A spin label study of the erythrocyte membranes in duchenne muscular dystrophy

183

CZinica Chimica Acta, 105 (1980) 183-194 @ Elsevier/North-Holland Biomedical Press

CCA 1422

A SPIN LABEL STUDY OF THE ERYTHROCYTE DUCHENNE MUSCULAR DYSTROPHY

MEMBRANES IN

MICHEL LAURENT a, DENIS DAVELOOSE a, FRANCOIS SIEGMUND FISCHER b and GEORGES SCHAPIRA b,*

LETERRIER

a,

a Division de Biophysique du Centre de Recherches du Service de Santk des Armies, 1 Bis, rue du Lieutenant Raoul Batany, F. 92141 Clamart (France) and b Institut de Pathologie Mokculaire, Universitk Rend-Descartes, CHU Cochin Port-Royal, 24, rue du Faubourg Saint-Jacques, F. 75014 Paris (France) (Received

November

2nd, 1979)

Summary Red blood cells and freshly prepared erythrocyte membranes of 15 patients with Duchenne muscular dystrophy (DMD) as well as age-matched controls were studied by the spin label method. No significant modifications appeared for spin-labelled proteins of ghost membranes. With the two fatty acid spin labels, 5-nitroxide stearate and 16-nitroxide stearate, we have confirmed previous results of Sato et al. concerning the thermal behaviour of the erythrocyte membranes, i.e. no change near the polar part probed by 5nitroxide stearate and a linearization of the fluidity versus temperature variation around 12”C, as explored by 16-nitroxide stearate. Furthermore we studied in the whole erythrocyte the amplitude of the 5-nitroxide stearate electron spin resonance signal as a function of the microwave power. This saturation effect was observed in 12 out of 15 controls and only in 1 out of 13 DMD cases studied. In erythrocyte membranes labelled with 16-nitroxide stearate the penetration of the label inside membranes was statistically different between DMD and controls. These new findings furnish further arguments in favour of a structural alteration of the phospholipid organization of erythrocyte membranes in DMD. Associated together, these different sets of tests obtained by spin labelhng permit good statistical discrimination between DMD and normal subjects.

* Correspondence should be addressed to Prof. G. Schapira, Institut de Pathologie MoKculaire, CHU Co&in Port-Royal, 24. rue du Faubourg Saint-Jaques. 75014 Paris, France.

INSERM-

184

Introduction Duchenne muscular dystrophy (DMD) is inherited in males as an X-linked recessive trait and is manifested by dystrophy of proximal muscles. From birth until the age of 3 years there are no apparent clinical symptoms; thereafter difficulties of squatting and frequent falling appear. The patients become progressively unable to walk and death occurs generally before the age of 25 years.. Numerous biochemical abnormalities have been described, related to defects of the muscle membranes [I] which result in an increase of muscular enzymes in the serum [2]. In erythrocyte membranes isolated from DMD patients, several findings suggest indeed a generalized membrane defect [ 361. In this paper, erythrocyte membranes were studied by electron spin resonance (ESR) using the spin label method [ 71. We first re-examined the original results of Butterfield et al. [S] on the proteins of erythrocyte ghosts and those of Sato et al. [9] on the phospholipids of intact erythrocytes with a more significant number of patients. Then we investigated the phospholipid structure of intact erythrocytes and of ghosts with labels which explore either the hydrophobic core of the membrane or the region near their polar part. Each sample was studied as a function of temperature and microwave power, and this comprehensive study allowed us to define a global test for identification of DMD patients with a restricted number of experiments using only a small amount of blood (150 ~1). Material and methods Fifteen patients (Clinique de Corbeil, France) from 4 to 14 years old were studied. The diagnosis of DMD was established by determination of creatine kinase activity, clinical history and physical examination. Fifteen age-matched subjects were used as controls. Heparinized blood samples arrived at the laboratory after approximately 2 h. They were washed three times in 150 mmol/l NaCl, 5 mmol/l sodium phosphate buffer, pH 8. Sato et al. [9] had previously shown that heparin did not affect the ESR spectra. Intact red blood cells were used between 24 and 48 h. Ghosts were prepared from red blood cells by the method of Dodge et al. [lo], modified by Fairbanks et al. [ 111. They were stored at 4°C no more than 5 days in the presence of 0.5 mmol/l NaN,. Protein concentration in ghosts was about 3 to 4 mg/ml. Spin labelled derivatives were obtained from Syva, Palo Alto, U.S.A. The general molecular formula of nitroxide stearic acid is as follows: CH,-(CH,),,>Cq(CH,),,-COOH 0

NO

Spin label I (m = 12, IZ = 3: 5-nitroxide stearic acid) was 2-(3-carboxypropyl) 2-tridecyl-4,4dimethyl-3oxazolidinyloxyl and spin label II (m = 1, IZ = 14: 16nitroxide stearic acid) was 2-( lo-carboxydecyl)-2-ethyl-4,4-dimethyl-3-oxazolidinyloxyl). Intact erythrocytes (7 * 10’ cells/ml) and ghosts (4 mg protein

185

per ml) were labelled by adding to the suspension 2% (v/v) of stock solutions (lo-* mol/l) of labels in ethanol. In some cases, ethanol was evaporated before incorporation under a stream of dry nitrogen gas, leaving a thin film of spin label. These control experiments demonstrated that such a concentration of ethanol in membranes had no effect on the experimental variables measured. In order to eliminate free nitroxide in solution, ghosts labelled with spin label II were incubated at 25°C for 30 min with 2.5 mmol/l ascorbate. After centrifugation (30 000 X g for 30 min), the pellet was washed in PBS and then resuspended in the same buffer. The membrane proteins were labelled with 3-maleimido-2,2,5,5,tetramethyl1-pyrrolidinyloxyl (spin label III) so that the final concentration of label did not exceed 1/50th of the total protein concentration. Ghosts were incubated at 4°C for 12 h and free label was eliminated by washing twice in 100 vol of PBS. Spin labelled membranes were introduced into a Varian quartz flat cell (sample volume 50 ~1) and the spectra were recorded with an E 3 Varian ESR spectrometer equipped with a laboratory-built temperature regulator, with an accuracy of +0.2”C. Results Spin label I (5-nitroxide stearic acid) The ESR spectra of intact erythrocytes and ghosts labelled with spin label I were recorded at 3°C intervals from 0 to 45°C. The maximum coupling constant 2 7’11(Fig. 1) was plotted against temperature. In all cases (control and

2 T II C.........................................

-----r;

Fig. 1. Electron spin resonance spectra of intact erythrocytes labelled with 5-nitroxide stearate at different temperatures. Definition of the maximum apparent coupling constant 2Tll : r,C:of the amplitude and half width of the low field line. The 2Tl1 parameters give information about the rigidity and the structural organization of the phospholipid fatty acid chains.

186

TABLE

I

2 ?‘/j COUPLING STANT

CONSTANTS

AGAINST

INTACT

AT

0%

TEMPERATURE

AND

SLOPES

MEASURED

OF

WITH

THE

PLOTS

5-NITROXIDE

OF

THE

COUPLING

STEARATE

SPIN

CON-

LABEL

ON

ERYTHROCYTES 2 T// at O’C

Slope

(Gauss)

(GaussPC)

_-mean

Iktrema

S.D.

mean

S.D. .._-__ .-

-_.-.-

.Normal DMD

patients

63.025

0.300

62.5-63.75

0.277

0.019

62.850

0.276

62.543.37

0.268

0.010 ,__...

---

extrema 0.255-0.921 0.257-0.288

_ __

DMD, whole erythrocytes and ghosts) approximately linear variations were observed. The 2 ‘7,) values at 0°C and the slopes of the plots are given in Table I. Very low individual variations are observed in each group; however no significant differences are seen between normal and DMD erythrocytes. Probe I was also used in a microwave power saturation study. Since relaxation times in ESR spectroscopy are generally long, especially for free radicals, the saturation of the signal is easy to observe when the microwave power is increased, i.e. the intensity of the signal does not increase linearly with the square root of the microwave power but becomes constant or even decreases after a maximum has been reached. Since the signals of spin labels included in membrane are composite, measurement of the amplitude of one of the spectrum lines as a function of the microwave power is not adequate for the study of the saturation phenomenon. The best way would be to integrate twice the whole spectrum. This is a diffi-

15

10 N x =l x 4 5

5

Fig.

2.

of

the

at

high

Plot

dial

of

of the low

microwave values

erythrocytes.

10

field

power of

The

JP

for

cavity

integrated

was not

power They

measurable.

intensity

(0 -0)

normal

is characteristic

microwave

the E 3 spectrometer.

ple in the

line

of

(A

and normal

intensity are thus

xed

(mW) only

X AH2,

DMD blood

cells.

corresponds

relative

see

Fig.

erythrocytes

values,

1) as a function (’ -m)_

It is generally to the values

since

the true

of the square

The not

observed

displayed power

plateau

root

observed with

DMD

on the attenuator

delivered

to the sam-

187 TABLE

II

SATURATION SPIN Total

number

cases

studied

Normal

13

ON

INTACT

EKYTHROCYTES

LABELED

WITH

5.NITROXIDE

STEARATE

of

With

Without saturation

effect

saturation

Saturation effect

effect

doubtful

subjects 11

3

15 DMD

EFFECTS

LABEL

1

patients 1

9 x2

= 12.25

3

p = 0.22”0

cult task when the ESR spectrometer is not directly coupled to a calculator. Therefore we have evaluated the intensity (I) of the low field line, which is characteristic of the more immobilized labels, by measuring its amplitude (A) and its half height width (AH) and by using the classical relation I = A X (AH)2. If no saturation occurs, this intensity I plotted against the square root of the microwave power must give a straight line. Fig. 2 shows such a plot in the case of normal and DMD erythrocytes. A saturation effect is observed with normal erythrocytes whereas the plot is continously linear for DMD red cells. The results obtained for all the samples studied are summarized in Table II. Similar measurements for isolated erythrocyte membranes did not show such a difference between DMD and normal subjects. Spin label II (16-nitroxide

stearic acid)

On in tact ery throcy te As with spin label I the spectra of intact erythrocytes incubated with spin label II were recorded at 3°C intervals from 0” to 45°C. Such spectra can be interpreted by measuring the amplitude ratio of lines ho and h_, as indicated in Fig. 3. This ratio is related to the fluidity of the local environment of the label. Examples of plots of log ho/h._, against the reciprocal of the temperature are shown in Fig. 4. For 12 out of 15 normal erythrocyte samples a break in the plot was clearly observed between 8.5 and 13”C, and its existence was uncertain for the three other cases. On the other hand, this break was observed only in 5 out of 15 DMD patients, and uncertain in 1 case (Fig. 5). These results are summarized in Table III). On ery throcy te ghosts The spectra obtained with this label on erythrocyte Fig. 5. It can be observed that on the high field line called (a) is superimposed. Thus we have measured on the amplitude of the lines (a/h _1), Table IV summarizes that this ratio is significantly higher in DMD cells than Spin label III Proteins of the ghost

membranes

were labelled

ghosts are shown in h_l another component each sample the ratio of the results which show in normal erythrocytes.

with maleimide

nitroxide

3.3

32

3.4 l/T

Fig.

3. ESR

Definition Fig. of

4.

spectra of

Plot

normal

two

the of

of

of

the

the

(see and

DMD

erythsocytes of

iog(hgjh_I)

(I!------“9

thirds

intact

amplitudes

definition

Duchenne

cases

labelled

medium

and on

Fig.

erythrocytes

studied,

whereas

with

high

IGnitroxidc

field

3)

lines

strara1.e and

as a function

fm----‘).

it was

ho

only

of

The doubtful

at

3.5

36

: 7

x 103 different

temperatures.

h-l. the

break

reciprocal in the

in 3 out

of

temperature

curve

was

15 control

not

in a case observed

in

samples.

209. Fig.

5.

ESR

spectrum

of

erythrocyte

ghosts

labelled

with

16

NS

at

0°C.

Definition

of

the

lines

hkl.

TABLE

III

TEMPERATURE STEARATE Total cases

Normal

number

15

of

studied

LABEL

011’ (Plot

INTACT

of log -_. -

Presence

of

in the plot

EHYTHROCYTES

a break around

LABELED

WITH

l&NITROXIDE

“fxms l/7‘) .._..-.-

(ho/h-l)

Break

not

observable

Break

1 IOG

subjects

1.5 DMD

STUDY SPIN

12

0

3

9

I

patients 5 x2

2 12.88

p = O.lGYo

doubtful

a and

189 TABLE

IV

RATIO

OF

THE

NITROXIDE Number

LINE

AMPLITUDE

STEARATE

ON

of cases

Normal

MEASURED

a/h_1

ERYTHROCYTE

Mean

S.D.

1.39

0.71

2.64

0.89

ON

THE

SPECTRA

OBTAINED

WITH

16-

GHOSTS

subjects

15 DMD

patients

14

t = 0.014

t = 4.179

III. The spectra show the simultaneous presence of strongly (line S) and weakly (line W) immobilized labels (Fig. 6). The W/S ratio versus temperature was plotted for the erythrocytes from eight control and eight DMD subjects. No significant differences were observed between the two series of samples (Fig. 6). Multifactorial

analysis of the results

With two out of the three labels used, three different types of experiments gave individually significantly different results between control and Duchenne

40

3c w I 2(

l(

D TEMPERATURE Fig. of

6.

ESR

lines

(open

W

spectrum and

symbols)

S. and

of

Plots DMD

erythrocyte

ghosts

of

ratio

the

(closed

W/S

symbols)

COCI labelled

with

as a function erythrocyte

maleimide of

temperature

ghosts

studied.

nitroxide in the

(spin

label

different

III). cases

Definition of

normal

Fig. 7. The multifactorial analysis calculates for each sample a characteristic value from the experimental data. These values are distributed along a discriminant axis, the unity of which depends on the coding of the data. With the coding procedure indicated in the text, the characteristic values were found to extend from -1.5 to +0.15 for normal cases and from -0.2 to +1.5 for DMD cases. The figures represent the histogram of the distribution of these values. Above the axis, the control cases, under the axis the DMD cases. The overlap concerns only one sample in each series.

intact erythrocytes and erythrocyte ghosts. Both groups of subjects have been compared by discriminant factorial analysis { 121. The three variables have been coded in the following way: the experimental values of the a/h_, ratios were multiplied by 100; for the plots of (A X AH2) against JP and log (ho/h_,) against l/T the value 200 was attributed when no break was observable, the value 0 when an inflection was present on the curve, and the value 100 in the case of doubtful results. The calculation was performed on an Intertechnique multi 20 computer, with a program written by Dr. Perrault. The histogram of the repartition of the 15 normal and 13 DMD samples along the calculated discriminant axis is represented in Fig. 7. It can be observed that only one DMD sample is found as a false negative and one normal sample as a false positive case. This discrimination was not modified by the choice of other values for the introduction of the expe~men~l data. Discussion By the use of spin label methods, modifications of both the protein and the lipid structure of the erythrocyte membrane have recently been described in several muscular diseases [8,9,13,14]. We consider it important to point out that in progressive muscular dystrophy of the Duchenne type conflicting reports on the structural or functional modifications of the erythrocyte membrane have been published: important erythrocyte deformability [15], which was not confirmed by others [16,17]; modification in the endogenous phospho~lation of erythrocyte membrane proteins, specifically spectrin or band II [f&-20], results which other groups have been unable to confirm [20,21]. For another membrane-bound enzyme,

adenylate cyclase activity increases have been found [22,23] which could not be repeated by us 1211. It is, therefore, important to ascertain whether the functional and structural changes observed in this disease can be repeated unequivocally. For these reasons, we undertook an extensive study of the changes in erythrocyte membrane structure by the use of the spin label method in order to confirm previously published results for DMD and to extend this type of investigation into several new areas. Protein

spin lubelling

Butterfield et al, [ 8,131 were the first to observe a modification of the erythrocyte ghosts spin-labelled on their proteins, They showed an increase in the IV/Z ratio in Duchenne muscular dystrophy and in dystrophic myotony. We were unable to confirm their results on DMD patients. In our study the temperature of the samples was carefully controlled (between 0” and 40°C). Large dispersion in the variation of the W/S ratio as a function of temperature in each series of samples was obtained (Fig. 6); therefore no significant differences could be shown between DMD and normal ghosts. Such a large variation from sample to sample has been previously observed in a study of erythrocyte ghost prepared from 50 different. normal blood donors. This study was undertaken to observe eventual modifications of the erythrocyte membrane protein structure during blood storage. It was also observed, that for a given sample of spinlabelled ghosts, the spectrum was very stable since the same ‘w/Svalues were recorded one week later (Leterrier, F., Daveloose, D., Chobert, A., unpublished results). Therefore, the variation observed in the present study seems to be independent of the methodology, but inherent in the samples studied. Phospholipid spin labelling Like Sato et al. [ 91 and Wilkerson et al. [ 141, we have used stearic nitroxide spin labels, which have previously given considerable information on artificial and natural membranes, particularly under the influence of drugs and in inherited diseases [ 7,22,24,25]. Spin lubeE i (S-nitroxide stearic acid) explores the phospholipids near their polar part. At low microwave power we confirm the results of Sato et al. [9] who did not find any differences between control and DMD erythrocytes. On the contrary, the power saturation study performed with this spin label shows that the peripheral part of the membrane lipid leaflet of the DMD erythrocyte seems to be modified. This is shown by the absence of saturation of the ESR signal in the majority of the DMD erythrocytes studied. Wilkerson et al. [14] have observed differences between DMD and normal erythrocytes labelled with probe I in a saturation transfer EPR (STEPR) experiment. This technique uses a high level of microwave power, and the out-ofphase detection of signal produces spectra which are much more sensitive to low frequency molecular motions of spin labels than classical EPR. These authors did not detect significant differences in the rate of tumbling of spin label I between DMD and erythrocyte ghosts, but they observed a significant decrease in the signal intensities of the principal STEPR lines recorded in DMD erythrocytes. This was explained by accumulation of spin labels in focal mem-

192

brane sites, leading to increasing spin-spin interactions between labels. Our results can be explained also by this hypothesis. As a matter of fact the saturation phenomenon in classical EPR is difficult to observe when the life time of the spin-excited state is short. The two principal ways for these excited states to lose their energy are, first, the interactions of the paramagnetic species with their environment (in the case of membrane spin labelling, the surrounding phospholipids), and, secondly the spin-spin exchange mechanism, which occurs when the paramagnetic species are separated by short distances. Since no differences in the spectral parameters (2T,,), nor in thermal behaviour, are observed in our study of 5-nitroxide stearate labelled membranes, the first mechanism can be excluded, and the difference in the saturation behaviour is more probably due to an exchange mech~ism, which implies a different repartition of the labels inside the DMD membrane, with high local label concentrations. Although this interpretation of the variations in saturation behaviour between DMD and controls is not certain, it is interesting to note that a highly significant difference is observed between both series of samples (Table II). The biological and clinical observations, made on the three cases of DMD where a saturation of the EPR signal was obtained, could not be differentiated from the other patients. 5’pin label II (16 NS). Sato et al. [9] have shown modifications in the thermal behaviour in intact erythrocytes from DMD patients, which resulted in the absence of a break in the plot of log(h,/h_,) against l/Y’. We have confirmed these findings but it is clear that this behaviour is not constant in all DMD erythrocytes, since for one third of our cases thermal transition as in normal subjects was detected. Here also, no particular biological or clinical fact distinguished these subjects from the others. If, on the contrary, erythrocyte ghosts are labelled with 16-nitroxide stearate one observes clear indications that the hydrophobic core of the DMD erythrocyte membrane is modified. The spectra are characterized by a higher value of the ratio (a)/L1 (Table IV). The width and position of the line (a) (Fig. 5) could indicate the presence of free label not incorporated within the membrane. However, successive washings of the preparation did not remove this signal, and furthermore it was recorded in the presence of 5 mmol/l ascorbic acid, which chemi~~ly reduces all free nitroxides in solution in a few seconds. Thus this signal is due to label inside the membranes, but in a very fluid and relatively hydrophilic environment. These areas are not accessible to ascorbic acid at O”C, a temperature at which this compound does not penetrate inside erythrocyte membrane. When the experiments are performed at 37”C, the intensity of the line (a) is slowly reduced, and after 1 h, the difference in the ratio a/h _1 between DMD and normal ghosts is no longer observed. It is possible to postulate as a consequence of the differential accessibility of ascorbate with temperature that some areas are more accessible to the reducing agent than the hydrophobic part of the membrane itself. Although we have no satisfactory explanation for the differences observed by EPR, and certainly further work is therefore needed, the multifactori~ analysis of our results shows an excellent discrimination between normal and DMD red blood cells. Even if each of the spectroscopic tests described in this study is

193

not sufficient alone to differentiate with a high degree of confidence DMD from normal subjects, their association has been shown to separate the two populations with a high degree of accuracy. It can be envisaged that this type of measurement could be used in the diagnosis of the illness in the newborn and even during the intra-uterine stage when there is a family history of DMD. Acknowledgements This work was supported by Contract 77/1106 from “Directions des Recherches et Etudes Techniques”, by a contract from the Muscular Dystrophy Association of the U.S.A., and by contracts from the “Institut National de la Sante et de la Recherche Medicale”, the “Centre National de la Recherche Scientifique”, the “Fondation Nationale de la Recherche Medicale” and the “Comite Francais de Soutien a la Recherche contre la Myopathie”. We wish to thank Dr. Perrault for his aid in the statistical analysis of the results. References 1 Dreyfus, J.C. and Schapira, G. (1962) Biochemistry of hereditary myopathies. Charles C. Thomas, Springfield, IL 2 Schapira, G., Dreyfus. J.C., Schapira. F. and Kruh. J. (1955) Glycogenolytic enzymes in human progressive muscular dystrophy. Amer. J. Phys. Med. 34. 313-319 3 Probstfield. J.L.. Wang, Y. and From. A.H.L. (1972) Cation transport in erythrocvtes of Duchenne muscular dystrophy. Proc. Sot. Exp. Biol. Med. 141, 479483 4 Percy. A.K. and Miller, M.E. (1976) Reduced deformability of erythrocyte membranes from patients with Duchenne muscular dystrophy. Nature (Land.) 258, 147-148 5 Pickard, N.A.. Grwmer. H.D.. Verril, H.L., Isaacs, E.R., Robinow, M.. Nance, W.E., Myers. E.C. and Goldsmith, B. (1978) Systematic membrane defect in the proximal muscular dystrophy. N. EngI. J. Med. 299, 841-846 6 Wakayama, Y., Hodson, A., Pleasure. D.. Bonilla, E. and Shortland, D.L. (1978) Alteration in erythrocyte membrane structure in Duchenne muscular dystrophy. Ann. Neural. 4, 253-256 7 Berliner. L.J. (1976) Spin labeling, theory and application. Academic Press, New York 8 Butterfield. D.A., Chesnut, D.B., Appel, S.H. and Roses, A.D. (1976) Spin label study of erythrocyte membrane fluidity in myotonic and Duchenne muscular dystrophy and congenital myotonia. Nature (Land.) 263.159-161 9 Sate, B.. Nishikida, K., Samuels, L.T. and Tyler, F.H. (1978) Electron spin resonance studies of erythrocytes from patients with Duchenne muscular dystrophy. J. CIin. Invest. 61, 251-259 10 Dodge, J.T., Mitchell, C. and Hanahan, D.J. (1963) The preparation and chemical characteristics of hemoglobin free ghosts of human erythrocytes. Arch. Biochem. Biophys. 100, 119-129 11 Fairbanks, G., Steck. T.L. and WaIIach, D.F.H. (1971) Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochemistry 10, 2602-2617 12 Lebard. L.. Morineau, A. et Tabard, N. (1977) Technique de la Description Statistique. Dunod. Paris 13 Butterfield, D.A., Roses, A.D., Appel. S.H. and Chesnut, D.B. (1976) Electron spin resonance studies of membrane proteins in erythrocytes in myotonic muscular dystrophy. Arch. Biochem. Biophys. 177. 226-234 14 Wilkerson, L.S., Perkins, R.E., Roeloes. R.. Swift, L., Dalton, L.R. and Park, J.H. (1978) Erythrocyte membrane abnormalities in Duchenne muscular dystrophy monitored by saturation transfer electron pammagnetic resonance spectroscopy. Proc. Natl. Acad. Sci. U.S.A. 75, 838-841 15 Matheson. D.W. and Howland, J.L. (1974) Erythrocyte deformation in human muscular dystrophy. Science 184, 165-166 16 Miale. T.D., Frias. J.F. and Lawson, D.L. (1975) Erythrocytes in human muscular dystrophy. Science 187.453 17 Garceau. C., Leblond, P.F., Lyonnais, J. and Beaupre, A. (1978) Morphologic et dhformabilite des Crythrocytes dans la dystrophie musculaire. Nouv. Rev. Franc. HBmatol. 20, 585-598 18 Roses, A.D., Herbstreith. M.H. and Appel. S.H. (1975) Membrane protein kinase alteration in Duchcnne muscular dystrophy. Nature (Land.) 254. 350-351

194 19 Roses, A.D. and Appel, S.H. (1976) Erythrocyte speetrin peak II phasphory~ation in Duchenne muscular dystrophy. J. Neural. Sci. 29, 185-193 20 Iyer. S.L., Hoenig, P.A., Sherblom, A.P. and Howland, J.L. (1977) Membrane function affected by genetic muscular dystrophy. Erythrocyte ghost protein kinase. Biochem. Med. 18. 384-391 21 Fischer. S., Tort&w, M.. Piau, J.P.. Delaunay, J. and Schapira, G. (1978) Protein kinase and adenyiate cyclase of erythrocyte membrane from patients with Duchenne muscular dystrophy. Clin. Chim. Acta 88.437440 of red cell 22 Falk. R.S., Campion, D.. Guthrie. D.. Sparkes, R.S. and Fox, C.F. (1979) Phosphorylation membrane proteins in Duchenne muscular dystrophy. N. En@. J. Med. 300, 258-259 23 Mawatari, S.. Schonberg, M. and Olarte, M. (1976) Biochemical abnormalities of erythrocyte membranes in Duchenne muscular dystrophy. Arch. Neurol. 33.489-493 24 Leterrier. F. (1979) L’utilisation des marqueurs de spin et de fluorescence pour 1’Btude de I’effet des m6dicaments sur 11%membranes. Mises Point Biochim. Pharmacol. 2, 50-93 25 Daveloose, D., Viret, J., Mofle, D. and Leterrier. F. (1980) Mise en &vidence, par marquage de spin d’une modification structurale de la membrane Crythracytaire du rat g~n~t~quen~ent hypertendu. C.R. Acad. Sci., in press