- Email: [email protected]
In vivo limb muscle differentiation in the absence of nerves: A quantitative study
EXPERIMEN’l’AI~
NEUROI,OCY
In Vivo
55, 160-172
Limb Muscle of Nerves:
(1977)
Differentiation A Quantitative
HEINZ
POPIELA
in the Absence Study 1
of
Drjartmcnt a&
Zoology, Mirhiyart State Unizwsity, East Lakrg. illickigatt Dcpartuacrrt of .ilwtoulg, Case ~~‘cstcrrt Rcscrz~c Urrivcrsity, School of Mcdicirtc, Ckvcland, Ohio 44106
Reccizwd
October
I?,
1976;
rczIisiorr
rccciz~cd
December
48,92#,
7, 1976
The entire presumptive nervous system of stage 24 ~mbZystorna mncnluembryos was surgically excised to produce aneurogenic larvae. For controls, embryos of the same age were operated in an identical fashion except that a segment of neural tube (about 1 mm) dorsal to the limb bud anlagen was left intact. The number of peripherally situated muscle nuclei (a measure of myofiber differentiation) peaked at the four-digit stage of aneurogenie limb development. The number of peripheral nuclei in the controls also peaked at the same time but remained at that maximum. With the exception of Day 13, the number of peripheral nuclei was always greater in control than in aneurogenic muscle. The total number of muscle nuclei per limb cross section gradually decreased in aneurogenic muscle to a number significantly lower than in controls at 31 days after the operation and thereafter. The cross-sectional size of aneurogenic muscle nuclei seemed to decrease at a faster rate than control nuclei. Reflecting nuclear density, the myonuclear area per muscle area in aneurogenic limbs was smallest at 21 days after the operation. After 21 days, aneurogenic nuclear densities progressively increased whereas control densities remained at the low, Day-21 value. From Day 13 to 21 the muscle nuclear density in control and aneurogenic limbs was statistically equal. Aneurogenic myofibers in an advanced state of degeneration incorporated more [“Hlleucine or ]“H]uridine than fibers where cross striations were still visible with the light microscope. With the electron microscope, acid phosphatase-stained lysosomes were found adjacent to disintegrating myofibrillar material in degenerating aneurogenic myofibers. Ruthenium red staining indicated that myofiber basement laminae developed normally in aneurogenic muscle and persisted.
tam
1 This research was supported by National Institutes of Health Grant No. AM15732 awarded to Dr. C. S. Thornton and Dr. S. Bromley and in part by a National Institutes of Health grant to Dr. M. Singer. I thank Dr. K. S. Warren (F. E. Rippel Foundation) for the use of the TTMC particle counter. This work was derived from a Ph.D. dissertation submitted to the Department of Zoology, Michigan State University. The author’s present address is Department of Biochemistry, University of Washington, Seattle, Washington 98195. 160 Copyright All rights
0 1977 by Academic Press. Inc. of reproduction in any form reserved.
ISSN
0014-4956
AiXEUROGEiTIC
MUSCLE
161
INTRODUCTIC)N If nerves are absent or greatly reduced prior to axonal outgrowth, muscle differentiation occurs almost normally (7-9, 19, 27, 31) at first hut muscle degenerates if ingrowth of nerves is prevented (3, 4, 10, 19, 21. 27, 2s). Similarly. innervated, differentiated muscle will degenerate after denervation (5, 6, 17, 18, 22, 26). A previous study (25) showed that among the tissues studied muscle is the most sensitive to the absence of nerves. Aneurogenic myotul)es initially grew somewhat and differentiated into late myotubes but subsequently degenerated in a characteristic ultrastructural pattern (30). In the present report the cytologic parameters of aneurogenic muscle development and degeneration are elaborated. MATERIALS
AND
METHODS
Surgical Proccdurc. The surgical procedure and tissue processing to produce aneurogenic and control animals were described previously (25). Briefly, A&/ysfo~a waczrlatum embryos had their entire presumptive nervous system removed at Harrison’s stage 24 ; subsequent development yielded aneurogenic larvae. In controls, the presumptive nervous system was removed also except that a segment of neural tube (about 1 mm) dorsal to the presumptive limb buds was left intact. Operated larvae were allowed to develop in sterile Steinberg’s saline until fixed and processed for electron microscopy (25). Histo~etvg. Plastic-embedded limbs were sectioned at 1 pm and midstylopodium section samples were histometrically measured with a TTMC particle measurement computer system (Millipore Corp.). To determine nuclear density, a region of a muscle bundle for each limb was selected. Myonuclear area measurements were made under oil (X 100) and the electronically selected muscle frame was placed at random near the periphery of a muscle bundle. The same muscle bundle was chosen for each limb. Lengths of myonuclei were measured in longitudinal sections with an ocular micrometer. Rllthenilrlll Red Staining (,Vyofibcv Basement Lamirtac). The method of Luft (15) was used except that ruthenium red was not added to the glutaraldehyde fixative solution. Prior to fixation with ruthenium redosmium tetroxide, the limb skin was peeled from fixed limbs becausethe skin blocks penetration of the dye. Acid Pltosphatasc Staining (Lysosomcs). Glutaraldehyde-fixed limbs denuded of their skin were incubated 15 min according to the method of Novikoff et al. (20). For control incubations, 10 mmol sodium fluoride was added to the incubation mixture or the substrate was omitted. Sodium fluoride was shown by quantitative histochemistry (23) and by biochemi-
162
HEINZ
POPIELA
Cal techniques (I, 2) to be an extremely potent (98%) inhibitor of acid phospliatase. AlltoradiograpAy. Aneurogenic animals, whose forelimbs were in the early four-digit stage of development, were treated with 20 &i tritiated uridine or 1 $Zi tritiated leucine in a volume of 0.02 ml. The tritiatecl precursors were injected intraperitoneally with a Hamilton syringe to which a glass micropipet was attached with paraffin to the needle. After 3 h of incorporation the animals were killed while immersed in the fixative and were processed for light microscopy and electron microscopy. Stained sections were dipped in Ilford K-5 emulsion diluted 3: 1 with distilled water. After a 3-week exposure time the stained slides were photographically processed. RESULTS In the aneurogenic limb musculature the number of peripheral myonuclei, as a measure of myofiber differentiation, reaches its peak at the four-digit stage of development in the upper arm (Fig. 1) . In the control 1.0
L
0
13
21
31
43
Days after stage 24 FIG. 1. Number of peripheral myonuclei per total number of myonuclei. Vertical arrows-statistics for aneurogenic vs control ; horizontal arrows-statistics for 31 days vs 43 days; n.d.-no statistical difference between means at the indicated level of significance. Other numbers within graph are statistical differences between means at the indicated levels of significance. Hand symbols signify the shape of the distal part of the forelimb at various times.
ANEUROGEKIC
hIUSCLE
163
Days after stage 24 FIG. 2. Total number of muscle nuclei per limb cross section. n.d.-No statistical difference between aneurogenic and control number of muscle nuclei at P < 0.05. Other numbers between curves are the statistical differences between aneurogenics and controls at indicated level of significance. Hand symbols signify the shape of the forelimb hand region at various times of development.
larvae, peripheral nuclear numbers plateau at the four-digit stage. The plateau is statistically significant taking into consideration that at the 5% level of significance the control 31- and 33-day means are equal whereas for aneurogenics they are not (Fig. 1, horizontal arrows). At 13 days after the operation there is no statistical difference between aneurogenic and control groups at the 1 and 5% levels of significance (Fig. 1). However, at 21, 31, and 13 days nuclear numbers are statistically significantly different between aneurogenic and control larvae. In addition to the gradual disappearance of aneurogenic muscle (25), the total number of muscle nuclei (nlyonuclei and satellite cell nuclei) also declines with time in control and aneurogenic limb muscle (Fig. 2). Initially, at Days 13 and 21 there is no statistical difference between aneurogenic and control mean numbers of muscle nuclei, but from Day 31 a separation between controls and aneurogenics at the 0.1% level of significance becomes established. Also, beginning with Day 31 the total number of muscle nuclei per limb section in aneurogenic limbs diminishes from less than two-thirds to as 1~uc1~ as one-fourth as compared to control limbs. The cross-sectional area of tmuscle nuclei also diminishes with time (Fig. 3). However, only on Days 43 and 54 is there a statistical difference between aneurogenic and control groups. Frotm Day 13 to 31 and on Day 65 there is no statistical difference between the two groups. The length of aneurogenic and control nlyonuclei, in longitudinal sections, at
164
HEINZ
POPIELA
“E 3 x60. aI 0 250.
340 m 230.
0’
13
21
31
43
54
65
Days after stage 24 l
FIG. 3. Cross-sectional area of muscle nuclei. -Aneurogenic; O-control; Anormal; n.d.-no statistical difference between control and aneurogenic means at P < 0.05. Numbers between curves are the statistical differences between aneurogenic and control means at the indicated levels of significance. Hand symbols signify the shape of the forelimb hand region at various times of development.
43 days of development is presented in Table 1. Analysis of variance indicates that from the 10 animals examined (control and aneurogenic) all myonuclei are statistically equal at the 1% level of significance. The control nuclear area per muscle area reaches a minimum level by 21 days TABLE Myonuclear Animal
a Average equal at P
Length myonuclei”
1 Lengths
of (pm)
1 2 3 4 5
46.408 37.430 36.694 40.820 39.641
zt f f f f
2.000 1.945 2.328 1.238 2.181
Aneurogenic
6 7 8 9 10
36.694 38.757 38.020 40.378 39.494
f f + f f
1.591 1.415 1.356 1.238 1.739
Control
length
< 0.01.
of 10 to 12 myonuclei
per animal
43-day-old
43-day-old
f
standard
animals
animals
error.
All
means
are
AKEUROGEKIC
M1:SCI.E
165
(Fig. 4)) whereas the aneurogenic muscle area attains a peak at 54 days. This peak sum nuclear area is approximately at the same level as in Day-13 muscle. At Days 13 and 21 no statistical difference between aneurogenic and control sum nuclear area is found. However, from Day 31 the difference between aneurogenics and controls is significant at the 1 and 57% levels. In a previous study (30) lysosomes were not detected in conventionally fixed aneurogenic muscle. Using a cytochemical stain for acid phosphatase, lysosomes were frequently found next to myofibrillar material within degenerating aneurogenic muscle in this investigation. Figure 5 shows a sample electron micrograph of suc11 a preparation. Here, a lysosome is situated adjacent to some myofibrillar material ; however, phagosomes were never found in aneurogenic myofibers or remnants of myofibers. Also, developing aneurogenic muscle or control muscle never exhibited acid phosphatase-stained lysosomes. Far from being metabolically inactive, degenerating aneurogenic limb muscle readily incorporated tritiated leutine (Fig. 6) or tritiated uridine. In fact, by visual inspection one notices in autoradiographic preparations the presence of more disintegration silver grains over partially degenerated muscle than over muscle where cross striations are still visible (Fig. 6. arrows ) Autoradiographs of uridine incorporation experiments are not shown because the grain distributions
I
0
Days after stage 24
FIG. 4. Sum mynonuclear area per selected muscle area (ordinate). n.d.-No statistical difference between control and aneurogenic means at P < 0.05. Numbers between curves are the difierences between aneurogenic and control means at indicated levels of significance. Hand symbols signify the shape of the forelimb hand region at various times of development.
166
HEINZ
POPIELA
FIG. 5. Acid phosphatase preparation of degenerating aneurogenic muscle. L-lysosome ; My-myofibrillar material ; N-myonucleus. Electron micrograph, ~31,200. FIG. 6. Autoradiograph of a longitudinal forelimb section from a Z-day-old larva. Grains are disintegration sites of incorporated tritiated leucine. Note that more grains are visible over degenerated myofibers (arrowheads) than over fibers where cross
ANEUROGENIC
167
MUSCLE
were similar to those in the leucine incorporation experiments shown in Fig. 6. In ruthenium red-stained muscle it was found that aneurogenic myofibers developed basement laminae similar in appearance to those in normal animals. In addition to limb musculature, the axial musculature also disappeared in time, although the degeneration of axial muscle was not specifically studied here. DISCUSSION In two previous reports (25, 30:) it was shown that aneurogenic muscle develops into early myofibers but subsequently disappears. It seemed that the myofibrillar material was primarily affected by the aneurogenic condition and that muscle nuclei persisted. The present study indicates that the total number of muscle nuclei per aneurogenic limb cross section also gradually declined (Fig. 3). In control muscle, such a decline during the period Day 13 to 65 was not evident in a time-series statistical analysis at the 1s level of significance. Thus, one may conclude that the total number of muscle nuclei diminished in aneurogenic limbs but not in control limbs. Statistical equality between aneurogenic and control numbers of muscle nuclei from 13 to 21 days suggests that the total numbers of muscle nuclei in aneurogenic and control larvae were the same during that time. Beginning with Day 31, however, the total numbers of muscle nuclei were different in control versus aneurogenic limbs (Fig. 2 ), suggesting that survival and possibly, proliferation of nuclei was interfered with more in aneurogenic than in control limbs. By counting the number of muscle nuclei in the denervated thorax musculature of the cricket, Thommen (39) also saw a reduction in the number of nuclei per individual muscle. For example, after denervation at stage 3 or 5, two of the thorax muscles contained 6.4 and 15% of normal numbers of nuclei measured at adult stages. However, a shorter denervation period led to a survival of a larger number of nuclei. Gutmann and Zelena (6) claim a temporary 40% increase in DNA during the first few months of denervation in adult rat limb muscle followed by a return to normal levels thereafter. In the present study, an increase in the total number of nuclei per aneurogenic limb, even from notch to the three-digit stage, a time when myotubes differentiate, is doubtful. The amount of nucleoplasm in cross section per aneurogenic muscle area returned from a minimum at Day 21 to notch-stage levels by Day 54 (Fig. 4 j This reversion in nuclear density does not seemto have occurred by swelling of nuclei (,6) because the size of muscle nuclei actually destriations x 1590.
are
still
visible.
E-epitlmuis
; H-lnuuerus
; M-muscle.
I’hotomicrograph,
168
HEINZ
POPIELA
creased with developmental age (Fig. 3). Apparently, the reversion in nuclear density of aneurogenic muscle is a reflection of the greater rate of myofibrillar disappearance relative to nuclear numbers combined with a greater survival rate for myonuclei. A similar survival of muscle nuclei relative to denervated muscle was hinted at by Gutmann and Zelena (6). Then too, if one rearranges the data obtained by Thommen (29) into a form similar to that used in this study, one perceives the same increased number of muscle nuclei per muscle area in all the denervated cricket thorax muscles. Figure 4, furthermore, shows a parallel decrease in nuclear density in control and aneurogenic limb muscle from 13 to 21 days of development. During this period, the difference between aneurogenic and control muscle is not statistically significant, allowing one to conclude that their nuclear densities decrease at the same rate and are the same from Day 13 to 21. Such a conclusion implies some growth of aneurogenic muscle during the early stages of limb development. Beginning at Day 21, control nuclear density reached a minimum level of about 15% nucleoplasmic area per muscle area whereas aneurogenic muscle showed a sharp increase in nuclear density, as mentioned previously. From the three-digit stage of development until termination of the experimental series, aneurogenie nuclear densities were distinctly and statistically significantly greater than control densities. It was previously mentioned that individual nuclear areas decreased with progressing developmental age (Fig. 3). Because no statistical difference between aneurogenic and control muscle is apparent until Day 43, it seems that nuclei decrease in cross-sectional size regardless of whether or not muscle is innervated during this time. However, at Days 43 and 54 aneurogenic nuclei were statistically smaller in cross section than control nuclei. A decrease in cross-sectional area, of course, does not necessarily mean a decrease in volume because it is well known that myonuclei elongate as muscle differentiates. However, Table 1 shows the lengths of muscle nuclei for aneurogenic and control larvae at 43 days of development, and analysis of variance indicates that, at the 1% level of significance, all nuclei (aneurogenic and control) are of the same length. Accordingly, one can infer that at least 43-day aneurogenic nuclei are smaller in volume than control nuclei, making it improbable that a differential elongation of aneurogenic vs control muscle nuclei occurs. With this information, one could interpret the steeper slope of aneurogenic nuclear size as compared to the control slope (Fig. 3) to mean a greater rate of nuclear shrinkage for aneurogenic muscle nuclei. At 65 days the number of aneurogenic, control, and normal muscle nuclei were statistically the same; the apparent size increase in aneurogenic nuclei from Day 54 to 65 was probably due to a general limb edema as discussed previously (25).
ANEUROGENIC
MUSCLE
169
Although a comparison of the experimental larvae with normal, unoperated, and feeding larvae was not the major purpose of this investigation, reference plots for normal animals at Days 43 and 65 indicate differences, presumably due to compounded effects of normalcy. For example, limbs from normal larvae indicated a two- to fourfold greater nuclear number than control limbs at Days 43 and 65, respectively (Fig. 2 ), a reflection of the greater amount of muscle present in normal limbs (25 ) ; the nuclear density (Fig. 4) was significantly lower in normal muscle than in control muscle at Day 65, an indication of a greater amount of sarcoplasmic components in normal muscle. When the peripherally situated nuclei were counted as a measure of myofiber differentiation (Fig. 1) , it was foulid that aneurogenic myofibers from the upper forelimb reached their peak of differentiation at the fourdigit stage. Control myofibers differentiated at the same time and remained differentiated whereas aneurogenic fibers appeared to revert to an undifferentiated state. The apparent return of aneurogenic myofibers to an undifferentiated state is possibly an illusion because contractile material disappeared at a greater rate than nuclei. leaving nuclei with a thin rim of cytoplasm in the latter part of the experimental series. That control fibers indeed reached a plateau in their differentiation and aneurogenic fibers did not is indicated by the statistical comparison of aneurogenic and control fibers (Fig. 1, horizontal arrows) during the 31- to 43-day period. Control fibers did not differ statistically from Days 31 to 43 at the 5% level of significance whereas aneurogenic fibers did during the same period and at the same level of significance. At notch stages of development, control and aneurogenic fibers were equally midifferentiated ; however, subsequent to Day 13, control fibers were always more differentiated than aneurogenic fibers. Both conclusions are confirmed by statistical inference (Fig. 1, vertical arrows). Even at their peak differentiation, the aneurogenic myofibers were less differentiated than control fibers. In a previous study (30) ultrastructural changes during aneurogenic muscle degeneration were described. In the present examination of aneurogenie muscle degeneration, it was fouiid that the ultrastructural characteristics of degeneration as described by Tweedle et al. (30) applied here also, although the methodology was different and larvae were totally aneurogenic. The most striking characteristics of aneurogenic muscle at its peak period of differentiation is the abnormal dilation of the sarcoplasmic reticulum. Otherwise, aneurogenic myofibers looked quite normal at that time. Examination of rutheninm retl-stained tissues showed that basement laminae developed in aneurogenic myofibers in a fashion similar to that of normal limbs, indicating that the absence of nerves did not affect the
170
HEINZ
POPJELA
development of fiber sheaths. Typical muscle satellite cells (16) were detected in aneurogenic, control, and normal A~~~blysfo~na muscle although satellite cells are characteristically found only in mammalian muscle and have only recently been shown to exist in salamanders (24). The mechanism by which muscle atrophies and contractile protein disappears has remained a mystery to this day. Thus, Kohn (11, 12) did not find enzymes capable of digesting myosin in denervated muscle nor has contractile protein been seen within lysosomal membranes. In the aneurogenic limb muscle stained for acid phosphatase, lysosomes were seen between distintegrating myofibrils (Fig. 5), but as others have noted (12, 13, 14), besides the presence of lysosomes, no evidence of lysosomal digestion of contractile material has been observed. Some macrophages were seen to ingest muscle fragments but did not seem numerous enough to account for the massive disappearance of aneurogenic muscle. It is interesting to note that aneurogenic muscle was far from metabolically inactive but incorporated tritiated leucine (Fig. 6) and tritiated uridine vigorously. In fact, by visual inspection more disintegration grains were found over myofibers where cross striations had disappeared than over fibers where cross striations were still clearly visible. Possibly, myofibers well underway in their atrophy synthesize more protein and RNA than relatively unatrophied fibers. REFERENCES 1. BOYER, P. D., H. LARDY, AND R. MYKB~CK, Eds. 1961. Tke En,-ytrtcs, I’ol. 5. Academic Press, New York. 2. BOYER, P. D., Ed. 1971. Ths Enzyrws, 1-01. 4. Academic Press, New York. 3. CRAIN, S. M., AND E. R. PETERSON. 1974. Development of neural connections in culture. ;Jutt. N.Y. Acad. Sri. 228 : 6-34. 4. EASTLICK, H. I,. 1943. Studies on transplanted embryonic limbs of the chick. I. The development of muscle in nerveless and innervated grafts. J. Exp. 2001. 93: 27-45. 5. ENCEI., A. G., ANU H. H. STONNINGTON. 1974. Morphological effects of denervation of muscle. A quantitative ultrastructural study. d~x. N.Y. Acad. Sri. 228: 68-88. 6. GUTMANN, E., AND J. ZELENA. 1962. Morphological changes in the denervated muscle. Pages 57-102 irz E. GUTMANN, Ed., The Demm~atcd Muscle. CzeckoSlovak Academy of Science, Prague. 7. HARRISON, R. G. 1904. An experimental study of the relation of the nervous system to the developing musculature in the embryo of the frog. Amcv. J. A~zat. 3: 197-220. 8. HOADLEY, L. 1925. The differentiation of isolated chick primordia in chorioallantoic grafts. II. The effect of the presence of the spinal cord, i.e. innervation on the differentiation of the somitic region. f. .!?xp. Zool. 4’2: 143-162. 9. HOOKER, D. 1911. The development and function of voluntary and cardiac muscle in embryos without nerves. J. Exp. Zool. 11: 159-186 .
ANELTROGEXIC 10.
E.
HUNT,
grafts, 11. KOHN,
Amrr. 12. KOHN, for
A. with R. R.
1932. The differentiation of chick limb buds special reference to the muscles. J. I:‘.rp. Zor~l. 1964. Mechanism of protein loss in dencrvation
J. I’atld.
R. R. 1966. Denervation autolysis iri ejitr0. =l~rzrr.
L’ltrasfrm-t.
in chorio-allantoic 62: 57-91. muscle atrophy.
45: 435-447.
muscle atrophy. Studies J. I’nthol. 48: 241-257. 13. I.ocrcsrr~~, R. A., AP\‘D J. BEAUI.ATOS. 1974. Programmed evidence for lysosomes during the normal hrcakdown muscles. J. l~7/frtrstrr~rf. Krs. 46 : 43-62. 14. I.OC.I;SHIX, R. A., AND J. BEAVI.ATON. 1974. Programmed appearance of lysosomes when death of the intersegmental
J.
171
AIUSCLE
of components
required
cell death. Cytochemical of the intersegmental cell death. muscles
Cytochemical is prevented.
Rcs. 46: 63-78.
15. LUFT, J. H. 1971. Ruthenium red and violet. I. Chemistry, purification, methods of use for electron microscopy and mechanism of action. :frmt. Rec. 171: 347368. 16. MAURO, A. 1961. Satellite cell of skeletal muscle fiber. J. Biophys. Biochcm. Cytol. 9: 493. 17. MILEDI, R., AIGD C. R. SI.ATEK. 1969. Electron-microscopic structure of denervated skeletal muscle. Proc. R. Sot. (Lord.) [Biol.] 174: 253-269. 18. MUSCATELLO, U., MARGRETH, A., ASD bi. XLOI~I. 1965. On the differential response of sarcoplasm and myoplasm to denervation in frog muscle. J. Cell Bid. 27: l-24. 19. MUCHMORE, W. B. 1968. The influence of neural tissue on the early development of somitic muscle in ventro-lateral implants in A-lrrrbystonw. J. Exp. Zool. 169: 251-258. 20. NOVIICOFF, P. M., A. B. NOVXOFF, N. QGINTASA, AXD J.-J. HAU\V. 1971. Golgi apparatus, GERL, and lysosomes of neurons in the rat dorsal root ganglia studied by thick and thin section cytochemistry. J. Cc/l Biol. 50: 859-886. 21. N~~ESCII, H. 1968. The role of the nervous system in insect morphogenesis and regeneration. dwm. Rczl. E~~fowd. 13: 2734. 22. PEI~IXGIGWIXO, C., AND C. FRANZINI. 1963. An electron microscopic study of denervation atrophy in red and white skeletal muscle fibers. J. Cell Biol. 17: 327-349. 23. PFEIFER, U., AND H. WITSTHEL. 1972. Zum zeitlichen Ablauf der fur die Elektronenmikroskopie modifizierten Gomori-Reaktion auf saure Phosphatase. Acta Hisforhrm. (Suppl.) 12 : 111-120. 24. POIJIELA, H. 1976. Muscle satellite cells in urodele amphihians: Facilitated identification of muscle satellite cells using ruthenium red staining. J. Exp. 2001. 198 : 57-64. 25. POPIELA, H. 1976. Itc zko limb tissue development in the absence of nerves: A quantitative study. E.rp. Ncrrval. 53 : 214-226. 26. REES, D., A~sD P. N. R. USHER\VOOD. 1972. Effects of denervation on the ultrastructure of insect muscle. J. Ccl/ Sri. 10: 667-682. 27. SHIMADA, Y., D. A. FISCHXIAN, AiYD A. A. MOSCONA. 1967. The fine structure of embryonic chick skeletal muscle cells differentiated irz vitro. J. Cell Biol. 35: 445-453. 28. SHIMADA, Y., AND D. A. FISCH~IAN. 1973. Morphological and physiological evidence for the development of functional neuromuscular junctions ilz vitro. Dezelop. Biol. 31 : 200-225. 29. TIIOMAIES, H. 1974. Untersuchungen iiber die Wirkung der Denervation auf das
172
HEINZ
POPIELA
C. D., H. PWIELA, AND C. S. THOHNTON. 1974. Ultrastructure of the development and subsequent breakdown of muscle in aneurogenic limbs (Ambystoma). J. Exp. Zool. 190: 155-166. 31. ZELENA, J. 1962. The effect of denervation on muscle development. Pages 103-126 irz E. GUTMANN, Ed., Tkc Drucrvuted Nlucle. Czechoslovak Academy of Science, Prague. 30. TWEEDLE,
NEUROI,OCY
In Vivo
55, 160-172
Limb Muscle of Nerves:
(1977)
Differentiation A Quantitative
HEINZ
POPIELA
in the Absence Study 1
of
Drjartmcnt a&
Zoology, Mirhiyart State Unizwsity, East Lakrg. illickigatt Dcpartuacrrt of .ilwtoulg, Case ~~‘cstcrrt Rcscrz~c Urrivcrsity, School of Mcdicirtc, Ckvcland, Ohio 44106
Reccizwd
October
I?,
1976;
rczIisiorr
rccciz~cd
December
48,92#,
7, 1976
The entire presumptive nervous system of stage 24 ~mbZystorna mncnluembryos was surgically excised to produce aneurogenic larvae. For controls, embryos of the same age were operated in an identical fashion except that a segment of neural tube (about 1 mm) dorsal to the limb bud anlagen was left intact. The number of peripherally situated muscle nuclei (a measure of myofiber differentiation) peaked at the four-digit stage of aneurogenie limb development. The number of peripheral nuclei in the controls also peaked at the same time but remained at that maximum. With the exception of Day 13, the number of peripheral nuclei was always greater in control than in aneurogenic muscle. The total number of muscle nuclei per limb cross section gradually decreased in aneurogenic muscle to a number significantly lower than in controls at 31 days after the operation and thereafter. The cross-sectional size of aneurogenic muscle nuclei seemed to decrease at a faster rate than control nuclei. Reflecting nuclear density, the myonuclear area per muscle area in aneurogenic limbs was smallest at 21 days after the operation. After 21 days, aneurogenic nuclear densities progressively increased whereas control densities remained at the low, Day-21 value. From Day 13 to 21 the muscle nuclear density in control and aneurogenic limbs was statistically equal. Aneurogenic myofibers in an advanced state of degeneration incorporated more [“Hlleucine or ]“H]uridine than fibers where cross striations were still visible with the light microscope. With the electron microscope, acid phosphatase-stained lysosomes were found adjacent to disintegrating myofibrillar material in degenerating aneurogenic myofibers. Ruthenium red staining indicated that myofiber basement laminae developed normally in aneurogenic muscle and persisted.
tam
1 This research was supported by National Institutes of Health Grant No. AM15732 awarded to Dr. C. S. Thornton and Dr. S. Bromley and in part by a National Institutes of Health grant to Dr. M. Singer. I thank Dr. K. S. Warren (F. E. Rippel Foundation) for the use of the TTMC particle counter. This work was derived from a Ph.D. dissertation submitted to the Department of Zoology, Michigan State University. The author’s present address is Department of Biochemistry, University of Washington, Seattle, Washington 98195. 160 Copyright All rights
0 1977 by Academic Press. Inc. of reproduction in any form reserved.
ISSN
0014-4956
AiXEUROGEiTIC
MUSCLE
161
INTRODUCTIC)N If nerves are absent or greatly reduced prior to axonal outgrowth, muscle differentiation occurs almost normally (7-9, 19, 27, 31) at first hut muscle degenerates if ingrowth of nerves is prevented (3, 4, 10, 19, 21. 27, 2s). Similarly. innervated, differentiated muscle will degenerate after denervation (5, 6, 17, 18, 22, 26). A previous study (25) showed that among the tissues studied muscle is the most sensitive to the absence of nerves. Aneurogenic myotul)es initially grew somewhat and differentiated into late myotubes but subsequently degenerated in a characteristic ultrastructural pattern (30). In the present report the cytologic parameters of aneurogenic muscle development and degeneration are elaborated. MATERIALS
AND
METHODS
Surgical Proccdurc. The surgical procedure and tissue processing to produce aneurogenic and control animals were described previously (25). Briefly, A&/ysfo~a waczrlatum embryos had their entire presumptive nervous system removed at Harrison’s stage 24 ; subsequent development yielded aneurogenic larvae. In controls, the presumptive nervous system was removed also except that a segment of neural tube (about 1 mm) dorsal to the presumptive limb buds was left intact. Operated larvae were allowed to develop in sterile Steinberg’s saline until fixed and processed for electron microscopy (25). Histo~etvg. Plastic-embedded limbs were sectioned at 1 pm and midstylopodium section samples were histometrically measured with a TTMC particle measurement computer system (Millipore Corp.). To determine nuclear density, a region of a muscle bundle for each limb was selected. Myonuclear area measurements were made under oil (X 100) and the electronically selected muscle frame was placed at random near the periphery of a muscle bundle. The same muscle bundle was chosen for each limb. Lengths of myonuclei were measured in longitudinal sections with an ocular micrometer. Rllthenilrlll Red Staining (,Vyofibcv Basement Lamirtac). The method of Luft (15) was used except that ruthenium red was not added to the glutaraldehyde fixative solution. Prior to fixation with ruthenium redosmium tetroxide, the limb skin was peeled from fixed limbs becausethe skin blocks penetration of the dye. Acid Pltosphatasc Staining (Lysosomcs). Glutaraldehyde-fixed limbs denuded of their skin were incubated 15 min according to the method of Novikoff et al. (20). For control incubations, 10 mmol sodium fluoride was added to the incubation mixture or the substrate was omitted. Sodium fluoride was shown by quantitative histochemistry (23) and by biochemi-
162
HEINZ
POPIELA
Cal techniques (I, 2) to be an extremely potent (98%) inhibitor of acid phospliatase. AlltoradiograpAy. Aneurogenic animals, whose forelimbs were in the early four-digit stage of development, were treated with 20 &i tritiated uridine or 1 $Zi tritiated leucine in a volume of 0.02 ml. The tritiatecl precursors were injected intraperitoneally with a Hamilton syringe to which a glass micropipet was attached with paraffin to the needle. After 3 h of incorporation the animals were killed while immersed in the fixative and were processed for light microscopy and electron microscopy. Stained sections were dipped in Ilford K-5 emulsion diluted 3: 1 with distilled water. After a 3-week exposure time the stained slides were photographically processed. RESULTS In the aneurogenic limb musculature the number of peripheral myonuclei, as a measure of myofiber differentiation, reaches its peak at the four-digit stage of development in the upper arm (Fig. 1) . In the control 1.0
L
0
13
21
31
43
Days after stage 24 FIG. 1. Number of peripheral myonuclei per total number of myonuclei. Vertical arrows-statistics for aneurogenic vs control ; horizontal arrows-statistics for 31 days vs 43 days; n.d.-no statistical difference between means at the indicated level of significance. Other numbers within graph are statistical differences between means at the indicated levels of significance. Hand symbols signify the shape of the distal part of the forelimb at various times.
ANEUROGEKIC
hIUSCLE
163
Days after stage 24 FIG. 2. Total number of muscle nuclei per limb cross section. n.d.-No statistical difference between aneurogenic and control number of muscle nuclei at P < 0.05. Other numbers between curves are the statistical differences between aneurogenics and controls at indicated level of significance. Hand symbols signify the shape of the forelimb hand region at various times of development.
larvae, peripheral nuclear numbers plateau at the four-digit stage. The plateau is statistically significant taking into consideration that at the 5% level of significance the control 31- and 33-day means are equal whereas for aneurogenics they are not (Fig. 1, horizontal arrows). At 13 days after the operation there is no statistical difference between aneurogenic and control groups at the 1 and 5% levels of significance (Fig. 1). However, at 21, 31, and 13 days nuclear numbers are statistically significantly different between aneurogenic and control larvae. In addition to the gradual disappearance of aneurogenic muscle (25), the total number of muscle nuclei (nlyonuclei and satellite cell nuclei) also declines with time in control and aneurogenic limb muscle (Fig. 2). Initially, at Days 13 and 21 there is no statistical difference between aneurogenic and control mean numbers of muscle nuclei, but from Day 31 a separation between controls and aneurogenics at the 0.1% level of significance becomes established. Also, beginning with Day 31 the total number of muscle nuclei per limb section in aneurogenic limbs diminishes from less than two-thirds to as 1~uc1~ as one-fourth as compared to control limbs. The cross-sectional area of tmuscle nuclei also diminishes with time (Fig. 3). However, only on Days 43 and 54 is there a statistical difference between aneurogenic and control groups. Frotm Day 13 to 31 and on Day 65 there is no statistical difference between the two groups. The length of aneurogenic and control nlyonuclei, in longitudinal sections, at
164
HEINZ
POPIELA
“E 3 x60. aI 0 250.
340 m 230.
0’
13
21
31
43
54
65
Days after stage 24 l
FIG. 3. Cross-sectional area of muscle nuclei. -Aneurogenic; O-control; Anormal; n.d.-no statistical difference between control and aneurogenic means at P < 0.05. Numbers between curves are the statistical differences between aneurogenic and control means at the indicated levels of significance. Hand symbols signify the shape of the forelimb hand region at various times of development.
43 days of development is presented in Table 1. Analysis of variance indicates that from the 10 animals examined (control and aneurogenic) all myonuclei are statistically equal at the 1% level of significance. The control nuclear area per muscle area reaches a minimum level by 21 days TABLE Myonuclear Animal
a Average equal at P
Length myonuclei”
1 Lengths
of (pm)
1 2 3 4 5
46.408 37.430 36.694 40.820 39.641
zt f f f f
2.000 1.945 2.328 1.238 2.181
Aneurogenic
6 7 8 9 10
36.694 38.757 38.020 40.378 39.494
f f + f f
1.591 1.415 1.356 1.238 1.739
Control
length
< 0.01.
of 10 to 12 myonuclei
per animal
43-day-old
43-day-old
f
standard
animals
animals
error.
All
means
are
AKEUROGEKIC
M1:SCI.E
165
(Fig. 4)) whereas the aneurogenic muscle area attains a peak at 54 days. This peak sum nuclear area is approximately at the same level as in Day-13 muscle. At Days 13 and 21 no statistical difference between aneurogenic and control sum nuclear area is found. However, from Day 31 the difference between aneurogenics and controls is significant at the 1 and 57% levels. In a previous study (30) lysosomes were not detected in conventionally fixed aneurogenic muscle. Using a cytochemical stain for acid phosphatase, lysosomes were frequently found next to myofibrillar material within degenerating aneurogenic muscle in this investigation. Figure 5 shows a sample electron micrograph of suc11 a preparation. Here, a lysosome is situated adjacent to some myofibrillar material ; however, phagosomes were never found in aneurogenic myofibers or remnants of myofibers. Also, developing aneurogenic muscle or control muscle never exhibited acid phosphatase-stained lysosomes. Far from being metabolically inactive, degenerating aneurogenic limb muscle readily incorporated tritiated leutine (Fig. 6) or tritiated uridine. In fact, by visual inspection one notices in autoradiographic preparations the presence of more disintegration silver grains over partially degenerated muscle than over muscle where cross striations are still visible (Fig. 6. arrows ) Autoradiographs of uridine incorporation experiments are not shown because the grain distributions
I
0
Days after stage 24
FIG. 4. Sum mynonuclear area per selected muscle area (ordinate). n.d.-No statistical difference between control and aneurogenic means at P < 0.05. Numbers between curves are the difierences between aneurogenic and control means at indicated levels of significance. Hand symbols signify the shape of the forelimb hand region at various times of development.
166
HEINZ
POPIELA
FIG. 5. Acid phosphatase preparation of degenerating aneurogenic muscle. L-lysosome ; My-myofibrillar material ; N-myonucleus. Electron micrograph, ~31,200. FIG. 6. Autoradiograph of a longitudinal forelimb section from a Z-day-old larva. Grains are disintegration sites of incorporated tritiated leucine. Note that more grains are visible over degenerated myofibers (arrowheads) than over fibers where cross
ANEUROGENIC
167
MUSCLE
were similar to those in the leucine incorporation experiments shown in Fig. 6. In ruthenium red-stained muscle it was found that aneurogenic myofibers developed basement laminae similar in appearance to those in normal animals. In addition to limb musculature, the axial musculature also disappeared in time, although the degeneration of axial muscle was not specifically studied here. DISCUSSION In two previous reports (25, 30:) it was shown that aneurogenic muscle develops into early myofibers but subsequently disappears. It seemed that the myofibrillar material was primarily affected by the aneurogenic condition and that muscle nuclei persisted. The present study indicates that the total number of muscle nuclei per aneurogenic limb cross section also gradually declined (Fig. 3). In control muscle, such a decline during the period Day 13 to 65 was not evident in a time-series statistical analysis at the 1s level of significance. Thus, one may conclude that the total number of muscle nuclei diminished in aneurogenic limbs but not in control limbs. Statistical equality between aneurogenic and control numbers of muscle nuclei from 13 to 21 days suggests that the total numbers of muscle nuclei in aneurogenic and control larvae were the same during that time. Beginning with Day 31, however, the total numbers of muscle nuclei were different in control versus aneurogenic limbs (Fig. 2 ), suggesting that survival and possibly, proliferation of nuclei was interfered with more in aneurogenic than in control limbs. By counting the number of muscle nuclei in the denervated thorax musculature of the cricket, Thommen (39) also saw a reduction in the number of nuclei per individual muscle. For example, after denervation at stage 3 or 5, two of the thorax muscles contained 6.4 and 15% of normal numbers of nuclei measured at adult stages. However, a shorter denervation period led to a survival of a larger number of nuclei. Gutmann and Zelena (6) claim a temporary 40% increase in DNA during the first few months of denervation in adult rat limb muscle followed by a return to normal levels thereafter. In the present study, an increase in the total number of nuclei per aneurogenic limb, even from notch to the three-digit stage, a time when myotubes differentiate, is doubtful. The amount of nucleoplasm in cross section per aneurogenic muscle area returned from a minimum at Day 21 to notch-stage levels by Day 54 (Fig. 4 j This reversion in nuclear density does not seemto have occurred by swelling of nuclei (,6) because the size of muscle nuclei actually destriations x 1590.
are
still
visible.
E-epitlmuis
; H-lnuuerus
; M-muscle.
I’hotomicrograph,
168
HEINZ
POPIELA
creased with developmental age (Fig. 3). Apparently, the reversion in nuclear density of aneurogenic muscle is a reflection of the greater rate of myofibrillar disappearance relative to nuclear numbers combined with a greater survival rate for myonuclei. A similar survival of muscle nuclei relative to denervated muscle was hinted at by Gutmann and Zelena (6). Then too, if one rearranges the data obtained by Thommen (29) into a form similar to that used in this study, one perceives the same increased number of muscle nuclei per muscle area in all the denervated cricket thorax muscles. Figure 4, furthermore, shows a parallel decrease in nuclear density in control and aneurogenic limb muscle from 13 to 21 days of development. During this period, the difference between aneurogenic and control muscle is not statistically significant, allowing one to conclude that their nuclear densities decrease at the same rate and are the same from Day 13 to 21. Such a conclusion implies some growth of aneurogenic muscle during the early stages of limb development. Beginning at Day 21, control nuclear density reached a minimum level of about 15% nucleoplasmic area per muscle area whereas aneurogenic muscle showed a sharp increase in nuclear density, as mentioned previously. From the three-digit stage of development until termination of the experimental series, aneurogenie nuclear densities were distinctly and statistically significantly greater than control densities. It was previously mentioned that individual nuclear areas decreased with progressing developmental age (Fig. 3). Because no statistical difference between aneurogenic and control muscle is apparent until Day 43, it seems that nuclei decrease in cross-sectional size regardless of whether or not muscle is innervated during this time. However, at Days 43 and 54 aneurogenic nuclei were statistically smaller in cross section than control nuclei. A decrease in cross-sectional area, of course, does not necessarily mean a decrease in volume because it is well known that myonuclei elongate as muscle differentiates. However, Table 1 shows the lengths of muscle nuclei for aneurogenic and control larvae at 43 days of development, and analysis of variance indicates that, at the 1% level of significance, all nuclei (aneurogenic and control) are of the same length. Accordingly, one can infer that at least 43-day aneurogenic nuclei are smaller in volume than control nuclei, making it improbable that a differential elongation of aneurogenic vs control muscle nuclei occurs. With this information, one could interpret the steeper slope of aneurogenic nuclear size as compared to the control slope (Fig. 3) to mean a greater rate of nuclear shrinkage for aneurogenic muscle nuclei. At 65 days the number of aneurogenic, control, and normal muscle nuclei were statistically the same; the apparent size increase in aneurogenic nuclei from Day 54 to 65 was probably due to a general limb edema as discussed previously (25).
ANEUROGENIC
MUSCLE
169
Although a comparison of the experimental larvae with normal, unoperated, and feeding larvae was not the major purpose of this investigation, reference plots for normal animals at Days 43 and 65 indicate differences, presumably due to compounded effects of normalcy. For example, limbs from normal larvae indicated a two- to fourfold greater nuclear number than control limbs at Days 43 and 65, respectively (Fig. 2 ), a reflection of the greater amount of muscle present in normal limbs (25 ) ; the nuclear density (Fig. 4) was significantly lower in normal muscle than in control muscle at Day 65, an indication of a greater amount of sarcoplasmic components in normal muscle. When the peripherally situated nuclei were counted as a measure of myofiber differentiation (Fig. 1) , it was foulid that aneurogenic myofibers from the upper forelimb reached their peak of differentiation at the fourdigit stage. Control myofibers differentiated at the same time and remained differentiated whereas aneurogenic fibers appeared to revert to an undifferentiated state. The apparent return of aneurogenic myofibers to an undifferentiated state is possibly an illusion because contractile material disappeared at a greater rate than nuclei. leaving nuclei with a thin rim of cytoplasm in the latter part of the experimental series. That control fibers indeed reached a plateau in their differentiation and aneurogenic fibers did not is indicated by the statistical comparison of aneurogenic and control fibers (Fig. 1, horizontal arrows) during the 31- to 43-day period. Control fibers did not differ statistically from Days 31 to 43 at the 5% level of significance whereas aneurogenic fibers did during the same period and at the same level of significance. At notch stages of development, control and aneurogenic fibers were equally midifferentiated ; however, subsequent to Day 13, control fibers were always more differentiated than aneurogenic fibers. Both conclusions are confirmed by statistical inference (Fig. 1, vertical arrows). Even at their peak differentiation, the aneurogenic myofibers were less differentiated than control fibers. In a previous study (30) ultrastructural changes during aneurogenic muscle degeneration were described. In the present examination of aneurogenie muscle degeneration, it was fouiid that the ultrastructural characteristics of degeneration as described by Tweedle et al. (30) applied here also, although the methodology was different and larvae were totally aneurogenic. The most striking characteristics of aneurogenic muscle at its peak period of differentiation is the abnormal dilation of the sarcoplasmic reticulum. Otherwise, aneurogenic myofibers looked quite normal at that time. Examination of rutheninm retl-stained tissues showed that basement laminae developed in aneurogenic myofibers in a fashion similar to that of normal limbs, indicating that the absence of nerves did not affect the
170
HEINZ
POPJELA
development of fiber sheaths. Typical muscle satellite cells (16) were detected in aneurogenic, control, and normal A~~~blysfo~na muscle although satellite cells are characteristically found only in mammalian muscle and have only recently been shown to exist in salamanders (24). The mechanism by which muscle atrophies and contractile protein disappears has remained a mystery to this day. Thus, Kohn (11, 12) did not find enzymes capable of digesting myosin in denervated muscle nor has contractile protein been seen within lysosomal membranes. In the aneurogenic limb muscle stained for acid phosphatase, lysosomes were seen between distintegrating myofibrils (Fig. 5), but as others have noted (12, 13, 14), besides the presence of lysosomes, no evidence of lysosomal digestion of contractile material has been observed. Some macrophages were seen to ingest muscle fragments but did not seem numerous enough to account for the massive disappearance of aneurogenic muscle. It is interesting to note that aneurogenic muscle was far from metabolically inactive but incorporated tritiated leucine (Fig. 6) and tritiated uridine vigorously. In fact, by visual inspection more disintegration grains were found over myofibers where cross striations had disappeared than over fibers where cross striations were still clearly visible. Possibly, myofibers well underway in their atrophy synthesize more protein and RNA than relatively unatrophied fibers. REFERENCES 1. BOYER, P. D., H. LARDY, AND R. MYKB~CK, Eds. 1961. Tke En,-ytrtcs, I’ol. 5. Academic Press, New York. 2. BOYER, P. D., Ed. 1971. Ths Enzyrws, 1-01. 4. Academic Press, New York. 3. CRAIN, S. M., AND E. R. PETERSON. 1974. Development of neural connections in culture. ;Jutt. N.Y. Acad. Sri. 228 : 6-34. 4. EASTLICK, H. I,. 1943. Studies on transplanted embryonic limbs of the chick. I. The development of muscle in nerveless and innervated grafts. J. Exp. 2001. 93: 27-45. 5. ENCEI., A. G., ANU H. H. STONNINGTON. 1974. Morphological effects of denervation of muscle. A quantitative ultrastructural study. d~x. N.Y. Acad. Sri. 228: 68-88. 6. GUTMANN, E., AND J. ZELENA. 1962. Morphological changes in the denervated muscle. Pages 57-102 irz E. GUTMANN, Ed., The Demm~atcd Muscle. CzeckoSlovak Academy of Science, Prague. 7. HARRISON, R. G. 1904. An experimental study of the relation of the nervous system to the developing musculature in the embryo of the frog. Amcv. J. A~zat. 3: 197-220. 8. HOADLEY, L. 1925. The differentiation of isolated chick primordia in chorioallantoic grafts. II. The effect of the presence of the spinal cord, i.e. innervation on the differentiation of the somitic region. f. .!?xp. Zool. 4’2: 143-162. 9. HOOKER, D. 1911. The development and function of voluntary and cardiac muscle in embryos without nerves. J. Exp. Zool. 11: 159-186 .
ANELTROGEXIC 10.
E.
HUNT,
grafts, 11. KOHN,
Amrr. 12. KOHN, for
A. with R. R.
1932. The differentiation of chick limb buds special reference to the muscles. J. I:‘.rp. Zor~l. 1964. Mechanism of protein loss in dencrvation
J. I’atld.
R. R. 1966. Denervation autolysis iri ejitr0. =l~rzrr.
L’ltrasfrm-t.
in chorio-allantoic 62: 57-91. muscle atrophy.
45: 435-447.
muscle atrophy. Studies J. I’nthol. 48: 241-257. 13. I.ocrcsrr~~, R. A., AP\‘D J. BEAUI.ATOS. 1974. Programmed evidence for lysosomes during the normal hrcakdown muscles. J. l~7/frtrstrr~rf. Krs. 46 : 43-62. 14. I.OC.I;SHIX, R. A., AND J. BEAVI.ATON. 1974. Programmed appearance of lysosomes when death of the intersegmental
J.
171
AIUSCLE
of components
required
cell death. Cytochemical of the intersegmental cell death. muscles
Cytochemical is prevented.
Rcs. 46: 63-78.
15. LUFT, J. H. 1971. Ruthenium red and violet. I. Chemistry, purification, methods of use for electron microscopy and mechanism of action. :frmt. Rec. 171: 347368. 16. MAURO, A. 1961. Satellite cell of skeletal muscle fiber. J. Biophys. Biochcm. Cytol. 9: 493. 17. MILEDI, R., AIGD C. R. SI.ATEK. 1969. Electron-microscopic structure of denervated skeletal muscle. Proc. R. Sot. (Lord.) [Biol.] 174: 253-269. 18. MUSCATELLO, U., MARGRETH, A., ASD bi. XLOI~I. 1965. On the differential response of sarcoplasm and myoplasm to denervation in frog muscle. J. Cell Bid. 27: l-24. 19. MUCHMORE, W. B. 1968. The influence of neural tissue on the early development of somitic muscle in ventro-lateral implants in A-lrrrbystonw. J. Exp. Zool. 169: 251-258. 20. NOVIICOFF, P. M., A. B. NOVXOFF, N. QGINTASA, AXD J.-J. HAU\V. 1971. Golgi apparatus, GERL, and lysosomes of neurons in the rat dorsal root ganglia studied by thick and thin section cytochemistry. J. Cc/l Biol. 50: 859-886. 21. N~~ESCII, H. 1968. The role of the nervous system in insect morphogenesis and regeneration. dwm. Rczl. E~~fowd. 13: 2734. 22. PEI~IXGIGWIXO, C., AND C. FRANZINI. 1963. An electron microscopic study of denervation atrophy in red and white skeletal muscle fibers. J. Cell Biol. 17: 327-349. 23. PFEIFER, U., AND H. WITSTHEL. 1972. Zum zeitlichen Ablauf der fur die Elektronenmikroskopie modifizierten Gomori-Reaktion auf saure Phosphatase. Acta Hisforhrm. (Suppl.) 12 : 111-120. 24. POIJIELA, H. 1976. Muscle satellite cells in urodele amphihians: Facilitated identification of muscle satellite cells using ruthenium red staining. J. Exp. 2001. 198 : 57-64. 25. POPIELA, H. 1976. Itc zko limb tissue development in the absence of nerves: A quantitative study. E.rp. Ncrrval. 53 : 214-226. 26. REES, D., A~sD P. N. R. USHER\VOOD. 1972. Effects of denervation on the ultrastructure of insect muscle. J. Ccl/ Sri. 10: 667-682. 27. SHIMADA, Y., D. A. FISCHXIAN, AiYD A. A. MOSCONA. 1967. The fine structure of embryonic chick skeletal muscle cells differentiated irz vitro. J. Cell Biol. 35: 445-453. 28. SHIMADA, Y., AND D. A. FISCH~IAN. 1973. Morphological and physiological evidence for the development of functional neuromuscular junctions ilz vitro. Dezelop. Biol. 31 : 200-225. 29. TIIOMAIES, H. 1974. Untersuchungen iiber die Wirkung der Denervation auf das
172
HEINZ
POPIELA
C. D., H. PWIELA, AND C. S. THOHNTON. 1974. Ultrastructure of the development and subsequent breakdown of muscle in aneurogenic limbs (Ambystoma). J. Exp. Zool. 190: 155-166. 31. ZELENA, J. 1962. The effect of denervation on muscle development. Pages 103-126 irz E. GUTMANN, Ed., Tkc Drucrvuted Nlucle. Czechoslovak Academy of Science, Prague. 30. TWEEDLE,