Correlative Light Optical, Scanning Electron, and Transmission Electron Microscopy of Skeletal Muscle in Muscular Dystrophy and Muscular Atrophy: A Pilot Study H. D. Geissinger, PhD, R. A. Vriend, MSc, C. A. Ackerley, and S. Yamashiro, MSc Department of Biomedical Sciences, University of Guelph, Guelph, Ontario, N1 G 2 W l , Canada

Biopsies of skeletal muscle from three different cases of muscular dystrophy and one case of spinal muscular atrophy that had been fixed with Karnovsky's fluid were either routinely prepared for scanning electron microscopy (SEM) or were frozen to -2OOC and sectioned on a steel knife in a cryostat a t 5-10 prn. The sections were coversiipped and examined using a light microscope equipped with polarizing optics (Pol). After areas were selected, the sections were prepared for SEM and thereby examined. The tissues on the slides that had been observed with light microscopy (LM) and SEM were prepared further for transmission electron microscopy (TEM) by infiltrating them with Epon and cutting sections a t approximately 100 nm on an ultramicrotome. i t i s shown that the stage of contraction in one pathologic myofiber may vary along i t s length. The following advantages may be realized by using correlative (Pol + SEM -tTEM) microscopy on skeletal muscle biopsies: 1) lesions can be differentiated from "normal" surrounding tissue; 2) doubtful structures can be reexamined with the SEM and TEM; and 3) the SEM image of different states of muscle contraction can be reinterpreted in the light of the Poi or TEM image.

KEY WORDS: muscular dystrophy, spinal atrophy, m y o fiber, hypercontraction, scanning electron microscopy, correlative microscopy.

INTRODUCTION The morphological pathology of human muscular dystrophies has been well described a t the light microscopic (LM) and transmission electron microscopic (TEM) levels.'-3 Up to the present time there have been three scanning electron microscopic (SEM) studies

We thank the Ontario Ministry of Agriculture and Food and the Natural Sciences and Engineering Research Council of Canada for financial support. The technical help of Ms. D. Markusic and the secretarial help of Ms. C. A. Thomson and Mrs. M.McGregor is very much appreciated.

of m y ~ p a t h i e s . ~ " Since these SEM studies were carried out on specimens that had not been critical point-dried (CPD) but air-dried, the results of these investigations may be misleading. Furthermore, these studies did not attempt to verify the unfamiliar SEM image by means of correlative techniques on the same area of the lesion. There is only one preliminary correlative microscopic (LM .+ SEM -+ TEM) published study of human myopathies.' The main reason for the following investigations is to augment these pilot results. I t was found feasible and informative to examine mouse skeletal muscle by the use of combined LM, SEM, and TEM techniques.8 These techniques

Ultrastructural Patholoav. 1:327-335. 1980 Copyright 0 1980 by Hemisphere Pcblishing Corporation 0191-3123 80/0303 7- 9 3 25

327

H. D. Geissinger e t al.

328 have been i m p r o ~ e d ~so ' ' ~that it is now quite possible t o assess the same ultrastructural details in the SEM and TEM.

MATERIALS AND METHODS The biopsy material was supplied to us through the courtesy of Dr. L. E. Becker of the Department of Pathology of the Hospital for Sick Children in Toronto. The specimens had been fixed in Karnovsky's fluid" for several hours and then were kept a t 4°C in 0.2 M phosphate buffer (pH 7.4). Small portions (0.5 mm3) were dissected out, postfixed in Os04, and routinely processed for TEM, using Epon for the embedding media. The blocks were sectioned on an ultramicrotome a t 1 pm. These sections were stained with toluidine blue and examined with LM. Ultrathin sections were also cut from these blocks and contrasted with 2% uranyl acetate in 70% alcohol and Reynold's lead citrate and examined and photographed on a JEOL JEM 100s transmission electron microscope. Half of the remaining tissue was dehydrated in an ascending series of alcohols, critical pointdried in COz,'2 and dry fractured with a razor blade.I3 The specimens were then examined in a JEOL JSM-35 scanning electron microscope after having been sputter-coated with gold-palladium (Au-Pd).14 The other half of the tissue was frozen t o -2OOC and 20 sections were cut from each specimen a t 5-10 p m on a cryostat. Some of these were stained with hematoxylin-eosin; the rest were left unstained. The slides were mounted in glycerol and examined with a light microscope equipped with polarizing optics (Pol) and photographed a t various magnifications in order to find specific areas to be examined later under the scanning electron microscope. The coverslips were floated off from the sections in three changes (20 min each) of 10% alcohol in order t o remove all traces of glycerol. Subsequently the specimens were prepared for SEM examination, using a modified thiocarbohydrazide (OTOTO) procedure" or by postfixing in 0 ~ 0 4 ,dehydrating in a graded series of alcohol, critical point-drying, and sputtercoating with Au-Pd. The specific areas on the sections observed by LM were then examined

and photographed under the scanning electron microscope. Subsequent Epon embedding, aligning of the block face for cutting on an ultramicrotome, and TEM examination have been described in detail previously.8

RESULTS Limb -Girdle Muscular Dystrophy Although this biopsy was from a 14-year-old boy who had been clinically diagnosed as limbgirdle muscular dystrophy, there were no recognizable lesions as seen by LM, SEM, TEM, or Pol +SEM +TEM. Since we did not have access to a completely normal biopsy from a child, the following correlative microscopic features are only included t o point out one major use of correlation (Pol + SEM + TEM) microscopy, i.e., the reinterpretation of an LM or SEM image. Figure l a (Inset) shows an area where there was doubt if this was of the nature of connect i v e tissue or muscle. The SEM micrograph (Fig. l a ) shows the surface structure but sheds no further light on the correct interpretation, while the TEM micrograph (Fig. Ib) of precisely the same area shows the typical appearance of a muscle fiber.

Atypical Muscular Dystrophy The biopsy was from an 8-year-old boy who had exhibited clinical signs of leg weakness. Light microscopy in cross sections showed a marked variation in the fiber diameter and a moderate increase in connective tissue of the endomysium, while a slight increase in the connective tissue of the perimysium was noted in only 1 of 20 sections. The lesions were focal in nature. Mild hyalinization, necrosis, and vacuolization were noted in some myofibers. There was very slight evidence of fiber regeneration (i.e., thin basophilic myofibers and "rowing" of centrally placed nuclei). Perivascular edema was moderate in most sections, and mast cells in the vicinity of capillaries were seen. Scanning electron microscopy of standard (dry fracture) preparations showed most myofibers to be of normal appearance (Fig. 2); however vacuoles were present in some fibers (Fig. 3). These vacuoles were interpreted as

FIG. 1 Limb-girdle muscular dystrophy. (a) SEM (thiocarbohydrazide slide preparation) showing two myofibers, a capillary (0, and the questionable structure (XI. X 1500. (Inset) L M of same area, H&E, X400. (b) TEM of correlated area showing X to be a portion of a longitudinally cut myofiber. X5400.

FIG. 2 Atypical muscular dystrophy. SEM (dry fracture) of an apparently normal myo fiber, with some mitochondria (M) running transversely across the m yo fibril. X 7800.

FIG. 3 Atypical muscular dystkophy. SEM (dry fracture) showing possible remnants o f lipid droplets ( L) between m yo fibrils. X 10,000.

329

H. D. Geissinger et al.

330 spaces that were left after the loss of lipid droplets due to processing for SEM. Hypercontracted myofibers (Fig. 4) were clearly identified by SEM. Adjacent areas of the same myofiber (Fig. 5) showed such poor definition of the surface of myofibrils that individual sarcomeres could not be distinguished. It seems reasonable to assume that these myofibrils were necrotic. Transmission electron microscopy confirmed the presence of lipid droplets in the sarcoplasm of affected fibers. Focal degeneration of myofibrils, including several sarcomeres, was noted in some myofibers. Some myofibrils appeared narrowed with large intermyofibrillar spaces, which appeared to be filled with granular sarcoplasm. The slight signs of regeneration of some myofibers was confirmed, as was the presence of mast cells.

FIG. 5 Atypical muscular dystrophy. SEM (dry fracture) showing a loss in definition of individual sarcomeres of the myo fibrils. X6000.

dystrophy. SEM FIG. 4 Atypical muscular

(dry fracture) showing focal point of hypercontraction in a myofiber.

X4000.

Correlation (Pol + SEM + TEM) microscopy results were as follows. In Fig. 6a,a Pol micrograph of a lesion shows a myofiber (d to g) in various stages of contraction. Figure 6b i s a survey SEM of the same area. Higher magnification SEM micrographs show the regular length of sarcomeres of a normally contracted myofiber (Fig. 6c) and a hypercontracted area in a different fiber (Fig. 6d). The following TEM micrographs of the same fiber that had been examined by SEM shows the progression of changes. Thus it is shown in Fig. 6e that although the myofiber appeared healthy, the sarcoplasmic reticulum (SR) appeared slightly dilated, a feature that could not have been appreciated in a previous SEM micrograph of approximately the same area in the fiber (not shown). Further on in the fiber, sarcomere

FIG. 6 Atypical muscular dystrophy. (a) Pol map of lesion showing A- and I-bands of individual myofibers. Areas that had been photographed with the scanning and the transmission electron microscopes are indicated by the appropriate letters. X400. (b) Low-power SEM (thiocarbohydrazide slide preparation) of the same entire area as a. X480. (c) SEM showing portion of normally contracted myofiber. X 18,000. (d) SEM showing portion of h ypercontracted m yo fiber. Two hypercontraction bands (HB) are indicated. X22,OOO. (e) TEM showing mild dilation of sarcoplasmic reticulum (SR), well-definedm yofibrils, mitochondria, and lipid droplets ( L) in sarcoplasm. X 70,000. (f) TEM showing hypercontraction bands ( HB) in myo fibrils and degenerating mitochondria in sarcoplasm. X7500. (9) TEM showing absence of sarcomere banding. X7500.

Correlative Microscopy of Muscle Lesions

331

H. D. Geissinger et al.

332 length shortens and the corresponding TEM micrograph (Fig. 6fl shows a condensation of contractile elements in the I-band region. Along the length of the fiber in the hypercontracted area, definition of individual sarcomeres is almost nonexistent and damage to focal areas in the sarcoplasm are apparent (Fig. 6g).This damage is so pronounced that it is impossible to decide whether the mitochondria, sarcoplasmic reticulum, lipid droplets, all, or any two of these structures are affected.

.." -.-

--- -

Duchenne Muscular Dystrophy The biopsy was from a 6-year-oldboy who had the typical signs of Duchenne muscular dystrophy (DMD) since the age of 2; years. On LM, lesions tended to be circumscribed. A marked variation in myofiber diameter and a marked increase in connective tissue of the endomysium and the perimysium, which contained many fibroblasts, macrophages, and lymphocytes, were seen. Edema was noted in the connective tissue, especially around arterioles. Many of the myofibers showed varying degrees of degeneration, such as hyalinization and vacuolization. Some necrotic fibers were also observed. Signs of myofiber regenera-

FIG. 7 Duchenne muscular dystrophy. SEM (dry fracture) of hypercontracted m yo fibrils showing bulging h ypercontraction bands (HB 1. X 70,000.

FIG. 8 Duchenne muscular dystrophy. SEM of an area adjacent to Fig. 7 showing hypercontraction state of m yo fibrils to be even more severe than that shown in Fig. 7. X7800. tion were seen. Some of these myofibers contained vesicular nuclei. On standard SEM many fibers appeared normal; however there were a few hypercontracted fibers (Figs. 7 and 8) or fibers that were presumably necrotic because they showed l i t t l e sarcomere distinction. Transmission electron microscopy showed that some myofibers had duplicated basal laminas. Other myofibers were completely necrotic. Less severe lesions of Z-line streaming were noted. A clumping of triads, intramyofibrillar rods, filamentous bodies, and a granular sarcoplasm were noted occasionally. In the endomysium and perimysium, mast cells, mononuclear cells, and an increase in collagen were found. The capillary endothelium was often swollen. Correlation (Pol -+ SEM + TEM) microscopy shpwed the hypercontracted myofibers in a DMD lesion to be almost identical in appearance to those shown earlier in lesions of atypical muscular dystrophy.

Spinal Muscular Atrophy The biopsy was taken from a 3-year-old girl who had been clinically diagnosed as having spinal muscular atrophy.

Correlative Microscopy of Muscle Lesions

333

FIG. 9 Spinal muscular atrophy. (a) Pol map of lesion. Area of following micrographs is indicated (b). X400. (b) SEM (thiocarbohydrazide slide preparation) showing randomly arrangedprojections on myofibrils (MF). X4000. (c) TEM showing large spaces between myo fibrils of one myofiber. X6000. (d) TEM showing spacing between myofibrils to contain irregularly arranged (fibrous or membranous?) substance. X I5,OOO. Light microscopy of the biopsy showed no visible lesions in most of the specimen. A few fibers showed focal areas of hypercontraction that were verified by SEM. Transmission electron microscopy showed similar, but less severe, pathological changes in the myofibers, as had previously been noted on standard TEM of the DMD biopsy. Specifically, a disorganization of many myofilaments, Z-line streaming in sarcomeres, and a prominent

duplication of the basal lamina of myofibers ("looping") were noted. Lipid droplets were observed in the sarcoplasm of myofibers, pericytes, and endotherial cells. Results of correlation (Pol + SEM + TEM) microscopy were as follows. The Pol micrograph (Fig. %) shows nonbirefringent longitudinal streaks in one fiber. The SEM micrograph (Fig. 9b) shows a peculiar ragged appearance of the surface of most myofibrils. The TEM

334 micrographs (Fig. 9, c and d ) are correlates of the area seen in Fig. 9b. Myofibrils showed changes that could be interpreted as either a lysis and distortion of myofibrils or a distortion of membranous components. These changes appear to coincide with the ragged surface that had been seen on SEM in Fig. 9b.

DISCUSSION Some of the findings reported in this paper have t o our knowledge not been reported previously. To this date, only three papers have been found that describe SEM of pathological skeletal muscle. In 1973, Makita e t al.4 described an increased number of lipocytes i n dystrophic murine muscle that had been examined by SEM. lntracellular abnormalities, however, were not described. Human and murine myopathic tissues were examined by Sakuragawa and Sato’ and Sakuragawa e t a1.6 In the normal specimens, openings on the surface of the myofibers were observed and interpreted as pores leading t o the transverse tubules. These were arranged in parallel rows and were usually found a t the level of the A-I junction. The openings and the external A-I banding pattern appeared t o have disappeared in focal areas of both dystrophic and neurogenic atrophic myofibers in early stages of the diseases. These investigators reported these openings t o be frequently observed in normal muscle. However, this may have been due t o preparing their specimens by air-drying as opposed to the presently accepted procedure of critical point-drying. I t is thought that perhaps the air-drying procedure may have accentuated these openings even though t he external lamina was still adhering to the sarcolemma1 surface. It is interesting, however, that these openings and the A-I banding patterns were absent in certain focal areas of diseased muscle. Perhaps this could have been due to contracted regions of myofibers, which showed very little external detail. Regretably, none of these three papers shows TEM micrographs of the corresponding areas that had been examined by SEM. A short paper7 describes the possible usefulness of the correlation method (LM + SEM + TEM) in the morphological diagnosis of myopathies. The biopsy material fixed in glutaraldehyde was subsequently embedded i n

H. D. Geissinger et al. paraffin. Although this resulted in interpretable SEM images, it made interpretation of TEM micrographs often difficult and uncertain. The methodology employed in the present study resulted in an acceptable level of the preservation of ultrastructural detail. However, a limiting factor lies in the skill of the microtomist because perfect serial ultrathin sections of a very thin tissue slice (5-10 pm) are required. Another disadvantage of the correlation procedure i s that it i s very time-consuming; therefore it would be recommended not as a routine screening procedure but only when reasonable doubts regarding the correct histopathologic diagnosis exist.16 Notwithstanding the disadvantages, we think that the method has several advantages that outweigh i t s shortcomings: 1. It is often easier to decide under the light microscope if a lesion is pathological and then to examine i t s ultrastructure than it is to decide this grossly with the naked eye or dissecting microscope and harvest the tissue accordingly for ultrastructural (SEM or TEM) examination. 2. The area of a section that has been prepared for LM (e.g., 8 mm’) is much larger than the corresponding surface area in a section that has been routinely prepared for TEM (e.g., 0.5 mm’). The advantage of the LM (Pol) sections prepared by methods cited in this paper i s that the LM observer can examine the entire section with the scanning electron microscope and pick out certain areas of doubt that then can be examined with the transmission electron microscope. 3. Images derived from Pol, SEM, and TEM often complement one another, and the SEM image (which is of the surface ultrastructure) often cannot be interpreted without the knowledge of the ultrastructure of the interior (TEM image).

Therefore these intermicroscopic correlation methods are recommended until such time that surface ultrastructural detail (SEM image) can be interpreted with the same assurance as the interior ultrastructural detail (TEM image). Present investigations proved the usefulness of the techniques employed on pathologic muscle. Further applications of this technique on muscle biopsies are in progress.

335

Correlative Microscopy of Muscle Lesions

REFERENCES 1. Bethlem J: Muscle Pathology. Introduction and Atlas. New York: American Elsevier, 1970. 2. Dubowitz V , Brooke MH: Muscle Biopsy: A Modern Approach. Philadelphia: W. B. Saunders, 1973. 3. Mair WGP, Tom6 FMS: Atlas of the Ultrastructure of Disease Human Muscle. Edinburgh. Churchill-Livingstone, 1972. 4. Makita T. Kiwaki S, Sandborn E: Scanning electron microscopy and cytochemical localization of carnitine acetyl-transferaseactivity in normal and dystrophic muscle of mice. Histochem J 5~335-342, 1973. 5. Sakuragawa N, Sat0 T : Surface structure of normal and myopathic muscles observed by scanning electron microscopy. Clin Neurol (Tokyo) 12: 529-538,1972. 6. Sakuragawa N, Sat0 T, Tsubaki T: Scanning electron microscopic study of skeletal muscle: Normal, dystrophic and neurogenic atrophic muscle in mice and humans. Arch Neurol 28:247-251, 1973. 7. Geissinger HD, Ackerley CA: Methods of intermicroscopic (LM, SEM, TEM) correlation of myopathies. In: Proceedings of 6th Annual Meeting of Microscopical Society of Canada, edited by JM Sturgess, SA Omar, CFA Culling, pp. 42-43. Toronto: Imperial Press, 1979. 8 . Geissinger HD, Yamashiro S, Ackerley CA: Preparation of skeletal muscle for intermicroscopic (LM, SEM, TEM) correlation. In: Scanning Electron Microscopy/l978, edited by RP Becker, 0 Johari, vol. 2, pp. 267-274. AMF O'Hare: SEM Inc., 1978. 9. Vriend RA: Attempts a t improving the procedure for direct (LM-SEM-TEM) correlation. In: Proceedings of 6th Annual Meeting of Microscopical

10.

11.

12.

13.

14.

15.

16.

Society of Canada, edited by JM Sturgess, SA Omar, CFA Culling, pp. 66-67. Toronto: Imperial Press, 1979. Vriend, RA, Geissinger HD: An improved direct intermicroscopic (LM + SEM -+ TEM) correlative procedure for the examination of mammalian skeletal muscle. J Microsc 119, 1980, in press. Karnovsky MJ: A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell Biol 27:137A-I38A, 1965. de Bault LE: A critical point drying technique for scanning electron microscopy of tissue culture cells grown on plastic substratum. In: Scanning Electron Microscopy/l973, edited by 0 Johari, I Corvin, pp. 317-324. Chicago: IIT Research Institute, 1973. Flood PR: Dry fracturing techniques for the study of soft internal biological tissues in the scanning electron microscope. In: Scanning Electron Microscopyll975, edited by 0 Johari, I Corvin, pp. 287-294. Chicago: IIT Research Institute, 1975. Echlin P : Sputter coating techniques for scanning electron microscopy. In: Scanning Electron Microscopy/l975, edited by 0 Johari, I Corvin, pp. 217-224, 332. Chicago: IIT Research Institute, 1975. Malick LE, Wilson RB: Modified thiocarbohydrazide procedure for scanning electron microscopy: Routine use for normal, pathological or experimental tissue. Stain Techno1 50:265-269, 1975. Geissinger HD: lntermicroscopic (LM, SEM, TEM) correlation. In: Principles and Techniques of Scanning Electron Microscopy. Biological Applications, edited by MA Hayat, vol. 5, pp. 94121. New York: Van Nostrand-Reinhold, 1976.

Request reprints from H. D. Geissinger.