muscles were incubated with a monoclonal antibody against embryonic myosin heavy chains. Intrafusal fiber types were identified by noting their staining ...
Histochemis
Histochemistry (1988) 88:262271
9 Springer-Verlag l~988tst~
Immunohistochemical demonstration of embryonic myosin heavy chains in adult mammalian intrafusal fibers A. Maier ~, B. Gambke 1, and D. Pette 2, i Department of Cell Biology and Anatomy, University of Alabama at Birmingham, Birmingham, AL 35294, USA 2 Fakultfit fiir Biologic, Universitfit Konstanz, Postfach 5560, D-7750 Konstanz, Federal Republic of Germany Accepted May 6, 1987
Summary. Serial cross sections of rat, rabbit and cat intrafusal fibers from muscle spindles of normal adult hindlimb muscles were incubated with a monoclonal antibody against embryonic myosin heavy chains. Intrafusal fiber types were identified by noting their staining patterns in adjacent sections incubated for myofibrillar ATPase after acid or alkaline preincubation. In rat and rabbit muscle spindles dynamic nuclear bag1 fibers reacted strongly at the polar and juxtaequatorial regions. Static nuclear bag2 fibers reacted weakly or not at all at the polar region, but showed a moderate a m o u n t of activity at the juxtaequator. At the equatorial region both types of nuclear bag fibers displayed a rim of fluorescence surrounding the nuclear bags, while the areas occupied by the nuclear bags themselves were negative. Nuclear chain fibers in rat and rabbit muscle spindles were unreactive with the specific antibody over their entire length. I n cat muscle spindles both types of nuclear bag fibers presented profiles which resembled those of the nuclear bag fibers in the other two species, but unlike in rat and rabbit spindles, cat nuclear chain fibers reacted as strongly as dynamic nuclear bag1 fibers.
Introduction M a m m a l i a n adult intrafusal fibers contain structural components which may be interpreted as being persistent immature features. These include a " m y o t u b e region" (Barker 1948), and the small size of intrafusal fibers relative to extrafusal fibers (Maier and Eldred 1974; Kozeka and Ontell 1981). h n m a t u r e elements may also persist in the contractile apparatus of intrafusal fibers. Recent immunohistochemical studies on adult muscle spindles have shown that static nuclear bag2 (bag2) fibers react with antibodies raised against slow twitch, slow tonic and embryonic myosins (Rowlerson et al. 1985). This can occur if the myosin in bag2 fibers has an antigenic site which is c o m m o n to the various forms of myosin, or if bag2 fibers contain more than one type of myosin. D y n a m i c nuclear bags (bag1) and static bag2 fibers present myofibrillar ATPase (mATPase) profiles which are u n c o m m o n in adult extrafusal muscle (Ovalle and Smith 1972; Maier et al. 1974). These profiles Dedicated to Professor Dr. T.H. Schiebler on the occasion of his 65th birthday * To whom offprint requests should be sent
may be an expression of specific contractile properties and, consequently, may indicate a type of myosin unique to intrafusal fibers. The purpose of the current study was to examine which intrafusal fiber types contain immature forms of myosin, and whether this is a constant feature a m o n g rat, rabbit and cat muscle spindles. Polar, juxtaequatorial and equatorial regions were inspected separately to determine whether or not the distribution of myosins is uniform along the length of the intrafusal fibers.
Materials and methods Fifteen or more normal half (polar region to equator) or full-length muscle spindles each from adult rat, rabbit and cat muscles were examined. Extensor digitorum longus, tibialis anterior, medial gastrocnemius and soleus muscles were obtained from animals killed with an overdose of sodium pentobarbital. The muscles were dissected and cut into small pieces, frozen in melting isopentane at - 1 5 9 ~ and stored at - 7 0 ~ until used. Alternate serial cross sections were cut for immunohistochemistry(4 ~tm) and histochemistry (8 gin) on a freezing microtome ( - 25~ C), transferred to cover slips and air-dried for 30 rain. Sections used for immunohistochemistry were incubated with a 4% solution of bovine serum albumin (BSA) for 10 rain. At that time the BSA was drained off and 10 gl of the primary antibody, diluted 1:1200 from a stock solution of 27 mg/ml in phosphate buffered saline (PBS) was applied to the sections. This monoclonal antibody, termed 2B6, was prepared and described by Gambke and Rubinstein (1984). It is specific for embryonic myosin heavy chains and does not cross-react with adult rat and cat slow or fast myosins (Gambke and Rubinstein 1984). Incubation was carried out for one hour. The incubation medium was then discarded and the sections were washed three times in PBS. Thereafter, 10 gl of the secondary antibody, fluorescein-conjugated sheep anti-mouse IgG (U.S. Biochemical Corp. Cleveland, OH 44128) was applied. The antibody was reconstituted with 1 ml sterile saline per 10 mg of lyophylized powder and diluted 1:50 with PBS. Sections were incubated with the secondary antibody for 30 rain. At the end of this period the sections were washed three more times with PBS and then mounted on glass slides with a 7: 3 mixture of PBS and glycerol. Control sections were treated like any other section, except that incubation with the primary antibody was omitted. All incubations were carried out in a moist chamber at 37~ C. Fluorescence was observed with a Leitz Ortholux II microscope, using an I2 filter block. Photographic images were recorded on Kodak Tri-X film rated at ASA 1600 and developed in KodakD-76 diluted 1 : 1 for 11 rain. Histochemical staining for acid-stable (pH 4.7 preincubation) and alkaline-stable (pH 10.1 preincubation) mATPase (Brooke and Kaiser 1970) was performed on sections which were in series with the sections incubated with the 2B6 antibody. This was done to permit typing of the intrafusal fibers that had been identified in the immunohistochemical sections.
268
269 Anterior latissimus dorsi (ALD) muscles from three eightweek-old chicks were also examined histochemically and immunohistochemically as described above. The ALD consists almost entirely of slow tonic extrafusal fibers (Pierobon Bormioli et al. 1980), thus providing a control as to whether the 2B6 antibody crossreacts with tonic myosin. Sections of chicken extensor digitorum longus (EDL) were processed along with the ALD to allow comparison with a more typical somatic muscle.
A
P
JE
E
JE
P
BI[
r
B2 Ch l
0
I
Results
Intrafusal fibers in control sections which had been incubated without the primary a n t i b o d y were non-reactive. Consequently, the various degrees o f staining which were visible when the specific a n t i b o d y was included in the incubation medium were taken to represent relative amounts o f embryonic myosin heavy chains. The occurrence of significant a m o u n t s of reaction p r o d u c t was restricted to intrafusal fibers, while other components o f the muscle spindle and extrafusal tissue were negative or stained only very weakly. Observations in intrafusal fibers were limited to their intracapsular portions. A t the p o l a r region of b o t h rat and rabbit muscle spindles bag1 fibers reacted strongly with the a n t i b o d y against embryonic myosin, but bag2 fibers reacted much less or not at all (Figs. 1-3). A t the j u x t a e q u a t o r bag2 fibers showed an increase in reactivity relative to bag~ fibers and to their own p o l a r level (Fig. 4). A t the equator b o t h types o f nuclear bag fibers presented rings o f staining which were o f equal or near equal intensity. The rings surrounded the nuclear bags; however, the nuclear bags themselves were negative (Figs. 5-7). The difference in staining between rat bag1 and bag2 fibers at the j u x t a e q u a t o r is seen in Fig. 8. Nuclear chain fibers were unreactive for their entire length (Figs. 3, 4, 7, 8), except that at the equator they might stain very faintly (Fig. 5). M o s t of the rat and rabbit spindles surveyed contained four intrafusal fibers, but the same pattern o f staining was observed regardless o f the size of the axial bundle. Cat muscle spindles deviated from this pattern in one significant respect. Nuclear chain fibers from the polar region to the equator reacted as strongly as the bagl fibers with the antibody against embryonic myosin, and at the equator they exhibited a ring o f reaction product a r o u n d the nuclear chains (Figs. 9 11). Otherwise, the reactions were similar to rat and rabbit spindles. The bag2 fiber was the weakest staining unit o f the axial bundle, and it reacted
B
P
JE
E
JE
P
BI[ B2
Ch
mm LESS
REACTIO.
PnOOUCr
MORE
Fig. 13. Schematic representation of the distribution of relative amounts of anti-embryonic myosin heavy chain reaction product in the intracapsular portions of intrafusal fibers in cat (A), and rat and rabbit (B) muscle spindles. The dark elliptic areas at the equator represent the nuclear bags of dynamic nuclear bag~ (B1), static nuclear bagz (B2) and nuclear chains (Ch), respectively. E, equator; JE, juxtaequator; P, pole much less (or was negative) at the p o l a r region than at the equator (Figs. 9 and 11). Both types o f nuclear bag fibers displayed the characteristic ring o f fluorescence at the equator (Fig. 11). Cross-sections o f chick A L D and E D L muscles along with r a n d o m l y chosen cross-sections o f rabbit and cat soleus muscle spindles were incubated on the same coverslip for the purpose o f comparing the reactivity with the antibody among these tissues, In the A L D only weak staining was observed in extrafusal fibers, while most o f intrafusal fibers fluoresced strongly (Fig. 12). The intrafusal fibers o f the cat and rabbit muscle showed their characteristic staining pattern. Chicken E D L extrafusal fibers stained very weakly or were negative, and intrafusal fibers either were negative or stained moderately (not shown). The reactions o f cat, rat and rabbit intrafusal fibers with the specific a n t i b o d y against the embryonic myosin heavy chain are summarized in Fig. 13.
Figs. 1-12. Histochemical staining for myofibrillar actomyosin ATPase (mATPase) and immunohistochemical reaction with anti-embryonic myosin heavy chain in muscle spindles of rabbit (Figs. 1-6), rat (Figs. 7 and 8) cat (Figs. 9--11) and chicken (Fig 12) muscles. Large arrowheads point to bag2 fibers, medium-sized arrowheads to bag1 fibers and small arrowheads to chain fibers Figs. 1-6. Serial cross sections of the same rabbit tibialis anterior muscle spindle incubated at the polar region for mATPase, pH 4.7 preincubation (Fig. 1) and pH 10A preincubation (Fig. 2), incubated with anti-embryonic myosin heavy chain at the polar (Fig. 3), juxtaequatorial (Fig. 4) and equatorial (Fig. 5) regions, and incubated for mATPase (pH 10.1 preincubation) near mid-equator (Fig. 6). Bar is 10 gm Figs. 7 aml 8. Serial cross sections of rat medial gastrocnemius muscle spindle at the equatorial (Fig. 7) and juxtaequatorial (Fig. 8) regions incubated with anti-embryonic myosin heavy chain. Bars are 10 gm Figs. 9-11. Serial cross sections of the same cat soleus muscle spindle at the polar region incubated with anti-embryonic myosin heavy chain (Fig. 9) and incubated for mATPase, pH 4.7 preincubation (Fig. 10), and at the equator incubated with anti-embryonic myosin heavy chain (Fig. 11). Bar is 10 gm Fig. 12. Cross-sectioned chicken anterior latissimus dorsi extrafusal fibers (EF) and intrafusal fibers (arrows) incubated with anti-embryonic myosin heavy chain. Bar is 20 pm
270 Discussion
Results of this and other studies show that in adult mammalian muscles isoforms of myosin occur in intrafusal fibers which are uncommon in extrafusal fibers of skeletal muscles. Depending on the species, the anti-embryonic myosin antibody used here reacts only with nuclear bag fibers (rat, rabbit) or with nuclear bag and nuclear chain fibers (cat). It does not react with adult slow and fast myosins in rat and cat extrafusal fibers (Gambke and Rubinstein 1984), and it only cross-reacts slightly with rabbit fast-twitch extrafusal fibers. The finding by Rowlerson et al. (1985) that cat nuclear chain fibers react strongly with anti-embryonic myosin agrees with our data. However, their observation that the bag~ fiber reacts very little or not at all with antiembryonic myosin is at variance with our results. The dissimilar findings may in part result from the respective specificity or non-specificity of the antibodies for one particular epitope, due to the different sources (rat muscle, bovine muscle) from which the antigens in the two studies were obtained (Sartore etal. 1982; Gambke and Rubinstein 1984). Or, separate epitopes may reside on different parts of the myosin heavy chains and are not equally accessible to the antibodies. Pierobon Bormioli et al. (1980) have demonstrated that in rodents at the encapsulated portion of the polar region bag2 fibers react less than bag~ fibers with antichicken ALD (tonic) myosin. The current study has shown that this also applies to anti-embryonic myosin heavy chain. However, the reaction reported by Pierobon Bormioli et al. (1980) and our results can not be attributed to the same myosin isoform because the antibody 2B6 cross-reacts very little with tonic extrafusal fibers (Gambke and Rubinstein 1984), but reacts strongly with most intrafusal fibers in chicken ALD (Fig. 12). Polyclonal antibodies against chicken heart myosin (te Kronnie et al. 1981, 1982) react with mammalian slow-twitch extrafusal fibers and produce in intrafusal fibers a staining pattern which is similar to the reactions which we observed with the anti-embryonic myosin heavy chain in rats and rabbits. Rowlerson et al. (1985) have proposed that bag2 fibers have a distinct form of myosin which is similar to that of slow-twitch extrafusal fibers, and Pierobon Bormioli et al. (1980) have suggested that nuclear bag fibers contain a myosin isoform which is similar to slow tonic myosin. The greater staining at the polar region of bag~ fibers relative to bag2 fibers after incubation with the 2B6 antibody would fit with its reported slower contractile spead (Boyd 1976) if this antibody recognizes an antigenic site on a slow myosin isoform. However, the occurrence in cat chain fibers of 2B6 reaction product is at odds with this presumed association because chain fibers are the fastest contracting intrafusal fibers (Boyd 1976), and because there was no significant crossreaction of 2B6 with fasttwitch extrafusal fibers. In addition, a strong positive reaction with a monoclonal anti-fetal myosin antibody has recently been reported by Schiaffino et al. (1986) for presumed nuclear chain fibers in spindles of normal adult muscles in the human. The primary sensory axon induces muscle spindle development at what will become the equator. The bag2 fiber appears first, the bagz fiber next and chain fibers last. Milburn (1973, 1984) has suggested that there is a gradual decrease in the morphogenetic influence of the primary sensory axon after it has induced spindle development, and
that this decrease over time is expressed by fewer central nuclei being present in the later appearing chain fibers than in the earlier formed nuclear bag fibers. It remains to be seen whether this presumed change in neural influence has an effect on the continuous expression of embryonic myosin throughout adult bag1 fibers, chain fiber of cats, and at the equator and juxtaequator of bag2 fibers. During development of extrafusal fibers there may be a sequential appearance and disappearance of embryonic and neonatal myosins before fast and slow adult myosin isoforms are expressed. Whalen et al. (1981) have suggested that the conversion from embryonic to adult myosin in extrafusal fibers is related to the elimination of polyneural innervation. If this concept holds for intrafusaI fibers, one would expect that single poles of cat chain fibers, and of cat, rat and rabbit bag1 fibers, have retained polyneural innervation in the adult stage because they contain embryonic myosin. Interestingly enough, embryonic myosin heavy chain expression has recently been shown at the m R N A and immunohistochemical level in in rat extraocular muscles that contain multiply innervated extrafusal fibers (Wieczorek et al. 1985); however, it has yet to be demonstrated if the existence of embryonic myosin in these fibers is related to polyneural innervation. Immunohistochemical studies performed on chronically stimulated fast-twitch rabbit muscle indicate that embryonic myosin may be expressed in regenerating fibers (Maier et al. 1986) as well as in existing fibers undergoing fast-to-slow conversion (Maier, Pette and Schiaffino, unpublished results). These observations fit with the immunohistochemical demonstration of embryonic myosin in IIC fibers of human dystrophic muscles (Schiaffino et al. 1986). Therefore, embryonic myosin, or an isoform sharing common antigenic epitopes with the latter, is not only transiently expressed during muscle fiber development (Whalen et al. 1981) and regeneration (Sartore et al. 1982), but may also exist in specific adult intrafusal and extrafusal fibers. Walro and Kucera (1984, 1985) have presented evidence that an appreciable number of rat fusimotor axons coinnervate bag1, and bag2 and/or chain fibers, thus leaving this species with an incomplete separation of static and dynamic motor systems. On the other hand, in cat receptors the dynamic and static motor innervation is virtually always confined to bag~ fibers, and to bag2 and chain fibers, respectively (Boyd 1981; Boyd and Gladden 1985; Arbuthnott et al. 1982). Since the embryonic myosin profiles of nuclear bag fibers do not vary between cats and rats, but their pattern of motor innervation does, dynamic or static motor innervation should not be a major factor in the persistence or absence of embryonic myosin in adult intrafusal fibers. Te Kronnie et al. (1982) have demonstrated that in 21-day old rats de-efferented at birth, bag1 and bag2 fibers react with anti-heart myosin, while in 21-day old control animals only bag1 fibers react. This suggests that motor innervation may play a role in the expression of myosin isoforms in mature intrafusal fibers. However, it is not clear if de-efferentation at birth will remove all influence from motor innervation on development because, at least in cats (Milburn 1984) and mice (Kozeka and Ontell 1981), fusimotor innervation of the spindle occurs prior to birth. The histochemical mATPase reaction at the intracapsular portions of the poles of nuclear chain fibers appears to be fairly uniform in mammals (Ovalle and Smith 1972; Maier et al. 1974; Maier et al. 1976; Kucera 1981). How-
271 ever, embryonic myosin heavy chain content at that region varies markedly, at least between rats and rabbits on one hand, and cats on the other (Figs. 3 and 9). It appears likely, therefore, that myosin polymorphism in intrafusal fibers is not detectable with the m A T P a s e reaction under the conditions used in this study. Previous immunohistochemical studies on intrafusal fibers have been primarily limited to observations at the polar region (Pierobon Bormioli et al. 1980; te K r o n n i e et al. 1981, 1982; Rowlerson et al. 1985). We have now shown that myosin reacting with an anti-embryonic myosin heavy chain antibody is also present in the ring of myofibrillae which surrounds nuclear bags, and in cat spindles in addition also in the ring of myofibrillae which surrounds nuclear chains. The presence of embryonic myosin and the absence of fusimotor endings at the equator thus appears to de-emphasize the importance of motor innervation to the expression of myosin isoforms. In this context, it should also be taken into account that additional factors such as stretch may play a regulatory role in the expression of myosin isoforms. K e n n e d y et al. (1986) have recently shown that stretch is an i m p o r t a n t factor in the control of slow myosin isoform expression in extrafusal fibers of the latissimus anterior dorsi muscle of chicken. In summary, adult extrafusal fibers of muscles of the limbs contain fast or slow isoforms of myosin and normally lack embryonic myosin. In contrast, adult intrafusal fibers clearly display embryonic myosin or myosin isoform(s) that strongly react with a monoclonal antibody directed against embryonic myosin heavy chain. However, the distribution of embryonic myosin varies between nuclear bag and chain fibers in a species-specific manner, and in a given species one fiber type (chain) or a fiber segment (bag2) may lack it or present only low a m o u n t s of it (bag2). These variations seem to be independent of differences in dynamic or static motor innervation.
Acknowledgements. This study was supported by Deutsche Forschungsgemeinschaft, Sonderforschungsbereich t56, "Mechanismen zellul/irer Kommunikation", a grant from the Muscular Dystrophy Association of America, and by the University of Alabama at Birmingham. References Arbuthnott ER, Ballard K J, Boyd IA, Gladden MH, Sutherland FI (1982) The ultrastructure of cat fusimotor endings and their relationship to foci of sarcomere convergence in intrafusal fibers. J Physiol 331:285-309 Barker D (1948) The innervation of the muscle-spindle. Q J Microsc Sci 89:143-186 Boyd IA (1976) The response of fast and slow nuclear bag fibres and nuclear chain fibres in isolated cat muscle spindles to fusimotor stimulation, .and the effect of intrafusal contraction on the sensory endings. Q J Exp Physiol 61:203-254 Boyd IA (198/) The action of the three types of intrafusal fiber in isolated cat muscle spindles on the dynamic and length sensitivities of primary and secondary endings. In: Taylor A, Prochazka A (eds) Muscle receptors and movements. Macmillan, London, pp 17-32 Boyd IA, Gladden MH (1985) Morphology of mammalian muscle spindles. In: Boyd IA, Gladden MH (eds) The muscle spindle. Macmillan, London, pp 3 22
Brooke MH, Kaiser KK (1970) Muscle fiber types: How many and what kind? Arch Neurol 23:369-379 Gambke B, Rubinstein NA (1984) A monoclonal antibody to the embryonic myosin heavy chain of rat skeletal muscle. J Biol Chem 259:12092-12100 Kennedy JM, Kamel S, Tambone WW, Vrbovfi G, Zak R (1986) The expression of myosin heavy chain isoforms in normal and hypertrophied chicken slow muscle. J Cell Biol 103:977-983 Kozeka K, Ontell M (1981) The three-dimensionalcytoarchitecture of developing murine muscle spindles. Dev Biol 87:133 147 Kronnie G te, Donselaar Y, Soukup T, van Raamsdonk W (1981) Immunohistochemical differences in myosin composition among intrafusal muscle fibres. Histochemistry 73 : 65-74 Kronnie G te, Donselaar Y, Soukup T, Zelenfi J (1982) Development of immunohistochemical characteristics of intrafusal fibres in normal and de-efferented rat muscle spindles. Histochemistry 74:355-366 Kucera J (1981) Histochemical profiles of cat intrafusal muscle fibers and their motor innervation. Histochemistry 73:397~418 Maier A, Eldred E (1974) Postnatal growth of extra- and intrafusal fibers in the soleus and medial gastrocnemius of the cat. Am J Anat 141:161-177 Maier A, Simpson DR, Edgerton VR (1974) Histochemical profiles of guinea-pig intrafusal fibers in normal muscle and after denerration, cordotomy and tenotomy. Histochem J 6:665-677 Maier A, Simpson DR, Edgerton VR (1976) Histological and histochemical comparisons of muscle spindles in three hind limb muscles of the guinea pig. J Morphol 148 : 185-191 Maier A, Gambke B, Pette D (1986) Degeneration-regeneration as a mechanism contributing to the fast to slow conversion of chronically stimulated fast-twitch rabbit muscle. Cell Tissue Res 244: 635-643 Milburn A (1973) The early development of muscle spindles in the rat. J Cell Sci 12:175-195 Milburn A (1984) Stages in the development of cat muscle spindles. J Embryol Exp Morphol 82:177-216 Ovalle WK, Smith RS (1972) Histochemical identification of three types of intrafusal fibers in the cat and monkey based on the myosin ATPase reaction. Can J Physiol Pharmacol 50:195 202 Pierobon Bormioli S, Sartore S, Vitadello M, Schiaffino S (1980) '~Slow" myosins in vertebrate skeletal muscle. An immunofluorescence study. J Cell Biol 85:672-681 Rowlerson A, Gorza L, Schiaffino S (1985) Immunohistochemical identification of spindle fibre types in mammalian muscle using type-specific antibodies to isoforms of myosin. In: Boyd IA, Gladden MH (eds) The muscle spindle. Macmillan, London, pp 29 34 Sartore S, Gorza L, Schiaffino S (1982) Fetal myosin heavy chains in regenerating muscle. Nature 298 : 29zV296 Schiaffino S, Gorza L, Dones I, Cornelio F, Sartore S (1986) Fetal myosin immunoreactivity in human dystrophic muscle. Muscle Nerve 9:51-58 Walro JM, Kucera J (1984) Nonselective motor innervation of intrafusal fibers in muscle spindles of the rat. Exp Brain Res 56:187 192 Walro JM, Kucera J (1985) Motor innervation of intrafusal fibers in rat muscle spindles : Incomplete separation of dynamic and static systems. Am J Anat 173:55 68 Whalen RG, Sell SM, Butler-Browne GS, Schwartz K, Bouveret P, Pinset-Hgrstr6m I (1981) Three myosin heavy-chain isozymes appear sequentially in rat muscle development. Nature 292:805-809 Wieczorek DF, Periasamy M, Butler-Browne GS, Whalen RG (1985) Co-expression of multiple, myosin heavy chain genes, in addition to a tissue-specific one, in extraocular musculature. J Cell Biol 101:618-629
