THE ANATOMICAL RECORD 229:298-304 (1991)
Regeneration of Skeletal Muscle in Streptozotocin-Induced Diabetic Rats ADARSH K. GULATI AND M.S. SWAMY Department of Anatomy (A.K.G.) and Department of Cell and Molecular Biology (M.S.S.), Medical College of Georgia, Augusta, Georgia
ABSTRACT The present study analyzes the regeneration of skeletal muscle in diabetic rats. Intravenous injection of streptozotocin (STZ) was used to induce diabetes. Six weeks later the extensor digitorum longus (EDL) muscles from diabetic rats were either transplanted into diabetic or normal hosts to initiate regeneration. Normal EDL muscle transplants in normal and diabetic hosts were also performed for comparison. One, 2,4, and 12 weeks after transplantation, the EDL regenerates were morphologically analyzed. Regeneration and formation of neuromuscular junctions were observed in all transplants, including diabetic regenerates in diabetic hosts. The overall mass and myofiber size of the diabetic EDL regenerate in the diabetic host was significantly reduced in spite of complete regeneration. Recovery of the diabetic muscle mass and the myofiber size was observed after transplantation into normal hosts. A reduction in mass and myofiber size was observed in normal EDL muscles transplanted into diabetic hosts. It is concluded that poor recovery of diabetic muscle is related to metabolic and structural alterations in the diabetic host, rather than to innate capacity of the muscle to per se undergo regeneration and reinnervation. The observed enhancement in recovery of diabetic muscle after transplantation in a normal host and deterioration of normal muscle after transplantation in a diabetic host shows that the host environment determines the success of muscle regeneration. A diabetic state in rats can be induced by the administration of streptozotocin (STZ), a n agent that causes destruction of insulin producing (3 cells in the pancreas. The resultant diabetic state is known to have profound effects on skeletal muscle morphology, physiology and metabolism (Armstrong et al., 1975; Chen and Ianuzzo, 1982; Paulus and Grossie, 1983; Feczko and Klueber, 1988; Klueber et al., 1989). Protein synthesis and metabolism in muscle, along with enzymatic activity and mitochondria1 function, are decreased in the diabetic condition (Wool et al., 1968; Manchester, 1970; Jefferson et al., 1974; Paulus and Grossie, 1983). Muscle mass is reduced due to elevated proteolysis, and associated increase in oxidation of free fatty acids and ketone bodies (Randle, 1966). Signs of degeneration and denervation have also been reported in skeletal muscle of murine diabetes (Feczko and Klueber, 1988). Thickening of capillary basal lamina within the skeletal muscle has been described in diabetics (Vracko, 1970). The objective of this study was to analyze the regenerative ability and recovery of skeletal muscle in a n experimentally induced diabetic state. An experimental model used extensively in muscle regeneration studies involves transplantation of extensor digitorum longus (EDL) muscle in rats (Carlson and Gutmann, 1975; Hansen-Smith and Carlson, 1979; Gulati, 1986, 1988). A predictable sequence of events follows transplantation, with a number of factors regulating the regenerative success. A majority of myofibers except a thin rim of peripheral myofibers undergo intrinsic degeneration a s a result of devascularization 0 1991 WILEY-LISS, INC
caused by the transplantation procedure. With the restoration of blood supply, the resident precursor myosatellite cells become active, proliferate, and differentiate into myoblasts. The myoblasts fuse to form myotubes that gradually increase in size and mature into myofibers. The regenerated muscle becomes innervated, recovers and becomes functional (Snow, 1977; Carlson, 1978; Gulati et al., 1982). In the present study, the EDL muscle transplantation model was used to examine muscle regeneration under diabetic and nondiabetic control conditions. The results demonstrate impaired recovery of muscle regenerates in diabetic rats as compared to regenerates in normal rats. MATERIALS AND METHODS
Inbred male Fischer rats weighing 150-200 g were used in this study to exclude the possibility of immune rejection of muscle transplants by their hosts. Rats were fed Purina rodent chow ad libitum and housed individually in wire cages. Rats were made diabetic by a single intravenous injection in the tail of a freshly prepared solution of STZ (65 mg/kg, in 0.1 M citrate buffer, pH 4.5) under ether anesthesia (Perry et al., 1987). To screen for diabetes plasma glucose levels were determined by the glucose oxidase assay of Raabo
Received August 8, 1989; accepted July 30, 1990. Address reprint requests to Dr. A.K. Gulati, Department of Anatomy, Medical College of Georgia, Augusta, GA 30912-2000.
299
MUSCLE REGENERATION IN DIABETES
and Terkildsen (1960) provided as a kit from Sigma Chemical Company (St. Louis, MO). Streptozotocintreated rats developed a markedly elevated plasma glucose levels (510-791 mgidl) that were measured a t 2 weeks after injection and a t the time of muscle regenerate analysis. A few rats that did not exhibit increase in plasma glucose level after STZ injection were excluded from the study. Six weeks after STZ injection the rats were divided into the following groups for muscle transplantation surgery. In a group of eight diabetic animals, the EDL muscle was autotransplanted (i.e., transplanted at the same site in the same diabetic animal) in order to study the regeneration pattern of diabetic muscle in the diabetic host. The EDL muscles from the second group of eight diabetic animals were isotransplanted (i.e., transplanted into another genetically identical Fischer rat) into the left side of nondiabetic normal rats in order to study regeneration of diabetic muscle in normal animals. The right EDL muscle of these normal animals was also autotransplanted as normal muscle regeneration controls. In addition a group of eight diabetic animals were bilaterally transplanted with normal EDL muscle. This surgical protocol resulted in four transplantation groups for comparison: diabetic muscle into diabetic hosts (DM-DH), diabetic muscle into normal host (DMNH), normal muscle into normal host (NM-NH), and normal muscle into diabetic host (NM-DH). For each of these groups 16 muscles were prepared with 4 muscles being analyzed at each of the time intervals. The transplantation procedure of the EDL muscle was similar to that described in detail earlier (Gulati, 1986, 1988). After chloral hydrate anesthesia (40 mgl 100 g body weight, i.p.) the EDL muscle was exposed by cutting the overlying skin and muscle. The EDL muscle was then removed from its bed by cutting the proximal and distal tendons. The muscle weight of each diabetic and normal control muscle was recorded for comparison with weights after regeneration. For transplantation into a diabetic host, the diabetic muscle was returned to its original bed and the cut tendons sutured back under slight tension. For transplantation of diabetic muscle into a normal host, the normal host EDL muscle was removed and replaced by the diabetic muscle. Similar procedure was followed for normal muscle autotransplants and for normal muscle transplants into diabetic hosts. All animals were allowed to recover and maintained until the time of muscle analysis. Four muscle transplants from each group were removed for morphological analysis a t intervals of 1 , 2 , 4 , and 12 weeks after transplantation. Under anesthesia the EDL transplants were removed, weighed, and frozen in liquid nitrogen. Cross sections, 8 and 24 pm thick were cut in a cryostat set at -20°C. Sections cut from different regions (i.e., throughout the length of the muscle) were then mounted on glass slides. The 8 pm sections were stained with periodic acid-Schiff hematoxylin (PASLhematoxylin) and used for general morphological and morphometric analysis. The 24 pm thick sections were stained with cholinesterase-silver (Toop, 1976) for determining the innervation status. Some muscles which had never been transplanted were also examined to define the morphological features of normal and diabetic muscles and myofiber size. For morphometric analysis, a Zeiss axiophot light micro-
lO0r
-
80 60
k-
-
I
U
W 3
40
20
n-
NM
NH-NH
NM-DH
DH
OM-DH
OM-NH
MUSCLE TYPES Fig. 1. A comparison of relative muscle weights in mg (t SEM) of normal (NM), 6 week diabetic (DM), and the 12 week EDL muscle regenerates from the various transplantation groups. A significant reduction in muscle weights was seen in 6 week diabetic EDL (DM) as compared to normal EDL (NM). The normal EDL muscle regenerates in normal host (NM-NH)and in diabetic host (NM-DH)are shown and are significantly different. The diabetic EDL muscle regenerates in diabetic (DM-DH)and normal hosts (DM-NH)are shown and are also significantly different. Student’s t test was used for statistical analysis between groups and significance was P < 0.05.
scope attached to a color camera and a n IBM computer was employed. Computer assisted quantification was accomplished by using the CoreSCAN Program (Extension Systems, Sebastopol, CAI. Four muscles from each experimental group were used and area of 50 myofibers from each was determined. Individual myofibers from different regions of the muscle were outlined, and the number of pixels within the outlined area was determined. The number of pixils was converted mathematically into area (pm’). Statistical analysis was done to compare, before and after regeneration, mean myofiber area using Student’s t test, with a probability less than 0.05 indicating significance. RESULTS Comparison of Normal (NM) and Diabetic (OM) Muscles
The muscle weights of normal and 6 week diabetic nontransplanted EDL muscles are shown in Figure 1. A significant reduction in the mass of diabetic muscle a s compared to normal muscle was observed (Fig. 1, NM and DM). Corresponding reduction in myofiber size was also observed between normal and diabetic muscle (Fig. 2, NM and DM). Normal rat muscle possessed polygonal myofibers of variable diameter and staining intensity. Individual myofibers had a thin pericellular endomysium, with groups of myofibers surrounded by a thicker perimysium (Fig. 3A). The myofibers in diabetic muscle were smaller in size, but like the normal muscle were polygonal and possessed peripheral nuclei. In the diabetic muscles, many small nerves and motor endplates were seen demonstrating innervation (Fig. 3B and C). Regeneration of Diabetic Muscle in Diabetic Host (DM-DH)
After transplantation of diabetic EDL into a diabetic host, the various events typical of regeneration were observed. However, the overall weight of the muscle
300
A.K. GULATI AND M.S. SWAMY
Regeneration of Diabetic Muscle in Normal Host (DM-NH)
3500~
Ix W m
1500
L0l 1000 > r
500 0 NU
OM
NU-NH
OR-NH
NU-OH
OM-OH
MUSCLE TYPES Fig. 2. A comparison of myofiber size in km2 ( i S E M ) normal (NM), 6 week diabetic (DM),and the 12 week EDL muscle regenerates from the various transplantation groups. A significant reduction in myofiber size is seen between NM and DM groups. The myofiber size of normal and diabetic EDL muscles transplanted into normal hosts (NM-NH and DM-NH, respectively) shows a significant recovery in the size of diabetic muscle after transplantation into normal host. On the other hand the myofiber sizes of normal and diabetic EDL into diabetic hosts (NM-DH and DM-DH, respectively) show a significant deterioration in the rnyofiber size. Student’s t test was used for statistical analysis between groups and significance was P < 0.05.
regenerate and myofiber size were much reduced (Figs. 1and 2, DM-DH) as compared with normal muscle regenerate (Figs. 1 and 2, NM-NH). One week transplants exhibited three morphologically distinguishable zones (Fig. 3D): a n outer zone of original surviving myofibers, a middle myogenic zone composed of proliferated myoblasts and small myotubes, and a n inner zone of degenerated myofibers. By 2 weeks the entire muscle was filled with small regenerated myotubes surrounded by a zone of small original myofibers (Fig. 3E). Myotubes were identifiable because of their round appearance and prominent central nuclei (Gulati, 1986, 1988).The myotubes matured into polygonal myofibers by 12 weeks, but the overall myofiber size remained small. Central and/or peripherally located nuclei were seen in these long-term diabetic muscle regenerates (Fig. 3F). The presence of nerves and motor endplates was also seen throughout the muscle regenerate (Fig. 3G).
Regeneration and recovery of muscle weight and myofiber size were observed in diabetic muscle after transplantation into normal host (Figs. 1 and 2, DM-NH). This increase was significant a s compared to diabetic muscle regenerates in diabetic host (Figs. 1and 2, DMDH). In fact, the recovery was similar to that observed in normal muscle transplants into normal hosts (Figs. 1 and 2, compare DM-NH and NM-NH). Morphologically, the regeneration pattern was similar to that described for transplants in diabetic hosts at different time intervals. Three zones were visible in 1 week transplants, with the formation of small myotubes in the myogenic zone (Fig. 5A). Maturation of myotubes followed, and by 12 weeks, a normal appearing muscle regenerate with small nerves was seen (Fig. 5B). The regenerated myofibers were larger than regenerated myofibers in diabetic hosts (compare DM-NH and DMDM, Fig. 2). Small nerves (Fig. 5B) and motor endplates were present in these long-term grafts (Fig. 5C). The results show that the changes in skeletal muscle that accompany diabetes are reversed when diabetic muscle is transplanted into normal host. Regeneration of Normal Muscle in Normal Host (NM-NH)
Regeneration after transplantation of normal muscle has been described previously (Carlson, 1978; Gulati et al., 1982), and was included here as a n additional control for comparison with the other experimental groups. As described earlier, there was a significant reduction in muscle weight and myofiber size after transplantation a s compared to the original weight of normal muscle (Figs. 1 and 2, NM and NM-NH). One week regenerates exhibited a small myogenic zone consisting of activated myoblasts and myotubes between the outer surviving myofibers and the inner degenerating myofibers (Fig. 5D). The entire muscle was filled with polygonal myofibers by 12 weeks, and these muscles were clearly innervated (Fig. 5E and F). The myofiber size of regenerates in this group matched the myofiber size of diabetic muscle regenerates in normal hosts (Fig. 2, compare DM-NH and NM-NH), and was considerably larger than the diabetic muscle regenerates in diabetic hosts (Fig. 2, DM-DH). DISCUSSION The present study describes morphological changes in diabetic muscle and its regeneration pattern after
Regeneration of Normal Muscle in Diabetic Host (NM-DH)
Regeneration pattern of normal muscle transplanted Fig. 3. Cross-section of normal (A) and 6 week diabetic (B and C ) into diabetic host was similar to that observed for dia- EDL muscles. Normal muscle consists of polygonal myofibers of variable size and staining intensity with a thin surrounding endomysium betic muscle transplanted into diabetic host described earlier (Fig. 3). The muscle regenerate weight and my- (A). The myofibers in 6 week diabetic muscle are considerably smaller than normal myofibers (compare A and B). Small nerves (arrows in B ofiber size at the 12 week stage in NM-DH group was and C )and motor endplates (long arrows in C) are seen in the diabetic also similar to t h a t seen in the DM-DH group (Figs. 1 muscle. The pattern of diabetic EDL muscle regeneration in the diaand 2). In 2 week regenerates the presence of small betic host is shown in D, E, and F. In l week diabetic muscle transregenerated myotubes was clearly seen (Fig. 4A). In plants (D), three zones are identifiable. These are, outer zone of surmyofibers (s), an intermediate myogenic zone (m), and an inner long-term regenerates (i.e., 12 weeks) the regenerated viving zone of degenerating myofibers (d). In 2 week transplants (E), small myofibers remained small in spite of establishment of round myotubes (m)with central nuclei are seen surrounded by a zone innervation (Fig. 4B and C). Morphological and mor- of original surviving myofibers (s). Groups of small regenerated myphometric data of muscle regenerates in this group ofibers are seen in 12 week diabetic transplants (F). Perimysial contissue (arrows in F) separates the groups of myofibers. Small showed a significant compromise in regenerative re- nective nerves and motor endplates (arrows in G ) are also seen in 12 week covery of normal muscle after transplantation into a diabetic transplants. (A,B,D-F PAShematoxylin stain; C,G: cholinesterase-silver stain; A-E, x 100; G, x 111.) diabetic host.
MUSCLE REGENERATION IN DIABETES
Fig. 3.
301
302
A.K. GULATI AND M.S. SWAMY
Fig. 4. The pattern of normal EDL muscle regeneration in diabetic hosts is shown. In 2 week transplants (A), many small round myotubes (m) surrounded by a zone of original surviving myofibers (s)are seen. The regenerated myotubesimyofibers remain small in 12 week transplants (B) and the outer rim of surviving myofibers also decrease
in size. The presence of nerves (long arrow in C) and motor endplates (short arrows in C) is seen in these 12-week regenerates. In B and C, n represents the normal muscle that is used to prepare blocks of the small regenerates in order to facilitate sectioning. (A,B: PAS-hematoxylin stain; C: cholinesterase-silver stain; A-C, x 100.)
transplantation. The reduction of muscle mass and of myofiber size in the diabetic state has been reported earlier. This is related to increased proteolysis and decreased protein synthesis (Flaim et al., 1980; Chen and Ianuzzo, 1982). Denervation and degeneration of myofibers and myoneural junctions have also been described in skeletal muscle in the diabetic state (Feczko and Klueber, 1988; Klueber et al., 1989). Features of denervation atrophy were also observed in muscles from 18 week diabetic rats during the course of this study (Gulati, unpublished observation). A reduction in muscle mass and myofiber size is known to occur after denervation of the muscle (Gulati, 1988, 1989; Carlson and Faulkner, 1988). Extensive fibrosis that occurs with denervation atrophy was not seen in the diabetic muscles. Regeneration of transplanted muscles occurs rapidly with formation of polygonal myofibers, a feature of mature muscle regenerates. The presence of many neuromuscular junctions meant that diabetic muscle was capable of replicating the different phases of the regeneration process as described for normal muscle regeneration. This implies that lack of recovery in regenerated muscle is not due to establishment of innervation or other abnormalities in the regeneration process, but must be related to the host metabolic environment. Changes in the vascular component of skeletal muscle during diabetes as shown by Vracko (1970) can also contribute to impaired recovery. Recovery of diabetic muscle after transplantation into normal host provides additional evidence that the muscle itself is not permanently affected by the diabetic state, and that the outcome of regeneration is primarily determined by the host environment. A reversal of diseased human muscle state to a more normal state has been described after transplanting such muscles into the normal environment of nude mice (Gulati et al., 1988). Similarly long-term denervated muscles
have been shown to regenerate and recover after transplantation into a normal host environment (Carlson and Faulkner, 1988; Gulati, 1989). Taken together, these various transplantation studies demonstrate that skeletal muscle preserves its ability to regenerate and can recover from a variety of experimental and metabolic insults. However, transplantation to a normal environment is crucial for this muscle recovery. A number of studies have reported neuropathic changes and reduced rate of nerve regeneration in diabetes (Schmidt and Scharp, 1982; Powell et al., 1986; Longo et al., 1986). The transplantation procedure used in the current study invariably results in injury to small nerves that innervate the transplanted muscle (Carlson, 1978). Such injured nerves seem to undergo rapid regeneration and form motor endplates on the regenerated myofibers even in diabetic hosts. Whether the nerve regeneration or the formation of motor endplates was delayed in regenerates of diabetic hosts as compared to normal hosts could not be determined by the present study. It appears, however, that these processes were not severely affected. In conclusion, the results provide morphological evidence of muscle regeneration in the diabetic state. The reduction in muscle mass and myofiber size in the diabetic state is not due to lack of reinnervation, but is probably related to metabolic and/or vascular alterations that are known to occur with diabetics. The effect of long-term diabetes on muscle regeneration and recovery remains to be established. ACKNOWLEDGMENTS
The authors thank Dr. Thomas H. Rosenquist for helpful comments, Ms. Paula Wade and Ms. Young Benyaghoub for technical assistance, and Ms. Carla Motes for preparation of the manuscript.
MUSCLE REGENERATION IN DIABETES
Fig. 5A, B, D-F (Fig. 5C and legend on overleaf)
303
304
A.K. GULATI AND M.S.SWAMY
Fig. 5. Diabetic EDL muscle transplants in normal hosts (A-C) and normal EDL muscle transplants in normal hosts (D-F) are shown. The regeneration pattern and recovery in these two types of transplants is similar. In 1 week transplants (A and D), three zones are again identifiable. An outer zone of original surviving myofibers (s), an intermediate myogenic zone (m), and a n inner zone of degen-
erating myofibers (d). Regenerated myofibers of large size are observed in 12 week transplants (B and E, also compare with F). Small nerves (arrows in B, C, and F) and motor endplates (long arrows in C and F) are also present. (A, B, D, E: PAShematoxylin stain; C,E: cholinesterase-silver stain; A-F, X 100.)
LITERATURE CITED
Jefferson, L.S., D.E. Rannels, B.L. Munger, and H.E. Morgan (1974) Insulin in the regulation of protein turnover in heart and skeletal muscle. Fed. Proc., 33t1098-1104. Klueber, K.M., J.D. Feczko, G. Schmidt, and J.B. Watkins, 111 (1989) Skeletal Muscle in the diabetic mouse: histochemical and morphometric analysis. Anat. Rec., 225t41-45. Longo, F.M., H.C. Powell, J. LeBeau, M.R. Gerrero, H. Heckman, and R.R. Myers (1986) Delayed nerve regeneration in streptozotocin diabetic rats. Muscle Nerve, 9t385-393. Manchester, K.L. (1970) The control by insulin of amino acid accumulation in muscle. Biochem. J., 117t457-465. Paulus, S.F., and J. Grossie (1983) Skeletal muscle in alloxan diabetes. A comparison of isometric contractions in fast and slow muscle. Diabetes, 32r1035-1039. Perry, R.E., M.S. Swamy, and E.C. Abraham (1987) Progressive changes in lens crystallin glycation and high molecular weight aggregate formation leading to cataract development in streptozotocin-diabetic rats. Exp. Eye Res. 44t269-282. Powell, H.C., F.M. Longo, J.M. LeBeau, and R.R. Myers (1986) Abnormal nerve regeneration in galactose neuropathy. J . Neuropathol. Exp. Neurol., 45t151-160. Raabo, E., and T.C. Terkildsen (1960) In the enzymatic determination of blood glucose. Scand. J. Clin. Lab Invest. 12t402-407. Randle, P.J. (1966) Carbohydrate metabolism and lipid storage and breakdown in diabetes. Diabetologia, 2,237-247. Schmidt, R.E., and D.W. Scharp (1982) Axonal dystrophy in experimental diabetic autonomic neuropathy. Diabetes, 31t761-770. Snow, M.H. (1977) Myogenic cell formation in regenerating rat skeletal muscle injured by mincing. I. A fine structural study. Anat. Rec., 188t181-200. Toop, J. (1976) A rapid method for demonstrating skeletal muscle motor innervation in frozen sections. Stain Technol., 51tl-6. Vracko, R. (1970) Skeletal muscle capillaries in diabetics, a quantitative analysis. Circulation, 41 t271-283. Wool, E.G., W.S. Stirewalt, K. Kurihara, R.B. Low, P. Bailey, and D. Oyer (1968) 11. Hormones and metabolic function-mode of action of insulin in the regulation of protein biosynthesis in muscle. Recent Progr. Hormone Res., 24t139-208.
Armstrong, R.B., P.D. Gollnick, and C.D. Ianuzzo (1975) Histochemical properties of skeletal muscle fibers in streptozotocin-diabetic rats. Cell Tissue Res., 162r387-394. Carlson, B.M. (1978) A review of muscle transplantation in mammals. Physiol. Bohemoslov., 27t387-400. Carlson, B.M., and E. Gutmann (1975) Regeneration in free grafts of normal and denervated muscles in the rat: Morphology and histochemistry. Anat. Rec., 183r47-62. Carlson, B.M., and J.A. Faulkner (1988) Reinnervation of long-term denervated rat muscle freely grafted into a n innervated limb. Exp. Neurol., 102t50-56. Chen, V., and C.D. Ianuzzo (1982) Metabolic alterations in skeletal muscle of chronically streptozotocin-diabetic rats. Arch. Biochem. Biophys., 21 7t131-138. Feczko, J.D., and K.M. Klueber (1988) Cytoarchitecture of muscle in a genetic model of murine diabetes. Am. J . Anat. 182t224-240. Flaim, K.E., M.E. Copenhaver, and L.S. Jefferson (1980) Effects of diabetes on protein synthesis in fast- and slow-twitch rat skeletal muscle. Am. J . Physiol., 239tE88-95. Gulati, A.K. (1986) Pattern of skeletal muscle regeneration after reautotransplantation of regenerated muscle. J. Embryol. Exp. Morphol., 92rl-10. Gulati, A.K. (1988) Long-term retention of regenerative capability after denervation of skeletal muscle, and dependency of late differentiation on innervation. Anat. Rec., 220t429-434. Gulati, A.K. (1989) Regeneration, reinnervation and recovery of denervated skeletal muscle after transplantation. Anat. Rec., 223; 46A-47A. Gulati, A.K., A.H. Reddi, and A.A. Zalewski (1982) Distribution of fibronectin in normal and regenerating skeletal muscle. Anat. Rec., 204t175-183. Gulati, A.K., M.H. Rivner, M. Shamsnia, T.R. Swift, and G.S. Sohal (1988) Growth of skeletal muscle from patients with amyotrophic lateral sclerosis transplanted into nude mice. Muscle Nerve., 11: 33-38. Hansen-Smith, F.M., and B.M. Carlson (1979) Cellular response to free grafting of the extensor digitorum longus muscle of the rat. J. Neurol. Sci., 41t149-173.
Regeneration of Skeletal Muscle in Streptozotocin-Induced Diabetic Rats ADARSH K. GULATI AND M.S. SWAMY Department of Anatomy (A.K.G.) and Department of Cell and Molecular Biology (M.S.S.), Medical College of Georgia, Augusta, Georgia
ABSTRACT The present study analyzes the regeneration of skeletal muscle in diabetic rats. Intravenous injection of streptozotocin (STZ) was used to induce diabetes. Six weeks later the extensor digitorum longus (EDL) muscles from diabetic rats were either transplanted into diabetic or normal hosts to initiate regeneration. Normal EDL muscle transplants in normal and diabetic hosts were also performed for comparison. One, 2,4, and 12 weeks after transplantation, the EDL regenerates were morphologically analyzed. Regeneration and formation of neuromuscular junctions were observed in all transplants, including diabetic regenerates in diabetic hosts. The overall mass and myofiber size of the diabetic EDL regenerate in the diabetic host was significantly reduced in spite of complete regeneration. Recovery of the diabetic muscle mass and the myofiber size was observed after transplantation into normal hosts. A reduction in mass and myofiber size was observed in normal EDL muscles transplanted into diabetic hosts. It is concluded that poor recovery of diabetic muscle is related to metabolic and structural alterations in the diabetic host, rather than to innate capacity of the muscle to per se undergo regeneration and reinnervation. The observed enhancement in recovery of diabetic muscle after transplantation in a normal host and deterioration of normal muscle after transplantation in a diabetic host shows that the host environment determines the success of muscle regeneration. A diabetic state in rats can be induced by the administration of streptozotocin (STZ), a n agent that causes destruction of insulin producing (3 cells in the pancreas. The resultant diabetic state is known to have profound effects on skeletal muscle morphology, physiology and metabolism (Armstrong et al., 1975; Chen and Ianuzzo, 1982; Paulus and Grossie, 1983; Feczko and Klueber, 1988; Klueber et al., 1989). Protein synthesis and metabolism in muscle, along with enzymatic activity and mitochondria1 function, are decreased in the diabetic condition (Wool et al., 1968; Manchester, 1970; Jefferson et al., 1974; Paulus and Grossie, 1983). Muscle mass is reduced due to elevated proteolysis, and associated increase in oxidation of free fatty acids and ketone bodies (Randle, 1966). Signs of degeneration and denervation have also been reported in skeletal muscle of murine diabetes (Feczko and Klueber, 1988). Thickening of capillary basal lamina within the skeletal muscle has been described in diabetics (Vracko, 1970). The objective of this study was to analyze the regenerative ability and recovery of skeletal muscle in a n experimentally induced diabetic state. An experimental model used extensively in muscle regeneration studies involves transplantation of extensor digitorum longus (EDL) muscle in rats (Carlson and Gutmann, 1975; Hansen-Smith and Carlson, 1979; Gulati, 1986, 1988). A predictable sequence of events follows transplantation, with a number of factors regulating the regenerative success. A majority of myofibers except a thin rim of peripheral myofibers undergo intrinsic degeneration a s a result of devascularization 0 1991 WILEY-LISS, INC
caused by the transplantation procedure. With the restoration of blood supply, the resident precursor myosatellite cells become active, proliferate, and differentiate into myoblasts. The myoblasts fuse to form myotubes that gradually increase in size and mature into myofibers. The regenerated muscle becomes innervated, recovers and becomes functional (Snow, 1977; Carlson, 1978; Gulati et al., 1982). In the present study, the EDL muscle transplantation model was used to examine muscle regeneration under diabetic and nondiabetic control conditions. The results demonstrate impaired recovery of muscle regenerates in diabetic rats as compared to regenerates in normal rats. MATERIALS AND METHODS
Inbred male Fischer rats weighing 150-200 g were used in this study to exclude the possibility of immune rejection of muscle transplants by their hosts. Rats were fed Purina rodent chow ad libitum and housed individually in wire cages. Rats were made diabetic by a single intravenous injection in the tail of a freshly prepared solution of STZ (65 mg/kg, in 0.1 M citrate buffer, pH 4.5) under ether anesthesia (Perry et al., 1987). To screen for diabetes plasma glucose levels were determined by the glucose oxidase assay of Raabo
Received August 8, 1989; accepted July 30, 1990. Address reprint requests to Dr. A.K. Gulati, Department of Anatomy, Medical College of Georgia, Augusta, GA 30912-2000.
299
MUSCLE REGENERATION IN DIABETES
and Terkildsen (1960) provided as a kit from Sigma Chemical Company (St. Louis, MO). Streptozotocintreated rats developed a markedly elevated plasma glucose levels (510-791 mgidl) that were measured a t 2 weeks after injection and a t the time of muscle regenerate analysis. A few rats that did not exhibit increase in plasma glucose level after STZ injection were excluded from the study. Six weeks after STZ injection the rats were divided into the following groups for muscle transplantation surgery. In a group of eight diabetic animals, the EDL muscle was autotransplanted (i.e., transplanted at the same site in the same diabetic animal) in order to study the regeneration pattern of diabetic muscle in the diabetic host. The EDL muscles from the second group of eight diabetic animals were isotransplanted (i.e., transplanted into another genetically identical Fischer rat) into the left side of nondiabetic normal rats in order to study regeneration of diabetic muscle in normal animals. The right EDL muscle of these normal animals was also autotransplanted as normal muscle regeneration controls. In addition a group of eight diabetic animals were bilaterally transplanted with normal EDL muscle. This surgical protocol resulted in four transplantation groups for comparison: diabetic muscle into diabetic hosts (DM-DH), diabetic muscle into normal host (DMNH), normal muscle into normal host (NM-NH), and normal muscle into diabetic host (NM-DH). For each of these groups 16 muscles were prepared with 4 muscles being analyzed at each of the time intervals. The transplantation procedure of the EDL muscle was similar to that described in detail earlier (Gulati, 1986, 1988). After chloral hydrate anesthesia (40 mgl 100 g body weight, i.p.) the EDL muscle was exposed by cutting the overlying skin and muscle. The EDL muscle was then removed from its bed by cutting the proximal and distal tendons. The muscle weight of each diabetic and normal control muscle was recorded for comparison with weights after regeneration. For transplantation into a diabetic host, the diabetic muscle was returned to its original bed and the cut tendons sutured back under slight tension. For transplantation of diabetic muscle into a normal host, the normal host EDL muscle was removed and replaced by the diabetic muscle. Similar procedure was followed for normal muscle autotransplants and for normal muscle transplants into diabetic hosts. All animals were allowed to recover and maintained until the time of muscle analysis. Four muscle transplants from each group were removed for morphological analysis a t intervals of 1 , 2 , 4 , and 12 weeks after transplantation. Under anesthesia the EDL transplants were removed, weighed, and frozen in liquid nitrogen. Cross sections, 8 and 24 pm thick were cut in a cryostat set at -20°C. Sections cut from different regions (i.e., throughout the length of the muscle) were then mounted on glass slides. The 8 pm sections were stained with periodic acid-Schiff hematoxylin (PASLhematoxylin) and used for general morphological and morphometric analysis. The 24 pm thick sections were stained with cholinesterase-silver (Toop, 1976) for determining the innervation status. Some muscles which had never been transplanted were also examined to define the morphological features of normal and diabetic muscles and myofiber size. For morphometric analysis, a Zeiss axiophot light micro-
lO0r
-
80 60
k-
-
I
U
W 3
40
20
n-
NM
NH-NH
NM-DH
DH
OM-DH
OM-NH
MUSCLE TYPES Fig. 1. A comparison of relative muscle weights in mg (t SEM) of normal (NM), 6 week diabetic (DM), and the 12 week EDL muscle regenerates from the various transplantation groups. A significant reduction in muscle weights was seen in 6 week diabetic EDL (DM) as compared to normal EDL (NM). The normal EDL muscle regenerates in normal host (NM-NH)and in diabetic host (NM-DH)are shown and are significantly different. The diabetic EDL muscle regenerates in diabetic (DM-DH)and normal hosts (DM-NH)are shown and are also significantly different. Student’s t test was used for statistical analysis between groups and significance was P < 0.05.
scope attached to a color camera and a n IBM computer was employed. Computer assisted quantification was accomplished by using the CoreSCAN Program (Extension Systems, Sebastopol, CAI. Four muscles from each experimental group were used and area of 50 myofibers from each was determined. Individual myofibers from different regions of the muscle were outlined, and the number of pixels within the outlined area was determined. The number of pixils was converted mathematically into area (pm’). Statistical analysis was done to compare, before and after regeneration, mean myofiber area using Student’s t test, with a probability less than 0.05 indicating significance. RESULTS Comparison of Normal (NM) and Diabetic (OM) Muscles
The muscle weights of normal and 6 week diabetic nontransplanted EDL muscles are shown in Figure 1. A significant reduction in the mass of diabetic muscle a s compared to normal muscle was observed (Fig. 1, NM and DM). Corresponding reduction in myofiber size was also observed between normal and diabetic muscle (Fig. 2, NM and DM). Normal rat muscle possessed polygonal myofibers of variable diameter and staining intensity. Individual myofibers had a thin pericellular endomysium, with groups of myofibers surrounded by a thicker perimysium (Fig. 3A). The myofibers in diabetic muscle were smaller in size, but like the normal muscle were polygonal and possessed peripheral nuclei. In the diabetic muscles, many small nerves and motor endplates were seen demonstrating innervation (Fig. 3B and C). Regeneration of Diabetic Muscle in Diabetic Host (DM-DH)
After transplantation of diabetic EDL into a diabetic host, the various events typical of regeneration were observed. However, the overall weight of the muscle
300
A.K. GULATI AND M.S. SWAMY
Regeneration of Diabetic Muscle in Normal Host (DM-NH)
3500~
Ix W m
1500
L0l 1000 > r
500 0 NU
OM
NU-NH
OR-NH
NU-OH
OM-OH
MUSCLE TYPES Fig. 2. A comparison of myofiber size in km2 ( i S E M ) normal (NM), 6 week diabetic (DM),and the 12 week EDL muscle regenerates from the various transplantation groups. A significant reduction in myofiber size is seen between NM and DM groups. The myofiber size of normal and diabetic EDL muscles transplanted into normal hosts (NM-NH and DM-NH, respectively) shows a significant recovery in the size of diabetic muscle after transplantation into normal host. On the other hand the myofiber sizes of normal and diabetic EDL into diabetic hosts (NM-DH and DM-DH, respectively) show a significant deterioration in the rnyofiber size. Student’s t test was used for statistical analysis between groups and significance was P < 0.05.
regenerate and myofiber size were much reduced (Figs. 1and 2, DM-DH) as compared with normal muscle regenerate (Figs. 1 and 2, NM-NH). One week transplants exhibited three morphologically distinguishable zones (Fig. 3D): a n outer zone of original surviving myofibers, a middle myogenic zone composed of proliferated myoblasts and small myotubes, and a n inner zone of degenerated myofibers. By 2 weeks the entire muscle was filled with small regenerated myotubes surrounded by a zone of small original myofibers (Fig. 3E). Myotubes were identifiable because of their round appearance and prominent central nuclei (Gulati, 1986, 1988).The myotubes matured into polygonal myofibers by 12 weeks, but the overall myofiber size remained small. Central and/or peripherally located nuclei were seen in these long-term diabetic muscle regenerates (Fig. 3F). The presence of nerves and motor endplates was also seen throughout the muscle regenerate (Fig. 3G).
Regeneration and recovery of muscle weight and myofiber size were observed in diabetic muscle after transplantation into normal host (Figs. 1 and 2, DM-NH). This increase was significant a s compared to diabetic muscle regenerates in diabetic host (Figs. 1and 2, DMDH). In fact, the recovery was similar to that observed in normal muscle transplants into normal hosts (Figs. 1 and 2, compare DM-NH and NM-NH). Morphologically, the regeneration pattern was similar to that described for transplants in diabetic hosts at different time intervals. Three zones were visible in 1 week transplants, with the formation of small myotubes in the myogenic zone (Fig. 5A). Maturation of myotubes followed, and by 12 weeks, a normal appearing muscle regenerate with small nerves was seen (Fig. 5B). The regenerated myofibers were larger than regenerated myofibers in diabetic hosts (compare DM-NH and DMDM, Fig. 2). Small nerves (Fig. 5B) and motor endplates were present in these long-term grafts (Fig. 5C). The results show that the changes in skeletal muscle that accompany diabetes are reversed when diabetic muscle is transplanted into normal host. Regeneration of Normal Muscle in Normal Host (NM-NH)
Regeneration after transplantation of normal muscle has been described previously (Carlson, 1978; Gulati et al., 1982), and was included here as a n additional control for comparison with the other experimental groups. As described earlier, there was a significant reduction in muscle weight and myofiber size after transplantation a s compared to the original weight of normal muscle (Figs. 1 and 2, NM and NM-NH). One week regenerates exhibited a small myogenic zone consisting of activated myoblasts and myotubes between the outer surviving myofibers and the inner degenerating myofibers (Fig. 5D). The entire muscle was filled with polygonal myofibers by 12 weeks, and these muscles were clearly innervated (Fig. 5E and F). The myofiber size of regenerates in this group matched the myofiber size of diabetic muscle regenerates in normal hosts (Fig. 2, compare DM-NH and NM-NH), and was considerably larger than the diabetic muscle regenerates in diabetic hosts (Fig. 2, DM-DH). DISCUSSION The present study describes morphological changes in diabetic muscle and its regeneration pattern after
Regeneration of Normal Muscle in Diabetic Host (NM-DH)
Regeneration pattern of normal muscle transplanted Fig. 3. Cross-section of normal (A) and 6 week diabetic (B and C ) into diabetic host was similar to that observed for dia- EDL muscles. Normal muscle consists of polygonal myofibers of variable size and staining intensity with a thin surrounding endomysium betic muscle transplanted into diabetic host described earlier (Fig. 3). The muscle regenerate weight and my- (A). The myofibers in 6 week diabetic muscle are considerably smaller than normal myofibers (compare A and B). Small nerves (arrows in B ofiber size at the 12 week stage in NM-DH group was and C )and motor endplates (long arrows in C) are seen in the diabetic also similar to t h a t seen in the DM-DH group (Figs. 1 muscle. The pattern of diabetic EDL muscle regeneration in the diaand 2). In 2 week regenerates the presence of small betic host is shown in D, E, and F. In l week diabetic muscle transregenerated myotubes was clearly seen (Fig. 4A). In plants (D), three zones are identifiable. These are, outer zone of surmyofibers (s), an intermediate myogenic zone (m), and an inner long-term regenerates (i.e., 12 weeks) the regenerated viving zone of degenerating myofibers (d). In 2 week transplants (E), small myofibers remained small in spite of establishment of round myotubes (m)with central nuclei are seen surrounded by a zone innervation (Fig. 4B and C). Morphological and mor- of original surviving myofibers (s). Groups of small regenerated myphometric data of muscle regenerates in this group ofibers are seen in 12 week diabetic transplants (F). Perimysial contissue (arrows in F) separates the groups of myofibers. Small showed a significant compromise in regenerative re- nective nerves and motor endplates (arrows in G ) are also seen in 12 week covery of normal muscle after transplantation into a diabetic transplants. (A,B,D-F PAShematoxylin stain; C,G: cholinesterase-silver stain; A-E, x 100; G, x 111.) diabetic host.
MUSCLE REGENERATION IN DIABETES
Fig. 3.
301
302
A.K. GULATI AND M.S. SWAMY
Fig. 4. The pattern of normal EDL muscle regeneration in diabetic hosts is shown. In 2 week transplants (A), many small round myotubes (m) surrounded by a zone of original surviving myofibers (s)are seen. The regenerated myotubesimyofibers remain small in 12 week transplants (B) and the outer rim of surviving myofibers also decrease
in size. The presence of nerves (long arrow in C) and motor endplates (short arrows in C) is seen in these 12-week regenerates. In B and C, n represents the normal muscle that is used to prepare blocks of the small regenerates in order to facilitate sectioning. (A,B: PAS-hematoxylin stain; C: cholinesterase-silver stain; A-C, x 100.)
transplantation. The reduction of muscle mass and of myofiber size in the diabetic state has been reported earlier. This is related to increased proteolysis and decreased protein synthesis (Flaim et al., 1980; Chen and Ianuzzo, 1982). Denervation and degeneration of myofibers and myoneural junctions have also been described in skeletal muscle in the diabetic state (Feczko and Klueber, 1988; Klueber et al., 1989). Features of denervation atrophy were also observed in muscles from 18 week diabetic rats during the course of this study (Gulati, unpublished observation). A reduction in muscle mass and myofiber size is known to occur after denervation of the muscle (Gulati, 1988, 1989; Carlson and Faulkner, 1988). Extensive fibrosis that occurs with denervation atrophy was not seen in the diabetic muscles. Regeneration of transplanted muscles occurs rapidly with formation of polygonal myofibers, a feature of mature muscle regenerates. The presence of many neuromuscular junctions meant that diabetic muscle was capable of replicating the different phases of the regeneration process as described for normal muscle regeneration. This implies that lack of recovery in regenerated muscle is not due to establishment of innervation or other abnormalities in the regeneration process, but must be related to the host metabolic environment. Changes in the vascular component of skeletal muscle during diabetes as shown by Vracko (1970) can also contribute to impaired recovery. Recovery of diabetic muscle after transplantation into normal host provides additional evidence that the muscle itself is not permanently affected by the diabetic state, and that the outcome of regeneration is primarily determined by the host environment. A reversal of diseased human muscle state to a more normal state has been described after transplanting such muscles into the normal environment of nude mice (Gulati et al., 1988). Similarly long-term denervated muscles
have been shown to regenerate and recover after transplantation into a normal host environment (Carlson and Faulkner, 1988; Gulati, 1989). Taken together, these various transplantation studies demonstrate that skeletal muscle preserves its ability to regenerate and can recover from a variety of experimental and metabolic insults. However, transplantation to a normal environment is crucial for this muscle recovery. A number of studies have reported neuropathic changes and reduced rate of nerve regeneration in diabetes (Schmidt and Scharp, 1982; Powell et al., 1986; Longo et al., 1986). The transplantation procedure used in the current study invariably results in injury to small nerves that innervate the transplanted muscle (Carlson, 1978). Such injured nerves seem to undergo rapid regeneration and form motor endplates on the regenerated myofibers even in diabetic hosts. Whether the nerve regeneration or the formation of motor endplates was delayed in regenerates of diabetic hosts as compared to normal hosts could not be determined by the present study. It appears, however, that these processes were not severely affected. In conclusion, the results provide morphological evidence of muscle regeneration in the diabetic state. The reduction in muscle mass and myofiber size in the diabetic state is not due to lack of reinnervation, but is probably related to metabolic and/or vascular alterations that are known to occur with diabetics. The effect of long-term diabetes on muscle regeneration and recovery remains to be established. ACKNOWLEDGMENTS
The authors thank Dr. Thomas H. Rosenquist for helpful comments, Ms. Paula Wade and Ms. Young Benyaghoub for technical assistance, and Ms. Carla Motes for preparation of the manuscript.
MUSCLE REGENERATION IN DIABETES
Fig. 5A, B, D-F (Fig. 5C and legend on overleaf)
303
304
A.K. GULATI AND M.S.SWAMY
Fig. 5. Diabetic EDL muscle transplants in normal hosts (A-C) and normal EDL muscle transplants in normal hosts (D-F) are shown. The regeneration pattern and recovery in these two types of transplants is similar. In 1 week transplants (A and D), three zones are again identifiable. An outer zone of original surviving myofibers (s), an intermediate myogenic zone (m), and a n inner zone of degen-
erating myofibers (d). Regenerated myofibers of large size are observed in 12 week transplants (B and E, also compare with F). Small nerves (arrows in B, C, and F) and motor endplates (long arrows in C and F) are also present. (A, B, D, E: PAShematoxylin stain; C,E: cholinesterase-silver stain; A-F, X 100.)
LITERATURE CITED
Jefferson, L.S., D.E. Rannels, B.L. Munger, and H.E. Morgan (1974) Insulin in the regulation of protein turnover in heart and skeletal muscle. Fed. Proc., 33t1098-1104. Klueber, K.M., J.D. Feczko, G. Schmidt, and J.B. Watkins, 111 (1989) Skeletal Muscle in the diabetic mouse: histochemical and morphometric analysis. Anat. Rec., 225t41-45. Longo, F.M., H.C. Powell, J. LeBeau, M.R. Gerrero, H. Heckman, and R.R. Myers (1986) Delayed nerve regeneration in streptozotocin diabetic rats. Muscle Nerve, 9t385-393. Manchester, K.L. (1970) The control by insulin of amino acid accumulation in muscle. Biochem. J., 117t457-465. Paulus, S.F., and J. Grossie (1983) Skeletal muscle in alloxan diabetes. A comparison of isometric contractions in fast and slow muscle. Diabetes, 32r1035-1039. Perry, R.E., M.S. Swamy, and E.C. Abraham (1987) Progressive changes in lens crystallin glycation and high molecular weight aggregate formation leading to cataract development in streptozotocin-diabetic rats. Exp. Eye Res. 44t269-282. Powell, H.C., F.M. Longo, J.M. LeBeau, and R.R. Myers (1986) Abnormal nerve regeneration in galactose neuropathy. J . Neuropathol. Exp. Neurol., 45t151-160. Raabo, E., and T.C. Terkildsen (1960) In the enzymatic determination of blood glucose. Scand. J. Clin. Lab Invest. 12t402-407. Randle, P.J. (1966) Carbohydrate metabolism and lipid storage and breakdown in diabetes. Diabetologia, 2,237-247. Schmidt, R.E., and D.W. Scharp (1982) Axonal dystrophy in experimental diabetic autonomic neuropathy. Diabetes, 31t761-770. Snow, M.H. (1977) Myogenic cell formation in regenerating rat skeletal muscle injured by mincing. I. A fine structural study. Anat. Rec., 188t181-200. Toop, J. (1976) A rapid method for demonstrating skeletal muscle motor innervation in frozen sections. Stain Technol., 51tl-6. Vracko, R. (1970) Skeletal muscle capillaries in diabetics, a quantitative analysis. Circulation, 41 t271-283. Wool, E.G., W.S. Stirewalt, K. Kurihara, R.B. Low, P. Bailey, and D. Oyer (1968) 11. Hormones and metabolic function-mode of action of insulin in the regulation of protein biosynthesis in muscle. Recent Progr. Hormone Res., 24t139-208.
Armstrong, R.B., P.D. Gollnick, and C.D. Ianuzzo (1975) Histochemical properties of skeletal muscle fibers in streptozotocin-diabetic rats. Cell Tissue Res., 162r387-394. Carlson, B.M. (1978) A review of muscle transplantation in mammals. Physiol. Bohemoslov., 27t387-400. Carlson, B.M., and E. Gutmann (1975) Regeneration in free grafts of normal and denervated muscles in the rat: Morphology and histochemistry. Anat. Rec., 183r47-62. Carlson, B.M., and J.A. Faulkner (1988) Reinnervation of long-term denervated rat muscle freely grafted into a n innervated limb. Exp. Neurol., 102t50-56. Chen, V., and C.D. Ianuzzo (1982) Metabolic alterations in skeletal muscle of chronically streptozotocin-diabetic rats. Arch. Biochem. Biophys., 21 7t131-138. Feczko, J.D., and K.M. Klueber (1988) Cytoarchitecture of muscle in a genetic model of murine diabetes. Am. J . Anat. 182t224-240. Flaim, K.E., M.E. Copenhaver, and L.S. Jefferson (1980) Effects of diabetes on protein synthesis in fast- and slow-twitch rat skeletal muscle. Am. J . Physiol., 239tE88-95. Gulati, A.K. (1986) Pattern of skeletal muscle regeneration after reautotransplantation of regenerated muscle. J. Embryol. Exp. Morphol., 92rl-10. Gulati, A.K. (1988) Long-term retention of regenerative capability after denervation of skeletal muscle, and dependency of late differentiation on innervation. Anat. Rec., 220t429-434. Gulati, A.K. (1989) Regeneration, reinnervation and recovery of denervated skeletal muscle after transplantation. Anat. Rec., 223; 46A-47A. Gulati, A.K., A.H. Reddi, and A.A. Zalewski (1982) Distribution of fibronectin in normal and regenerating skeletal muscle. Anat. Rec., 204t175-183. Gulati, A.K., M.H. Rivner, M. Shamsnia, T.R. Swift, and G.S. Sohal (1988) Growth of skeletal muscle from patients with amyotrophic lateral sclerosis transplanted into nude mice. Muscle Nerve., 11: 33-38. Hansen-Smith, F.M., and B.M. Carlson (1979) Cellular response to free grafting of the extensor digitorum longus muscle of the rat. J. Neurol. Sci., 41t149-173.
