Skeletal muscle sarcoplasmic reticulum phenotype in myotonic dystrophy

Neuromusc. Disord., Vol. 6. No. 1, pp. 33--47, 1996

Pergamon

Copyright © 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0960-8966/96 $15.00 + .00

0960--8966(95) 00016-X

SKELETAL

MUSCLE

SARCOPLASMIC

IN MYOTONIC

RETICULUM

PHENOTYPE

DYSTROPHY

E. DAMIANI,* C. ANGELINI,t M. PELOSI,* R. SACCHETTO,* E. BORTOLOSO* and A. MARGRETH:~ *National Research Council Unit for Muscle Biologyand Physiopathology,Department of Biomedical Sciences, University of Padova, via Trieste 75, 35121 Padova, Italy and tDepartment of Neurology, University of Padova, via Giustiniani 5, 35128 Padova, Italy (Received 4 December 1994; revised 13 April 1995; accepted 3 May 1995)

Abstract--In this study we investigated the sarcoplasmic reticulum (SR), alongside myofibrillar phenotype, in muscle samples from five Myotonic Dystrophy (DM) patients and five control individuals. DM muscles exhibited as a common feature, a decrease in the slow isoform of myosin heavy chain (MHC) and of troponin C in myofibrils. We observed a match between myofibrillar changes and changes in SR membrane markers specific to fiber type, i.e. the fast (SERCA1) Ca2+-ATPase isoform increased concomitantly with a decrease of protein phospholamban (PLB), which in native SR membranes colocalizes with the slow (SERCA2a) SR Ca2+-ATPase, and regulates its activity depending on phosphorylation by protein kinases. Our results outline a cellular process selectively affecting slow-twitch fibers, and non-degenerative in nature, since neither the total number of Ca2+-pumps or of ryanodine receptor/Ca 2÷release channels, or their ratio to the dihydropyridine receptor/voltage sensor in junctional transverse tubules, were found to be significantly changed in DM muscle. The only documented, apparently specific molecular changes associated with this process in the SR of DM muscle, are the defective expression of the slow/cardiac isoform of Ca2+-binding protein calsequestrin, together with an increased phosphorylation activity of membrane-bound 60 kDa Ca2+-calmodulin (CAM) dependent protein kinase. Enhanced phosphorylation of PLB by membrane-bound Ca2+-CaM protein kinase also appeared to be most pronounced in biopsy from a patient with a very high CTG expansion, as was the overall 'slow-to-fast' transformation of the same muscle biopsy. Animal studies showed that endogenous Ca2+-CaM protein kinase exerts a dual activatory role on SERCA2a SR Ca2*-ATPase, i.e. either by direct phosphorylation of the Ca2+-ATPase protein, or mediated by phosphorylation of PLB. Our results seem to be consistent with a maturational-related abnormality and/or with altered modulatory mechanisms of SR Ca2+-transport in DM slow-twitch muscle fibers. Key words: Skeletal muscle, myotonic dystrophy, sarcoplasmic reticulum, Ca2+-pump, phospholamban, calmodulin-dependent protein kinase.

INTRODUCTION

protein kinase (DM-PK) or myotonin kinase gene [1]. Myotonic dystrophy (DM) or Steinert's disease It has been demonstrated [1] that tissueis an autosomal dominant multisystem dis- specific isoforms of the DM-PK gene exist, that order, connected to expansion of a CTG trinu- appear to be generated through alternative cleotide repeat at chromosome region 19q13.3 splicing of this gene. Furthermore, evidence has [1]. The clinical features involve cardiac and accumulated recently that, in human skeletal smooth muscle and other tissues and organs, in muscle, like in adult muscles of other addition to skeletal muscle [2, 3]. The muta- mammalian species, there is a selective or tional expansion of the unstable [CTG]n repeat predominant expression of the muscle-type of is in the 3'-terminal exon of a gene, i.e. DM- DM-PK gene product in type I, slow-twitch fibers [4]. :~Author to whom correspondence should be addressed The primary effects of dynamic mutations in at: Departimento di Scienze Biomediche Sperimentali, the DM-PK gene are still controversial, since Universita di Padova, via Trieste 75, 35121 Padova, Italy. 33

34

E. Damiani et al.

both down-regulations and up-regulations were found in studies of DM-PK mRNA and protein levels in DM tissues [1]. However, a decreased level in DM muscle of the 53 kDa muscle-type protein product of DM-PK gene is further supported by a recent study [5]. Early histopathological studies outlined the selective atrophy of type I fibers as a distinctive, early change [6]. According to a recent study [7], the smallness of type I fibers in DM would also have to be attributed to a delay in postnatal maturation of the same type of fibers. The hypothesis of maturation-related membrane abnormalities in DM muscle has been advanced recently, in connection with evidence suggesting a specific deficiency in sarcoplasmic reticulum (SR) Ca2+-pumps and Na+-K+-ATPase protein [8], as well as to explain the persistence in DM muscle of apamin-receptors/Ca2+-activated K +channels [9]. There is, however, quite controversial evidence from earlier biochemical studies, concerning changes in SR Ca2+-uptake activity in DM muscle, since both decreases and increases were described [3], depending on the stage of the pathological process [10]. Fasttwitch and slow-twitch fibers are quite heterogeneous in SR Ca2+-transport properties [11], and in a number of matched properties. Thus analysis of whole muscle homogenates, or of isolated membranes from DM muscle, unless complemented by analysis of isomyosins [12, 13], cannot easily distinguish between primary changes and those dependent on changes in DM fiber population. In the present study, we investigated muscle specimens from adult patients with DM and from normal individuals. As a new approach we carried out side-by-side analysis in each muscle specimen of: (i) myosin heavy chain (MHC) and troponin C isoforms; (ii) specific markers of junctional transverse tubules (TT) (dihydropyridine receptor, DHPR), and of the SR junctional membrane domain (ryanodine receptor (RyR), calsequestrin (CS) and histidine-rich Ca2+-binding protein (HRC); (iii) marker proteins of the SR free domain (Ca 2+pump, and regulatory protein phosp'holamban, (PLB)). These skeletal membrane markers differ between muscle fiber types, either quantitatively or qualitatively [14-16]. Human H R C gene was mapped to chromosome 19q13.3, in the vicinity of the DM locus [17].

CASE REPORTS

All patients were admitted to the Department of Neurology of the University of Padova.

Case No. 1 ( B.L. ) (Biopsy 1111) The father had myotonia in cold weather, frontal baldness and steppage during gait. The patient (female, age of onset 25 yr) with a large pedigree of myotonic dystrophy, was characterized by association with a chromosomal translocation [18]. She did not complain of any symptom, and the neurological examination was normal, except that the face showed a narrowing of the lower segment, giving the appearance of 'hatched face'. ECG, ophthalmological examination, CPK levels and EMG were normal.

Case No. 2 (S.L.) (Biopsy 1913) The father had bilateral cataracts and cognitive impairment at age of 65 yr. This female patient at age 23 yr (age of onset) complained of hand myotonia, and occasional difficulty in speech. At neurological examination, she presented myotonia in the hands, mild atrophy and weakness of distal muscles. She had a nasal voice. The CPK values, cognitive and ocular examinations were normal. She presented oligomenorreha, but all the hormonal levels were in the normal range. The ECG disclosed the presence of left anterior bundle branch block. The EMG disclosed myotonia and myopathic signs.

Case No. 3 (KS.) (Biopsy 2649) The father (V.G., case No. 4) was affected by myotonic dystrophy. At age 13 yr (age of onset), this female patient complained of hand myotonia. She also presented irregular menses. At neurological examination, she presented marked grip myotonia, slight distal atrophy and mild distal weakness. ECG, CPK value, cognitive and ocular examination were normal. The EMG disclosed myotonic signs.

Case No. 4 (V.G.) (Biopsy 2651) The father had bilateral cataracts. This male patient at age 38 yr (age of onset) started to notice myotonia in hands and occasional difficulty in speech. At neurological examination, he showed marked myotonia in the hands,

SR in Myotonic Dystrophy

bilateral ptosis, and frontal baldness. He had a nasal voice. The cognitive function was normal. The andrological evaluation revealed oligospermia. The hormonal levels were normal. The ocular examination by slit lamp disclosed bilateral cataracts. The ECG disclosed the presence of atrio-ventricular block. The CPK values were slightly elevated (229 IU 1-1; v.n. 0-190). The EMG disclosed myotonic, myopathic and neurogenic signs. Case No. 5 (S.P.) (Biopsy 2629)

The father was affected by myotonic dystrophy. This male patient, at age 20 yr (age of onset), started to complain of myotonia in the hands. At neurological examination, he presented myotonia in the hands, mild hypotrophy and weakness of distal muscles. He had a nasal voice and frontal baldness. The endocrinological, cognitive and ocular examination were normal. The CPK values were elevated (494 IU l-l). The ECG disclosed a conduction defect. The EMG disclosed myotonic, myopathic and neurogenic signs. DNA analysis

DNA analysis was carried out on venous blood samples from each of four patients, as reported in Results. The number of CTG trinucleotide repeats was established for each patient by length variation on Southern blotting and polymerase chain reaction, using the procedures described in detail by Melacini et aL [19]. MATERIALS AND METHODS

DM muscle specimens used in biochemical studies were obtained from diagnostic biopsies. DM samples were either from the biceps brachii (biopsy 1111; biopsy 1913), or from the vastus lateralis (biopsy 2649; biopsy 2651; biopsy 2629). Normal adult samples were taken from thigh muscles (controls 1,2,3,4) or from only the vastus lateralis (control 5), in the course of orthopaedic surgery. All muscle samples were frozen with liquid nitrogen and stored at -80°C, until used. Representative fast-twitch and slow-twitch muscles of the rabbit were the same as used in previous studies [14-16]. All biochemical analyses on muscle samples carried out at the Center for Muscle Biology and Physiopathology of

35

the Department of Biomedical Sciences of the University of Padova. Preparative procedures

Individual muscle biopsies were homogenized in low-ionic strength medium containing 10 mM Hepes, pH 7.4, 20 mM KCI, 100 I.tm PMSF, 1 ~tg ml-l leupeptin, and were fractionated by differential centrifugation, as described by Damiani et al. [20] and Margreth et al. [16], to yield a purified myofibrillar and a total membrane fraction. In addition, in the case of rabbit and rat muscles, we used SR subfractions, obtained by isopycnic sucrose density centrifugation according to Saito et al. [21], with slight modifications [14]. The human SR subfractions used in the present study were the stored frozen subfractions from a previous study [22]. Protein concentration was determined according to Lowry et al. [23], using bovine serum albumin as standard. Biochemical assays

Ca2+-Mg2+-dependent ATPase activity was measured at 37°C, in the presence of Ca 2+ionophore A23187, using an enzyme-coupled spectrophotometric assay [24]. For investigating the Ca2+-dependence of Ca2+-ATPase activity, the free Ca2+-concentrations varied from 0.026 ~tM to 0.34 mM (0.0026-0.63 mM total CaCI2), by using a Ca2+-EGTA buffer. Free Ca2+-concentrations were calculated as described by Fabiato [25]. The binding of [3H]ryanodine (New England Nuclear, 77-87 Ci mmo1-1) and [3H]PN200-110 (New England Nuclear, 72 Ci mmol-l) to membranes was carried out as described by Margreth et al. [16]. Membrane protein phosphorylation by exogenous cAMP-dependent protein kinase was carried out, as described by Damiani and Margreth [26], at 30°C for 5 min, in the presence of 50 ~tM [y-32p]ATP (New England Nuclear, 3000 Ci mmol-l), 2.5 ~tM cAMP and of type II protein kinase A (50 Ixg ml-l) (Sigma Chemical Co, St Louis, MO, U.S.A., 1-2 units per ~tg of protein). Synthetic PLB, prepared as described in Vorherr et al. [27], was a generous gift of Prof. E. Carafoli. Endogenous Ca2+-calmodulin (CaM)-dependent protein kinase was assayed by incubating membranes for 1 min at 0-4°C in 0.1 ml of 20 mM Tris-Mops, pH 7.0, 5 mM MgCI 2, 80 mM

36

E. Damianiet al.

KCI, 0.5 mM EGTA, 0.49 mM CaCI 2 without or with 6 ~tM CaM (from hog brain, Boeringher, Mannheim). The final free Ca 2÷concentration was 10 I.tM. The concentration of [y-32p]ATP was 500 Ixm. The reaction was stopped by adding SDS-solubilization buffer.

Gel electrophoresis, western blotting and ligand overlay experiments One-dimensional SDS-polyacrylamide gel electrophoresis (SDS-PAGE) of muscle membranes was carried out according to Laemmli [28]. Two-dimensional electrophoresis was carried out as described by Damiani and Margreth [26], in the presence of 0.1 mM EGTA in the first dimension, and of 1 mM CaCI 2 in the second dimension. For separating the fast and slow isoform of MHC, myofibrils were subjected to electrophoresis in 6% polyacrylamide SDS-gels [29]. Gels were stained either with Coomassie blue and then Stains All, or by a silver nitrate method. Protein electrophoretic transfer onto nitrocellulose was also as described [15]. Blots, after staining with Ponceau red, were used for different types of ligand overlay:

Immunoblotting. Blots were incubated, as described by Damiani and Margreth [15], with: (i) chicken antibody to rabbit, fast-twitch Ca 2+ATPase [30], 5 ktg/ml; (ii) mouse antibody to dog cardiac CS [15], (5 Ixg ml-l), (iii) mouse monoclonal antibody (Chemicon International Inc., MAB427, 1:3000 diluted) to DHPR alsubunit and (4) mouse monoclonal antibody to bovine cardiac troponin C [31]. Immunoenzymic staining of blots was carried out, using either nitroblue tetrazoliurn/5-bromo-4-chloro-3indolylphosphate (toluidine salt) or diaminobenzidine, in the case of alkaline phosphatase and of horseradish peroxidase conjugates, respectively.

carried out, using a Bio Rad Model GS-670 densitometer, equipped with Molecular AnalystTWPC image analysis software. RESULTS

Type I fiber changes in D M muscle All muscle biopsies demonstrated type I fiber atrophy, of mild to medium (biopsy 2649), to severe degree (biopsy 2629). Histochemical type I fibers were 38% (biopsy 1913), and 50% (biopsy 1111), respectively in biceps brachii biopsies; and 44% (biopsy 2651), 55% (biopsy 2649) and 60% (biopsy 2629) in biopsies from vastus lateralis. Histochemical fiber-typing could not be carried out on stored-frozen normal samples used for biochemical analysis. Also for this reason, but mainly because it was considered that analysis of the isoprotein composition of myofibrils could provide a more reliable index of the fractional volume of type I (slow-twitch) and type II (fast-twitch) fibers, we relied mainly on this alternative approach for comparing control with DM samples, as well as between DM samples.

Isoprotein composition of myofibrils with respect to myosin and troponin C

The iso-MHC composition of DM and normal muscle specimens was analysed by SDS-PAGE of isolated myofibrils, under conditions which are known to be optimal for separating the fast (MHC-II) from the slow (MHC-I) isoform (see Methods). The results shown in Fig. 1A, demonstrate an excess of MHC-II to be a common feature of DM specimens, either from biceps brachii (Fig. 1A, biopsy 1913) or from vastus lateralis muscle (Fig. 1A, biopsies 2649, 2651, 2629), and an apparently larger excess of the same myosin isoform, in the case of DM sample 2629, in [125I]LDL overlay. Binding of human 125I- which atrophy of type I fibers was more labelled LDL (Biomedical Technologies Inc., pronounced. Stoughton MA, U.S.A.) 0.122 #Ci /.tg-I on The slow-cardiac isoform of thin filament blots was performed under conditions identical troponin C was detected on western blots of to those described previously [26], after SDS- myofibrils (Fig. 1B), by using a monoclonal gel electrophoresis in the absence of fl-mercap- antibody specific to this protein isoform [31], as toethanol. demonstrated by its ability to recognize the Autoradiography of SDS gels after protein corresponding troponin C isoform in isolated phosphorylation (12 h exposure at-80°C) and myofibrils from rabbit slow-twitch muscle (Fig. of blots after overlay with ~25I-labelled LDL 1B). On qualitative evidence, myofibrils from (90 min exposure at -80°C), were done by using two normal specimens, including the vastus Beta-max films (Amersham). Densitometry was lateralis (Fig. 1B, controls 5 and 4), appeared

37

SR in Myotonic Dystrophy

Correlations between CTG expansion and myofibrillar changes

A

¢

¢¢,

,

All patients had a significant increase in the number of CTG triplets. Patients with the highest CTG repeat lengths were patient No. 2 (biopsy 1913), and patient No. 5 (biopsy 2629), i.e. 833 and 935, respectively. CTG repeat lengths for patient No. 3 (biopsy 2649) and patient No. 4 (biopsy 2651) were 564 and 265 respectively.

II I I

II

I

DM

N

B

Dihydropyridine receptor and ryanodine receptor I

i

I

DM

The DHPR content of skeletal muscle membranes, correlating with their content in junctional TT [32], was estimated using immunoblot techniques and specific antibody to 170 kDa DHPR a-1 subunit (Fig.2), and/or [3H]PN200-110 binding measurements (Table 1). Both kinds of results indicated considerable similarities in DHPR membrane density, between DM and control samples. High-affinity binding of [3H]ryanodine to the RyR present in skeletal muscle membranes, and reflecting their content in junctional SR, also gave similar Bma x values (Table 1). For estimating the overall DHPR/RyR ratio in DM and normal muscle samples, we measured high affinity binding of tritiated PN200-110 and ryanodine to myofibrils, in addition to the isolated membrane fraction [16]. The estimated total amounts (Bmax value per g muscle) for vastus lat. of two DM patients (biopsies 2629 and 2561) were 50 and 54 pmol for the DHPR and 28 and 37 pmol for the

I

~/

N

Fig. 1. Characterization of the isoprotein composition of human myofibrils of SDS-PAGE and immunoblot. Myofibrils were isolated and purified from homogenates of control and D M muscles, as described in Methods. Panel A: 6% SDS-PAGE [29], after staining with silver nitrate. Only the region of the gels according to MHC bands is shown. Band I: slow M H C isoform; Band II: fast M H C isoform. About 1 gg of protein was loaded per lane. Panel B: Immunoblot with monoclonal antibody to cardiac/slow troponin C, after electrophoresis in 12.5% SDS-PAGE. Only the region of the gels corresponding to troponin C bands is shown. About 100 gg of protein was loaded per lane.

to be richer in slow-cardiac troponin C, compared to DM muscle myofibrils of three individuals DM muscles investigated (Fig. I B, biopsies 2649, 2651 and 2629).

Table 1. Binding characteristics of tritiated PN200-110 and ryanodine and kinetic parameters of Ca2+-dependent ATPase activity of isolated membranes from human muscle Membrane preparation

Yield (mg g.m)

[3H]PN200-110 binding Bm~x Kd (pmol mg-lpr.) (nM)

Normal 1 2 3 4 5

10.6 9.9 8.7 9.8 8.6

2.06 2.39 3.63 ---

DM 2629 2649 2651

13.5 10.5 11.3

2.79 . 2.40

[3H]Rya, binding Bm~x Kd (pmol mf-lpr.) (nM)

0.60 0.57 0.55 --0.73 .

. 0.46

Ca2+-ATPase activity Vm~ KS0 Hill (~tmol min-~ mglpr) (~tm) coefficient

1.15 1.70 0.83 1.02 0.65

6.0 10.9 7.9 15.4 11.5

0.25 0.35 0.49 0.83 0.33

0.25 0.15 0.22 0.17 0.20

1.5 1.0 1.5 1.7 1.7

4.61

13.2

1.10

6.3

0.85 0.39 0,64

0.20 0.15 0.26

1.6 1,5 1.5

.

Skeletal muscle membranes were isolated as reported in Methods. Specific binding of [~H]PN200-110 and [~H]ryanodine was calculated from the difference between total binding and that measured in the presence of 1000-fold excess of cold PN200-110 and ryanodine. Bm~x and Kd values were obtained by Scatchard plot analysis of binding curves. Ca2+-ATPase activity was measured at 37°C, in the presence of Ca2+-ionophore A23187, as reported in Materials and Methods. Kinetic parameters were calculated from the experimental curves shown in Fig. 5, by Hill equation [VeXn/(Xn + Kn)], where V is the maximal value of ATPase activity, X is the free [Ca2+], n is the Hill coefficient and K the free [Ca2.] required for half-maximal stimulation.

38

E. Damiani et aL

kDa

1

2

3

4

5

6

200 116 97 77 55 42

I

II

N

I

DM

Fig. 2. Immunodetection of D H P R a-1 subunit in skeletal muscle membranes. Electrophoresis was carried out in a 5-10% polyacrylamide linear gradient gel. Blots were incubated with a mouse monoclonal antibody to rabbit a-l subunit of the DHPR. Lane l : TT-enriched light membrane fraction, isolated from rabbit fast muscle (Rl fraction, [21]), 10 lag; lanes 2-6: human total membrane fraction (200 ~tg); lane 2: control 4; lane 3: control 5: lane 4: D M 2651; lane 5: D M 2649; lane 6 2629.

RyR, respectively. The corresponding values, including one reported previously [16] for normal adult muscle were 50-57 pmol and 23-27 pmol g.m, in the stated order.

slow/cardiac isoform of CS, in addition to the 60 kDa fast-skeletal isoform, both in the case of control and of DM muscles (Fig. 3A). To this purpose, we used immunoblot techniques and antibody to dog cardiac CS, already shown to be highly cross-reactive with the homologous isoform of rat [14] and the rabbit [15] slowtwitch muscle, and as further shown in Fig. 3B (lane 6). In immunoblots of human control muscle membranes (Fig. 3B, lanes 4, 5), there was a visibly immunoreactive thin protein band at 52 kDa, i.e. having a mobility corresponding to that of purified slow/cardiac CS from human muscle [14], while only trace amounts of this protein were detected in blots of DM muscles (Fig. 3B, lanes 1-3). We tried to confirm the deficiency of slow/cardiac CS in DM muscle membranes, by using the two-dimensional SDS-gel system already described [26]. This electrophoretic method is best adequate for separating skeletal and cardiac CS isoforms, when present in complex protein mixtures, as confirmed by the results obtained with normal human muscle (Fig. 3C, panel N). Under the same experimental conditions, we obtained negative results for the presence of slow-cardiac CS in three individual DM samples [biopsies 2629, 2649 and 2651). The results obtained with biopsy 2651 are illustrated in Fig. 3C (panel DM).

Calsequestrin CS is the major lumenal CaZ+-binding protein of the SR and is confined mainly to junctional terminal cisternae (TC). It exists in two distinct isoforms, the fast skeletal and the slow-cardiac isoform, which are the products of two different genes. The gene for the skeletal isoform, which is the only one expressed in mammalian fast-twitch fibers [14, 15], was mapped to human chromosome lq21 [33]. The skeletal and the slow-cardiac isoforms are present in a ratio of approximately 3:1 in homogeneously slowtwitch muscle of the rabbit [15]. In previous work, we used SDS-PAGE of isolated TC from adult human muscle, in combination with 45Ca overlay techniques [34] and chromatography on phenyl-Sepharose [15], to identify the slow/cardiac isoform of CS as a 52 kDa blue Stains All-stained Ca 2+- binding protein. However, using crude SR membrane preparations, Stains All-staining of Laemmli's slab gels did not demonstrate enough sensitivity to be able to identify unambiguously the

Histidine-rich Ca2+-binding protein Protein analysis of SR subfractions from rabbit fast-twitch muscle was the first to demonstrate [26] that 170 kDa Ca 2+- and [125I]LDL-binding H R C protein, which had been previously characterized and cloned by Hofmann et al. [35], colocalizes with the RyR/CaZ+-release channel to the same SR subcompartment involved in Ca z+ storage and Ca2+-release. Human H R C protein was cloned and mapped to human chromosome 19q at region 13, 3, and thus in close proximity to the DM-PK gene [17]. Although differing from its rabbit homologous in apparent molecular mass (140 kDa), on SDS-PAGE [34], it was shown to share in several significant biochemical properties [34]. We investigated for the presence of 140 kDa [~25I]LDL-binding protein on ligand blots of total skeletal muscle membranes from DM patients and control individuals, using for reference the HRC-enriched junctional TC

SR in Myotonic Dystrophy

39

B

A

kDa

1

2

3

4

5

6

1

2

3

4

5

6

i i 116 97 77 55 42

C

DM

N

1st

Fig. 3. Identification of calsequestrin in skeletal muscle membranes by SDS-PAGE and immunoblotting. Panel A: Electrophoresis was carried out in a 5-10% SDS-PAGE. Proteins were stained with Coomassie blue and then with Stains All. Lane 1: DM 2651; lane 2: DM 2629; lane 3: DM 2649; lane 4: control 5; lane 5: control 4; lane 6: control 3. About 100 lag of protein was loaded per lane. Asterisk indicates 60 kDa, blue-staining CS (fast skeletal isoform). Arrow indicates the RyR monomer. Panel B: immunoblot with mouse polyclonal antibody to dog cardiac CS. Lane 1: DM 2649, 100 lag; lane 2: DM 2651, 100 lag; lane 3: DM 2629, 100 lag; lane 4: control 4, 100 lag; lane 5: control 5, 100 lag; lane 6: Isolated TC fraction from rabbit slow-twitch muscle, 30 I~g. Arrow: human cardiac CS. Double arrow: human skeletal CS. Arrowhead: rabbit cardiac CS. Double arrowhead: rabbit skeletal CS. Panel C: Two dimensional SDS-gel electrophoresis [26]. Protein was electrophoresed in the presence of EGTA in the first dimension (lSt), and of CaCI2 in the second dimension (2nd). Gels were stained with Stains All. N. control muscle (4); DM; biopsy 2651. Arrow: cardiac CS isoform. Double arrow: fast skeletal CS isoform. About 60 lag of protein was loaded per lane.

fraction [34] from a d u l t h u m a n skeletal muscle (Fig. 4A, lane 5). O n qualitative evidence, the 140 k D a [I25I]-LDL b i n d i n g p r o t e i n a p p e a r e d to be present in similar a m o u n t s in n o r m a l (Fig. 4A, lane 6, a n d Fig. 4B, lane 4) a n d D M (Fig. 4A, lanes 7 a n d 8, a n d Fig. 4B, lanes 5 a n d 6) muscle m e m b r a n e s .

Sarcoplasmic reticulum Ca2+-A TPase A n t i g e n i c [30] a n d c D N A c l o n i n g studies [36] of SR Ca 2+ A T P a s e d e m o n s t r a t e d the existence of two distinct isoforms for r a b b i t fast-twitch and slow-twitch adult muscles (named SERCA 1 and SERCA2a, respectively).

40

E. Damiani et al. A kDa

1

2

3

4

5

6

7

8

B kDa

I

2

3

4

5

6

200 200

t 116 97

116 97

66 66 43 43 31

Ponceau red

t25I-LDL overlay

Ponceau red

tT"5I-LDL overlay

Fig. 4. Identification of the [125I]LDL-binding protein on ligand blots after SDS-PAGE. S D S - P A G E was carried out as in Fig. 3, panel A, except that ~ m e r c a p t o e t h a n o l was omitted. Blots were incubated with [125I]LDL, and [1251]LDL-binding proteins were detected by autoradiography. Panel A: lane 1 and 5: purified terminal cisternae from normal h u m a n muscle, 60 R-g; total muscle membranes; lanes 2 and 6: control 1,200 p.g; 3: lane 7: D M 1913, 200 p.g; lanes 4 and 8: D M 1111, 135 Ixg. Panel B: muscle membranes, lanes 1 and 4) control 1,150 lag; lanes 2 and 5: D M 2651, 150 ~tg; lanes 3 and 6: D M 2629, 150 p.m.

Although being the products of two separate genes, and antigenetically dissimilar [30], the two isoforms have similar enzymatic properties and are activated by Ca 2÷ in a cooperative manner [37]. The data from experiments measuring the Ca 2+ dependency of Mg2÷-dependent ATPase activity of human skeletal muscle membrane are shown in Fig. 5 and Table 1. These data demonstrate extensive similarities in apparent Ca 2+ affinity (K50 = about 0.2 laM), as well as in Hill coefficient values (approximately 2, with only one exception), for the SR Ca 2÷ATPase of human control and DM samples, as well as with rabbit fast-twitch muscle SR Ca 2+ATPase (K50 = 0.4 /.tM) [37]. Vmax values, i.e.

the maximum activity at optimum Ca 2+ concentrations, for DM samples were in the mediumupper control range of values (Table 1). On immunoblotting a purified longitudinal SR subfraction from human normal muscle with specific antibody to rabbit fast-skeletal SR Ca2+-ATPase (Fig. 6, lane 1), a prominent Ca2+-ATPase protein band, corresponding to the human homologous isoform (i.e. homologous to rabbit SERCA1), was readily detected, in agreement with the enrichment in Ca 2+ATPase protein of this SR subfraction [R2, see Ref. 21]. Although less prominent, the fast Ca2+-ATPase isoform was detectably present in crude SR membranes. A consistent finding was that membranes from DM muscles (Fig. 6,

Table 2. Membrane content of [32p]labelled phospholamban pentamer and monomer in normal and DM muscles Sample

Pentamer densitometric area (OD x mm)

CPM

Monomer densitometric area (OD x mm)

CPM

4 5

0.88 1A9

1411 1724

0.69 0.52

1087 980

DM 2629 2649 2651

0.48 0.85 0.43

770 1152 842

0.41 0.50 0.59

653 869 1063

Normal

Except for control 4, all muscles biopsies were from Vastus lateralis. Densitometry was carried out on autoradiographic film (see Fig. 7). After autoradiography, the protein bands corresponding to pentameric and monomeric PLB were cut out the dried gel and counted by liquid scintillation counting.

SR in Myotonic Dystrophy 1,0

CONTROL

09

v

08 0,7

b

£

:.s

0.6 0.5 0.4

<

"8 E

kDa

41

1

2

3

4

5

6

200 116 97 77

03 02

55 42

0.1

O0 10 -9 10 -8 10

7

10-6 10-5 10-4

10-3

~!~iiiiiii?!:i~i'~¸¸ I

%,~"

08 0.7

o

0.6

:.S o ~ ®

0.5

~

0.4

0.1 O. 0

I

N

Fig. 6. Identification of the fast-twitch (SERCA1) Ca 2÷ATPase isoform in human skeletal muscle membranes by immunoblotting with specific polyclonal antibody. Panel A: Electrophoresis was carried out in 5-10% SDS-PAGE. Immunoblots were incubated with chicken polyclonal antibodies to rabbit fast Ca2+-ATPase. Lane 1: purified longitudinal (R2) SR fraction (see Methods) from thigh muscles of normal individual; lanes 2-6: total membrane fractions: lane 2: D M 2651; lane 3: D M 2629; lane 4: control 5; lane 5: control 1; lane 6: control 2. About 25 lag of protein was loaded per lane.

0.9

"~

DM

I

DM

1.0

>,

I

v

f

I

I

10 9 1 0 - 8

____J

I

I

10 7 10 6 1 0 - 5

J

[]

10-4

I

10 S

f r e e Ca 2+, [M]

Fig. 5. Ca2+-dependence of Ca2+-ATPase activity of human skeletal muscle membranes. ATPase activity was assayed as reported in Methods, and at the free Ca 2÷ concentrations indicated on the abscissa (see Methods). Each point represents the mean value of duplicate determinations carried out on individual preparations. Ks0 was calculated by Hill plot analysis (see Table 1). Upper panel: Symbols: o: control 1; n: control 2; 0: control 3; A: control 4; V: control 5. Lower panel: o: D M 2629; V: D M 2651; D: D M 2649.

lanes 2, 3) appeared to be enriched in immunostained protein, when compared with normal samples (1 and 2, see Fig. 6, lanes 5, 6). The higher was the membrane content in SERCA 1 isoform, as evidenced by immunoblotting, the higher appeared to be the specific Ca2+-ATPase activity in the same membrane preparation (see Table 1).

Phospholamban In slow-twitch fibers, the SERCA2a Ca >ATPase isoform and regulatory protein PLB,

have been co-localized to the same membrane area in longitudinal SR [38]. In contrast, there is evidence from analysis of the PLB specific m R N A [39], that PLB is never expressed in SR Ca2+-pump membrane of fasttwitch fibers. We tested for the presence of PLB in human skeletal control and DM samples, by phosphorylating membranes in comparison with purified synthetic PLB [26], with [?'-32p]ATP for 5 rain at 30°C, in the presence of added cAMP and of exogenous protein kinase A, followed by SDS-PAGE and autoradiography (Fig. 7). Identification of radiolabelled PLB pentamer rested on its dissociation into 5 kDa protein monomers on heating the phosphorylated samples [27, 40] before electrophoretic analysis (Fig. 7). The relative amount of radioactivity incorporated into PLB pentamers and monomers (Fig. 7, lanes 1, 3, 5, 7, 9), was estimated both by densitometry of the corresponding areas on autoradiographs and by scintillation counting of the excised gel segments. The results obtained by the two kinds of measurements appeared to be in good agreement (Table 2). On comparing the radioactivity incorporated into the PLB monomer containing one protein kinase A

42

E. Damiani et al. kDa

1

2

3

4

5

6

7

8

9

I0

1

2

3

4

5

6

7

8

9

10

I!

12

97

3!

o

14

9hi Boiling

+

+

+

+

+

P o n c e a u red

+

+

+

+

+

+

Autoradiography

Fig.7. Pattern of membrane protein phosphorylation by exogenous cyclic AMP-dependent protein kinase. Membranes were phosphorylated as described in Methods, in the presence of cAMP and of type II cyclic AMP-dependent protein kinase, electrophoresed by 10-20% SDS-PAGE, and transferred onto nitrocellulose. Phosphorylated proteins were detected by autoradiography. About 50 lag of membrane protein and 1 lag of synthetic PLB was loaded per lane. Where indicated, samples were incubated for 3 min in boiling water before SDS-PAGE. Lanes 1 and 2: control 4; lanes 3 and 4: control 5; lanes 5 and 6: D M 2629; lanes 7 and 8; DM 2649; lanes 9 and 10: D M 2651; lanes 11 and 12: synthetic PLB.

phosphorylation site [40], differences between DM and normal samples appeared to be small (Table 2). However, as assessed from the overall amount of radioactivity incorporated into the protein monomer and pentamer, the membrane content of PLB was about 55% of the average control value in the case of biopsy 2629 (see also Section 1). Noticeably, type I fiber atrophy and predominance of the fast isoform of MHC were most marked and, conversely, Ca2+-ATPase activity was found to be highest in the same biopsy (see data in Table 1). By comparison, the percentage content of PLB was calculated to be 78% and 73% of controls for biopsies 2649 and 2651, respectively.

prominent for DM samples, compared to normal samples. On qualitative evidence, the 60 kDa phosphoprotein appeared to be most abundant in biopsy 2629, from patient No. 3 (Fig. 8, lane 5). PLB, in addition to being a major substrate of the CaM-kinase, appeared to be more effectively phosphorylated in membranes derived from Vastus lateralis biopsies of three DM patients (lanes 5-10), compared to control samples (lanes 1-4), and thus regardless of differences in PLB phosphorylation by exogenous protein kinase A (compare with data in Table 2).

Phosphorylation of phospholamban by 60 kDa, endogenous Ca2+-CaM-dependent protein protein kinase

This paper describes the characterization in DM muscle biopsies of specific protein components of SR membranes, which are known to differ between mammalian fast-twitch and slow-twitch fibers, either quantitatively or qualitatively [14-16]. Within this general context, our main findings can be so recapitulated.

PLB is phosphorylated, not only by cAMPdependent protein kinase, but also by a Ca 2+CaM-dependent protein kinase [40]. The Ca2+-CaM protein kinase family, comprises an SR membrane-bound isoform, also referred to as 60 kDa protein kinase, which on binding of CaM becomes self-phosphorylated [41]. We examined the phosphorylation of human skeletal muscle membranes at pCa 5.0, in the absence or the presence of 6 ~tM CaM. CaMdependent phosphorylation, as evidenced by autoradiography, was already maximal after 1 min incubation at 0-4°C (Fig. 8). The 60-kDa, self-phosphorylating component responsible for endogenous CaM-protein kinase activity was detected in skeletal muscle membranes as a radioactive band of corresponding mobility which was consistently more

DISCUSSION

Junctional SR membrane domain Skeletal RyR (RyR1), as investigated at the protein level, is expressed as a single isoform in mammalian (rabbit and rat) fast-twitch and slow-twitch fibres, although its density differs several fold, depending on fiber type, in correlation with differences in the pattern of CS isoforms [14]. There is, apparently, a lack of correlation in what we found with DM muscle, in which the total number of high-affinity ryanodine binding sites, corresponding to the density of functional RyR/Ca2+-release channels does not significantly change, while the slow/cardiac protein isoform of CS is defec-

SR in Myotonic Dystrophy C o o m a s s i e blue kDa

I

2

3

4

5

6

43 Autoradiography

7

8

9

10

I

2

3

4

5

6

7

8

9

10

2OO

116 55

43 31 21 14

+

Fig. 8.

+

+

+

+

CaM

Pattern of membrane protein phosphorylation by endogenous Ca2+-CaM-dependent protein kinase. Membranes

were phosphorylatedin the presenceof Ca2÷and in the presenceor absence of CaM (see Methods). Phosphorylatedproteins were detected by autoradiography. About 100 p.g of protein was loaded per lane. Lanes 1 and 2: control 4; lanes 3 and 4: control 5: lanes 5 and 6: DM 2629; lanes 7 and 8: DM 2649; lanes 9 and 10: DM 2651. PLB: Phospholamban; 60 kDa PK: self-phosphorylatedCa2+-CaM-dependent protein kinase.

tively or not at all expressed, compared to normal muscle samples. Apart from CS, the simplest explanation may be that in the human species, in contrast to other mammalian species, there are little differences in SR volume [42] and overall SR surface area [43] between fast-twitch and slow-twitch fibers, so that differences in fiber Ca2+-channel density are also predictably small. There are also less differences in [3H]ryanodine binding sites between isolated TC from human and rabbit fast-twitch muscle [22], than between fast and slow muscle TC [14]. While then the membrane density of [3H]ryanodine binding sites does not provide sufficient clues to the existence of changes in the fiber population of the muscle, the absence of any marked change in the DM muscle samples examined is equally remarkable. That in itself suggests that membrane degenerative changes did not take place. The same conclusion is suggested by the finding that the overall muscle D H P R / R y R ratio did not change. We were also unable to detect changes in the membrane content of [125I]LDL binding protein HRC, a junctional SR specific protein [26, 34]. In addition to the absence of the slow/cardiac isoform of CS, the only other main difference noted for DM muscle was the increased membrane content of 60 kDa Ca2+-CaM dependent protein kinase. The membranebound endogenous protein kinase, although selectively associated with junctional SR of rabbit fast-twitch muscle [41], in the case of cardiac and slow-twitch muscle was shown to localize to the C a > - p u m p free domain of the SR [44]. The distribution between junctional

and free SR of 60 kDa CaZ+-CaM dependent protein kinase, in the case of human skeletal muscle remains to be clarified (see below). S R Ca2+-A TPase

Two distinct protein isoforms of Ca 2+ATPase, the major component of longitudinal SR and of the extra-junctional portion of S R T C [21, 45], are normally segregated in fast-twitch and slow-twitch fibers of adult mammals, i.e. the SERCA1 and SERCA2a isoform, respectively [36]. In this study we investigated for the presence of SR Ca2+-ATPase SERCA1 (fast) isoform in DM muscle membranes, using immunoblot techniques and specific polyclonal antibody to the corresponding rabbit isoform [30, 46]. Our results indicate that the SERCA1 SR Ca 2+ATPase protein isoform was more abundantly present in D M samples, when compared with normal samples. Thus, such changes, although not as pronounced, appear to be in line with the concomitant isoform transition of CS in D M muscle from a mixed to a pure fast pattern. Using antibodies raised against a mixture of SERCA1 and SERCA2a SR Ca2+-ATPase isoforms and ELISA techniques, together with phosphoenyzme measurements, Benders et al. [9] recently claimed a marked decrease of SR Caa+-ATPase protein in D M muscle. On the contrary, characterization of SR Ca2+-ATPase kinetic parameters in the present study, did not demonstrate any significant difference in either Caa+-dependency or in Vrnax values, between DM and normal muscle specimens. There is,

44

E. Damiani et al.

however, an ill-defined relationship in human mixed muscles, regardless of whether normal or diseased, between maximum Ca2+-ATPase activity (even when measured under optimal conditions, i.e. in the presence of Ca 2+ionophore A23187), and the total number of Ca2÷-pumps. Unlike in fast-twitch fibers in which the SERCA1 Ca2+-ATPase isoform is always in a high active state, there is a wide discrepancy between the density of Ca2+-pump units, as determined in the intact SR of slow-twitch fibers using morphometric methods [47], and that deduced from ATPase activity measurements in the isolated SR [14]. The discrepancy seems to arise mainly from the fact that the SERCA2a SR Ca2+-ATPase is in a partially inhibited state in slow-twitch fibers, due to its interaction with endogenous regulatory protein PLB, when the latter is in a non-phosphorylated state [40, 48]. PLB, although potentially able to interact with either the SERCA1 or SERCA2a isoforms of SR Ca2+-ATPase [48], is uniquely expressed in the SR of slow-twitch fibers, in which it co-localizes with the corresponding Ca2+-ATPase isoform [38]. Phosphorylation of PLB removes the inhibitory interaction with SR CA2+-ATPase, as confirmed by studies with human chemicallyskinned fibers [42]. On the assumption, which seems to be valid, that PLB is present in a fixed amount to SERCA2a SR Ca2÷-ATPase in their native membrane environment, the amount of membrane-bound PLB, in DM muscle should mirror-image changes of the SERCA1 Ca z+ATPase isoform, if these reflected the atrophy of type I fibers.

PLB The membrane content of PLB in DM muscle samples, as estimated by phosphorylation experiments with exogenous cAMP-dependent protein kinase A, was found to range from about 70-80% to 55% of control values. Significantly, PLB content was lowest in biopsy 2629 (patient No. 5), in which the SERCA1 isoform of SR CaZ+-ATPase, based on immunoblot data, appeared to be more highly expressed.

PLB phosphorylation by membrane-bound Ca2÷-CaM dependent protein kinase From knowledge that cAMP-dependent protein kinase and Ca2+-CaM protein kinase

act coordinately in/3-adrenergic stimulation of heart muscle, via phosphorylation of PLB [40], we have investigated Ca2+-CaM-dependent phosphorylation of human skeletal muscle membranes. Our results demonstrate the presence of an endogenous 60 kDa protein kinase, that was autophosphorylated in the presence of Ca 2÷ and CaM, and which phosphorylated PLB, in addition to other membrane components, whose identification, however, must be postponed to future work with purified SR membrane fractions. The significant finding is that, based on both the enhanced selfphosphorylation of 60 kDa protein kinase, and on the extent of 32p-labelling from []t_32p] ATP of PLB, DM muscle membranes appear to be invariably enriched in the endogenous Ca2÷-CaM PK, the change again being greatest in the case of biopsy 2629 (patient No. 5).

Relationship between SR membrane changes, myofibrillar changes and CTG trinucleotide length Direct comparative SR and myofibrillar data were available for three muscle biopsies (2649, 2651 and 2629), and these were from patients No. 3, 4, 5, respectively. SR changes, such as the decrease of membrane-bound PLB in DM muscle, which could be best quantified, appeared to match closely the decrease of the slow isoform of MHC and of troponin C. A relationship between the overall slow-to-fast transformation of the muscle and CTG trinucleotide repeat length, also seemed to be supported on comparing patient No. 5, having the highest CTG number (935), with the other two patients, who had intermediate CTG repeat lengths (564 and 265). In proximal muscles, object of the present study (biceps brachii and vastus lateralis), which are also the less affected in DM, the smallness of type I fibers, either due to delayed maturation [7] or to true atrophy [6, 12], was not very marked, except in muscle biopsy 2629, from patient No. 5. Thus, our findings which first provide unequivocal biochemical evidence that slow-twitch muscle fibers are selectively affected in DM, at the same time lend themselves to speculation that this may underlie the existence of a specific trophic influence from the product of the DM-PK gene.

45

SR in Myotonic Dystrophy

Specific molecular changes of slow-twitch muscle fibers in relation to the primary genetic defect in DM Our results outline a cellular process not affecting the TT and SR membrane components of E-C coupling at junctional triads, and not basically affecting the total number of Ca 2+ -channels and of Ca2+-pumps. Thus changes affecting the slow fiber population of DM muscle do not appear to be degenerative in nature, they rather seem to fit the view of a maturation-related abnormality, specific to these fibers. A 53 kDa protein product of DMPK gene, also characterized by tyrosine-kinase activity [49], was immunolocalized in adult human and mouse skeletal muscle at the neuromuscular, as well as the myotendinous junction, and shown to be selectively distributed in the same type of fibers [4]. There is likewise evidence from a recent study in the rabbit and the rat, that membranes derived from slowtwitch muscle are correspondingly enriched in the DM-PK gene product [50]. A main difficulty in testing the hypothesis of a fiber-type specific differentiating influence from the DM-PK gene product during postnatal ontogenesis, is that developmental changes in the expression of genes coding for specific isoforms of muscle proteins, concern mainly muscle fibers destined to become fasttwitch fibers in the adult stage. Slow/cardiac CS, which is the only CS isoform expressed by fetal muscle [39], is down-regulated in developing skeletal muscle, at critical periods after birth [51], but only in fast-twitch muscle is its synthesis turned off completely with the attainment of the adult stage. Significantly, the slowcardiac isoform of CS is co-expressed with major fast CS isoform in junctional TC of mature slow-twitch muscle fibers [15, 52], after muscle denervation [14], as well as after thyroid hormone treatment [53]. This seems to imply the presence of a permanent, predominantly myogenic influence over the continued expression of slow-cardiac CS in slow-twitch fibers, while, for instance, PLB, which is not expressed normally in rabbit fast-twitch muscle [11], does however become expressed after long-term eloctrical stimulation at low frequency [54]. Within this conceptual framework, the seemingly blocked expression of the slow-cardiac isoform of CS in DM muscle, is of considerable interest, as it could be also its possible functional effects on SR Ca2÷-release [55].

Concerning modulatory mechanisms of Ca :+transport and Ca2+-release, whose alterations might affect more profoundly Ca2+-transients, our finding of an enhanced Ca2÷-CaM PKdependent phosphorylation of PLB in DM muscle, is most intriguing. Membrane-associated Ca2+-CaM protein PK of cardiac and slow-twitch muscle SR exerts a dual activation role on the Ca2÷-pump, i.e. either mediated by phosphorylation of PLB [40], or by isoformspecific (SERCA2a) phosphorylation of the Ca2+-ATPase [44]. In addition, inactivation of the SR2+Ca2+-release channel, through the action of endogenous Ca2÷-CaM protein kinase, has been shown in fast-twitch muscle of both the frog [56] and the rabbit [57], and it seems to involve phosphorylation of junctional SR specific proteins, such as triadin and histidine-rich, LDL and Ca2÷-binding protein H R C [58]. Clearly, up-regulation in DM muscle slowtwitch fibers of SR membrane-bound Ca 2÷CaM PK, because of its ability to phosphorylate regulatory protein PLB, as well as to phosphorylate and activate the Ca 2÷pump, would conceivably lead to a faster rate of decline in the falling phase of Ca2+-transient initiated during depolarization-induced Ca 2÷release from SR Ca2÷-channels [59]. Under these conditions, the Ca2÷-pump would overpower Ca2÷-release. Might that relate to characteristic muscle weakness in DM, and could that, in turn, secondarily lead to the atrophy of slow-twitch fibers? The answer to this and other fundamental questions, concerning the proximate effects to DM-PK gene mutations in skeletal muscle, must be postponed to future studies. Acknowledgements--M. Pelosi, R. Sacchetto and E. Bortoloso participated in this work in partial fulfillment of the requirements for the PhD degree. The technical assistance of Mr G. Tobaldin, in some phases of this work, is gratefully acknowledged. This work was supported by grant funds to A. Margreth from M.U.R.S.T. (40% Progetto Nazionale Biologia e Patologia del muscolo striato), and from Telethon (project No. 512).

REFERENCES 1. Wieringa B. Commentary: Myotonic dystrophy reviewed: back to the future? Human Molec Gen 1994; 3: 1-7. 2. Harper P S, Rudel R. Myotonic Dystrophy. In: Engel A G, Banker B Q, eds. Myology. New York: McGraw-Hill Book Company, 1994; 1192-1219. 3. Engel A G, Schotland D L. Myotonic disorders. In:

46

E. Damiani et al. Goldensohn E S, Appel S H, eds. Scientific Approaches to Clinical Neurology. Philadelphia: Lea &

4.

5.

6.

7.

8.

9.

10.

11. 12. 13. 14.

15. 16.

17.

18. 19.

20.

21.

Febiger, 1977: 1656-1666. van der Ven P, Jansen G, van Kuppevelt T, et al. Myotonic dystrophy kinase is a component of neuromuscular junctions. Human Molec Gen 1993; 2: 1889-1894. Koga R, Nakao Y, Kurano Y, et al. Decreased myotonin-protein kinase in the skeletal and cardiac muscles in myotonic dystrophy. Biochem Biophys Res Comm 1994; 202: 577-585. Brooke M H, Engel W K. The histographic analysis of human muscle biopsies with regard to fiber types. 3. Myotonias, myastenia gravis and hypokalemic periodic paralysis. Neurology 1969; 19: 469-472. Tohgi H, Kawamorita A, Utsugisawa K, Yamagata M, Sano M. Muscle histopathology in myotonic dystrophy in relation to age and muscular weakness. Muscle Nerve 1994; 17: 1037-1043. Benders A G, Timmermans A H, Oosterhop A, et al. Deficiency of Na÷/K÷-ATPase and sarcoplasmic reticulum Ca2+-ATPase in skeletal muscle and cultured muscle cells of myotonic dystrophy. Biochem J 1993; 293: 269-274. Renaud J F, Desnuelle C, Schmid-Antonmarchi H, et al. Expression of apamine receptor in muscles of patients with myotonic muscular dystrophy. Nature 1986; 319: 678-680. Mussini I, Di Mauro S, Angelini C. Early ultrastrucrural and biochemical changes in muscle in Dystrophia Myotonica. J Neurol Sci 1970; 10: 585-604. Pette D, Staron R. Cellular and molecular diversities of mammalian skeletal muscle fibers. Rev Physiol Biochem Pharmacol 1990; 116: 1-76. Volpe P, Damiani E, Margreth A. Immunological properties of myosin in myotonic dystrophy. Acta Neurol Belg 1982; 82: 145-149. Salviati G, Betto R, Danieli-Betto D, et al. Myosin light chains and muscle pathology. Neurology 1986; 36: 693-607. Damiani E, Margreth A. Characterization study of the ryanodine receptor and of calsequestrin isoforms of mammalian skeletal muscles in relation to fiber types. J Musc Res Cell Motil 1994; 15: 86-101. Damiani E, Volpe P, Margreth A. Coexpression of two isoforms of calsequestrin in rabbitslow-twitch muscle. J Musc Res Cell Motil 1990; 11: 522-530. Margreth A, Damiani E, Tobaldin G. Ratio of dihydropyridine to ryanodine receptors in mammalian and frog twitch muscles in relation to the mechanical hypothesis of excitation-contraction coupling. Biochem Biophys Res Comm 1993; 197:1303-1311. Hofmann S, Topham M, Hsieh C L, Francke U. cDNA and genomic cloning of HRC, a human sarcoplasmic reticulum protein, and localization of the gene to human chromosome 19 and mouse chromosome 7. Genomics 1991; 9: 656-669. Neri G, Serra A, Tedeschi B, et al. Myotonic dystrophy and chromosome translocation segregating in the same family. J Neurogenetics 1987; 4: 47-56. Melacini P, Villanova C, Menegazzo E, et al. Correlation between cardiac involvements and CTG trinucleotide repeat length in Myotonic Dystrophy. J Am Coil Cardiol 1995; 25: 239-254. Damiani E, Tobaldin G, Volpe P, Margreth A. Quantitation of ryanodine receptor of rabbit skeletal muscle, heart and brain. Biochem Biophys Res Comm 1991; 175: 858--865. Saito A, Seiler S, Chu A, Fleischer S. Preparation and morphology of sarcoplasmic reticulum cisternae from rabbit skeletal muscle. J Cell Biol 1984; 99: 875-885.

22.

23. 24.

25.

26.

27.

28. 29. 30. 31. 32.

33.

34.

35.

36.

37.

38.

39.

Damiani E, Barillari A, Tobaldin G, Pierobon S, Margreth A. Biochemical characteristics of free and junctional sarcoplasmic reticulum and of transverse tubules in human skeletal muscle. Muscle Nerve 1989; 12: 323-331. Lowry O H, Rosebrough N J, Farr A L, Randall R J. Protein measurements with the Folin phenol reagent. J Biol Chem 1951; 193: 265-275. Salviati G, Volpe P, Betto R, et aL Biochemical heterogeneity of skeletal muscle microsomal membranes. Membrane origin, membrane specificity, and fiber types. Biochem J 1982; 202: 289-301. Fabiato A. Computer programs for calculating total from specified free and free from specified total ionic concentrations in aqueous solution containing multiple metals and ligands. Methods Enzymol 1988; 157: 378-401. Damiani E, Margreth A. Subcellular fractionation to junctional sarcoplasmic reticulum and biochemical characterization of 170 kDa Ca 2÷- and low-densitylipoprotein-binding protein in rabbit skeletal muscle. Biochem J 1991; 277: 825-832. Vorherr T, Wroszek A, Chiesi M, Carafoli E. Total synthesis and functional properties of the membraneintrinsic protein phospholamban. Protein Science 1993; 2: 339-347. Laemmli U K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970; 227: 680~85. Biral D, Betto R, Danieli-Betto D, Salviati G. Myosin heavy chain composition of single fibres from normal human muscle. Biochem J 1988; 259: 307-308. Damiani E, Betto R, Salvatori S, et al. Polymorphism of sarcoplasmic reticulum adenosine triphosphatase. Biochem J 1981; 197: 245-248. Saggin L, Gorza L, Ausoni S, Schiaffino S. Troponin C switching in the developing heart. J Biol Chem 1989; 2,64: 16299-16302. Salvatori S, Damiani E, Barhanin J, et al. Co-localization of the dihydropyridine receptor and the cyclic AMP-binding subunit of the intrinsic protein kinase to the junctional membrane of the transverse tubules of skeletal muscle. Biochem J 1990; 267: 679-687. Otsu K, Fuji J, Periasamy M, et al. Chromosome mapping of five human cardiac and skeletal muscle sarcoplasmic reticulum protein genes. Genomics 1993; 17: 507-509. Picello E, Damiani E, Margreth A. Low-affinity Ca 2÷binding sites versus Zn2÷-binding sites in histidine-rich Ca2÷-binding protein of skeletal muscle sarcoplasmic reticulum. Biochem Biophys Res Comm 1992; 186: 659-667. Hofmann S L, Goldstein J L, Orth K, et al. Molecular cloning of a histidine-rich Ca2÷-binding protein of sarcoplasmic reticulum that contains highly conserved repeated elements. J Biol Chem 1989; 264: i 8083-18090. Brandl C, Green N, Korczack B, MacLennan D H. The two Ca2÷-ATPase genes: homologies and mechanistic implications of the deduced amino acid sequence. Cell 1986; 44: 597-607. Lytton J, Westlin M, Burk S, Shull G, MacLennan D H. Functional comparison between isoforms of the sarcoplasmic reticulum or endoplasmic reticulum family of calcium pumps. J Biol Chem 1992; 267: 14483-14489. Jorgensen A O, Jones L. Localization of phospholamban in slow but not fast canine skeletal muscle fibers. An immunocytochemical and biochemical study. J Biol Chem 1986; 261: C3775-3781. Arai M, Otsu K, MacLennan D, Periasamy M. Regulation of sarcoplasmic reticulum gene expression

SR in Myotonic Dystrophy

40.

41.

42.

43.

44.

45.

46.

47. 48. 49.

during cardiac and skeletal muscle development. Am J Physiol 1992; 262: C614-620. Wegener A, Simmerman H, Lindemann J, Jones L. Phospholamban phosphorylation in intact ventricles. Phosphorylation of serine 16 and threonine 17 in response to fl-adrenergic stimulation. J Biol Chem 1989; 264:11468-11474. Leddy J, Murphy B, Qu Y, Doucet J P, Pratt C, Tuana B. A 60 kDa polypeptide of skeletal muscle sarcoplasmic reticulum is a calmodulin-dependent protein kinase that associates with and phosphorylates several membrane proteins. Biochem J 1993; 295: 849-856. Salviati G, Sorenson M, Eastwood A. Calcium accumulation by the sarcoplasmic reticulum in two populations of chemically skinned human muscle fibers. Effects of calcium and cyclic AMP. J Gen Physiol 1982; 79: 603~532. Hayashi K, Miller R, Brownell K. Three dimensional architecture of sarcoplasmic reticulum and T-system in human skeletal muscle. Anat Record 1987; 218: 275-283. Hawkins C, Xu A, Narayanan N. Sarcoplasmic reticulum calcium pump in cardiac and slow twitch skeletal muscle but not fast twitch skeletal muscle undergoes phosphorylation by endogenous and exogenous Ca2+/calmodulin dependent protein kinase. J Biol Chem 1994; 269:31198-31206. Jorgensen A O, Shen C, MacLennan D, Tokuyasu K. Ultrastructural localization of the Ca 2÷ - Mg2÷-depen dent ATPase of sarcoplasmic reticulum in rat skeletal muscle by immunoferritin labeling of ultrathin frozen sections. J Cell Biol 1982; 92: 409-416. Salviati G, Pierobon-Bormioli S, Betto R, et al. Tubular aggregates: sarcoplasmic reticulum origins, calcium storage ability and functional implications. Muscle Nerve 1985; 8: 299-306. Franzini-Armstrong C, Jorgensen A O. Structure and development of E-C coupling units in skeletal muscle. Ann Rev Physiol 1994; 56: 509-534. James P, Inui M, Tada M, Carafoli E. Nature and site of phospholamban regulation of the Ca2÷-pump of sarcoplasmic reticulum. Nature 1989; 342: 90-92. Mayer P E, Faure R, Bouchard J-P, Thibault M-C,

50.

51.

52.

53.

54. 55. 56. 57.

58.

59.

47

Puymirat J. The myotonin protein kinase phosphorylates tyrosine residues in normal human skeletal muscle. Biochem Biophys Res Comm 1994; 199: 89-92. Salvatori S, Biral D, Furlan S, Marin O. Identification and localization of the myotonic dystrophy gene product in skeletal and cardiac muscles. Biochem Biophys Res Comm 1994; 203: 1365-1370. Sacchetto R, Volpe P, Damiani E, Margreth A. Postnatal development of rabbit fast-twitch skeletal muscle: accumulation, isoform transition and fibre distribution of calsequestrin. J Musc Res Cell Motil 1993; 14: 646~53. Biral D, Volpe P, Damiani E, Margreth A. Coexistence of two calsequestrin isoforms in rabbit slow-twitch skeletal muscle fibers. FEBS Lett 1992; 299: 175-178. Arai M, Otsu K, MacLennan D, Alpert N, Periasamy M. Effect of thyroid hormone on the expression of mRNA encoding sarcoplasmic reticulum proteins. Circ Res 1991; 69: 266-276. Pette D, Vrbov~ G. Adaptation of mammalian skeletal muscle fibers to chronic electrical stimulation. Rev Physiol Biochem Pharmacol 1992; 120:115-202. Kawasaki T, Kasai M. Regulation of calcium channel in sarcoplasmic reticulum by calsequestrin. Biochem Biophys Res Comm 1994; 199:1120-1127. Wang J, Best P. Inactivation of the sarcoplasmic reticulum calcium channel by protein kinase. Nature 1992; 359: 739-741. Hain J, Nath S, Mayrleitner M, Fleischer S, Schindler H. Phosphorylation modulates the function of the calcium release channel of sarcoplasmic reticulum from skeletal muscle. Biophys J 1994; 67: 1823-1833. Damiani E, Picello E, Saggin L, Margreth A. Identification of triadin and of histidine-rich Ca 2÷binding protein as substrates of 60 kDa calmodulindependent protein kinase in junctional terminal cisternae of sarcoplasmic reticulum of rabbit fast muscle. Biochem Biophys Res Comm 1995; 209: 457-465. Garcia J, Schneider M F. Calcium transients and calcium release in rat fast-twitch skeletal muscle. J Physiol 1991; 463: 709-728.