Umeå University Medical Dissertations

Department of Integrative Medical Biology, Section for Anatomy, Umeå University and Centre for Musculoskeletal Research, Gävle University, Umeå, Sweden

HUMAN MUSCLE SPINDLES

Complex morphology and structural organisation

Jing-Xia Liu

Umeå 2004

Copyright © by Jing-Xia Liu, 2004 ISSN 0346-6612 ISBN 91-7305-762-2 Printed by VMC, KBC, Umeå University, Umeå 2004

2

In memory of my father

3 TABLE OF CONTENTS

ABBREVIATIONS 5

ABSTRACT 6

ORIGINAL PAPERS 7

INTRODUCTION 9

The Muscle Spindle 9 Skeletal muscle structure and sarcomeric proteins 13 Sarcoplasmic reticulum and SERCA isoforms 16 Aging 17

AIMS OF THE STUDY 20

MATERIAL AND METHODS 21

Muscle specimens 21 Tissue processing 21 Antibodies and labeling 21 Biochemistry 22 Enzymehistochemistry and immunocytochemistry 23 Statistical analyses 23

RESULTS 24

Intrafusal fiber composition 24 Complex muscle spindles 25 mATPase activity 26 MyHC composition of isolated muscle spindles 26 MyHC composition of intrafusal fibers 26 SERCA composition of intrafusal fibers 29

DISCUSSION 30

Human muscle spindle identity: BB versus DN 30 MyHC composition of isolated muscle spindles 31 MyHC composition of intrafusal fibers 32 Effects of aging on muscle spindles 33 SERCA isoforms 34 Future perspectives 35

SUMMARY 36

CONCLUSIONS 37

ACKNOWLEDGEMENTS 38

REFERENCES 41

PAPER I PAPER II PAPER III PAPER IV

4 ABBREVIATIONS

ALD Anterior latissimus dorsi BB Biceps brachii CSA Cross-sectional area DN Deep muscles of the neck EDTA Ethylene diamine tetra-acetic acid mAb Monoclonal antibody mATPase Myosin adenosine triphosphatase MyHC Myosin heavy chain MyLC Myosin light chain MyHCα-c α-cardiac MyHC MyHCemb Embryonic MyHC MyHCfet Fetal MyHC MyHCeom Extraocular MyHC MyHCm Masticatory MyHC MyHCsto Slow tonic MyHC NT-3 Neurotrophin-3 PAP Peroxidase-antiperoxidase PBS Phosphate buffer saline SERCA Sarco/endoplasmic reticulum ATPase SERCA1 SERCA fast isoform SERCA2 SERCA slow isoform SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis SR Sarcoplasmic reticulum

5 ABSTRACT

Muscle spindles are skeletal muscle mechanoreceptors that mediate the stretch reflex and provide axial and limb position information to the central nervous system. They have been proposed to play a major role in the pathophysiology of muscle pain. Knowledge about the normal human muscle spindles is needed in order to understand their role in muscle disease or dysfunction.

The aim of this study was to investigate the fiber content and MyHC composition of the muscle spindles in the human biceps brachii (BB) and deep muscles of the neck (DN); to determine whether there are age-related changes in human muscle spindles with respect to structure and MyHC composition; to investigate the distribution of SERCA isoforms and to evaluate whether there is a coordinated expression of SERCA and MyHC isoforms in intrafusal fibers. The myosin heavy chain (MyHC) content correlates to contraction velocity and force and the sarcoplasmic reticulum Ca2+ ATPase (SERCA) is a major determinant of muscle fiber relaxation velocity.

Muscle specimens obtained from young and old subjects were serially sectioned and the pattern of distribution of different proteins along the length of the intrafusal fibers was revealed by immunocytochemistry. The MyHC content of single muscle spindles was assessed with SDS-PAGE and immunoblots.

There were clear differences between BB and DN with regard to the morphology and MyHC composition of muscle spindles. Virtually each muscle spindle in the BB, but not in the DN, had a unique allotment of numbers of bag1, bag2 and chain fibers. In DN, a number of muscle spindles lacked either bag1 or bag2 fibers. Four major MyHC isoforms (MyHCI, IIa, α-cardiac and intrafusal) were detected by SDS-PAGE. In both BB and DN, immunocytochemistry revealed co-expression of several MyHC isoforms in each intrafusal fiber and regional heterogeneity. Both nuclear bag1 and bag2 fibers contained slow tonic MyHC uniformly and MyHCI, α-cardiac, embryonic and fetal with regional variations. Nuclear chain fibers contained MyHCIIa, embryonic and fetal and in the BB also MyHCIIx.

The total number of intrafusal fibers per spindle decreased significantly with aging, due to a significant reduction in the number of nuclear chain fibers. The patterns of MyHC expression were also affected by aging.

The bag1 fibers predominantly contained both SERCA isoforms in the encapsulated region. The bag2 fibers were more heterogeneous in their SERCA composition and 16-27% of them lacked both isoforms. Chain fibers contained SERCA1. There was a poor correlation between the MyHC and SERCA isoforms in nuclear bag fibers, whereas a strong correlation existed between MyHCIIa and SERCA1 in the nuclear chain fibers.

Human muscle spindles, each being unique, proved to be more complex than anticipated. The clear differences shown between the BB and DN muscle spindles suggest functional specialization in the control of movement among different human muscles. Aging apparently had profound effects on intrafusal fiber content and MyHC composition. The age-related changes in muscle spindle phenotype may reflect deterioration in sensory and motor innervation and are likely to have a detrimental impact on motor control in the elderly.

Key words: human, muscle spindle, intrafusal fiber, aging, biceps brachii, deep muscles of the neck, MyHC, mATPase, SERCA.

6

ORIGINAL PAPERS

This thesis is based on the following papers, which are referred to in the text by their Roman numerals:

I. Liu J-X, Eriksson P-O, Thornell L-E, Pedrosa-Domellöf F. Myosin heavy chain composition of muscle spindles in human biceps brachii. J Histochem Cytochem 50: 171-84, 2002.

II. Liu J-X, Thornell L-E, Pedrosa-Domellöf F. Muscle spindles in the deep muscles of the human neck. A morphological and immunocytochemical study. J Histochem Cytochem 51: 175-86, 2003.

III. Liu J-X, Eriksson P-O, Thornell L-E, Pedrosa-Domellöf F. Fiber content and myosin heavy chain composition of muscle spindles in aged human biceps brachii. J Histochem Cytochem 52: 2004.

IV. Liu J-X, Thornell L-E, Pedrosa-Domellöf F. Distribution of SERCA isoforms in human intrafusal fibers. Histochem Cell Biol 120: 299-306, 2003.

7

8

INTRODUCTION

The Muscle Spindle

Muscle spindles, proprioceptive organs in skeletal muscles, were first described in the middle of the nineteenth century (Kühnne, 1863; Weismann, 1861) and our understanding of their functional role has increased markedly over the last 40 years. Muscle spindles have been postulated to be of importance for the control of movement and posture in mammals (Matthews, 1981). This has recently been firmly verified in neurotrophin-3-deficient mice which were unable to support their weight and had an abnormal posture due to the absent proprioceptive afferents and muscle spindles in their limbs (Ernfors et al., 1994; Walro and Kucera, 1999). Muscle spindles have also been proposed to play a central role in regulating muscle tone and in the pathophysiology of work-related chronic muscle disorders (Johansson et al., 1999; Johansson and Sojka, 1991; Matre et al., 1998; Thornell et al., 2003a). Previous human studies have demonstrated that muscles rich in spindles including the small muscles of the neck are particularly associated with chronic muscle pain (Johansson and Sojka, 1991). In order to understand the possible roles of muscle spindles in pathological conditions, basic knowledge of their structure and composition is essential. In particular, it is important to elucidate the molecular basis of the particular muscle spindle phenotype.

Figure 1. Muscle spindle and its innervation

Muscle spindles are located within muscles and run in parallel with the extrafusal fibers. Muscle spindles consist of a bundle of small-diameter muscle fibers which are surrounded by a multi-layered spindle shaped capsule derived from the perineurium. These muscle

9 fibers are therefore referred to as intrafusal fibers (for review see Barker and Banks, 1994). Two types of intrafusal fibers are distinguished according to their size and the arrangement of their central nuclei. The larger, often longer fibers displaying a wide equatorial accumulation of nuclei are called nuclear bag fibers whereas the smaller, usually shorter fibers containing a central row of nuclei are called nuclear chain fibers.

Three types of intrafusal fibers, namely nuclear bag1 (dynamic bag1), nuclear bag2 (static bag2) and nuclear chain (static chain) fibers, have been identified on the basis of both their mATPase activities (Ovalle and Smith, 1972) and their physiological properties (Barker and Banks, 1994; Boyd, 1962; Matthews, 1981). Three regions have been distinguished in a given muscle spindle (Figure 1). They are defined as follows: the A region, including the equator (EQ) and the juxtaequatorial parts (Aj), containing the periaxial fluid space; the B region, extending from the end of the periaxial fluid space to the end of the capsule, and the extracapsular C region (Barker and Banks 1994). Each region can be further subdivided into an inner (proximal to the equator) and outer (distal to the equator) portion.

Muscle spindles are sensory mechanoreceptors that relay information about muscle length and changes in muscle length to the central nervous system (Barker and Banks, 1994; Matthews, 1981). Intrafusal fibers receive both sensory and motor innervation (Figure 1). One single primary afferent (Ia) commonly innervates all intrafusal fibers in the spindle. In addition, one or more secondary afferent nerve fibers (II) may be present. The sensory innervation is distributed along the equatorial and juxtaequatorial regions whereas the motor endings innervate the polar regions towards or beyond the end of the capsule. Two types of motoneurons innervate intrafusal fibers, the gamma motoneurons that are specific for muscle spindles and the beta motoneurons that innervate both intra- and extrafusal fibers. From a physiological point of view, these two types of motoneurons can be classified either as dynamic or static. The nuclear bag1 fibers are innervated by dynamic motoneurons while the nuclear bag2 and chain fibers are innervated by static motoneurones (Boyd, 1980). The complex innervation of intrafusal fibers is in contrast to that of extrafusal fibers which in general are innervated by only one motoneuron and have a single motor endplate. The presence of sensory and gamma motor innervation is unique for intrafusal fibers.

Muscle spindle density and distribution

Muscle spindles are found practically in all mammalian skeletal muscles and the number of muscle spindles varies markedly between different muscles (Boyd and Smith, 1984). Muscle spindle density, expressed as the number of spindles per gram of tissue, has been measured in different muscles of different species (Bakker and Richmond, 1982; Kulkarni et al., 2001; Peck et al., 1984; Voss, 1971). The studies in humans have shown that the spindle density is in general highest in hand, foot and neck muscles, lowest in the shoulder and thigh muscles and intermediate in the more distal muscles of the arm and the leg (Voss, 1971). The spindle density can be regarded as an indicator of functional differences between muscles. In general, high spindle density characterizes muscles initiating fine movements (e.g. lumbrical muscles) or maintaining posture (e.g. neck muscles) whereas low spindle density is characteristic of muscles initiating gross movements (e.g. biceps brachii).

10

The localization of muscle spindles varies within the muscles. In some muscles the spindles are confined to circumscribed regions, although in many other muscles the spindles are more evenly distributed in close association with the nerve supply. In the human masseter, for instance, the spindle density was about 1.5 times greater in the deep portion than in the superficial part of the muscle (Eriksson and Thornell, 1987). Likewise, in human thyroarytenoid muscle, which can be divided into a lateral muscularis compartment and a medial vocalis compartment, muscle spindles are concentrated in the superior vocalis subcompartment (Sanders et al., 1998). This finding suggests that the superior vocalis subcompartment is functionally different from the remainder of the thyroarytenoid muscle and that the tension in the superior vocalis subcompartment can be controlled independently, which is of importance for the biomechanics of vocal fold vibration during phonation (Sanders et al., 1998).

Intrafusal fiber composition and spindle categories

The number of intrafusal fibers per spindle varies between different muscles and species. Human muscle spindles generally contain more intrafusal fibers than those of smaller species. The average number of intrafusal fibers per spindle is 12.4 (range 8-24) in the first lumbrical muscle (Soukup et al., 2003), 8.6 (range 4-17) in large trunk and limb muscles (Sahgal et al., 1985), 8.1 (range 4-10) in intercostal muscles (Kucera and Dorovini-Zis, 1979) and 6.7 (range 3-14) in the first dorsal interossei muscles of adult males (Thornell et al. unpublished results). Muscle spindles usually contain four to five intrafusal fibers in rat hind limb muscles, and as a rule they usually contain one nuclear bag1, one nuclear bag2 and two to three nuclear chain fibers (Boyd and Smith, 1984; Zelena and Soukup, 1973). The spindle composition is rather diverse in cat limb muscles (Eldred et al., 1974; Kucera, 1982, 1983). Most commonly cat muscle spindles contain six to nine fibers, but some may possess as few as three or as many as twelve fibers (Banks et al., 1979; Eldred et al., 1974). An occasional spindle may possess either a larger or a smaller number of bag fibers than the usual two (Banks et al., 1979; Eldred et al., 1974). One-bag-fiber spindles have been specially noticed, since a great number of such spindles has been found in both human and cat neck muscles and, interestingly, they always lack bag1 fibers (Abrahams and Richmond, 1988; Bakker and Richmond, 1981; Cooper and Daniel, 1963; Richmond et al., 1986). Most of the one-bag2-fiber spindles could be traced into tandem linkages with larger spindles containing at least two nuclear bag fibers (Abrahams and Richmond, 1988; Bakker and Richmond, 1981; Richmond et al., 1986). Thus, two categories of muscle spindles are distinguished according to their complement: spindles with bag1, bag2 and chain fibers (b1b2c spindle) and spindles lacking bag1 fibers (b2c spindle).

Muscle spindles in relation to muscle architecture and function

The vast majority of the studies have dealt with the muscle spindles as single independent units. Much less attention has been paid to the collective operation of muscle spindles, which together compose the sensory apparatus of the entire muscle. In fact, the main roles that muscle spindles play in different muscles may be different (Richmond and Abrahams, 1979b; Scutter and Turker, 2001). In some limb muscles like soleus, the muscle spindles have been demonstrated to be of importance in the maintenance of static posture. This role

11 is supported by (i) the close proximity of the muscle spindles to motor units that develop small forces and are fatigue-resistant; and (ii) the input from muscle spindles is more effective on the small motoneurons supplying these motor units (Mendell and Henneman, 1971; Schmied et al., 1997; Scutter and Turker, 2001). In the masseter, the afferent input is more effective on larger motoneurones (Scutter and Turker, 1999, 2001), suggesting that muscle spindles in the masseter may be more important for the development of larger, fast forces, rather than for the maintenance of static posture.

The biceps brachii (BB) has been chosen as a reference limb muscle in the present study. The BB has a relatively ordered architecture in which the muscle fibers run generally parallel to the muscle’s longitudinal axis (Lieber and Friden, 2000). This muscle has two heads, with different origins and insertions. The combined muscle performs a variety of functions i.e. flexes the elbow, supinates the forearm and abducts the arm if the latter is laterally rotated. The BB has been shown to have neuromuscular compartments (Segal, 1992) and within the long head of the biceps brachii, motor units from different locations could perform different tasks (ter Haar Romeny et al., 1982). The muscle spindle density in the whole BB is 2.0 (Peck et al., 1984; Voss, 1971).

The neck is regarded as an important proprioceptive organ for postural control (Richmond and Abrahams, 1979b). Disturbance of gait has been demonstrated in animal experiments interfering with sensory inputs from the upper cervical region, either by damaging or anesthetizing, excising or sectioning of neck muscles, or by cutting upper cervical dorsal roots (Kulkarni et al., 2001). Similarly, in man, dizziness and ataxia are seen following damage or anesthesia of neck muscles, whiplash injuries, altered cervical vascularity or disturbances of cervical sympathetic tone (Richmond and Abrahams, 1979b).

Figure 2. The location of the deep muscles of the neck

The rectus capitis posterior major, the rectus capitis posterior minor, the obliquus capitis inferior and the obliquus capitis superior are located deeply in the human suboccipital region (Figure 2) and are here collectively called the deep muscles of the neck (DN). These muscles are relative small and short and perform fine rotatory movements of the head and help maintain the stability of the cervical spine. The deep neck muscles have an unusually high muscle spindle density which is nearly five times higher than that of the large splenius

12 capitis and three times that of the semispinalis capitis muscle (Cooper and Daniel, 1963; Kulkarni et al., 2001; Peck et al., 1984; Voss, 1971). In adults, the spindle density is about 30.0 in both the rectus capitis posterior major and the obliquus capitis inferior; 36.0 in the rectus capitis posterior minor and 40~43 in the obliquus capitis superior (Peck et al., 1984; Voss, 1971). The convergence of sensory afferents from deep neck muscles with vestibular and ocular inputs at various levels of the neuroaxis is well recognized (Cooper and Daniel, 1963; Kulkarni et al., 2001; Richmond and Abrahams, 1975, 1979a). The complex integrative mechanisms involved in head positioning in relation to vestibular and visual control probably demand finely tuned proprioceptive inputs from the deep neck muscles, therefore requiring such a high spindle density.

The BB and DN were chosen since they are different with regard to the anatomical and functional properties and differ in their nerve supply. The present study focused on the characterization of the muscle spindle's identity in human BB and DN with respect to fiber content, MyHC and SERCA composition.

Skeletal Muscle Structure and Sarcomeric Proteins

Striated skeletal muscles are characterized by a very precise organization of contractile and structural proteins into repetitive units arranged in series, the sarcomeres. A sarcomere is defined as the region between two Z-discs that anchor the thin (actin) filaments, with the thick (myosin) filaments giving rise to the I- and A-bands. In the middle of the A-band there is the so-called M-band. At the junction of the A- and I-bands there are the sarcoplasmic reticulum (SR) and the transverse (T) tubules. According to the sliding filament theory, the thick filament protein myosin attaches to actin, a thin filament, and contraction can occur as the result of ATP-dependent movement of the two filaments. There are numerous other proteins in the sarcomere whose roles are related to modulation of contraction and relaxation, maintenance of the structure, or both (Schiaffino and Reggiani, 1996).

Myosin and MyHC isoforms

The myosins are the best studied molecular motor and constitute a large superfamily of proteins. Myosins have so far been divided into 17 classes. Sarcomeric myosin, the most abundant protein in skeletal muscle, is a conventional class II myosin (for review see Bottinelli and Reggiani, 2000; Pette and Staron, 1990; Schiaffino and Reggiani, 1994). It has a large and complex structure with two globular heads attached to a long tail (Figure 3). It consists of two heavy chains (MyHC; molecular weight 200-220 kD) and two pairs of light chains (MyLC; molecular weight 16-20kD). MyHC is both a structural protein and Figure 3. Sarcomeric myosin II an enzyme.

13 The myosin head contains the ATP-binding site and the actin-binding site. Thus, MyHC functions as a "motor" and converts the chemical energy into mechanical energy. Single muscle fiber contractility studies and in vitro motility assays have shown that MyHC isoforms are the major determinants of the maximum shortening velocity in a muscle cell, and a close relationship between the maximum shortening velocity and MyHC composition has been documented in different species (for review see Bottinelli et al., 1996; Bottinelli and Reggiani, 2000; Moss et al., 1995; Schiaffino and Reggiani, 1996). MyHCs are therefore considered as the best markers of the functional properties of skeletal muscle fibers (Larsson and Moss, 1993).

Sarcomeric MyHCs are encoded by multigene families. Eight genes of sarcomeric class II MyHCs have been identified in the human genome (Table 1), and seven corresponding isoforms (MyHCI, MyHCIIa, MyHCIIx, MyHCemb, MyHCfet, MyHCIIα-c, MyHCeom) have so far been identified in human muscle fibers (Berg et al., 2001; Weiss et al., 1999). The MyHC genes and their isoforms are typically expressed in a tissue-specific and development stage-specific manner. Three genes code for MyHCI, MyHCIIa and MyHCIIx, which are the predominant isoforms expressed in human limb and trunk skeletal muscles, and the relative proportions of these isoforms vary from muscle to muscle (Harridge et al., 1996). Two genes code for MyHCemb and MyHCfet, which are typically expressed during early muscle development (Schiaffino et al., 1986b) and during regeneration in adult life (Schiaffino et al., 1986a). They are also found in special human adult muscles: the masticatory muscles (Butler-Browne et al., 1988; Korfage and Van Eijden, 1999; Monemi et al., 1999b; Stal et al., 1994), the extraocular muscles (Kjellgren et al., 2003b; Pedrosa-Domellof et al., 2000), and in intrafusal fibers (Eriksson et al., 1994; Pedrosa-Domellof and Thornell, 1994). In human heart, MyHCα-c represents the major isoform in the atrial myocardium, and MyHCI (also called MyHC β-cardiac) is mainly found in the ventricular myocardium (Gorza et al., 1984). MyHCα-c has also been identified in the human extraocular (Pedrosa-Domellof et al., 1992) and masseter (Monemi et al., 1999b) muscles as well as in human muscle spindles (Eriksson et al., 1994; Pedrosa- Domellof and Thornell, 1994). The MyHCeom isoform has been demonstrated in extraocular muscle (Kjellgren et al., 2003b; Pedrosa-Domellof et al., 2000; Weiss et al., 1999) and in laryngeal muscles (Briggs and Schachat, 2000; Sciote et al., 2002). The gene for MyHCIIb has been sequenced in human (Weiss et al., 1999) and unpublished data indicate that this isoform is present in human extraocular and laryngeal muscles (Andersen et al. in preparation, personal communication of Pedrosa-Domellöf). In a recent study, four MyHC isoforms have been revealed by SDS-PAGE in the human EOMs (Kjellgren et al., 2003b). In addition to three clearly identified isoforms i.e. MyHCI, MyHCIIa and MyHCemb, the fourth band, migrating between MyHCI and MyHCIIa, has been tentatively identified as MyHCeom. Since the genes for the human MyHCeom and MyHCIIb have been sequenced from human extraocular muscle samples (Weiss et al., 1999), the corresponding MyHC isoforms are expected to occur in the EOMs. To date antibodies purely against MyHCIIb are not available and it remains to be determined whether the band named MyHCeom really corresponds to the product of the MyHCeom gene or whether it corresponds to the MyHCIIb, or both (Kjellgren et al., 2003b).

14 An additional isoform, MyHCsto, is also present in special muscles, although its gene has not been identified in the human genome. It is not known whether MyHCsto derives from a unique gene or from post-transcriptional processing of one of the other MyHC genes

Table 1 MyHC isoforms and their genes in different adult human musclesa Isoform Gene Locus Limb/trunk Jaw-closer EOMs MyHCα-c MYH6 14q12 − + + MyHCI MYH7 14q12 + + + MyHCemb MYH3 17p13.1 − + + MyHCfet MYH8 17p13.1 − + + MyHCIIa MYH2 17p13.1 + + + MyHCIIx MYH1 17p13.1 + + + MyHCeom MYH13 17p13.1 − − + MyHCIIb MYH4 17p13.1 − − ? a: Based on Bottinelli and Reggiani 2000; Berg et al. 2001; Hoh 2002.

(Schiaffino and Reggiani, 1994). MyHCsto has been reported in extraocular muscles (Bormioli et al., 1979; Pedrosa-Domellof et al., 2000), laryngeal muscles (Han et al., 1999; Shiotani et al., 1999), and in intrafusal fibers (Pedrosa et al., 1989; Thornell et al., 1988, 1989). Finally, a MyHC isoform with its expression restricted to masticatory muscles and therefore named MyHCm or superfast has been identified in several mammalian species (for review see Hoh, 2002). The human homologue gene has been localized to 7q22, but its expression in human muscles has not been demonstrated (Berg et al., 2001; Bottinelli and Reggiani, 2000). This gene is presumably in the process of being lost from the human genome due to anatomical changes in the human lineage (Berg et al., 2001).

During development, muscle fiber diversification and myosin gene expression are regulated by multiple mechanisms including myoblast pre-determination and neuronal influences. In adults, the expression of MyHC genes can be further modulated by extrinsic factors such as electrical and neural stimulation, different mechanical loading conditions and hormones (for review see Pette and Staron, 1997; Schiaffino and Reggiani, 1996).

Sarcomeric MyHC isoform diversity is common to all striated muscles. The MyHC diversity is generated and maintained by the development of large and complex multigene families. It is generally accepted that each MyHC isoform is the product of one single gene since no other sarcomeric MyHC genes have been found to give rise to more than a single mRNA transcript (Bandman, 1999; Weiss and Leinwand, 1996), with the exception of Drosophila melanogaster, in which a single MyHC gene produces different transcripts through alternative splicing events (Rozek and Davidson, 1983). However, several lines of evidence suggest the existence of more than one MyHCI isoform. At the mRNA level, two different β-cardiac MyHC cDNAs, pHMC3 and pSMHCZ, have been found in human muscle (Jandreski et al., 1987). Using three different mAbs (A4.840, A4.951 and N2.261) recognizing distinct slow MyHC epitopes, Hughes et al. (Hughes et al., 1993) have reported the sequential appearance of three MyHCI isofoms during human and rat muscle development. Likewise, different phenotypes among MyHCI-containing fibers have also been reported in adult rabbit masseter muscle (English et al., 1998).

15 MyHCs in intrafusal fibers

Although there is an extensive body of knowledge concerning the myosin heavy chain composition of human extrafusal muscle fibers, corresponding information on human muscle spindles is scarce (Eriksson et al., 1994; Pedrosa-Domellof et al., 1993; Pedrosa- Domellof and Thornell, 1994; Thornell et al., 1988, 1989). Data from other species (e.g. cat, rat, rabbit) have shown that the mammalian nuclear bag and nuclear chain fibers have a complex MyHC composition that differs from that of extrafusal fibers (Kucera and Walro, 1989; Kucera et al., 1992; Maier et al., 1988a; Pedrosa et al., 1989, 1990; Pedrosa and Thornell, 1990; Pedrosa-Domellof et al., 1991; Rowlerson et al., 1988; Soukup et al., 1995; Walro and Kucera, 1999; Walro et al., 1997; Wang et al., 1997). In rat limb muscle spindles, for example, the bag1 and bag2 fibers preferentially react with antibodies against MyHCsto, MyHCI, MyHCα-c and to a variable extent with MyHCemb. In addition, the nuclear bag2 fibers react with antibodies against fast twitch MyHCII and neonatal MyHCneo, whereas the nuclear chain fibers mainly react with antibodies specific for MyHCneo and MyHCemb.

In contrast to the uniformity of the MyHC distribution in extrafusal fibers, the MyHC composition is not constant along the length of intrafusal fibers. In rat muscle spindles, bag1 fibers contain MyHCI towards the end of the B and in the C regions, and MyHCα-c for a short distance near the end of the capsule (Pedrosa et al., 1989). Likewise, bag2 fibers also contain MyHCs which vary along the fiber length.

Sarcoplasmic Reticulum and SERCA Isoforms

The sarcoplasmic reticulum (SR) is the highly specialized endoplasmic reticulum of muscle cells. It contains a very high concentration of Ca2+, and it is a system of parallel tubules running between the myofibrils (Franzini-Armstrong, 1994). It is composed of two continuous, yet heterogeneous compartments, namely the junctional SR and the free SR (Martonosi, 1994). The free SR contains the Ca2+ transport ATPase as the principle Figure 4. SR and T-tubules of skeletal muscle fiber component. It is usually divided into lateral sacs (or terminal cisternae) and longitudinal SR, that differ in protein composition. The junctions in which a T tubule is sandwiched between two SR terminal cisternaes are called triads. Another arrangement, in which one SR cisternae forms a junction with the T tubule is called dyad. The terminal cisternae and

16 longitudinal SR differ in protein composition (Martonosi, 1994). The special disposition of the SR relative to myofibrils indicates a role in the contractile activity of the muscle fibers. This role in fact is the active pumping and sequestering of calcium and thereby regulation of the muscle fiber contraction and, particularly, relaxation time.

The SERCA is the calcium pump responsible for recharging SR stores and for bringing calcium levels down to baseline following release, so that the muscle fibers can relax. The relaxation speed of a muscle fiber depends primarily on the rate at which Ca2+ is resequestered within the sarcoplasmic reticulum (SR), and a key element in its regulation is SERCA (Moller et al., 1996). Thus, SERCA is the major determinant of the relaxation rate of the muscle fibers.

SERCA is an integral membrane protein which belongs to the P-type transport ATPases with a molecular mass around 110 KD. SERCAs are encoded by three genes, to give rise to SERCA1, 2 and 3 (MacLennan et al., 1985; Moller et al., 1996). Transcripts from these genes undergo alternative splicing giving rise to at least seven isoforms, SERCA1a/b, SERCA2a/b, SERCA3a/b/c (for review see Lompre, 1998; Shull, 2000). Particular isoforms are associated with certain cell types. In skeletal muscle, SERCA1a is the major isoform of fast-twitch muscles and SERCA1b is detected transiently in neonatal muscles. SERCA2a predominates in slow-twitch muscles and cardiac muscles whereas SERCA2b is found in almost all non-muscle cells and is therefore considered to be a housekeeping calcium pump. SERCA3 is rarely found alone and is usually found in combination in a restricted range of cells along with SERCA2b (Moller et al., 1996).

Since the rate of muscle contraction and relaxation of a skeletal muscle is largely determined by the MyHC and SERCA isoforms, respectively, and since the slow-twitch muscle fibers predominantly express MyHCI and SERCA2 whereas fast-twitch muscle fibers express MyHCII and SERCA1, a rational correlation exists between the expression of MyHC and SERCA in skeletal muscles (Dux, 1993; Quiroz-Rothe and Rivero, 2001). Furthermore, the adaptations in MyHC expression are accompanied by commensurate adaptations in SERCA isoform in muscle fibers after functional overload (Talmadge et al., 1996) or chronic neuromuscular stimulation (Hamalainen and Pette, 1997). The neuromuscular activity has been proposed to play a crucial role in the coordinated expression of SERCA and MyHC isoforms (Hamalainen and Pette, 1997). Such correlations, however, are not applicable to all skeletal muscles. In rabbit extraoccular muscles (EOMs), for instance, the orbital multi-innervated fibers do not express fast MyHC but do express fast SERCA isoform and this is also the case in the EOMs of rat (Jacoby and Ko, 1993) and human (Kjellgren et al., 2003a).

Aging

Normal aging in human is associated with the loss of skeletal muscle strength, the loss of muscle mass known as sarcopenia, and increased muscle fatigability. These age-related changes in the skeletal muscle system can substantially reduce the amount and intensity of physical activity performed by the elderly, which in return result in a loss of independence. Since the average human lifespan has increased significantly in most parts of the world, the proportion of the population above the age of 60 years is growing rapidly. Therefore, age-

17 related sarcopenia is a major public health problem. A better understanding of the physiological and biochemical aspects of ageing, as well as of the pathophysiology of sarcopenia are the first steps towards developing rational therapeutic or preventive measures to address this problem.

Morphological changes in aging muscles

Loss of muscle strength or muscle weakness is one of the most obvious age-related changes seen in the elderly. Porter et al. (1995) reported that healthy people in their seventh and eighth decades are weaker than young adults by an average 20-40% during isometric and concentric knee extensor tests, and the very old (> 80 years) showed even greater reductions (50% or more). The loss of muscle strength is partly caused by morphological changes in skeletal muscles, one of the most marked being the loss of muscle mass, which is associated with the decrease in total muscle cross-sectional area (CSA). However, the decline in CSA is accompanied by accumulation of body fat and connective tissue and thus the decrease of muscle mass does not typically result in weight loss in the elderly (Overend et al., 1992; Rice et al., 1989).

An obvious consequence of aging is a marked atrophy of skeletal muscles. A recent longitudinal study with a biopsy interval of 12 years showed a 1.4%/yr reduction in muscle CSA (Frontera et al., 2000). Data obtained from vastus lateralis, tibialis anterior as well as biceps brachii have shown that fast-twitch type 2 fiber size is reduced with aging whereas the size of slow-twitch type 1 fibers remains much less affected (Aniansson et al., 1986; Grimby et al., 1982; Larsson et al., 1978; Lexell and Taylor, 1991; Lexell et al., 1988; Scelsi et al., 1980; Tomonaga, 1977). Brooke and Engel (1969) reported that in the third or fourth decade of life, the mean CSA of individual type 2 fibers in the quadriceps femoris muscle exceeds that of type 1 fibers by proximately 20%, however, by the age of 85, the area of individual type 2 fibers is less than 50% of that of type 1 fibers (Tomonaga, 1977).

The atrophy of aging skeletal muscles is caused by both a reduction in the fiber size and a loss of muscle fibers. The total number of muscle fibers decreases with aging, beginning already at about the age of 25 and proceeding at an accelerated rate thereafter. Unlike the consistent overall conclusions that age-related fiber atrophy is milder for type 1 fibers than for type 2 fibers, the findings for selective fiber number loss are somewhat conflicting, as both increased (Larsson et al., 1978), decreased (Frontera et al., 2000) and unchanged (Grimby et al., 1984; Lexell et al., 1988) relative proportion of type 1 fiber numbers have been reported to occur with aging in humans.

There is accumulating evidence showing that different muscles behave differently upon aging (for review see Thornell et al., 2003b). For instance, the age-related changes in the human masseter, a jaw closing muscle, and the lateral pterygoid, a jaw stabilizing muscle, are opposite to those reported for limb and trunk muscles (Monemi et al. 1998, 1999a, 1999c). On the contrary, age-related changes in the anterior and posterior bellies of the digastrics, a jaw opening muscle, resemble those of limb and trunk muscles (Monemi et al., 2000, 1999c). When the morphology of leg and arm muscles (i.e. vastus lateralis vs biceps brachii) of old men and women were examined, the mean area of the type 2 fibers was significantly smaller in the leg than in the arm (Grimby et al., 1982). This could indicate

18 differences in the aging process and/or a difference in the activity pattern between limb and jaw muscles or between leg and arm muscles (Porter et al., 1995).

Muscle spindles in aged muscle

Advanced age is also associated with significant changes in the physiological properties of the muscle spindle. Both dynamic and static position sense deteriorate with advanced aging (Ferrell et al., 1992; Meeuwsen et al., 1993; Robbins et al., 1995; Verschueren et al., 2002) and movement detection thresholds appear to increase in the elderly (Gilsing et al., 1995; Skinner et al., 1984; Thelen et al., 1998). A decrease in the dynamic sensitivity of muscle spindle primary endings has also been described in aged rats (Miwa et al., 1995).

Morphological changes in the innervation and structure have been reported in aged muscle spindles. Changes in the fine structure of the muscle spindle innervation consisting of spherical axonal swellings and expanded, abnormal motor end-plates have been described in some old muscles (Swash and Fox, 1972). Additionally, an increase in the spindle capsular thickness and a decrease in the number of intrafusal fibers per spindle have also been observed in aging muscles (Swash and Fox, 1972).

Although there are numerous studies on the effect of aging on human skeletal muscles as mentioned above, none of these have described the effects of aging on human muscle spindles with respect to MyHC isoforms.

19 AIMS OF THE STUDY

The present study aimed at characterizing human muscle spindles with respect to their structure and contractile proteins. The aims were:

• To investigate the morphological characteristics, intrafusal fiber content and MyHC composition of the muscle spindles in human biceps and deep muscles of the neck (Papers I, II).

• To explore whether intrafusal fiber content and MyHC composition changed with aging in human muscle spindles (Paper III).

• To investigate the presence and distribution of SERCA isoforms and to evaluate whether there was a correlation between SERCA and MyHC isoforms in human intrafusal fibers (Paper IV).

20 MATERIAL AND METHODS

Muscle Specimens

All human muscle samples were collected at autopsy shortly after death, a delay that does not hamper reliable muscle fiber typing (Eriksson et al., 1980). The samples were collected in accordance with the Swedish transplantation Law, with the approval of the Medical Ethical Committee, Umeå University.

BB samples were obtained from four females (age 48, 38,19 and 15) and six males (age 38, 37, 27, 25, 22 and 19; Paper I). In addition, adult human extraocular, the first lumbrical, heart muscle, fetal BB samples and adult chicken ALD muscles were also collected and used for biochemical analysis in Paper I. DN samples (rectus capitis posterior major, rectus capitis posterior minor, obliquus capitis inferior and obliquus capitis superior) were collected from two females (age 26 and 17) and three males (age 55, 21 and unknown; Paper II and IV). The muscle samples used in Paper III were collected from five old subjects (4 males and 1 female, ages 69, 78, 79, 82 and 83). None of the subjects was known to suffer from neuromuscular disease.

Tissue Processing

The muscle samples were mounted, rapidly frozen in propane chilled with liquid nitrogen and stored at −80°C until analysis.

Antibodies and Labeling

Table 2 Panel of MyHC antibodies used for immunocytochemical detection Antibodya Specific isoforms Short name Paper ALD19 slow tonic Anti-MyHCsto I-IV F88 α-cardiac Anti-MyHCα-c I-III A4.840 slowb Anti-MyHCI/1st I, II A4.951 slowb Anti-MyHCI/2nd I-III N2.261 slow, fast IIab Anti-MyHCI/3rd+IIa I, II A4.74 fast IIa Anti-MyHCIIa I, II BF35 all except IIx Anti-MyHC“all except IIx” I-IV VII1H3 slow, fast IIa, fast IIx Anti-MyHCI+IIa+IIx I 2B6 embryonic Anti-MyHCemb I-III NN5 fetal Anti-MyHCfet I NCL-MHCn fetal Anti-MyHCfet II, III NCL-SERCA1 fast SERCA Anti-SERCA1 IV NCL-SERCA2 slow SERCA Anti-SERCA2 IV a. All antibodies are monoclonal except NN5. b. More detailed descriptions of individual antibodies are given in the original papers I-IV.

Monoclonal and polyclonal antibodies specific for different MyHC and SERCA isoforms proteins were used in the present study. The short names of the antibodies are listed in Table 2. All antibodies were diluted in 0.01M PBS containing 0.1% bovine serum albumin (BSA) and used at their optimal dilution.

21 Biochemistry (Paper I)

Dissection of individual muscle spindles

One hundred micrometer thick sections were taken from the BB and first lumbrical muscles and freeze-dried overnight, starting with dry ice surrounding the samples and terminating at room temperature. Individual muscle spindles were then dissected free from the surrounding capsule and muscle tissue under stereomicroscopy (Wild, Heerbrugg, Switzerland), transferred into 0.2 ml plastic tubes and then lysed in Laemmli sample buffer (Bio-Rad Laboratories; Hercules, CA, USA) containing 5% β-mercaptoethanol. The sample solution was stored at -20°C until analysis.

Whole muscle extracts were prepared from adult human extraocular, heart and biceps muscles, fetal limb muscle (20 weeks of gestation) and chicken ALD muscle as described by Kadi and Thornell (1999). In brief, 60 µm thick muscle sections were extracted in four volumes of sample buffer: 0.3 M NaCl, 0.1 M NaH2PO4, 0.05 M Na2HPO4, 1 mM MgCl2, 10 mM Na4P2O7, 10 mM EDTA and 0.1% mM β-mercaptoethanol (pH 6.5). The tissue was scissor minced and extracted for 1 hr on ice. The extracts were centrifuged and the supernatants containing myosin were then diluted 1:1 in glycerol and stored at -20°C.

SDS-PAGE

The MyHC composition of the isolated muscle spindles and the whole muscle extracts was determined by SDS-PAGE (Laemmli, 1970). Five percent separating and 4.7% stacking gels were prepared from a solution containing 30% acrylamide-0.8% bisacrylamide. The separating gel contained 35% glycerol and the stacking gel contained 18% glycerol. The separating and the stacking gels were allowed to fully polymerize at room temperature for 1 and 3 hours, respectively. SDS-PAGE was run at 75 V for 22 hours, with the lower two- thirds of the gel unit surrounded by a 7°C water bath. The gels were then silver stained (Oakley et al., 1980) and photographed.

Western blot analysis

After SDS-PAGE, proteins were transferred to 0.45 µm nitrocellulose membrane (Bio-Rad Laboratories; Hercules, CA, USA) in a Tris/glycine buffer system using a Mini Trans-Blot apparatus (Bio-Rad Laboratories; Hercules, CA, USA) surrounded by a 15°C water bath unit for 17 hours at 30 V constant voltage. The membranes were washed twice for 5 min with 20 ml distilled water to remove gel and transfer buffer components and weakly bound proteins. The western blots were immunostained according to the WesternBreeze Chromogenic Immunodetection Protocol (WesternBreezeTM kit; Novex, CA, USA). In brief, the membranes were rinsed in blocking solution for 30 min and washed in water twice for 5 min. Then membranes were incubated at room temperature for 60 min in the following primary antibodies: N2.261 (diluted 1:250) or BF35 (1:3000). After 20 min washing in diluted Antibody Wash, the membranes were incubated in secondary antibody solution for 30 min. Following a 20-min washing in diluted Antibody Wash and a 6-min rinse in water, bands on the membranes were developed in chromogenic substrate solution.

22 Enzymehistochemistry and Immunocytochemistry (Papers I-IV)

Frozen specimens were serially sectioned at −23°C using a Reichert Jung cryostat (Leica, Nussloch, Germany). Sections to be used for demonstration of mATPase activity after preincubation at pH 10.4, 4.6 and 4.3 (Dubowitz, 1985) were 8 µm thick, those for immunocytochemistry were 5 µm thick.

Immunocytochemistry was performed using indirect peroxidase-anti-peroxidase (PAP) technique (Sternberger, 1979) with the antibodies listed in Table 1. In brief, sections were air-dried for 20-30 min, incubated with 5% normal rabbit (for mAbs) serum (Dakopatts, Glostrup, Denmark) for 15 min. Sections were then incubated with the appropriate primary antibody for 60 min at 37°C. After washing in 0.01M PBS for 15 min, sections were incubated again with normal serum for 15 min, followed by incubation with 2% rabbit anti- mouse antibody (Dakopatts, Glostrup, Denmark) for 30 min at room temperature. After washing in 0.01M PBS for 15 min, the sections were incubated with 1% mouse peroxidase- antiperoxidase (Dakopatts, Glostrup, Denmark) for 30 min and then washed in 0.01M PBS for 15 min. Normal swine serum (5%), swine anti-rabbit antibody (1%) and 1% rabbit peroxidase-antiperoxidase (Dakopatts) were used on the sections treated with the polyclonal antibody. Peroxidase developing was obtained by applying a solution containing diaminobenzidine and hydrogen peroxide for 5-10 min, followed by rinsing with running water for 5 min. Finally, the sections were dehydrated in graded dilutions of ethanol followed by xylene and mounted with DPX (BDH, Limited Poole, England). Control sections were processed as above, except that the primary antibody was omitted.

All sections were examined and photographed with a Zeiss Axiophot microscope (Carl Zeiss, Obercochen, Germany) equipped with a CCD camera (Dage-MTI, Inc., Michigan City); or a Nikon microscope (Eclipse, E800; Tokyo, Japan) equipped with a Spot camera (RT colour, DIANOSTIC, Instruments Inc. MI). Figures were prepared with Adobe Photoshop (version 5.5 and 7.0) software program.

Statistical Analyses

The StatView software (3rd edition, SAS Institute Inc., Cary, NC) was used for statistical analysis. Statistic comparisons with regard to differences in the intrafusal fiber numbers between males and females (Paper I), between BB and DN muscles (Paper II) and between young and old subjects (Paper III) were performed with an unpaired t-test. Differences were considered significant at p<0.05. Group data were presented as mean and standard deviations (SD).

23 RESULTS

Intrafusal Fiber Composition

Biceps brachii (Paper I)

The fiber type composition of the BB muscle spindles varied considerably. On average, there were 9.1 (range 4-16) intrafusal fibers per spindle, 2.2 (range 1-4) nuclear bag1, 1.3 (range 0-4) nuclear bag2 and 5.7 (range 2-12) nuclear chain fibers in each spindle.

Among 22 muscle spindles examined in the A or B regions, only two spindles had exactly the same fiber composition whereas the remaining spindles had a unique allotment of fibers (Table 2; Paper I). Sixty-eight percent of the spindles contained more bag1 than bag2 fibers whereas only 9% of the muscle spindles contained more bag2 than bag1 fibers. Fourteen percent of the muscle spindles lacked bag2 fibers (b1c spindles) and all of these spindles contained two bag1 fibers and a variable number of chain fibers.

Deep muscles of the neck (Paper II)

The number of intrafusal fibers per spindle varied also remarkably in the DN. Up to fifteen intrafusal fibers were seen in the large muscle spindles and there were as few as four fibers in small spindles. On average, there were 7.2 (range 4-15) fibers per spindle and in each spindle there were in average 1.3 (range 0-3) bag1, 1.0 (range 0-4) bag2 and 4.9 (range 2- 11) chain fibers.

Twenty out of 116 spindles studied in the A and/or B regions presented a unique allotment of numbers of bag1, bag2 and chain fibers but there were as many as 21 muscle spindles containing one bag1, one bag2 and four chain fibers (Table 2; Paper II). More than half of the spindles (58%) contained one bag1, one bag2 and a variable number of chain fibers. Fifteen percent of the spindles contained two bag1, one bag2 and several chain fibers. The remaining spindles contained three or more bag fibers, and up to six bag fibers were found in a single spindle. Twenty-nine percent of the spindles contained more bag1 than bag2 fibers. In contrast, only 10% of the spindles had fewer bag1 than bag2 fibers. Although most spindles studied contained all three fiber types (b1b2c spindle), 13% of the spindles contained only one type of bag fibers; 10.4% lacked bag2 fibers (b1c spindle) whereas 2.6% lacked bag1 (b2c spindle). Two out of twelve b1c spindles belonged to two pairs of complexes and the remaining were single receptors.

The average number of intrafusal fibers per spindle in the BB was significantly larger than that found in the DN (p=0.0002), and the difference was mainly attributed to the significantly higher number of bag1 fibers in BB (p<0.0001). No significant differences were found in the number of bag2 and chain fibers between BB and DN (Figure 9; Paper II).

We compared the intrafusal fiber content in muscle spindles of BB and DN with that of the first lumbrical muscles (Soukup et al., 2003), and found that the muscle spindles in the lumbrical muscle have the highest number of intrafusal fibers (Figure 9; Paper II). The

24 difference between BB and lumbrical muscles was mainly due to a lower number of bag1 and bag2 fibers in BB, whereas the difference between DN and the first lumbrical muscle was due to variation in the number of all three intrafusal fiber types.

Biceps brachii from old subjects (Paper III)

The intrafusal fiber type composition was studied in twenty-one muscle spindles encountered in either the A or inner B region, where the full complement of intrafusal fibers is present. On average, there were 6.2 intrafusal fibers per spindle, including 1.8 (range 1-3) bag1, 1.7 (0-5) bag2 and 2.8 (0-8) chain fibers (Table 2; Paper III). Nineteen percent of the muscle spindles contained one bag1, one bag2 and 2-4 chain fibers. The remaining spindles contained three or more bag fibers except for one muscle spindle that contained only one bag1 and one chain fibers (Figure 7E; Paper III). Sixteen out of 21 muscle spindles had a unique allotment of numbers of nuclear bag1, bag2 and chain fibers (Table 2; Paper III).

Strikingly, 3 out of 21 muscle spindles examined in the A or inner B region did not contain any nuclear chain fibers (Table 2; Figure 1; Paper III). Moreover, we found no nuclear chain fibers extending into the outer B or the C region, and very few chain fibers were observed in the inner B region. A number of nuclear chain fibers transiently appeared and ended in the A or even in the inner A region. The muscle spindles in the elderly had a lower total number of intrafusal fibers (p=0.0004) and a lower number of chain fibers (p<0.0001) per spindle in comparison with the muscle spindles from young adults (Table 2; Paper III).

Complex Muscle Spindles

All the spindles studied in the BB were single and isolated. In the DN, most spindles also appeared as single isolated receptors. However, 22% of the total number of DN muscle spindles occurred in different forms of linkage. The linked forms observed in the present study generally fitted previous descriptions of muscle spindles in the cat neck muscles (Bakker and Richmond, 1982, 1981; Richmond and Abrahams, 1975). The first were named "paired spindles" and occurred when two or three single spindles were in close apposition and had some form of mechanical connection. This complex arrangement was the most commonly found in the present study. The second complex form consisted of "parallel spindles", the spindle units still lying in close apposition but each one remaining as a separate entity in the polar region. In the equatorial region, however, the capsular division between the two spindles was lost. Two out of 36 spindles obtained from serial consecutive sections showed the third type of complex form. Each of these two spindles was first encountered in the C region where it had one bag1 and one bag2 fibers. The bag1 fiber became smaller and smaller and finally disappeared still in the C region. The remaining bag2 fiber continued and a new bag1 fiber appeared either in the C region or in the B region. A new spindle containing the new bag1 fiber together with the common bag2 fiber and five new chain fibers was then seen along the B and A regions, suggesting that the spindles were associated in a serial complex form. In addition to these complex forms, one spindle had an extension of connective tissue from the spindle capsule which enclosed three extrafusal fibers, starting in the outer A region, for a certain distance (Figure 3; Paper II).

25 mATPase Activity

In general, the mATPase activity of the bag fibers but not of the chain fibers showed variation along the fiber length, as described in detail in Papers I, II and summarized in Figure 5.

Figure 5. Schematic illustration of the mATPase activity after acid (pH 4.3) and alkaline (pH 10.4 preincubations along the length of three types of intrafusal fibers.

In the DN, the mATPase staining features of nuclear bag1, bag2 and chain fibers at different pH values were generally as described above in the BB, but some exceptions were observed. Nuclear chain fibers were always darkly stained over their entire length at pH 4.6 (Figure 5; Paper II) and were darkly stained in the A and inner B regions but lightly stained in the remaining parts at pH 10.4 (Figure 5; Paper II). In general, chain fibers displayed low or very low mATPase activity at pH 4.3, but 28% of the chain fibers showed intermediate or high activity in the A and B regions (Figure 5; Paper II).

MyHC Composition of Isolated Muscle Spindles (Paper I)

Four MyHC bands corresponding to MyHCI, MyHCIIa, MyHCemb and MyHCif could be identified on SDS-PAGE of muscle spindle extracts (Figure 1; paper I). Identical results were obtained with muscle spindles isolated from the biceps and from the lumbrical muscles and the latter were therefore used for immunoblotting as they were more easily available. The prominent band migrating between MyHCI and MyHCIIa, tentatively named MyHCif (Pedrosa-Domellof et al., 1993), was found only in intrafusal fibers and was not recognized by mAbs N2.261, BF35 (Figure 2; Paper I) in the present study or by anti- MyHCsto, anti-MyHCemb, anti-MyHCII in a previous study (Pedrosa-Domellof et al., 1993).

MyHC Composition of Intrafusal Fibers

Biceps brachii (Paper I)

Nuclear bag1, nuclear bag2 and nuclear chain fibers showed distinct staining patterns with

26 the different antibodies (Figures 5-8; Paper I), as summarized in Figure 6.

Figure 6 Schematic illustration of the staining profiles of three type of intrafusal fibers from adult BB.

27 Deep muscles of the neck (Paper II)

The staining patterns observed in the A, B and C regions of nuclear bag1, bag2 and chain fibers with a battery of nine antibodies are summarized in Figure 7 and shown in Figures 1- 5, paper II.

Figure 7. Schematic illustration of the staining patterns of bag1, bag2 and chain fibers in the DN.

Muscle spindles in BB of old subjects (Paper III)

Although most MyHC isoforms typically expressed by intrafusal fibers were apparently unaffected, the patterns of distribution of MyHCα-c, slow and fetal were altered in the BB of old subjects.

The majority of both bag1 and bag2 fibers were unstained with anti-MyHCα-c either in the A region or in the B region and no bag fibers were stained in the C region (Figure 7; Paper III). The staining intensities with anti-MyHCα-c varied from strong to weak. Nuclear chain fibers were generally unstained with anti-MyHCα-c, yet six chain fibers belonging to two muscle spindles showed weak staining in the A or inner B regions (Figure 7F; Paper III).

28 The bag2 fibers were not stained with anti-MyHCfet regardless of the region (Figures 4H and 5G; Paper III), except for two bag2 fibers which were weakly stained in the A region and in the C region, respectively (Figure 6B; Paper III).

The mAbs A4.840 and A4.951 against MyHCI stained the bag1 fibers similarly. They unstained the bag1 fibers in the equator and stained them weakly in the inner A region (Figures 4C and 4D Paper III). However, the staining intensity gradually increased throughout the outer A and the inner B regions, and was moderate to strong in the outer B and the C regions (Figure 5C; Paper III).

SERCA Composition of Intrafusal Fibers (Paper IV)

The vast majority of Table 3 Coexpression frequencies (%) of SERCA2 and nuclear bag1 fibers were SERCA1 in intrafusal fibers stained with both mAbs SERCA2/SERCA1 against SERCA1 and Fibers Region +/+ +/- -/+ -/- SERCA2 in the A and B Bag1 A 91.0 5.3 0 3.8 regions, and only with B 80.0 20.0 0 0 C 29.2 70.8 0 0 anti-SERCA2 in the C Bag2 A 62.0 20.0 2.0 16.0 region (Table 3; Figure B 26.9 38.5 7.7 26.9 1-3, paper IV). Sixty-two C 15.2 63.6 3.0 18.2 percent of the bag2 fibers Chain A 0.6 0 98.3 1.1 were also labeled by both B 0 0 100 0 anti-SERCA1 and anti- C 5.0 0 95.0 0 SERCA2 in the A region. Twenty percent of the bag2 fibers in the A region were only stained with anti-SERCA2 and 16% were unstained with both SERCA mAbs. A small number of bag2 fiber stained only by anti-SERCA1 were occasionally observed in the three regions (Table 3). Nuclear chain fibers were generally stained with anti-SERCA1 and the corresponding mAb against fast MyHCIIa. Nuclear chain fibers unstained with anti-SERCA2 were unstained with anti-SERCAI as well (Table 3).

Unlike nuclear chain fibers, such good correlation between SERCA and MyHC isoforms was not observed in bag fibers (Figure 4, paper IV). In fact, a great number of nuclear bag1 and bag2 fibers unstained with anti-MyHC mAbs were stained with the corresponding mAbs against slow or fast SERCA isoforms. In contrast, some bag fibers unstained with anti-SERCA were stained with anti-MyHCI. Moreover, the fast SERCA1 isoform was generally present over a much longer distance of the fiber length compared to the presence of MyHCIIa in nuclear bag1 fibers and that SERCA2 was seen over a shorter distance compared to the presence of MyHCI in bag2 fibers. In addition, the patterns of variation in staining intensity with mAbs against the SERCA isoforms were distinct from those observed with mAbs against the MyHC isoforms.

29 DISCUSSION

The present study disclosed fundamental features of the human muscle spindles, including their special fiber type composition, distinct patterns of co-expression of several MyHC isoforms in each fiber type as well as co-expression of both fast and slow SERCA isoforms in bag fibers. There were important differences between muscle spindles from the BB and the DN. Aging appears to have an important impact on muscle spindles with regard to intrafusal fiber content and MyHC composition.

Human Muscle Spindle Identity: BB versus DN

The total number of intrafusal fibers and the number of bag1 fibers were significantly higher in muscle spindles in BB than in DN. The present data showed that virtually each muscle spindle in BB had a different allotment of numbers of bag1, bag2 and chain fibers. In a rather large muscle with low muscle spindle density like the biceps, the position and function of each muscle spindle is likely to be of great importance. It is therefore not surprising that each muscle spindle presents unique morphological features. These structural variations fit with physiological data showing that each gamma-motoneuron has a response profile of its own (Johansson and Sojka, 1991).

The high muscle spindle density and the special features of the muscle spindles in the deep neck muscles allow not only great precision of movement but also adequate proprioceptive information needed both for control of head position and movements and for eye/head movement coordination (Bakker et al., 1984; Edney and Porter, 1986; Porter, 1986). In a small muscle with high spindle density and a postural role it is possible that several muscle spindles cover a particular muscle volume and thereby have similar functional requirements, reflected in a high frequency of muscle spindles with identical allotment of fiber types. On the other hand, the presence of b1c and b2c spindles and a wide variation in the patterns distribution of MyHCs along the fiber length indicate complex sensorimotor organization and might reflect higher plasticity. This study clearly revealed that the muscle spindles of the deep neck muscles have distinct morphological features and MyHC composition. The data further support the concept that each muscle is unique (Thornell et al., 2003a) both with regard to fiber type composition and sensorimotor organization.

The identification of such distinctive features confirming a particular identity to each muscle spindle reflect a high level of complexity for these receptors which is likely to reflect their spatial and functional context in the muscles as well as important plastic adaptation.

Muscle spindles lacking bag1 fibers

Three categories of muscle spindles, b1b2c, b1c and b2c, were identified in DN. Our findings suggest that human neck muscle spindles can lack either bag1 or bag2 fibers. Moreover, ten out of twelve b1c spindles existed as single isolated receptors and four contained two bag1 fibers, indicating that the b1c spindles in human neck muscles are rather common and do not represent a part of complexly linked spindles or a developmental deviation. B1c spindles have also been observed in BB.

30

The ability to distinguish between b1c and b2c spindles depends on the precise observation of differences in alkaline- and acid-ATPase activity between bag1 and bag2 fibers. In the outer A and B regions, the bag2 fiber can be clearly distinguished from bag1 fiber by its higher staining intensity under mild acid (pH 4.6) and acid (pH 4.3) pre-incubation (Figure 1, Paper II). Moreover, immunocytochemistry can also be used as an additional manner of identifying fiber types. For instance, bag2 fibers are strongly stained by anti-MyHC"all except IIx" in the A and B regions whereas bag1 fibers are unstained or weakly stained. The use of all these criteria allowed a reliable classification of nuclear bag1 and bag2 fibers in the present study.

One-bag-fiber spindles have also been found in cat muscles (Banks et al., 1979; Kucera, 1982; Kucera and Walro, 1987), particularly in the cat neck muscles (Abrahams and Richmond, 1988; Bakker and Richmond, 1982, 1981; Richmond and Abrahams, 1975; Richmond et al., 1986). However, these one-bag-fiber spindles selectively lack bag1 fibers with a single exception reported (Kucera, 1982). The b2c spindles usually contain only one bag2 fiber and appear as part of complexes, such as tandem spindles (Bakker and Richmond, 1982, 1981; Richmond et al., 1986) or dyads in which a b2c spindle and a Golgi tendon organ lie side-by-side (Abrahams and Richmond, 1988; Richmond and Abrahams, 1979b). Bakker and Richmond (1981) have reported that the distribution of b2c spindles is restricted to the regions near tendons or tendinous inscriptions in dorsal neck muscles in cat. However, we did not find any preferential location for the b1c spindles in our samples of the human deep neck muscles. Rather, the b1c units were located in connective tissue septa within the muscles.

In a conventional spindle, the bag1 fiber is known to play a major role in the production of the velocity-sensitive (dynamic) response in the primary ending whereas the bag2 fiber mediates the length (static) sensitivity along with the chain fibers (Barker and Banks, 1994; Boyd, 1981; Matthews, 1981). The lack of bag1 fibers in the b2c spindles implies that these spindles are exclusively length (static) sensitive. Because of the lack of data on human subjects with regard to b1c spindles, the functional significance of these muscle spindles remains to be elucidated.

MyHC Composition of Isolated Muscle Spindles

Four MyHC isoforms were present in intrafusal fibers in amounts sufficient to be detected by SDS-PAGE. The present study confirmed the presence of MyHCI, MyHCif and MyHCemb (Pedrosa-Domellof et al., 1993). The band migrating slightly faster than the band of MyHCemb has now been characterized by immunoblotting as MyHCIIa rather than MyHCfet as previously assumed (Pedrosa-Domellof et al., 1993).

When sufficient spindles from the biceps or from the first lumbrical muscle were pooled together for SDS-PAGE, similar migration patterns were observed, indicating that MyHCIIa, MyHCemb, MyHCif and MyHCI are the major isoforms present in human muscle spindles. However, it should be noted that clear differences in the relative amounts of these four isoforms are detectable by SDS-PAGE when comparing distinct single muscle spindles or the different regions of a single muscle spindle (Pedrosa-Domellof et al., 1993).

31 This further supports the detailed data obtained with immunocytochemistry (see below). SDS-PAGE is not sensitive enough to detect other isoforms present in very small amounts nor does it reflect the full complexity of the distribution of these MyHC isoforms among and along the spindle fibers.

MyHC Composition of Intrafusal Fibers

The study shows that the three types of intrafusal fibers differ from extrafusal fibers by co- expressing several MyHC isoforms, including MyHCsto, MyHCα-c, MyHCfet and MyHC emb, not typically seen in extrafusal fibers of limb and trunk muscles.

Human nuclear bag fibers of both BB and DN were strongly stained with the antibody against MyHCsto along their entire length. In contrast, bag2 fibers are typically less stained in the inner B region, and unstained in the outer B and C regions in rat hindlimb muscle spindles, (Pedrosa-Domellof et al., 1991; Wang et al., 1997). Even in human masseter muscle spindles (Eriksson et al., 1994), the reactivity and regional variability of bag2 fibers resembles the findings in rat with respect to MyHCsto. MyHCsto has been demonstrated in rat chain fibers for a short distance in the equatorial region (Kucera and Walro, 1989; Pedrosa-Domellof et al., 1991; Wang et al., 1997). However, it was never found in any of the 127 chain fibers of human BB and DN investigated here. These findings stress the presence of species differences and the need for more data on human muscle spindles.

It is obvious from the current results that nuclear chain fibers lacked MyHCsto, MyHCα-c and MyHCI but contained MyHCIIa. Although immunocytochemistry is not a quantitative method, the binding patterns of mAb A4.74 indicate that the amount of MyHCIIa varied along the fiber length, being considerably less in the juxtaequatorial region. Thus, in the juxtaequatorial region, chain fibers likely contained MyHCIIx isoform according to their strong and even binding with mAb VII1H3, absent staining with the pure anti-MyHCI as well as their weaker staining/unstaining with A4.74 mAb. Confirmation of whether MyHCIIx was present only in the juxtaequatorial region or in other regions as well was not possible in this study because of the unavailability of a pure anti-MyHCIIx antibody. The presence of MyHCIIa and MyHCIIx in chain fibers in the present study is in contrast with reports that neither of these two MyHCs were found in rat chain fibers (Kucera et al., 1992; Walro et al., 1997; Wang et al., 1997), although rat chain fibers do contain a specific form of fast twitch myosin (Kucera et al., 1992).

Two developmental MyHC isoforms, MyHCemb and MyHCfet, were found in all three intrafusal fiber types, although reactivity in bag fibers was lower than that in chain fibers, or might even be absent for a short distance. Maier et al. (1988a) reported that chain fibers of cat spindles reacted strongly with anti-MyHCemb, but those of rat and rabbit did not react at all. In contrast, Wang et al. (1997) using a distinct anti-MyHCemb antibody, found that rat chain fibers contained MyHCemb in their intracapsular region. MyHCfet is present in chain and bag2 fibers, but lacking in bag1 fibers of rat muscle spindles (Kucera and Walro, 1989; Pedrosa et al., 1989; Wang et al., 1997). This is in contrast with our finding of MyHCfet in bag1 fibers.

32 MyHCα-c was present in both bag1 and bag2 fibers but lacking in chain fibers, as previously reported on both human and rat muscle spindles (Pedrosa et al., 1990; Pedrosa- Domellof and Thornell, 1994; Wang et al., 1997). However, the distribution of MyHCα-c in intrafusal fibers of human biceps was distinct from that of rat and it remains to be determined whether motor innervation regulates the expression of this isoform (Pedrosa et al., 1990) in human nuclear bag fibers as well.

The immunocytochemical staining profiles of the intrafusal fibers of DN generally paralleled those of BB, although with some important exceptions. The most striking difference was the variation in MyHCemb and MyHCfet content of nuclear chain fibers. The distribution of the embryonic and fetal myosin varied between the chain fibers from different spindles, and along the length of a given chain fiber. These variations seemed to correlate to the variability in mATPase activity noted along the length of individual nuclear chain fibers in the neck muscle spindles. Nuclear chain fibers usually have uniform mATPase activity and a homogeneous MyHC composition along their entire length in muscle spindles of human biceps brachii (Liu et al., 2002) and the first lumbrical muscles (Soukup et al. unpublished data).

There was wide variation in MyHC composition among the intrafusal fibers studied. In fact, we found more variation in the staining patterns along the intrafusal fibers in the DN than in the BB. The widely variable staining features are likely to reflect a wider range of contractile properties of the intrafusal fibers and the more complex architecture and functions of these small deep neck muscles. The deep neck muscles have limited lever action because of their small size (Kamibayashi and Richmond, 1998) and in the rhesus monkey they are preferentially recruited during small turns whereas larger, multiarticular muscles became active additionally during larger faster turns (Corneil et al., 2001).

Effects of Aging on Muscle Spindles

A decrease in the number of intrafusal fibers and increased capsule thickness with increasing age have previously been reported in human muscle spindles (Swash and Fox, 1972). Here we showed that the average number of both nuclear bag1 and bag2 fibers per spindle did not change with increasing age whereas the nuclear chain fibers decreased significantly in number or were completely absent in some muscle spindles. Furthermore, the chain fibers also became much shorter in the aged spindles as they were seldom seen in the inner B region and were not found in the outer B or C regions.

Further studies are needed to determine the implications of the lack/decreased number of nuclear chain fibers. One would expect those muscle spindles to have a relatively less static sensitivity compared to the muscle spindles from young adults. This assumption is supported by previous physiological data that revealed a decline in static position sense at the ankle joint in the elderly (Meeuwsen et al., 1993; Robbins et al., 1995). A deterioration in the dynamic position sense has also been observed with aging (Verschueren et al., 2002), but it was six-fold smaller than the deterioration observed in static position sense (Robbins et al., 1995; Verschueren et al., 2002).

33 The age-related increase in the MyHCI isoform in bag1 fibers, decrease in the MyHCα-c and MyHCfet isoforms in bag fibers likely reflect a decrease in both the sensory and motor activities in muscle spindles with aging. However, the number of intrafusal fibers does not change following deafferentation and deefferentation in the adult rat (Walro et al., 1997; Wang et al., 1997) indicating that the changes seen with aging in human are more complex and may be multifactorial. We suggest that deterioration in the motor and sensory innervation might cause the phenotypic changes observed in aged muscle spindles and are likely to have an impact in motor control in the elderly.

SERCA Isoforms

Generally, the slow SERCA2 isoform was absent in nuclear chain fibers but present in nuclear bag fibers whereas the fast SERCA1 isoform was expressed in all three types of intrafusal fibers but differed considerably in its distribution. The staining intensity with the SERCA1 mAb varied in the order of chain>bag1>bag2, which is in accordance with the findings by Maier et al. (Maier et al., 1988b) in rodent muscle spindles.

Previous studies showed a greater abundance of the SR system in nuclear chain fibers (Ogata and Yamasaki, 1991, 1992; Ovalle, 1971, 1972; Sohn et al., 1999), which matches well the faster relaxation and contraction speed of this intrafusal fiber type (Boyd, 1980). In addition, a rather homogeneous and strong staining for fast SERCA1 and MyHCIIa was observed in nuclear chain fibers in the present study. As for the MyHC composition, the nuclear bag fibers had a more complex and variable SERCA composition. In the rat, the basic structure of the SR of the bag2 fibers resembles that of the chain fibers, although the terminal cisternae and SR networks are less well developed in the bag2 fibers than in the chain fibers (Ogata and Yamasaki, 1991). Immunoelectronmicroscopy studies will be needed to elucidate and correlate the morphology and the composition of the SR in human intrafusal fibers.

The low level of reactivity observed with both anti-SERCA antibodies in the B region of the nuclear bag2 fibers is in agreement with the electron microscopy findings that the SR is rare in bag2 fibers in the sleeve region (Ogata and Yamasaki, 1991).

The dissociation between SERCA and MyHC isoforms seen in the bag1 and bag2 fibers is not likely to be related to their motor innervation, as the nuclear bag1 fibers are innervated by dynamic efferents while the nuclear bag2 and chain fibers are innervated by static efferents (Barker and Banks, 1994; Boyd, 1980). Instead, we proposed that the presence of slow tonic MyHC, (MyHCsto) might be a relevant factor in this dissociation since MyHCsto is present in both the bag1 and bag2 fibers but absent in the chain fibers. This speculation can be supported by findings that in the extraocular muscles of the rat and rabbit (Jacoby and Ko, 1993), as well as of human (Kjellgren et al. 2003a), there were fibers containing MyHCsto and mixtures of SERCA isoforms.

34 Future Perspectives

Determining the MyHC composition of the intrafusal fibers is of importance because the contractile proteins mediate the sensitivity to stretch. Further characterization of the normal adult human muscle spindles with respect to their intra- and extrasarcomeric cytoskeleton, extracellular matrix and additional sarcomeric proteins is needed in order to determine the structure and function of these receptors and how they are affected in disease. The M-band proteins are of utmost importance for the early assembly of myofibrils and they link the contractile apparatus with the sarcomeric cytoskeleton. Ultrastrutural data indicate that intrafusal fibers differ from extrafusal muscle fibers regarding the structure or even the presence of M-band (Kucera and Dorovini-Zis, 1979; Ovalle, 1971; Thornell et al., 1995). A study of the M-band composition of human muscle spindles is underway.

The patterns of MyHC expression that occur in myotubes after they are innervated by Ia afferents have been well studied in both human (Pedrosa-Domellof and Thornell, 1994; Thornell et al., 1988, 1989) and rats (for review see Soukup et al., 1995; Walro and Kucera, 1999). There is no doubt that both sensory and motor innervation play key roles in regulating the formation and maintenance of muscle spindles as well as of expression of MyHC isoforms at different stages. The zinc-finger transcription factor Egr3 has been demonstrated to have an essential role in regulating genes required for the transformation of undifferentiated myotubes into intrafusal fibers, and hence for the phenotypic differentiation of muscle spindles (Tourtellotte et al., 2001; Tourtellotte and Milbrandt, 1998). Meanwhile, spindle-derived neurotrophin-3 (NT-3) has been identified as a the principle factor regulating the synaptic connectivity between muscle sensory and motor neurons (Chen et al., 2002). However, the molecular mechanisms involved in the morphogenesis of muscle spindles, such as the signaling molecules mediating afferent- myotube trophic interactions, the corresponding intracellular signal transduction mechanisms, or the patterns of gene expression leading to differentiation of intrafusal myotubes, are unknown.

35 SUMMARY

BB differed from DN in i) the intrafusal fiber type composition; ii) the total number of intrafusal fibers per spindle, and iii) the MyHC composition. Nearly each muscle spindle in the BB had a unique allotment of numbers of nuclear bag1, bag2 and chain fibers. In contrast the incidence of spindles with a unique fiber complement was rather low in the DN. A novel category of muscle spindles containing only nuclear bag1 and chain fibers was found in addition to the muscle spindles containing bag1+bag2+chain or bag2+chain fibers. Four major MyHC isoforms, MyHCI, MyHCIIa, MyHCfet and MyHCif could be identified in isolated muscle spindles by SDS-PAGE. Several MyHC isoforms were co-expressed in intrafusal fibers in both the BB and DN, and the patterns of MyHC expression varied along the length of the fibers, with the exception of MyHCsto. There were differences in the pattern of MyHC expression between the BB and the DN. MyHCfet, detected in nuclear bag1 fibers in the BB, was not observed in nuclear bag1 fibers in the DN. The expression of MyHCemb and MyHCfet in nuclear chain fibers showed regional variation in the DN only. The staining patterns along the fiber length varied more in the DN than in the BB.

The number of nuclear chain fibers of BB muscle spindles was reduced with aging. The expression of MyHCα-c in nuclear bag1 and bag2 fibers was greatly decreased in the A region; the expression of MyHCI was increased in nuclear bag1 fibers and that of MyHCfet decreased in bag2 fibers, whereas the patterns of distribution of the remaining MyHC isoforms were generally not affected by aging.

Nuclear bag1 fibers contained both SERCA1 and SERCA2 isoforms, with predominant staining seen with SERCA2 in the encapsulated regions. Most nuclear bag2 fibers also contained SERCA1 and SERCA2 isoforms and their coexistence was more frequent in the A region. SERCA1 was present whereas SERCA2 was generally absent in the nuclear chain fibers. There was no correlation between the presence of fast or slow MyHCs and SERCA1 or SERCA2 isoforms, respectively, in nuclear bag fibers whereas there was a strong correlation between the MyHCIIa and SERCA1 content in the nuclear chain fibers.

36 CONCLUSIONS

The present thesis further advances our knowledge of the human muscle spindles and provides a framework for future studies of these receptors in muscular dysfunction and neuromuscular disease.

Human muscle spindles proved to be more complex than anticipated.

The unique features of each muscle spindle in terms of i) fiber number and fiber type composition; ii) MyHC and SERCA content probably reflect individual specialization and minute adaptation to meet a variety of functional demands.

The differences between BB and DN muscle spindles suggest functional specialization in the control of movement among different human muscles

The MyHC and SERCA phenotype revealed for the human nuclear bag fibers was very intricate and adapted to attain a wide range of contraction and relaxation velocities.

The age-related changes in muscle spindle fiber composition and phenotype may reflect deterioration in sensory and motor innervation and are likely to have a detrimental impact on motor behavior in the elderly.

37 ACKNOWLEDGEMENTS

I have been waiting for the moment to write this part for a long time. For a person like me who is not good at expressing his/herself orally, there are no better places than here to express his/her great gratitude to the people who have helped for the past years. I sincerely thank all the people in the Department of Integrative Medical Biology for creating such a pleasant, creative and inspiring environment to work in. I enjoy my every single day there. In particular I would like to thank those who have provided their specific help to this thesis.

First of all, I would like to express my gratitude to my supervisor Fatima Pedrosa-Domellöf. Thank you for believing in me and supporting my ideas since day one, for your encouragement and understanding which have been an important support in making me feel confident to finish this project, for your patience to revise my writing word by word, for the numerous wonderful parties at your home and for teaching me how to bake the delicious apple pie. Over the past six years I have learned a great deal from you, both scientific and general. I am still amazed by your never-ending energy and enthusiasm.

Professor Lars-Eric Thornell, my co-supervisor. Thank you not only for providing me the opportunity to join your research group but also for giving me your constant support and encouragement, for taking care of me and my family when we first came to Umeå, for those nice dinners in every summer vacation and Julmat in winter so that we never felt left out in Umeå.

Professor Per-Olof Eriksson, my co-supervisor. When I first met you in your lab you explained nearly one hour how the masticatory muscle system functions. From you I also got my first lesson of how to distinguish nuclear bag fibers from nuclear chain fibers so that I could easily step into the new field. Thanks for updating my knowledge by constantly sending scientific reprints which I desperately needed.

I would also like to express my gratitude to late Professor Håkan Johansson, at the Center for Musculoskeletal Research, for burning enthusiasm for muscle spindles, kindness and financial support. Although you are not with us physically, you are very much alive in our mind and heart.

All the members of “muscle group” for sharing wonderful time together. I am grateful for the practical help at the lab bench from Anna-Karin Olofsson (my first experimental tutor, very patient with me), Margaretha Enerstedt (your laugh makes me feel that life is joyful) and Mona Lindström (I learned a lot of lab rules from you). Fawzi Kadi, for teaching me how to run the gel. Thanks also to Bengt Johansson, for the pleasant time at your summerhouse. Lena Carlsson, for your e-mails, for answering my questions and correcting my Swedish. Per Stål, for always presenting a happy face to cheer people up, for the laugh and jokes you brought to us. Eva Carlsson, for your many help and advice. Christer Malm, for nice chat during coffee break. Daniel Kjellgren, for your candy and chocolate, especially for your “intensive care” by closing the door behind you while I was still sitting in the office. My fellow doctoral students, Anders Eriksson, when will you finish your thesis? Good Luck! Dina Radovanovic, I admire your full energy to take care of your three children, especially when your husband was on business trip. When will you come back again?

My come-and-go office roommates, Dr. Tomas Soukup from Czech, Dr. Eva Ponten from Uppsala, and Ophthalmologist Berit Byström in Umeå for the fresh air you brought in. Masae Morishima for your timely companion when I was just about to sit there alone.

38 Thanks to the members of P-O’s lab at the Department of Clinical Oral Physiology: Inga Johansson, for your skillful technical assistance. Mehrdad Monemi, for collaboration and being co-author. Hamayun Zafar and Catharina Österlund, for always welcoming me whenever I showed up.

Ingrid Nilsson and Anna-Lena Tallander, the secretaries of the Department, for administrative support. Anna-Lena Tallander, for fixing everything for my flying back to Sweden and helping me with my thesis cover. Tomas Carlsson and Göran Dahlgren, for keeping my computer running.

Thanks also to Gunnel Folkesson, Lena Jonsson, Luda Novikova, Maria Brohlin, Maria Jonsson, and Ulla Hedlund, Ulla-Stina Spetz, Sigrid Kilter, for those unforgettable “tjej kvällen/lunchen”. Jianjun Ma and Chengang Zhang for nice chat in and out of the Anatomy building.

To all my Chinese friends in Umeå, for sharing time and feelings with me. Special thanks to the families of Chen Sa, Jiang Wei, Jian-Xi, Li-Min, Su-Yan, Tong-Yun, Xiao-Hong and Zhi-Yun for Sunday Chinese school, family gathering and New Year party. Zhi-Yun and your parents, for support and help before and after I moved to Chicago. Jiang Wei for parties and ready-to-help. Li- Min, Tong-Yun for the fun and laugh we families have had together. Ai-Hong and Jiang Wei for taking care of Zhuohang whilst I was in Paris.

My biggest regret is that my father, Lujia Liu, passed away two years before the completion of the thesis. It would have been a special treat for you if....

My mother, Wenrei Fu, for your love and support no matter what. I owe you more than words can state. My brother Tiejun and sisters Jingjun and Meijun together with your families, Haijing, Wenjian and Li Wei, for being important persons in my life, for just being there and sharing your lives with me, for taking care of our parents whilst I was away. Your love and support, demonstrated in so many ways at so many times, which make this thesis possible. My niece Liang Meng and nephew Guanyu, for sharing your happy childhood with Zhouhang so that he can remember it vividly forever. My one-month-old niece Tiantian, welcome to the world.

I also want to thank my mother-in-law, Guilan Sun, for your selflessness, for being a great caregiver, for setting a role model not only as a mother but also as a person for your three younger generations. My brother- and sister-in-laws, for taking most of the responsibility to take care of my mother-in-law and for being nice uncles and aunts to Zhuohang.

To my husband, Jiguo. To say thanks seems so small....but THANK YOU, for listening to me daily through good and bad. We have now shared almost two decades of friendship and love, and each day I feel luckier and luckier to have found you in my life. To my son, Zhuohang, for the happiness and joy you brought to my life. Over the last eight months, when we moved to Chicago, leaving behind all your friends and good memories, yet you still love us and you have done an excellent job in the new school. This is really what it means to live rich. Thank you for making me realize how special life is.

This work was supported by grants from The Medical Faculty, Umeå University, The Swedish Society for Medical Research, The Swedish Medical Research Council (12X-03934) and Magnus Bergvall Foundation.

39

40 REFERENCES

Abrahams, V. C. and Richmond, F. J. (1988). Specialization of sensorimotor organization in the neck muscle system. Prog Brain Res, 76, 125-135. Aniansson, A., Hedberg, M., Henning, G. B. and Grimby, G. (1986). Muscle morphology, enzymatic activity, and muscle strength in elderly men: a follow-up study. Muscle Nerve, 9, 585-591. Bakker, D. A. and Richmond, F. J. (1982). Muscle spindle complexes in muscles around upper cervical vertebrae in the cat. J Neurophysiol, 48, 62-74. Bakker, D. A., Richmond, F. J. and Abrahams, V. C. (1984). Central projections from cat suboccipital muscles: a study using transganglionic transport of horseradish peroxidase. J Comp Neurol, 228, 409-421. Bakker, G. J. and Richmond, F. J. (1981). Two types of muscle spindles in cat neck muscles: a histochemical study of intrafusal fiber composition. J Neurophysiol, 45, 973-986. Bandman, E. (1999). Functional properties of myosin isoforms in avian muscle. Poult Sci, 78, 729-734. Banks, R. W., Barker, D. and Stacey, M. J. (1979). Sensory innervation of cat hind-limb muscle spindles. J Physiol, 293, 40P-41P. Barker, D. and Banks, R. W. (1994). The muscle spindle. In: ENgel AG, Franzini- Armstrong C, eds. Myology.New York, McGraw-Hill, 333-360. Berg, J. S., Powell, B. C. and Cheney, R. E. (2001). A millennial myosin census. Mol Biol Cell, 12, 780-794. Bormioli, S. P., Torresan, P., Sartore, S., Moschini, G. B. and Schiaffino, S. (1979). Immunohistochemical identification of slow-tonic fibers in human extrinsic eye muscles. Invest Ophthalmol Vis Sci, 18, 303-306. Bottinelli, R., Canepari, M., Pellegrino, M. A. and Reggiani, C. (1996). Force-velocity properties of human skeletal muscle fibres: myosin heavy chain isoform and temperature dependence. J Physiol (Lond), 495, 573-586. Bottinelli, R. and Reggiani, C. (2000). Human skeletal muscle fibres: molecular and functional diversity. Prog Biophys Mol Biol, 73, 195-262. Boyd, I. and Smith, R. (1984). The Muscle Spindle. In: Dyck PJ, Thomas PK, Lombert EH, Bungr R, ed. Peripheral Neuropathy. WB Saunders Company, 171-202. Boyd, I. A. (1962). The struction and innervation of the nuclear bag muscle fiber system and the nuclear chain muscle fiber system in mammalian muscle spindles. London Royal Society Philosophical Transactions 245: 81-136. Boyd, I. A. (1980). The isolated mammalian muscle spindle. Trends in neurosciences, Nov 3, 258-265. Briggs, M. M. and Schachat, F. (2000). Early specialization of the superfast myosin in extraocular and laryngeal muscles. J Exp Biol, 203, 2485-2494. Brooke, M. H. and Engel, W. K. (1969). The histographic analysis of human muscle biopsies with regard to fiber types. 1. Adult male and female. Neurology, 19, 221- 233. Burk, S. E., Lytton, J., MacLennan, D. H. and Shull, G. E. (1989). cDNA cloning, functional expression, and mRNA tissue distribution of a third organellar Ca2+ pump. J Biol Chem, 264, 18561-18568.

41 Butler-Browne, G. S., Eriksson, P. O., Laurent, C. and Thornell, L. E. (1988). Adult human masseter muscle fibers express myosin isozymes characteristic of development. Muscle Nerve, 11, 610-620. Chen, H. H., Tourtellotte, W. G. and Frank, E. (2002). Muscle spindle-derived neurotrophin 3 regulates synaptic connectivity between muscle sensory and motor neurons. J Neurosci, 22, 3512-3519. Cooper, S. and Daniel, P. M. (1963). Muscle spindles in man: Their morphology in the lumbricals and the deep muscles of the neck. Brain, 86, 563-586. Corneil, B. D., Olivier, E., Richmond, F. J., Loeb, G. E. and Munoz, D. P. (2001). Neck muscles in the rhesus monkey. II. Electromyographic patterns of activation underlying postures and movements. J Neurophysiol, 86, 1729-1749. Dubowitz, V. (1985). Muscle biopsy: A practical approach. 2nd ed. London, Bailliere Tindall. Dux, L. (1993). Muscle relaxation and sarcoplasmic reticulum function in different muscle types. Rev Physiol Biochem Pharmacol, 122, 69-147. Edney, D. P. and Porter, J. D. (1986). Neck muscle afferent projections to the brainstem of the monkey: implications for the neural control of gaze. J Comp Neurol, 250, 389- 398. Eldred, E., Maier, A. and Bridgman, C. F. (1974). Differences in intrafusal fiber content of spindles in several muscles of the cat. Exp Neurol, 45, 8-18. English, A. W., Eason, J., Pol, M., Schwartz, G. and Shirley, A. (1998). Different phenotypes among slow/beta myosin heavy chain-containing fibres of rabbit masseter muscle: a novel type of diversity in adult muscle. J Muscle Res Cell Motil, 19, 525-535. Eriksson, O., Eriksson, A., Ringqvist, M. and Thornell, L. E. (1980). The reliability of histochemical fibre typing of human necropsy muscles. Histochemistry, 65, 193- 205. Eriksson, P. O., Butler-Browne, G. S. and Thornell, L. E. (1994). Immunohistochemical characterization of human masseter muscle spindles. Muscle Nerve, 17, 31-41. Eriksson, P. O. and Thornell, L. E. (1987). Relation to extrafusal fibre-type composition in muscle-spindle structure and location in the human masseter muscle. Arch Oral Biol, 32, 483-491. Ernfors, P., Lee, K. F., Kucera, J. and Jaenisch, R. (1994). Lack of neurotrophin-3 leads to deficiencies in the peripheral nervous system and loss of limb proprioceptive afferents. Cell, 77, 503-512. Ferrell, W. R., Crighton, A. and Sturrock, R. D. (1992). Age-dependent changes in position sense in human proximal interphalangeal joints. Neuroreport, 3, 259-261. Franzini-Armstrong, C. (1994). The sarcoplasmic reticulum and the transverse tubules. In: ENgel AG, Franzini-Armstrong C, eds. Myology.New York, McGraw-Hill, 176- 199. Frontera, W. R., Hughes, V. A., Fielding, R. A., Fiatarone, M. A., Evans, W. J. and Roubenoff, R. (2000). Aging of skeletal muscle: a 12-yr longitudinal study. J Appl Physiol, 88, 1321-1326. Gilsing, M. G., Van den Bosch, C. G., Lee, S. G., Ashton-Miller, J. A., Alexander, N. B., Schultz, A. B. and Ericson, W. A. (1995). Association of age with the threshold for detecting ankle inversion and eversion in upright stance. Age Ageing, 24, 58-66.

42 Gorza, L., Mercadier, J. J., Schwartz, K., Thornell, L. E., Sartore, S. and Schiaffino, S. (1984). Myosin types in the human heart. An immunofluorescence study of normal and hypertrophied atrial and ventricular myocardium. Circ Res, 54, 694-702. Grimby, G., Danneskiold-Samsoe, B., Hvid, K. and Saltin, B. (1982). Morphology and enzymatic capacity in arm and leg muscles in 78-81 year old men and women. Acta Physiol Scand, 115, 125-134. Grimby, G., Aniansson, A., Zetterberg, C. and Saltin, B. (1984). Is there a change in relative muscle fibre composition with age? Clin Physiol, 4, 189-194. Hamalainen, N. and Pette, D. (1997). Coordinated fast-to-slow transitions of myosin and SERCA isoforms in chronically stimulated muscles of euthyroid and hyperthyroid rabbits. J Muscle Res Cell Motil, 18, 545-554. Han, Y., Wang, J., Fischman, D. A., Biller, H. F. and Sanders, I. (1999). Slow tonic muscle fibers in the thyroarytenoid muscles of human vocal folds; a possible specialization for speech. Anat Rec, 256, 146-157. Harridge, S. D., Bottinelli, R., Canepari, M., Pellegrino, M. A., Reggiani, C., Esbjornsson, M. and Saltin, B. (1996). Whole-muscle and single-fibre contractile properties and myosin heavy chain isoforms in humans. Pflugers Arch, 432, 913-920. Hoh, J. F. (2002). 'Superfast' or masticatory myosin and the evolution of jaw-closing muscles of vertebrates. J Exp Biol, 205, 2203-2210. Hughes, S. M., Cho, M., Karsch-Mizrachi, I., Travis, M., Silberstein, L., Leinwand, L. A. and Blau, H. M. (1993). Three slow myosin heavy chains sequentially expressed in developing mammalian skeletal muscle. Dev Biol, 158, 183-199. Jacoby, J. and Ko, K. (1993). Sarcoplasmic reticulum fast CA(2+)-pump and myosin heavy chain expression in extraocular muscles. Invest Ophthalmol Vis Sci, 34, 2848-2858. Jandreski, M. A., Sole, M. J. and Liew, C. C. (1987). Two different forms of beta myosin heavy chain are expressed in human striated muscle. Hum Genet, 77, 127-131. Johansson, H., Sjolander, P., Djupsjobacka, M., Bergenheim, M. and Pedersen, J. (1999). Pathophysiological mechanisms behind work-related muscle pain syndromes. Am J Ind Med, Suppl 1, 104-106. Johansson, H. and Sojka, P. (1991). Pathophysiological mechanisms involved in genesis and spread of muscular tension in occupational muscle pain and in chronic musculoskeletal pain syndromes: a hypothesis. Med Hypotheses, 35, 196-203. Kadi, F. and Thornell, L. E. (1999). Training affects myosin heavy chain phenotype in the trapezius muscle of women. Histochem Cell Biol, 112, 73-78. Kamibayashi, L. K. and Richmond, F. J. (1998). Morphometry of human neck muscles. Spine, 23, 1314-1323. Kjellgren, D., Ryan, M., Ohlendieck, K., Thornell, L. E. and Pedrosa-Domellof, F. (2003a). Sarco(endo)plasmic reticulum Ca2+ ATPases (SERCA1 and -2) in human extraocular muscles. Invest Ophthalmol Vis Sci, 44, 5057-5062. Kjellgren, D., Thornell, L. E., Andersen, J. and Pedrosa-Domellof, F. (2003b). Myosin heavy chain isoforms in human extraocular muscles. Invest Ophthalmol Vis Sci, 44, 1419-1425. Korfage, J. A. and Van Eijden, T. M. (1999). Regional differences in fibre type composition in the human temporalis muscle. J Anat, 194, 355-362. Kucera, J. (1982). One-bag-fiber muscle spindles in tenuissimus muscles of the cat. Histochemistry, 76, 315-328.

43 Kucera, J. (1983). Multiple-bag-fiber muscle spindles in tenuissimus muscles of the cat. Histochemistry, 79, 457-476. Kucera, J. and Dorovini-Zis, K. (1979). Types of human intrafusal muscle fibers. Muscle Nerve, 2, 437-451. Kucera, J. and Walro, J. M. (1987). Heterogeneity of spindle units in the cat tenuissimus muscle. Am J Anat, 178, 269-278. Kucera, J. and Walro, J. M. (1989). Nonuniform expression of myosin heavy chain isoforms along the length of cat intrafusal muscle fibers. Histochemistry, 92, 291- 299. Kucera, J., Walro, J. M. and Gorza, L. (1992). Expression of type-specific MHC isoforms in rat intrafusal muscle fibers. J Histochem Cytochem, 40, 293-307. Kühnne, W. (1863). Ueber die Endigung der Berven im den Muskeln. Virchows Arch 27: 508-533. Kulkarni, V., Chandy, M. J. and Babu, K. S. (2001). Quantitative study of muscle spindles in suboccipital muscles of human foetuses. Neurol India, 49, 355-359. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680-685. Larsson, L. and Moss, R. L. (1993). Maximum velocity of shortening in relation to myosin isoform composition in single fibres from human skeletal muscles. J Physiol Lond, 472, 595-614. Larsson, L., Sjodin, B. and Karlsson, J. (1978). Histochemical and biochemical changes in human skeletal muscle with age in sedentary males, age 22--65 years. Acta Physiol Scand, 103, 31-39. Lexell, J. and Taylor, C. C. (1991). Variability in muscle fibre areas in whole human quadriceps muscle: effects of increasing age. J Anat, 174, 239-249. Lexell, J., Taylor, C. C. and Sjostrom, M. (1988). What is the cause of the ageing atrophy? Total number, size and proportion of different fiber types studied in whole vastus lateralis muscle from 15- to 83-year-old men. J Neurol Sci, 84, 275-294. Lieber, R. L. and Friden, J. (2000). Functional and clinical significance of skeletal muscle architecture. Muscle Nerve, 23, 1647-1666. Liu, J. X., Eriksson, P. O., Thornell, L. E. and Pedrosa-Domellof, F. (2002). Myosin heavy chain composition of muscle spindles in human biceps brachii. J Histochem Cytochem, 50, 171-184. Lompre, A.-M. (1998). The sarco(endo)plasmic Reticulum Ca2+-ATPases in the Cardiovascular System During Growth and Proliferation. Trends Cardiovasc Med, 8, 75-82. MacLennan, D. H., Brandl, C. J., Korczak, B. and Green, N. M. (1985). Amino-acid sequence of a Ca2+ + Mg2+-dependent ATPase from rabbit muscle sarcoplasmic reticulum, deduced from its complementary DNA sequence. Nature, 316, 696-700. Maier, A., Gambke, B. and Pette, D. (1988a). Immunohistochemical demonstration of embryonic myosin heavy chains in adult mammalian intrafusal fibers. Histochemistry, 88, 267-271. Maier, A., Leberer, E. and Pette, D. (1988b). Distribution of sarcoplasmic reticulum Ca- ATPase and of calsequestrin at the polar regions of rat, rabbit and cat intrafusal fibers. Histochemistry, 88, 273-276. Martonosi, A. (1994). Regulation of calcium by the sarcoplasmic reticulum. In: ENgel AG, Franzini-Armstrong C (eds) Myology., McGraw-Hill, New York, pp. 553-606.

44 Matre, D. A., Sinkjaer, T., Svensson, P. and Arendt-Nielsen, L. (1998). Experimental muscle pain increases the human stretch reflex. Pain, 75, 331-339. Matthews, P. B. (1981). Evolving views on the internal operation and functional role of the muscle spindle. J Physiol, 320, 1-30. Meeuwsen, H. J., Sawicki, T. M. and Stelmach, G. E. (1993). Improved foot position sense as a result of repetitions in older adults. J Gerontol, 48, P137-141. Mendell, L. M. and Henneman, E. (1971). Terminals of single Ia fibers: location, density, and distribution within a pool of 300 homonymous motoneurons. J Neurophysiol, 34, 171-187. Miwa, T., Miwa, Y. and Kanda, K. (1995). Dynamic and static sensitivities of muscle spindle primary endings in aged rats to ramp stretch. Neurosci Lett, 201, 179-182. Moller, J. V., Juul, B. and le Maire, M. (1996). Structural organization, ion transport, and energy transduction of P-type ATPases. Biochim Biophys Acta, 1286, 1-51. Monemi, M., Eriksson, P. O., Eriksson, A. and Thornell, L. E. (1998). Adverse changes in fibre type composition of the human masseter versus biceps brachii muscle during aging. J Neurol Sci, 154, 35-48. Monemi, M., Eriksson, P. O., Kadi, F., Butler-Browne, G. S. and Thornell, L. E. (1999a). Opposite changes in myosin heavy chain composition of human masseter and biceps brachii muscles during aging. J Muscle Res Cell Motil, 20, 351-361. Monemi, M., Kadi, F., Liu, J. X., Thornell, L. E. and Eriksson, P. O. (1999b). Adverse changes in fibre type and myosin heavy chain compositions of human jaw muscle vs. limb muscle during ageing. Acta Physiol Scand, 167, 339-345. Monemi, M., Liu, J. X., Thornell, L. E. and Eriksson, P. O. (2000). Myosin heavy chain composition of the human lateral pterygoid and digastric muscles in young adults and elderly. J Muscle Res Cell Motil, 21, 303-312. Monemi, M., Thornell, L. and Eriksson, P. (1999c). Diverse changes in fibre type composition of the human lateral pterygoid and digastric muscles during aging. J Neurol Sci, 171, 38-48. Moss, R. L., Diffee, G. M. and Greaser, M. L. (1995). Contractile properties of skeletal muscle fibers in relation to myofibrillar protein isoforms. Rev Physiol Biochem Pharmacol, 126, 1-63. Oakley, B. R., Kirsch, D. R. and Morris, N. R. (1980). A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal Biochem, 105, 361-363. Ogata, T. and Yamasaki, Y. (1991). Ultra-high-resolution scanning electron microscopic studies on the sarcoplasmic reticulum and mitochondria of the rat intrafusal muscle fibers. Part I. The capsular sleeve region. Arch Histol Cytol, 54, 471-490. Ogata, T. and Yamasaki, Y. (1992). Ultra-high-resolution scanning electron microscopic studies on the sarcoplasmic reticulum and mitochondria of the rat intrafusal muscle fibers. Part II: The extracapsular region. Arch Histol Cytol, 55, 117-124. Ovalle, W. K., Jr. (1971). Fine structure of rat intrafusal muscle fibers. The polar region. J Cell Biol, 51, 83-103. Ovalle, W. K., Jr. (1972). Fine structure of rat intrafusal muscle fibers. The equatorial region. J Cell Biol, 52, 382-396. Ovalle, W. K. and Smith, R. S. (1972). Histochemical identification of three types of intrafusal muscle fibers in the cat and monkey based on the myosin ATPase reaction. Can J Physiol Pharmacol, 50, 195-202.

45 Overend, T. J., Cunningham, D. A., Paterson, D. H. and Lefcoe, M. S. (1992). Thigh composition in young and elderly men determined by computed tomography. Clin Physiol, 12, 629-640. Peck, D., Buxton, D. F. and Nitz, A. (1984). A comparison of spindle concentrations in large and small muscles acting in parallel combinations. J Morphol, 180, 243-252. Pedrosa, F., Butler-Browne, G. S., Dhoot, G. K., Fischman, D. A. and Thornell, L. E. (1989). Diversity in expression of myosin heavy chain isoforms and M-band proteins in rat muscle spindles. Histochemistry, 92, 185-194. Pedrosa, F., Soukup, T. and Thornell, L. E. (1990). Expression of an alpha cardiac-like myosin heavy chain in muscle spindle fibres. Histochemistry, 95, 105-113. Pedrosa, F. and Thornell, L. E. (1990). Expression of myosin heavy chain isoforms in developing rat muscle spindles. Histochemistry, 94, 231-244. Pedrosa-Domellof, F., Eriksson, P. O., Butler-Browne, G. S. and Thornell, L. E. (1992). Expression of alpha-cardiac myosin heavy chain in mammalian skeletal muscle. Experientia, 48, 491-494. Pedrosa-Domellof, F., Gohlsch, B., Thornell, L. E. and Pette, D. (1993). Electrophoretically defined myosin heavy chain patterns of single human muscle spindles. FEBS Lett, 335, 239-242. Pedrosa-Domellof, F., Holmgren, Y., Lucas, C. A., Hoh, J. F. and Thornell, L. E. (2000). Human extraocular muscles: unique pattern of myosin heavy chain expression during myotube formation. Invest Ophthalmol Vis Sci, 41, 1608-1616. Pedrosa-Domellof, F., Soukup, T. and Thornell, L. E. (1991). Rat muscle spindle immunocytochemistry revisited. Histochemistry, 96, 327-338. Pedrosa-Domellof, F. and Thornell, L. E. (1994). Expression of myosin heavy chain isoforms in developing human muscle spindles. J Histochem Cytochem, 42, 77-88. Pette, D. and Staron, R. S. (1990). Cellular and molecular diversities of mammalian skeletal muscle fibers. Rev Physiol Biochem Pharmacol, 116, 1-76. Pette, D. and Staron, R. S. (1997). Mammalian skeletal muscle fiber type transitions. Int Rev Cytol, 170, 143-223. Porter, J. D. (1986). Brainstem terminations of extraocular muscle primary afferent neurons in the monkey. J Comp Neurol, 247, 133-143. Porter, M. M., Vandervoort, A. A. and Lexell, J. (1995). Aging of human muscle: structure, function and adaptability. Scand J Med Sci Sports, 5, 129-142. Quiroz-Rothe, E. and Rivero, J. L. (2001). Co-ordinated expression of contractile and non- contractile features of control equine muscle fibre types characterised by immunostaining of myosin heavy chains. Histochem Cell Biol, 116, 299-312. Rice, C. L., Cunningham, D. A., Paterson, D. H. and Lefcoe, M. S. (1989). Arm and leg composition determined by computed tomography in young and elderly men. Clin Physiol, 9, 207-220. Richmond, F. J. and Abrahams, V. C. (1975). Morphology and distribution of muscle spindles in dorsal muscles of the cat neck. J Neurophysiol, 38, 1322-1339. Richmond, F. J. and Abrahams, V. C. (1979a). Physiological properties of muscle spindles in dorsal neck muscles of the cat. J Neurophysiol, 42, 604-617. Richmond, F. J. and Abrahams, V. C. (1979b). What are the proprioceptors of the neck? Prog Brain Res, 50, 245-254. Richmond, F. J., Bakker, G. J., Bakker, D. A. and Stacey, M. J. (1986). The innervation of tandem muscle spindles in the cat neck. J Comp Neurol, 245, 483-497.

46 Robbins, S., Waked, E. and McClaran, J. (1995). Proprioception and stability: foot position awareness as a function of age and footwear. Age Ageing, 24, 67-72. Rowlerson, A., Mascarello, F., Barker, D. and Saed, H. (1988). Muscle-spindle distribution in relation to the fibre-type composition of masseter in mammals. J Anat, 161, 37- 60. Rozek, C. E. and Davidson, N. (1983). Drosophila has one myosin heavy-chain gene with three developmentally regulated transcripts. Cell, 32, 23-34. Sahgal, V., Subramani, V., Sahgal, S. and Kochar, H. (1985). Morphology and morphometry of human muscle spindles. In Boyd IA, Gladden MH (eds.) The muscle spindle. London, MacMillan, p 107-114. Sanders, I., Han, Y., Wang, J. and Biller, H. (1998). Muscle spindles are concentrated in the superior vocalis subcompartment of the human thyroarytenoid muscle. J Voice, 12, 7-16. Scelsi, R., Marchetti, C. and Poggi, P. (1980). Histochemical and ultrastructural aspects of m. vastus lateralis in sedentary old people (age 65--89 years). Acta Neuropathol (Berl), 51, 99-105. Schiaffino, S., Gorza, L., Dones, I., Cornelio, F. and Sartore, S. (1986a). Fetal myosin immunoreactivity in human dystrophic muscle. Muscle Nerve, 9, 51-58. Schiaffino, S., Gorza, L., Sartore, S., Saggin, L. and Carli, M. (1986b). Embryonic myosin heavy chain as a differentiation marker of developing human skeletal muscle and rhabdomyosarcoma. A monoclonal antibody study. Exp Cell Res, 163, 211-220. Schiaffino, S. and Reggiani, C. (1994). Myosin isoforms in mammalian skeletal muscle. J Appl Physiol, 77, 493-501. Schiaffino, S. and Reggiani, C. (1996). Molecular diversity of myofibrillar proteins: gene regulation and functional significance. Physiol Rev, 76, 371-423. Schmied, A., Morin, D., Vedel, J. P. and Pagni, S. (1997). The "size principle" and synaptic effectiveness of muscle afferent projections to human extensor carpi radialis motoneurones during wrist extension. Exp Brain Res, 113, 214-229. Sciote, J. J., Morris, T. J., Brandon, C. A., Horton, M. J. and Rosen, C. (2002). Unloaded shortening velocity and myosin heavy chain variations in human laryngeal muscle fibers. Ann Otol Rhinol Laryngol, 111, 120-127. Scutter, S. D. and Turker, K. S. (1999). Muscle spindle afferent input to motoneurons in human masseter. J Neurophysiol, 82, 505-507. Scutter, S. D. and Turker, K. S. (2001). The role of the muscle spindles in human masseter. Hum Mov Sci, 20, 489-497. Segal, R. L. (1992). Neuromuscular compartments in the human biceps brachii muscle. Neurosci Lett, 140, 98-102. Shiotani, A., Westra, W. H. and Flint, P. W. (1999). Myosin heavy chain composition in human laryngeal muscles. Laryngoscope, 109, 1521-1524. Shull, G. E. (2000). Gene knockout studies of Ca2+-transporting ATPases. Eur J Biochem, 267, 5284-5290. Skinner, H. B., Barrack, R. L. and Cook, S. D. (1984). Age-related decline in proprioception. Clin Orthop 184, 208-211. Sohn, A., Kaidoh, T. and Inoue, T. (1999). Scanning electron microscopic observation on the equatorial region of the rat intrafusal muscle fibres, especially on the subsarcolemmal sarcoplasmic reticulum. J Electron Microsc (Tokyo), 48, 659-663.

47 Soukup, T., Pedrosa-Domellof, F. and Thornell, L. E. (1993). Differentiation of supernumerary fibres in neonatally deefferented rat muscle spindles. Differentiation, 53, 35-43. Soukup, T., Pedrosa-Domellof, F. and Thornell, L. E. (1995). Expression of myosin heavy chain isoforms and myogenesis of intrafusal fibres in rat muscle spindles. Microsc Res Tech, 30, 390-407. Soukup, T., Pedrosa-Domellof, F. and Thornell, L. E. (2003). Intrafusal fiber type composition of muscle spindles in the first human lumbrical muscle. Acta Neuropathol (Berl), 105, 18-24. Stal, P., Eriksson, P. O., Schiaffino, S., Butler Browne, G. S. and Thornell, L. E. (1994). Differences in myosin composition between human oro-facial, masticatory and limb muscles: enzyme-, immunohisto- and biochemical studies. J Muscle Res Cell Motil, 15, 517-534. Sternberger, L. (1979). Immunohistochemistry. 2nd ed. New York, Wiley. Swash, M. and Fox, K. P. (1972). The effect of age on human skeletal muscle. Studies of the morphology and innervation of muscle spindles. J Neurol Sci, 16, 417-432. Talmadge, R. J., Roy, R. R., Chalmers, G. R. and Edgerton, V. R. (1996). MHC and sarcoplasmic reticulum protein isoforms in functionally overloaded cat plantaris muscle fibers. J Appl Physiol, 80, 1296-1303. ter Haar Romeny, B. M., Denier van der Gon, J. J. and Gielen, C. C. (1982). Changes in recruitment order of motor units in the human biceps muscle. Exp Neurol, 78, 360- 368. Thelen, D. G., Brockmiller, C., Ashton-Miller, J. A., Schultz, A. B. and Alexander, N. B. (1998). Thresholds for sensing foot dorsi- and plantarflexion during upright stance: effects of age and velocity. J Gerontol A Biol Sci Med Sci, 53, M33-38. Thornell, L.-E., Eriksson, P.-O., Fischman, D. A., Grove, B. K., Butler-Browne, G. S. and Virtanen, I. (1988). Human muscle spindle development. In: Hnik P, Soukup T, Vejsada R, Zelena J, ed. Mechanoreceptors, development, structure and function. New York , Plenum Press, 39-44. Thornell, L.-E., Grove, B. K., Pedrosa, F., Butler-Browne, G. S., Dhoot, G. K. and Fischman, D. A. (1989). Expression of slow tonic myosin in muscle spindle fibers early in mammalian development. In Stockdale, Kedes I, eds. Cellular and Molecular Biology of Muscle Development. New York, Alan R Liss, 471-480. Thornell, L.-E., Carlsson, L., Friden, J., F, P.-D., U, R. and T, S. (1995). Ultrastructure of bag and chain fibres and 3-D reconstruction of the sensory innervation in human lumbrical spindles. In Alpha and Gamma Moroe Systems, eds Taylor A, Gladden MH and Durbaba R. Plenum Press, New York and London, P222-229. Thornell, L.-E., Lindstrom, M., Renault, V., Mouly, V. and Butler-Browne, G. S. (2003b). Satellite cells and training in the elderly. Scand J Med Sci Sports, 13, 48-55. Thornell, L.-E., Kadi, F., Lindman, R. and Pedrosa-Domellof, F. (2003a). Morphological features related to muscle pain and muscle overload. In: Chronic work-related Myalgia: Neuromuscular mechanisms behind work-related chronic muscle pain syndromes, eds Johansson H, Windhorst U, Djupsjöbacka and M, Passatore M. Gävle University Press, pp. 95-109. Tomonaga, M. (1977). Histochemical and ultrastructural changes in senile human skeletal muscle. J Am Geriatr Soc, 25, 125-131.

48 Tourtellotte, W. G., Keller-Peck, C., Milbrandt, J. and Kucera, J. (2001). The transcription factor Egr3 modulates sensory axon-myotube interactions during muscle spindle morphogenesis. Dev Biol, 232, 388-399. Tourtellotte, W. G. and Milbrandt, J. (1998). Sensory ataxia and muscle spindle agenesis in mice lacking the transcription factor Egr3. Nat Genet, 20, 87-91. Verschueren, S. M., Brumagne, S., Swinnen, S. P. and Cordo, P. J. (2002). The effect of aging on dynamic position sense at the ankle. Behav Brain Res, 136, 593-603. Voss, H. (1971). Tabulation of the absolute and relative muscular spindle numbers in human skeletal musculature. Anat Anz, 129, 562-572. Walro, J. M. and Kucera, J. (1999). Why adult mammalian intrafusal and extrafusal fibers contain different myosin heavy-chain isoforms. Trends Neurosci, 22, 180-184. Walro, J. M., Wang, J. and Story, G. M. (1997). Afferent-inherent regulation of myosin heavy chain isoforms in rat muscle spindles. Muscle Nerve, 20, 1549-1560. Wang, J., McWhorter, D. L. and Walro, J. M. (1997). Stability of myosin heavy chain isoforms in selectively denervated adult rat muscle spindles. Anat Rec, 249, 32-43. Weismann. (1861). Über das Wachsen der quergestreiften Muskeln nach Beobachtungen am Frosch. Zeitschrift für rationelle Medizin. 10, 263-284. Weiss, A. and Leinwand, L. A. (1996). The mammalian myosin heavy chain gene family. Annu Rev Cell Dev Biol, 12, 417-439. Weiss, A., Schiaffino, S. and Leinwand, L. A. (1999). Comparative sequence analysis of the complete human sarcomeric myosin heavy chain family: implications for functional diversity. J Mol Biol, 290, 61-75. Zelena, J. and Soukup, T. (1973). Development of muscle spindles deprived of fusimotor innervation. Z Zellforsch Mikrosk Anat, 144, 435-452. Zelena, J. and Soukup, T. (1993). Increase in the number of intrafusal muscle fibres in rat muscles after neonatal motor denervation. Neuroscience, 52, 207-218.

49