Specific Metabolic Properties of Rat Oculorotatory Extraocular Muscles Can Be Linked to Their Low Force Requirements

Abstract

purpose. To gain insight into the metabolic pathways of oculorotatory extraocular muscle (EOM) fiber types at the cellular level to explain their high fatigue resistance, rapid contraction, and low force output.

methods. In consecutive sections of adult rat EOMs, the cross-sectional area (CSA) was calculated, and the activities of succinate-dehydrogenase (SDH) and α-glycerophosphate dehydrogenase (GPDH) were measured by quantitative histochemistry of different fiber types classified by the myofibrillar adenosine triphosphatase (mATPase) staining pattern.

results. In the orbital regions, type 1 (fast) fibers were present, showing small CSA, medium SDH, and low GPDH activity. The type 2 (slow) fibers exhibited extremely small CSA and low SDH and GPDH activity. In the global region, fast types 3 to 5 fibers were found, forming a continuum with an inverse correlation between CSA and enzyme activity. SDH and GPDH activity showed an unusual positive relationship in contrast to the skeletal muscle fiber types. The type 6 fibers were slow forming a close and clearly separated group with medium CSA and extremely low SDH and low GPDH activity.

conclusions. Muscle fibers in adult rat EOMs show unique metabolic properties not seen in other skeletal muscles, covering their extraordinary functional demands. It can be speculated that the EOMs embedded within the orbit material do not need to perform with high force, and therefore they could develop extensive systems that ensure both fatigue resistance (many mitochondria) and fast contraction with high mATPase activity (a well-developed sarcotubular system).

The specialization of skeletal muscle fibers is reflected by their contractile and metabolic properties (for review and references, see Refs. ^{1 2 3 4} ). On the basis of mechanical, biochemical, enzyme, and immunohistochemical studies, mammalian skeletal muscle fibers are subdivided into four discrete fiber types: slow-twitch oxidative type I, fast-twitch oxidative glycolytic IIA, fast-twitch glycolytic IIB, and IIX/D fibers possessing mixed characteristics between IIA and IIB fibers. As demonstrated by quantitative enzyme histochemistry, fibers with a predominantly oxidative metabolism are more or less resistant against fatigue, and they have smaller diameters, the opposite being true of fibers with a high glycolytic capacity. ^{3 4}  

The six oculorotatory extraocular muscles (EOMs) are structurally and functionally unique among the cross-striated muscles of vertebrates. They exhibit a diverse repertoire of actions, including steady eyeball fixation, slow vergence movements, pursuit movements at various speeds, and high-speed saccades over a wide range of angles. ⁵ In comparison to isolated fast-twitch limb muscles, ^{6 7} isolated mammalian EOMs show a low force output and are characterized by an exceptionally high shortening velocity, a much more rapid time course of isometric twitch contractions, a higher fusion frequency, and a lower twitch-to-tetanus ratio combined with a higher fatigue resistance (cat, ^{8 9 10 11 12} rabbit, ^{13 14 15} monkey, ¹⁶ rat, ^{17 18} and mouse ¹⁹ ). Beside fast-twitch fibers, EOMs contain a minority of multiply innervated slow-tonic muscle fibers, common in skeletal muscles of lower vertebrates, but very rare in those of mammals ^{20 21} (except muscle spindles ²² ) that produce only slow local contractions and contract extremely slowly. Tonic fibers are possibly responsible for the fact that EOMs display some properties normally observed only in muscles of lower vertebrates or in neonatal or denervated mammalian muscles (e.g., acetylcholine contractures). ²³ EOMs also possess some unusual structural features: They contain very thin fibers organized in layers with abundant capillaries, ^{24 25 26} their motor unit size is very small, ^{10 11 12} and many fibers branch and form muscle-to-muscle junctions. ²⁷ With electron microscopy, it has been shown that EOMs unusually contain many mitochondria and an extensive sarcotubular (SR-T) system. ²⁵ These functional and structural intricacies are reflected in complex muscle fiber types described in mammalian EOMs. Their fibers have been classified into six different types on the basis of their diameters, their arrangement in the muscle, their enzyme- and immunohistochemistry, and their innervation (for review, see Refs. ²⁸ , ²⁹ ). This system of classification differs markedly from that normally used to classify functionally different limb muscle fibers (described earlier). Therefore, some (or perhaps all) fiber types in EOMs do not correspond to the fiber types normally observed in skeletal muscles (for review, see Ref. ³⁰ ). Recently, it was shown in the rat that the expression of a lot of genes in EOMs is quite different in comparison to that of the anterior tibialis muscle. ^{31 32} Furthermore, EOMs exhibit a distinct allotype and a different sensitivity to disease. The muscles are spared in Duchenne’s muscular dystrophy, despite the widespread involvement of other skeletal muscles. On the other hand, they are early and prominent targets in chronic progressive external ophthalmoplegia, myasthenia gravis, Graves’ ophthalmopathy, and mitochondrial myopathies. ^{33 34}  

Early investigations of extracts of whole EOMs have shown that they have high activities of oxidative and glycolytic enzymes. ^{35 36 37} However, there is no detailed information at the cellular level about the metabolic pathways in individual fibers in the different fiber types of EOMs. The first purpose of this study was therefore to correlate the muscle fiber types in six oculorotatory EOMs with their metabolic (oxidative and glycolytic capacity) and structural (fiber size) characteristics. The second purpose was to discuss our hypothesis that EOMs exhibit specific metabolic, ultrastructural, and contractile characteristics in comparison with limb skeletal muscles, because they do not need high tension (force output). 

Methods

Animals and Tissue Processing

The Ethical Principles and Guidelines for Scientific Experiments on Animals were adhered to throughout the study, and the animals were treated according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The experiments were approved by the Regierungspräsidium Leipzig. 

Eight Wistar rats of both sexes (2 months old; body weight, 170 ± 30 g), anesthetized with ether and euthanatized by decapitation, were used in the study. The whole orbital content, including the eyeball, the optic nerve, and the EOMs (the oculorotatory recti and obliqui, the levator palpebrae, and the retractor bulbi muscles), without the Harderian gland were excised, stretched to the in situ length, quickly frozen in liquid nitrogen and stored at −80°C until further use. 

Enzyme Histochemistry and Morphometric Evaluations

The composed frozen blocks were transferred into a cryostat (model 1800; Reichert Jung, Vienna, Austria), and serial cross-sections (10 μm) were cut from the muscle mid-belly at −20°C through the composite muscle specimens immediately behind the eye ball, perpendicular to the optic nerve (Fig. 1) , stained, and analyzed by cytophotometry. This procedure ensures that the thickness of the section and the staining procedure is much the same in all muscles and all fibers of the block. To compare the staining of sections on different slides, a standardization factor was used by setting the absorbance of an unspecific background staining at 0. ⁴ The sections were stained for myofibrillar actomyosin adenosine triphosphatase (mATPase, EC 3.6.1.3; as a marker of contractility) ⁴ at pH 9.4 ³⁸ or using acid (pH 4.3, 4.4, or 4.5) and alkaline (pH 10.4) preincubations. ³⁹ Consecutive sections were stained for succinate dehydrogenase (SDH, EC 1.3.5.1; as a marker of oxidative metabolism) and glycerophosphate dehydrogenase (GPDH, EC 1.1.99.5; as a marker of glycolytic metabolism). ⁴⁰ A control reaction was performed on serial sections in absence of the substrate. The muscle fibers were classified according to their qualitative histochemical properties, ²⁸ and their mATPase and their metabolic enzyme activities were measured cytophotometrically. The cross-sectional area (CSA) of the fibers was determined planimetrically (Imaging System KS 100; Kontron, Eching, Germany). All data are expressed as the mean ± SD. Commercial software was used for graphic demonstrations and statistical analysis (Sigma-Plot; Systat Software, Inc., San Jose, CA). 

Cytophotometry

End point measurements were made with a computer-controlled microscope photometer with a scanning table (MPM 200; Carl Zeiss Meditec, Oberkochen, Germany). The mean absorbance (MA) of the final reaction product of the SDH or GPDH enzyme reaction was determined and taken as a measure of the relative enzyme activity. The correlation of the MA with corresponding enzyme activity was already shown, and the cytophotometric method was established and described as a tool in metabolic fiber typing. ^{4 40} A measuring spot 2.5 μm in diameter, scanning steps of 3 μm, and objective magnification ×20 were used. In each muscle section, we selected randomly 150 fibers arranged in groups (lateral rectus, medial rectus, superior rectus, and a few of the superior oblique) including all fiber types. On serial cross-sections we determined fiber area and the relative enzyme activity of mATPase, SDH, and GPDH. In each muscle and every enzyme reaction, three to five measurements were made in the same fiber for averaging and up to five serial sections were analyzed to exclude staining variability. 

Results

The oculorotatory EOMs of the rat—the recti and obliqui muscles—are, as in other vertebrates, organized in two layers (Fig. 1 ; Table 1 ). The outer orbital region (OR), composed of very thin fibers, is adjacent to the bony walls of the orbit and the inner global region (GR), containing mostly fibers of larger caliber, is oriented toward the eyeball (globe and optic nerve). According to their diameter and location, enzyme histochemistry, and pattern of innervation, as well as ultrastructure and MyHC isoform expression, the muscle fibers in EOMs have been classified into six (Table 1) ^{28 29} (and in one study, into seven ²⁰ ) fiber types. 

Orbital Region

There are two fiber types in the OR of the EOMs of the rat (Figs. 1 2A 2B ; Table 1 ). Most of the fibers in this region are type 1 (fast), showing a small CSA, combined with medium SDH (Fig. 3A)or mATPase (Fig. 3D)levels and low GPDH activity (Fig. 3C) . The rest were type 2 fibers (slow) exhibiting an extremely small CSA (the smallest of all EOM fibers in the rat) combined with low SDH, GPDH, and mATPase activity (Figs. 3A 3C 3D ; Table 1 ). 

Both fiber types formed distinctly separate groups, as shown by the relation between the parameters measured except the relation between CSA and GPDH activity (Fig. 3C) . Also according to the relation between SDH or GPDH and mATPase activity, the fiber types were distinctly separate, but there was no significant correlation of these parameters within or between the groups (data not shown). 

Global Region

In the GR of the rat EOMs, four fiber types were found (Figs. 1 2C 2D ; Table 1 ). According to their mATPase staining patterns (Fig. 1C) , the types 3 to 5 fibers, forming approximately 90% of all fibers, belonged to the fast category (for details, see Table 1 ). However, they exhibited marked differences in the CSA and the activity of the mATPase and that of the metabolic enzymes. Type 3 fibers were characterized by a small CSA and a lower activity of the mATPase combined with a high activity of both SDH and GPDH enzymes, whereas the opposite was true of type 5 fibers. Type 4 fibers possessed intermediate properties (Table 1) . The two-dimensional scatterplots of relationships between these parameters shows that fiber types 3, 4, and 5 were arranged in more or less compact groups, although with considerable overlap (Figs. 4) . 

For fiber types 3 to 5, there was an inverse correlation between the CSA and the SDH activity (r = −0.85, P < 0.001; Fig. 4A ) and, in contrast to limb skeletal muscles, also between the CSA and the GPDH activity (r = −0.62, P < 0.001; Fig. 4C ). The relationship between the SDH and the GPDH activity was positive (r = 0.68, P < 0.001; Fig. 4B ), again in contrast to limb skeletal muscle fiber types. This means that in one and the same cell, a high oxidative activity is combined with a high glycolytic activity. There was also a positive relation between the mATPase and the CSA (r = 0.64, P < 0.001; Fig. 4D ) and a negative one between the mATPase and the SDH or the GPDH activity (r = 0.55, P < 0.001 or 0.32; P < 0.01, respectively). 

According to their histochemical mATPase-reactions (Fig. 1) , approximately 10% of the fibers of the GR were type 6 (slow) fibers (Table 1) . They were characterized by a medium CSA and extremely low SDH, GPDH, and mATPase activity. According to their metabolic properties and in comparison to the fast fibers of the same region, they formed a very compact and clearly separate group (Fig. 4) . Also, the relation between metabolic enzymes (SDH and GPDH) and mATPase activity showed the distinct separation of the fast group (types 3–5) and the slow type 6 fibers. 

Our results show, in summary, the existence of interrelationships between CSA and SDH or GPDH activity. They demonstrate striking metabolic differences among muscle regions and different fast and slow fiber types and, on the other hand, the close interrelationships among the different fast fiber types. 

Discussion

General Considerations

Our investigation shows, for the first time, the quantitatively measured characteristics of oxidative and glycolytic metabolism of EOM fibers at the cellular level. It demonstrates that the high levels of oxidative and glycolytic pathways can be combined in one and the same fiber of an EOM. Our investigation thus lifts the semiquantitative histochemical data to the level of biochemistry, as there is a close correlation between the cytophotometric data and biochemistry. ⁴ These findings enabled us to hypothesize about the specific relation among muscle fiber force, fatigue resistance, and shortening velocity. Such characteristic adaptation can then subserve the specific physiological requirements of the EOM function. 

Levels of oxidative and glycolytic metabolism were studied earlier by biochemical and histochemical methods. The biochemical studies of the whole muscle extracts have shown that EOMs combine a high glycolytic and a high oxidative metabolism, ^{35 36 37} but they did not correlate measured activity with individual fiber types. On the other hand, histochemical studies correlated metabolic enzyme characteristics and fiber type, but they gave only semiquantitative data (from − to ++++, see Refs. ²⁸ , ²⁹ ). 

Investigations of whole rabbit muscle extracts ^{35 36 37} have demonstrated that the oxidative capacity of the EOM is about nine times higher than in the fast-twitch adductor magnus muscle and two times higher than in the slow-twitch soleus muscle. Only the heart muscle has higher oxidative enzyme activity. This property is linked to an extraordinary high blood flow in the EOMs. It has been recorded ⁴⁶ in the cat that the average blood flow in the six EOMs exceeded that of all other skeletal muscles investigated and was surpassed only by blood flow in the myocardium. Furthermore, it was demonstrated in primates and sheep that the blood flow per gram of tissue of the EOM was 10-fold higher than in gastrocnemius and soleus muscles. ⁴⁷  

In this study, we tried to demonstrate the different metabolic pathways of EOM fibers on the cellular level using the method of cytophotometry, as it was used in rat and llama skeletal muscles, ^{3 48 49 50} but not in EOMs. For that purpose, we used the classification scheme (see Table 1 ) that is generally accepted for fiber typing of mammalian EOM at present and that is based on the pioneering works in rats, cats, rabbits, and sheep ^{24 51 52} (for review, see Refs. ²⁸ , ²⁹ ). 

Fibers of the Orbital Region

We confirm that there are two muscle fiber types in this part of the EOM (Table 1) , the focally innervated type 1 (expressing embryonic/neonatal and EOM-specific MyHCs) and the multiply innervated type 2 (expressing embryonic/neonatal and slow-twitch MyHC). Both fiber types are very thin, they lack correlation between enzyme activities and the fiber diameter and, except for GPDH activity, they are metabolically clearly separate. A better developed capillary network in the OR than in the global one was found in cats ⁵³ and rats. ⁵⁴ Corresponding quantitative data are available from cats, ²⁴ rabbits, ²⁴ and humans. ²⁵ In conformity with the high amount of type 1 fibers (85%), rich in mitochondria and therefore with a high oxidative metabolism, the vascular density of the OR is approximately 1.4 times higher than in the GR. 

Electromyographic recording in humans have shown that the fibers of the OR are almost continuously active throughout the entire oculomotor range, whereas most of the fibers of the GR become silent only slightly out of the field of action. ⁵⁵ This difference in physiological activity may account for the high mitochondrial content, the high oxidative enzyme activity and the higher vascularity of the OR. Although the six fiber types were found in EOM of all analyzed mammalian species, it was found that the fibers of the OR exhibit a higher mitochondrial content in those species with a more complex eye movement (oculomotor behavior). ²⁸ Therefore, the rat has correspondingly less developed pulleys at the EOM and fibers of the OR with a lower mitochondrial content than observed in primates. 

Fibers of the GR

We distinguished four muscle fiber types in the GR of EOMs: the focally innervated types 3, 4, and 5 and the multiply innervated type 6 (Table 1) . Some of the type 3 to 5 fibers expressed MyHC isoforms of skeletal muscles (IIa, IId/x, and IIb), but all fibers exhibited positive reactions with antibodies against the specific EOM-MyHC. The fiber type 6 expressed the slow-twitch MyHC only. Metabolically the fiber types 3 to 5 formed a continuum with considerable overlapping and a negative correlation of SDH and GPDH activity and the fiber caliber. On the other hand, the type 6 fibers formed clearly separate groups according to their enzyme activity and diameters. 

One of the most important findings of this study is the positive correlation between the SDH and the GPDH activity. This finding means that in one and the same cell, a high oxidative and a high glycolytic metabolism is observable, a characteristic never found in skeletal muscle fibers, where a negative correlation is found—a high oxidative metabolism is combined with a low glycolytic one and vice versa. ^{41 56} Therefore, the EOMs combine an ability of high energy consumption necessary for a fast isometric contraction or a high shortening velocity with a high resistance against fatigue enabled by their oxidative metabolism. ^{14 15} However, this exceptional combination is traded for a lower force output. ¹⁵  

Further speculation leads to the following conclusion: For its proper function, each muscle needs force, velocity, and fatigue resistance. Of note, all three of these properties are never found in one fiber. The combination of a well-developed force and a high shortening velocity leads to a fast-twitch muscle with high fatigability, like the mammalian extensor digitorum longus muscle. ^{41 56} The combination of a well-developed force and a high fatigue resistance is followed by a slow contraction time or shortening velocity, as observed in a slow-twitch muscle, like mammalian soleus muscle, ^{41 56} or, as in the case of EOM, the combination of a high shortening velocity with a high fatigue resistance is followed by a lower force output. However, as was mentioned earlier, we can suppose that the EOMs do not need a high force output, because they are suspended by connective tissue pulleys in the orbital fatty tissue. ¹⁵  

It is generally accepted that an oxidative metabolism and the presence of numerous mitochondria are necessary for high fatigue resistance of muscles, whereas a high glycolytic release of energy and a well-developed sarcoplasmic reticulum (required for release and reaccumulation of Ca²+ ions) are necessary for high muscle shortening velocity. A quantitative electron microscopy study of the mitochondria and of the sarcotubular system volume in EOMs of rats supports our suggestion. ²⁷ Comparison of these and our data show a striking correlation (r = 0.99) between the SDH activity and the mitochondrial volume in the fiber types 3 to 5. On the other hand, the fiber types 1, 2, and 6 do not show such correlation. Similarly, the same relation between the GPDH activity (this article) and the volume of the sarcotubular system (r = −0.98) ²⁷ is present. 

We can conclude that EOMs of adult rats show unique fiber type characteristics with highly specialized metabolic arrangements not seen in other skeletal muscles, subserving their extraordinary functional demands. It can be speculated that the EOMs, suspended by a very complex system of connective tissues and pulleys at the orbit, do not have to produce high force, but they have to be fast-contracting and fatigue resistant. Therefore, they have developed an extensive mitochondrial system that ensures them fatigue resistance and a well-developed sarcotubular system enabling their fast contraction and high mATPase activity. We therefore suggest that the abilities of EOMs for faster contraction, their high fatigue resistance, and high recruitment of motor units engaged in various types of eye movements are reflected by their high energy demands, which is enabled by their low force requirements compared to limb skeletal muscles. 

 Supported by grants from MYORES LSH-CT-2004-511978, GAČR 304/05/0327 and 304/08/0256, and the Deutsche Forschungsgemeinschaft, Schwerpunkt Muskelforschung (As 74/1-2); and Research Project AV0Z 50110509.

 Submitted for publication December 10, 2007; revised January 25 and June 17, 2008; accepted September 5, 2008.

 Disclosure: G. Asmussen, None; K. Punkt, None; B. Bartsch, None; T. Soukup, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

(A) SDH activity in the retractor bulbi (RB), the medial rectus (RM), the superior rectus (RS), the superior oblique (OS), and the levator palpebrae (LP) muscles sectioned through the superior part of the orbit of the rat. Consecutive sections were stained for GPDH activity (B) and for myosin-ATPase activity after acid (pH 4.3) preincubation (C). Scale bar, 500 μm. ON, optic nerve; OMN, oculomotor nerve; OR, orbital region; GR, global region.

Table 1.

 

Morphological (Area), Enzyme Histochemical (SDH, GPDH, mATPase) Quantitative Characteristics and Percentage of Fiber Types 1–6 in the Extraocular Oculorotatory Recti and Obliqui Muscles

	Orbital Region			Global Region
		Type 1	Type 2	Type 3	Type 4	Type 5	Type 6
Area (μm2)	252 ± 92	108 ± 22	212 ± 53	325 ± 64	422 ± 61	399 ± 27
Percentage	85 ± 7	15 ± 7	35 ± 4	25 ± 4	30 ± 4	10 ± 2
SDH (MA)	0.37 ± 0.06	0.13 ± 0.03	0.71 ± 0.09	0.45 ± 0.05	0.27 ± 0.04	0.10 ± 0.02
GPDH (MA)	0.18 ± 0.03	0.15 ± 0.03	0.68 ± 0.09	0.57 ± 0.07	0.51 ± 0.08	0.13 ± 0.02
mATPase (MA)	0.64 ± 0.07	0.39 ± 0.06	0.68 ± 0.08	0.76 ± 0.08	0.81 ± 0.07	0.37 ± 0.06
Classification by Qualitative Enzyme Histochemistry and Innervation 28 29
Alkaline ATPase	+++	++	+++	+++	+++	+/−
Acid ATPase	+/−	+++	+/−	+/−	+/−	+++
SDH	+++	++	++++	+++	++	+/−
GPDH	++	+	++	+++	++++	+/−
AChE	Focal	Multiple	Focal	Focal	Focal	Multiple
Classification by Quantitative Ultrastructural Properties 27
Myofibrils (%)	60	78	55	65	71	83
SR-volume (%)	9	6	10	14	16	4
Mitochondria (%)	20	6	24	13	5	5
Other structures	11	10	11	8	8	8
Classification by MyHC Expression 34 41 42 43 44 45
MyHC isoforms
Embryonic	+	+	−	−	−	−
EOM	+	−	+	+	+	−
Slow type I	−	+*	−	−	−	+
Fast type II	−	−	+(IIA)	+(IIX/D)	+(IIB)	−

 Cross-sectional area, fiber type distribution, and levels of the SDH or the GPDH of the various fiber types in rat oculorotatory EOMs. The enzyme activities were determined cytophotometrically as the MA and they are used as markers of oxidative or glycolytic metabolism, respectively. Data are presented as mean ± SD. For comparison, previously published semiquantitative data, ^{27 28 29 34 41 42 43 44 45} as well as other features used for the classification are indicated. Alkaline ATPase, myosin adenosine triphosphatase, at pH 9.4 or after preincubation at pH 10.4; acid ATPase, myosin adenosine triphosphatase, after preincubation at pH 4.3 or 4.5; AChE, acetylcholinesterase (staining of motor nerve endings); SR, sarcoplasmic reticulum; I, slow type MyHC, EOM/extraocular “super fast” MyHC isoform; IIA, IIX/D, and IIB, fast MyHC isoforms. Note that type 1 fibers correspond to orbital singly innervated, type 2 to orbital multiply innervated, type 3 to global red (slow oxidative) singly innervated, type 4 to global intermediate (fast mixed) single innervated, type 5 to global pale (fast glycolytic) singly innervated, and type 6 to global multiply innervated (slow) fibers according to Yu Wai Man, et al. ³⁴ Alternative classification differentiates two populations of type 1 fibers. ²⁰ Qualitative enzyme histochemistry: −, negative: +/−, very low; +, low; ++, intermediate; +++, high; ++++, very high.

*  Fibers also stained with antibodies against slow-tonic and cardiac-α MyHCs.

Figure 2.

 

Enzyme histochemical reactions of consecutive cross-sections through the orbital (A, B) and global (C, D) regions of the rat superior rectus muscle stained for SDH (A, C) and GPDH (B, D) activity. Some individual fibers of different types are indicated by the numbers 1 to 6. (Note that not all classified fibers are marked).

Figure 3.

 

Scatterplots showing the relationship between cross-sectional area and (A) SDH, (C) GPDH, and (D) mATPase activities and between GPDH and SDH activity (B) expressed as the cytophotometrically determined MA of the slow-twitch type 1 (•) and the slow-tonic type 2 (○) muscle fibers in the OR of rat oculorotatory muscles. There are no significant correlations between the analyzed parameters within or between the groups.

Figure 4.

 

Scatterplots showing the relationship between cross-sectional area and SDH (A), GPDH (C) and mATPase (D) activities and between GPDH and SDH activity (B) expressed as the cytophotometrically determined MA of the fast twitch types 3 (▪), 4 (▾), and 5 (♦) and the slow tonic type 6 (□) muscle fibers of the GR of rat oculorotatory muscles. The regression lines of the type 3 to 5 twitch fibers are indicated.

Authors thank Rita Rätze and Cornelia Wenzel for excellent technical help and Pavel Hník for critical reading and valuable comments during the preparation of the manuscript.