there are exclusively type II fibers (E: ac-mATPase, pH 4.3), it is possible to note a split fiber (asterisk, D–F) with an AChE reaction (arrowhead, F). Scale bar ⫽ 20 µm. REGENERATION OF FAST AND SLOW MURINE MUSCLES Fig. 4. Serial cross sections in the deep region of the right gastrocnemius muscle 8 months after ACLMT injection (A,C–F). (A) Toluidine blue staining shows skeletal muscle fibers with a centralized nucleus (arrowheads). (C) Note a small type I split fiber (A, star) with AChE reaction (C, star). Type I muscle fiber shows an intense reaction after ac-mATPase at pH 4.3 (E) but no reaction after alc-mATPase at pH 10.3 (F). (D) The split 525 type I fiber (star) has lower SDH activity. When A, E and F are compared, it is possible to identify the presence of a centralized nucleus in both type I and II fibers. B: Left control gastrocnemius after toluidine blue staining shows a normal morphologic aspect, with the nucleus located in the periphery of the muscle fibers (arrowhead). Scale bar ⫽ 20 µm. 526 SALVINI ET AL. soleus and gastrocnemius muscles of mice, 8 months after induced damage by ACLMT. MATERIALS AND METHODS ACL Myotoxin ACLMT was purified from the crude venom of the broad-banded copperhead (Agkistrodon contortrix laticinctus), as previously described (Johnson and Ownby, 1993). Briefly, this process consists of fractionation of crude venom by anion exchange chromatography followed by final purification using cation exchange chromatography. Animal Care and Experimental Groups Eight male mice (white Swiss), weighing 30–35 g, were used. The animals were housed in groups in standard plastic cages in an animal room under controlled environmental conditions (12 hr dark/12 hr light cycle; temperature 22.5°C). Mice received standard food and had access to food and water ad libitum. All animals were given one dose of ACLMT (5 mg/kg) into the subcutaneous lateral region of the right hind limb, near the Achilles tendon. The injection was made in the distal to proximal direction in the middle line between the insertion of the Achilles tendon and the distal surface of the lateral malleolus of fibula. A similar region of the contralateral left muscle was injected with saline and used as a control. Histology and Histochemistry Fig. 5. Schematic representation of serial cross sections along 210 µm obtained in the deep region of the right gastrocnemius muscle 8 months after ACLMT injection. There are fibers with a centralized nucleus (fibers 1 and 2) along the 210 µm. Split fibers are also present (stars). Scale bar ⫽ 40 µm. fibers in both soleus (slow twitch) and gastrocnemius (fast twitch) muscles of mice several months after such an induced injury? The purpose of the experiments reported here was to contribute to the knowledge of the longtime regeneration characteristics of muscle type fibers after injury induced by a snake venom toxin. The primary aim was to analyze the regenerated skeletal muscle types I and II fibers of Eight months after ACLMT injection, the animals were weighed under deep ethyl ether anesthesia. Afterward, right and left gastrocnemius and soleus muscles were removed, weighed, immediately frozen in melting isopentane, and stored in a freezer at ⫺56°C. Frozen muscles were cut through the proximal to distal region using a cryostat (10-µm cross sections). Alternate serial sections were obtained in the middle region of both muscles to evaluate all muscle fibers of both soleus and gastrocnemius muscles. Histologic cross sections were stained with 1% toluidine blue / 1% borax or for acid phosphatase (AcPase; Lojda et al., 1976), myofibrillar ATPase activity (mATPase) after alkali (alc-mATP, pH 10.3; Guth and Samaha, 1969 and Butler and Cosmos, 1981) or acid pre-incubations (ac-mATP, pH 4.3; Brooke and Kaiser, 1970), succinate dehydrogenase (SDH; Nachlas et al., 1957), and acetylcholinesterase (AChE; Karnovsky and Roots, 1964). Longitudinal sections (10 µm), which were stained with toluidine blue, were obtained from two right soleus and gastrocnemius muscles. The incidence of damaged fibers was evaluated using videoprint montages of single serial cross sections of the middle region of muscles stained with toluidine blue or submitted to mATPase reactions (pH 4.3 and 10.3). This region was chosen because it contains the highest number of muscle fibers. All damaged fibers of the serial cross sections were identified. Furthermore, serial cross sections submitted to mATPase (pH 4.3 and 10.3) were used to Fig. 6. Schematic representation of serial cross sections along 420 µm obtained from the superficial region of the gastrocnemius muscle 8 months after ACLMT injection. There are muscle fibers with centralized nuclei in all sections evaluated (arrowheads). Note also the presence of split fibers. Scale bar ⫽ 80 µm. REGENERATION OF FAST AND SLOW MURINE MUSCLES Figure 6. 527 ⫺ ⫺ ⫺ ⫺ 01 (0.2%) 0.0 02 (0.5%) 01 (0.2%) 0.2% ⫾ 0.2 428 491 404 465 447 ⫾ 39 ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ 288 (51%) 353 (52%) 270 (50%) 231 (56%) 52% ⫾ 3** 561 669 552 411 548 ⫾ 106 1 2 3 4 X ⫾ SD ⫺ ⫺ ⫺ ⫺ 6,868 6,136 5,750 4,286 5,760 ⫾ 1,086 ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ ⫹⫹⫹ 1 2 3 4 X ⫾ SD 1,548 (21%) 1,184 (17%) 1,329 (20%) 1,216 (18%) 19% ⫾ 1.6* 7,548 7,055 6,590 6,748 6,985 ⫾ 422 Animal *P ⫽ 0.00009 (paired Student’s t test). **P ⫽ 0.00001 (paired Student’s t test). aSplit fiber: ⫹⫹⫹, frequent; ⫺, rarely observed. Animal Split fibera Centralized nucleus 19 (0.3%) 07 (0.1%) 07 (0.1%) 12 (0.3%) 0.2% ⫾ 0.09 Centralized nucleus Left muscle Number of fibers Split fibera Centralized nucleus Right muscle (ACLMT) Number of fibers Soleus Left muscle Number of fibers Split fibera Eight months after the injection of ACLMT, the right gastrocnemius muscles contained small regenerated muscle fibers, large numbers of centralized nuclei, and split fibers (Fig. 1). Longitudinal sections showed that centralized nuclei were arranged as isolated nuclei or together in a row (Fig. 2). Longitudinal sections also showed that some fibers branched (Fig. 2B) and, in some of them, the branches reunited at varying distances. Muscle fibers with centralized nuclei and split fibers were identified in both deep (Fig. 3A–C) and superficial (Fig. 3D–F) regions of the muscle. The deep region of the gastrocnemius muscle of mice is composed of types I and II fibers (Fig. 3A–C), whereas the superficial region contains exclusively type II fibers (Fig. 3D–F). Several small regenerated and differentiated type II (Fig. 3D–F) and type I (Fig. 4A,C–F) fibers showed positive AChE reactions. Both types I and II fibers had signs of previous injury, such as a centralized nucleus and split fibers. Serial cross sections taken in the proximal, middle, and distal regions of the muscle had these signs in all regions. The left gastrocnemius had normal morphologic features in all regions analyzed (Fig. 4B). Using serial cross sections, bundles of muscle fibers of both the deep and superficial regions of the right gastrocnemius were photographed and reconstructed along 210 and 420 µm, respectively (Figs. 5 and 6). Split fibers were present in both deep and superficial regions of the muscle. This schematic representation revealed that all muscle fibers showed at least a centralized nucleus, and some fibers contained a centralized nucleus in all sections evaluated. The number of muscle fibers with one or more centralized nuclei was assessed applying previously established criteria (Wernig et al., 1991a) to one series of cross sections through the medial region of muscle. The incidence of fibers with a centralized nucleus was significantly increased in the right muscle (19% ⫾ 1.6% versus 0.2% ⫾ Centralized nucleus Gastrocnemius Muscle Right muscle (ACLMT) The average weight of the animals was 76 ⫾ 4 g. There was no difference between the average weights of the right and left gastrocnemius muscles (0.26 g ⫾ 0.02 g versus 0.26 g ⫾ 0.02 g, respectively) or the right and left soleus muscles (0.016 g ⫾ 0.002 g versus 0.016 g ⫾ 0.002 g, respectively) 8 months after ACLMT injection. Number of fibers RESULTS Body, Soleus, and Gastrocnemius Weights Gastrocnemius identify the incidence of muscle fiber types (I, II, IIC) in the deep region of the gastrocnemius. Each split muscle fiber was counted as one fiber. Muscle fiber types were quantified only in the deep region of the muscle because the superficial region showed the exclusive presence of type II fibers. To locate the deep region of the gastrocnemius muscle, the distance from the deep to the superficial region of each cross section was measured by light microscopy, and the middle point of the cross section was identified. Only muscle fibers of the deep region were analyzed, also using videoprint montages. The muscle fiber types were evaluated using computer software (Vinspec) connected to a microscope. Chronic signs of muscle fiber injury were identified by the presence of split fibers and centralized nuclei, without acute signs of damage, such as necrosis, basophilia, and cellular infiltration (Carpenter and Karpati, 1984, Wernig et al., 1991a, Morini et al., 1998). Split fibera SALVINI ET AL. TABLE 1. Incidence of centralized nucleus and split fibers in the right and left gastrocnemius and soleus muscles 8 months after ACLMT injection 528 529 REGENERATION OF FAST AND SLOW MURINE MUSCLES TABLE 2. Incidence of skeletal muscle type fibers (I, II, IIC) in the right and left gastrocnemius 8 months after ACLMT injection Gastrocnemius Right muscle (ACLMT) Left muscle Animal Type I Type II Type IIC Type I Type II Type IIC 1 2 3 4 X ⫾ SD 468 (19%) 686 (24%) 471 (18%) 634 (21%) 21% ⫾ 03* 1,923 (79%) 2,206 (76%) 2,095 (81%) 2,331 (77%) 78% ⫾ 02** 46 (1.8%) 18 (0.6%) 19 (0.7%) 44 (1.5%) 1.2% ⫾ 0.6*** 130 (09%) 548 (20%) 152 (10%) 152 (8%) 12% ⫾ 06 1,269 (90%) 2,117 (79%) 1,386 (90%) 1,698 (92%) 88% ⫾ 06 04 (0.2%) 12 (0.4%) 00 (0.0%) 02 (0.1%) 0.2% ⫾ 0.2 *P ⫽ 0.009 (paired Student’s t test). **P ⫽ 0.01 (paired Student’s t test). ***P ⫽ 0.02 (paired Student’s t test). 0.09%, P ⫽ 0.00009, paired Student’s t test; Table 1). The degree of injury is probably underestimated in these studies because of its segmental occurrence and the evaluation at only one level. When more levels along the length of the muscles are evaluated, the total estimate of damaged fibers increases (see Figs. 5 and 6). The presence of split fibers was also high in the injured muscle but rare in the left control gastrocnemius (Table 1). The right gastrocnemius muscle always showed an increased number of muscle fibers when compared with the left one (Table 1). It was interesting that 8 months after ACLMT injection, there is still a significant change in the incidence of types I and II muscle fibers (Table 2). The incidence of type I fibers increased in the right muscle (21% ⫾ 03% versus 12% ⫾ 06%, P ⫽ 0.009, paired Student’s t test; Table 2), whereas that of type II decreased (78% ⫾ 02% versus 88% ⫾ 06%, P ⫽ 0.01, paired Student’s t test; Table 2) compared with the left one. Although there was also an increased percentage of type IIC fibers in the right muscle (1.2% ⫾ 06% versus 0.2% ⫾ 02%, P ⫽ 0.02, paired Student’s t test; Table 2), this percentage is still in the normal range for mamma–an skeletal muscles. The area analysis of types I and II muscle fibers of the gastrocnemius revealed that injured muscles on the right have a higher percentage of small regenerated fibers of both types I and II (0–1,500 µm2, Fig. 7) than the control left muscles, which have predominantly larger normal fibers (1,500–3,500 µm2; Fig. 7). Larger fibers observed in the right muscles (1,750–4,000 µm2) were probably recovered fibers, whereas the smaller fibers were regenerated fibers. Soleus Muscle Morphologic patterns similar to the longitudinal and cross sections of the right gastrocnemius were observed in the skeletal muscle fibers of the right soleus 8 months after ACLMT injection. All regions of the muscle showed large numbers of fibers with a centralized nucleus and split fibers (Fig. 8A). Despite the presence of a centralized nucleus, most of the fibers were differentiated into type I or type II (Fig. 8C,D). Type IIC fibers were rare. The control left soleus presented a normal morphologic aspect (Fig. 8B). When the soleus muscle fibers were analyzed to quantify the presence of centralized nuclei using only a serial cross section, it was observed that 52% ⫾ 3% of the right muscles had a centralized nucleus, while the left muscle had 0.2% ⫾ 0.2% (P ⫽ 0.00001, paired student’s t test; Table 1). DISCUSSION We previously reported (Morini et al., 1998) a significant decrease in the percentage of type II fibers and an increase in the percentage of type IIC fibers in the deep region of the murine gastrocnemius 21 days after damage induced by ACLMT. There was no change in the superficial region of the muscle, which contained only regenerated type II fibers. These results led us to suggest that there was a change in muscle fiber type in the deep region of gastrocnemius from type II to type I, through type IIC. Twenty-one days after ACLMT injection, 17% of muscle fibers identified in the deep region of the gastrocnemius were type IIC (Morini et al., 1998), whereas 8 months later the incidence of type IIC was 1.2% (see Table 2). This result, associated with an increased percentage of type I and decreased percentage of type II, confirms the hypothesis that ACLMT induced muscle fiber type transformation from type II to type I, through type IIC. The results reported here also indicate similar morphologic characteristics in both types I and II muscle fibers of regenerated soleus and gastrocnemius murine muscles, 8 months after induced injury by ACLMT. Despite their metabolic and physiologic differences, both muscles had significant numbers of muscle fibers with one or more centralized nuclei and split fibers in all regions (proximal, middle, and distal) evaluated. It is also interesting that 8 months after induced damaged, there was a large number of small regenerated types I and II muscle fibers with centralized nuclei. This suggests that despite the differentiation that took place in the small regenerated new fibers, hypertrophy and migration of the centralized nucleus to the periphery of muscle fibers did not accompany this process. Mouse soleus muscle contains an approximately equal number of slow-twitch (type I) and fast-twitch (type IIA) fibers (Lewis et al., 1982; Desypris and Parry, 1990; Wernig et al., 1991a), whereas the gastrocnemius muscle contains predominantly type II muscle fibers (Morini et al. 1998). In addition, the gastrocnemius muscle has two distinct regions, which can be identified by mATPase—a deep region composed of types I and II fibers and a superficial region composed of types IIA and IIB fibers only (Armstrong and Phelps, 1984; Morini et al., 1998). No fibrosis was observed in the evaluated muscles, indicating that, in general, all muscle fibers regenerated after induced injury by ACLMT. This observation confirms the efficiency of ACLMT as a model to induce muscle injury 530 SALVINI ET AL. Fig. 7. Average area (µm2) of types I and II fibers in the right (ACLMT) and left (control) gastrocnemius muscle 8 months after ACLMT injection. REGENERATION OF FAST AND SLOW MURINE MUSCLES 531 Fig. 8. Right soleus muscle 8 months after ACLMT injection (A: toluidine blue; C: alc-mATPase at pH10.3; D: SDH). Serial cross sections show that muscle fibers with a centralized nucleus (arrowheads, A) could be type I or type II (type II fibers are indicated by asterisks, A, C, D). (B) Left soleus muscle (toluidine blue) shows a normal morphologic aspect, with the nucleus located at the periphery of the muscle fibers (arrowheads). Scale bar ⫽ 20 µm. and muscle regeneration in slow- and fast-twitch skeletal muscles. Fibrosis is common after muscle necrosis and can be induced by different procedures, such as ischemia (Ownby et al., 1990), several types of myopathies (for review see Engel and Banker, 1986), and periodic contusions (Minamoto et al., 1999). The presence of fibrosis, a centralized nucleus, and split fibers were also observed in the soleus muscle of rat 6 months after induced injury by injection of Ringer solution at 60–70°C (Schmalbruch, 1976). Serial cross sections obtained from soleus and gastrocnemius muscles showed that both types I and II fibers had centralized nuclei 8 months after induced injury by ACLMT. It is important to note that all muscle fibers evaluated by serial cross sections in the deep and superficial regions of the gastrocnemius had centralized nuclei (see Figs. 5 and 6). The presence of a centralized nucleus and split fibers are well described in several conditions, such as denervated muscles (Lu et al., 1997; Rodrigues and Schmalbruch, 1995), regenerated muscles after induced injury by toxins (Davis et al., 1991; Morini et al., 1998) and running exercise (Wernig et al., 1991a), muscle graft (Carlson and Faulkner, 1983), and after muscle contusion (Minamoto et al., 1999). The proportion of type IIC muscle fibers is higher than normal after muscle injury because actively regenerating and recently regenerated fibers are histochemically type IIC, and this type of fiber is considered to be in the process of changing from one fiber type to another (Pette and Staron, 1990; Wernig et al., 1991a, b). Such change probably occurs after denervation and re-enervation, when a muscle fiber becomes connected to a new motoneuron of a different type (Wernig et al., 1991b). Another possibility is that the temporary disconnection of muscle from nerve results in the reprogramming of the motor neurons (Davis et al., 1991). Type IIC fibers are classified as undifferentiated and are usually rare, but not absent, in the normal muscles of adult mammals (Pette and Staron, 1990). This is in agreement with the low incidence of type IIC fibers 532 SALVINI ET AL. observed in the present study in the right and left murine gastrocnemius, thus indicating a low degree of ongoing changes in both muscles. An increased incidence of types IIC and I fibers and a decrease of type II fibers were also identified in mouse soleus muscle after muscle damage induced by running exercise, freezing, or overuse (Wernig et al., 1991a,b). Predominance of type I fibers in regenerated muscles were found in the soleus rat muscle 56 days after local injection of the crude venom of Notechis scutatus (Davis et al., 1991). Our hypothesis for the transformation process of fiber types is that in the presence of muscle damage or muscle paralysis, the motor neuron terminals responsible for the innervation of type I muscle fibers produce axonal sprouts faster than motor neurons of type II motor units (Brown et al., 1981; Desypris and Parry, 1990). This difference in the speed of production of axonal sprouts between fast and slow motor units was also observed in the skeletal muscles of mice (Duchen, 1970; Brown et al., 1980). It could explain why only the deep region of gastrocnemius showed an increased incidence of type IIC muscle fibers 21 days after damage by ACLMT injection (Morini et al., 1998) and an increased incidence of type I fibers 8 months after induced injury by ACLMT. Perhaps it occurs because only this region of the muscle has type I motor units. The absence of type IIC fibers in the superficial region of the gastrocnemius, where there are only type II motor units, confirms this hypothesis. The increased number of type I fibers and the presence of AChE activity in the small regenerated fibers of the gastrocnemius muscle 8 months after ACLMT injection suggest that the injury produced by ACLMT induced axonal sprouts and rearrangement of the motor units. The probable mechanism for the rearrangement of motor units after injury is denervation of the necrotized muscle fibers and subsequent re-enervation of regenerated muscle fibers by axonal sprouting (Brown et al., 1981; Wernig et al., 1991a,b). In conclusion, the results of this work indicate that ACLMT induced muscle injury in types I and II fibers in both soleus and gastrocnemius murine muscles; chronic signs of previous injury, such as centralized nuclei, split fibers, and small regenerated fibers, are seen 8 months after ACLMT injection. Moreover, the regenerated characteristics of both types I and II fibers are similar, and muscle injury induced muscle type fiber transformation from type II to type I, through type IIC. Thus, ACLMT can be considered an excellent model for inducing muscle damage and muscle regeneration in both fast- and slowtwitch muscles of mice, stimulating the rearrangement of the motor units and transformation of muscle fiber types. ACKNOWLEDGMENTS C.C. Morini was the recipient of a Master Fellowship from Coordenacão de Aperfeiçoamento de Pessoal de Ensino Superior (CAPES) and Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP). This work was supported by Conselho Nacional de Ensino i Pesquisa (CNPq) and FAPESP (Brazilian agencies). Tereza F.F. Piassi was the recipient of a technical fellowship from CNPq, and we thank her for technical assistance. LITERATURE CITED Armstrong RB, Phelps RO. 1984. Muscle fiber type composition of rat hindlimb. Am J Anat 171:259–272. Brown MC, Holland RL, Ironton R. 1980. Nodal and terminal sprouting from motor neurons in fast and slow muscles of mouse. J Physiol 306:493–510. Brown MC, Holland RL, Hopkins WG. 1981. Motor nerve sprouting. Ann Neurol Neurosci 4:17–42. Brooke MH, Kaiser KK. 1970. Muscle fiber types: how many and what kind? Arch Neurol 23:369–379. Butler J, Cosmos E. 1981. Enzyme markers to identify muscle-nerve formation during embriogenesis: modified myosin ATPase and silvercolinesterase histochemical reactions. Exp Neurol 73:831–836. Carlson BM, Faulkner JA. 1983. The regeneration of skeletal muscle fibers following injury: a review. Med Sci Sports Exerc 15(3):187– 198. Carpenter S, Karpati G. 1984. Pathology of skeletal muscle. New York, Edinburgh, London, Melbourne: Churchill Livingstone. Davis CE, Harris JB, Nicholson LVB. 1991. Myosin isoform transitions and physiological properties of regenerated and re-innervated soleus muscles of the rat. Neuromuscul Disord 1(6):411–421. Desypris G, Parry DJ. 1990. Relative efficacy of slow and fast motoneurons to reinnervate mouse soleus muscle. Am J Physiol 258(Cell Physiol 27):C62–C70. Duchen LW. 1970. Hereditary motor endplate disease in the mouse: light and electron microscope studies. J Neurol Neurosurg Psychiatry 33:238–250. Engel AG, Banker BQ. 1986. Myology. New York: McGraw-Hill. Grubb BD, Harris JB, Schofield IS. 1991. Neuromuscular transmission at newly formed neuromuscular junctions in the regenerating soleus muscle of the rat. J Physiol 441:405–421. Guth L, Samaha, FJ. 1969. Qualitative differences betweem altomyosin ATPase of slow and fast mammalian muscle. Exp Neurol 25:138–162. Holtzer H, Jones KW, Yaffe D. 1975. Research group on neuromuscular diseases: a report on various aspects of myogenic cell culture with particular reference to studies on the muscular dystrophies. J Neurol Sci 26:115–124. Johnson EK, Ownby CL. 1993. Isolation of myotoxin from the venom of Agkistrodon contortrix laticinctus (broad-banded copperhead) on the pathogenesis of myonecrosis induced by it in mice. Toxicon 31:243– 245. Karnovsky MJ, Roots LA. 1964. A ‘‘direct-coloring’’ thiocholine method for choline esterase. J Histochem Cytochem 12:219–221. Lewis DM, Parry DJ, Rowlerson A. 1982. Isometric contractions of motor units and immunohistochemistry of mouse soleus muscle. J Physiol (Lond) 25:393–402. Lojda Z, Gossrau R, Schiebler TH. 1976. Enzym-histochemische Methoden. Berlin and Heildelberg: Springer-Verlag. p 300. Lu DX, Huang SK, Carlson BM. 1997. Electron microscopic study of long-term denervated rat skeletal muscle. Anat Rec 248:355–365. Mauro A. 1979. Muscle regeneration. New York: Raven Press. Mauro A, Shafiq SA, Milhorat AT. 1970. Regeneration of striated muscle and myogenesis. Amsterdam: Excerpta Medica. Minamoto VB, Grazziano CR, Salvini TF. 1999. Effect of single and periodic contusion on the rat Soleus muscle at different stages of regeneration. Anat Rec 254:281–287. Morini CC, Pereira ECL, Selistre de Araújo HS, Ownby CL, Salvini TF. 1998. Injury and recovery of fast and slow skeletal muscle fibers affected by ACL myotoxin isolated from Agkistrodon contortrix laticinctus (broad-banded copperhead) venom. Toxicon 36:1007– 1024. Nachlas MM, Tsou KC, De Sousa E, Cheng CS, Seligman AM. 1957. Cytochemical demonstration of succinic dehydrogenase by the use of a new -nitrophenyl substituted ditetrazole. J Histochem Cytochem 5:420–436. Ownby CL. Locally acting agents: miotoxins, hemorragic toxins and dermonecrotic factors In: Shier WT, Mebs D, editors. Handbook of toxinology. New York: Marcel Dekker, Inc. p 601–654. Pette D, Staron RS. 1990. Cellular and molecular diversities of mammalian skeletal muscle fibers. Rev Physiol Biochem Pharmacol 16:2–63. REGENERATION OF FAST AND SLOW MURINE MUSCLES Rodrigues AC, Schmalbruch H. 1995. Satellite cells and myonuclei in long-term denervated rat muscle. Anat Rec 243:430–437. Schmalbruch H. 1976. The morphology of regeneration of skeletal muscle in the rat. Tissue Cell 8(4):673–692. Selistre de Araujo HS, White SP, Ownby CL. 1996. cDNA cloning and sequence analysis of a Lys49 phospholipase A2 myotoxin from Agkistrodon contortrix laticinctus snake venom. Arch Biochem Biophys 326:21–30. Snow MH. 1977. Myogenic cell formation in regenerating rat skeletal muscle injured by mincing. II. Na autoradiographic study. Anat Rec 188:201–218. 533 Snow MH. 1978. An autoradiographic study of satellite cell differentiation into regenerating myotubes following transplantation of muscles in young rats. Cell Tissue Res 186:535–540. Wernig A, Salvini TF, Irintchev A. 1991a. Axonal sprouting and changes in fibre types after running-induced muscle damage. J Neurocytol 20:903–913. Wernig A, Salvini TF, Langenfeld-Oster B, Irintchev A, Dorlochter M. 1991b. Endplate and motor unit remodelling in vertebrate muscles. In: Wernig A, editor. Motoneuronal plasticity: restorative neurology. Volume 5. Amsterdam: Elsevier. p 85–100.