- Email: [email protected]
Studies on model membrane. I. Effects of Ca2+ and antibiotics on permeability of cardiolipin liquid-crystalline vesicles
IO
BIOCHIMICA ET BIOPHYSICA ACTA
J~F,A 75376 STUDIES ON MODEL MEMBRANES I. EFI:ECTS OF (;a ~1 AND ANTIBIOTICS ON P E R M E A B I L I T Y OF CARI)I OLI PIN LI !~)UI1)-('RYSTALLINE VESICLES J. S:\HA, D. 1 ) A I ' A H A D J O P ( ) V L ( ) S axa) C. ]'i. \ \ : E N N E I ¢
Ne~, York State Department of Health, leoswell l)arl¢ 5[emodal Institute, Bl~jfalo, A'. Y. r42o 3 (c:.&:~ .i (Received August 4th, I969)
SUMMARY
Pure cardiolipin forms closed liquid-crystalline vesicles when suspended in aqueous salt solutions. Electronmicrographs of unsonicated suspensions indicate multilamellar vesicles and after sonication unilamellar vesicles are demonstrated by OsO~ fixation techniques. Both types of vesicles are relatively impermeable to sucrose. The self-diffusion rate of 22Na- efflux in sonicated vesicles in o.o37 /mlole Na= per ffinole cardiolipin per h and for ~"K- is about 0.o47 ,umole K~ per #mole cardiolipin per h. This rate of cation efflux by cardiolipin is typical of other vesicles prepared from acidic phospholipids. The efflux rate particularly of the sonicated particles is markedly increased in the presence of air, presumably because of the high percentage of unsaturated fatty acids which are susceptible to oxidation. Once prepared under N2, the permeability of the particles does not increase with time. Ca 2~ at low concentrations (z mM or less) is not effective in increasing permeability of cardiolipin vesicles in contrast to other acidic phospholipids. Valinomycin increases 42K rate 8-fold at o. 4 nil. Gramieidin increases a"K ~ rate 5-fold at 2o nM and also accelerates 22Na~ rate I6-fold at 2o nM.
INTRODUCe[ION
Only recently has it been possible to isolate inner and outer membranes of mitochondria in a relatively pure form 1. Analysis of the two layers indicates that cardiolipin is mainly present in inner membrane 2. It is now well established that inner membrane is the site of oxidative pbosphor3qationa. Furthermore, a role has been suggested for phospholipids in oxidative phosphorylation and electron transport in terms of maintaining electron transport activity< a. Since mitochondrial functions are intimately associated with ion movements and the recent "chemiosmotic hypothesis" of MITCHELL6 emphasizes the role of the membrane in separating H ~ and other cations, it is essential to have a better understanding of the permeability properties of inner membrane constituents. ]~'LEISCHFR et aI. 5 have shown that mitochondria can maintain their structural integrity after removal of lipids. This, of course, does not rule out the possibility that lipids might be necessary to maintain Biochim. Biophys. Acta, i96 (I970) IO--I 9
PERMEABILITY OF CARDIOLIPIN VESICLES
II
a tight membrane. In fact, in analogy with model membrane studies 7, it would appear that lipids of mitochondria might be partly responsible for separating ions and maintaining a potential difference. The discovery by BANCHA,~Iet al. s that phospholipids (amphipathic molecules) can form closed vesicles when suspended in aqueous salt solutions, has led to the study of physico-chemical properties of m a n y phospholipids except cardioliping-lL Permeability properties of cardiolipin liquid crystalline vesicles to 22Na+, aSK+ and uniformly l~C-labeled sucrose and the effects of gramicidin, valinomycin and Ca 2+ were studied in order to find the role played by cardiolipin as an inner men> brane constituent. MATERIALS
All reagents were analytical grade. The water was glass distilled. Cardiolipin was obtained from Supelco, Bellefonte, Pa., in sealed ampules under N 2 and kept at - I O ° when not in use. Microanalysis of ions present in cardiolipin was carried out by Referred Chemistry, North Hollywood, Calif., and Na + and K + contents were verified in our own laboratory. The amount of ions present in cardiolipin (expressed as /mmles per #mole of cardiolipin) are: Na + -: 2.3, K + = 0.05, Ca 2+ ~ o.I. 22Na+ was supplied by Amersham and Searle, DesPlanes, Ill., as 22NaC1 in aqueous solution (o.I mC/ml). 42K+ was obtained from Nuclear Science Division of ICN as 4eKC1 in aqueous solution (I 7 me/ml). Uniformly '4C-labeled sucrose was obtained from Nuclear Chicago in crystalline form (o. 5 mC per ampule). Gramicidin (a mixture of A and B) was obtained from Nutritional Biochemical Corp., and valinomycin was received from California Biochemical Corp. The dialysis bags (o.125 inch diameter) were a product of Union Carbide. Sephadex G-5o (coarse) was purchased from Pharmacia Fine Chemicals, Picataway, N.J. N 2 used to store cardiolipin after opening the ampule and for experiments is 99.996 % pure N~ of Liquid Carbonic Division of General Dynamics. Tris was Trizma base of Sigma. METHODS
The procedure used for the preparation of the cardiolipin vesicles and subsequent estimation of rates of diffusion was a minor modification of previous methods s,n. In most experiments, 4/~moles of cardiolipin stored in ethanol were dried under vacuum at 30-35 °. During evaporation, the tube was rotated by hand to make a thin layer at the b o t t o m of the tube. Evaporation was continued for i o - I 5 rain. Once the evaporation was complete, N 2 was allowed to flow over the dried cardiolipin layer. While N 2 was flushing over the thin shell of cardiolipin, an aqueous salt solution with radioactive solute was added to the test tube and mixed until all sides of the tube were clear. In cases where sonication was used, the tube after shaking on Vortex mixer was placed in a bath-type sonicator (Heat Systems Ultrasonic), so that the cardiolipin suspension was just below the water surface. N2, after passing through a gas-washing bottle containing distilled water, was flushed continuously during mixing and also during sonication. Sonication was carried out' for a total of 15 rain, in three 5-min intervals, each followed by a 5-min rest period to prevent elevation of temperature. The temperature during this procedure was elevated only Biochim. Biophys. Acta, I96 (I97 o) t o - ~ 9
12
j. SAHAet al.
slightly above room temperature (22 25'~). The tube with unsonicated or sonicated vesicles was closed under N 2 and left at room temperature 1-2 h for equilibration. Removal of excess radioactive material outside the vesicles was accomplished by passing through Sephadex G-5o (coarse)as described by PAPAHADJOPOULOSAND WATKINS~1. Dialysis was done at room temperature (22 25 °) in stoppered tubes with 2o inl dialysate for each bag containing o.5 ml of suspension (approx. o. 4 ~mole of cardiolipin). The salt solution contained I35 mM NaC1 or KC1 and IO mM TrisHC1 buffer adjusted to pH 7.4- For studies to deternline the activation energy of 22Na+ et'flux, bags were first dialysed at room temperature for 3o nlin, then the bags were transferred to different tubes containing 2o ml salt solution preadjusted to desired temperature. Bags were then transferred to other tubes containing 2o ml medium equilibrated at the sanle temperature for additional time intervals (iates over a period of 2-3 h were similar). Gramicidin and valinomyein in ethanol solution were added to cardiolipin suspension before pipetting into dialysis bags in io-#l quantities and controls had equivalent amount of ethanol. Ca 2~ was added to the dialysis medium. The rate of diffusion was expressed as percent of radioactivity captured per h. in this paper the term "vesicles" is used for both unsonicated and sonicated cardiolipin particles. 22Na+ and 42t( ~ were counted in Packard y-scintillation counter and rl4C]sucrose was counted in an ambient type liquid-scintillation counter (Nuclear Chicago). Cardiolipin was estimated from its phosphate content (i mole of cardiolipin per 2 moles of phosphate). Organic phosphate was estimated according to the method of LOHMANNAND JENDRASSEK TM. Electronmicrographs were prepared in the following way. Unsonicated cardiolipin vesicles were centrifuged at IOOooo x g for 9° rain. The sonicated vesicles were centrifuged at the same speed for 16 h. The pellets were then treated with OsO4, dehydrated and the photographs of lead citrate-stained thin sections were taken. RESULTS AND DISCUSSION
Physical characteristics of cardiolipin liquid-ct2vstalline vesicles Initially the electronmicrographs of cardiolipin vesicles were examined to correlate their physical characteristics to permeability properties. Fig. I (the electronmicrograph of unsonicated vesicles) shows heterogeneous nlulti-concentric, lamellar structures which vary greatly in size and shape. These vesicles are similar to other phospholipid particles as described by BANGHAM AND HORNE 13, and PAPAHAI)JOPOULOS ANt) MILLERTM. After sonieation (Fig. 2), the particles become smaller and most of them appear to be composed of one lamella of either spheroidal or crescentlike shape. Tile appearance of two lamellae is probably due to a cut through the infolded crescent-like particles. The transverse dimension of a single lamella in both sonicated and unsonicated vesicles is in agreement with bimolecular leaflet arrangement as suggested by earlier work reviewed recently ~,~5. Nevertheless, it is not possible to establish from this study involving purified cardiolipin how cardiolipin or other phospholipids are arranged in mitochondrial membrane with large amount of proteins.
Diffusio~z of ~14C]sucrose Normally sucrose is impermeable to most cell membranes and it is usually used to measure intercellular space. KI.INGENBERGAND PFAFFTM have shown that sucrose
Biochim. Biophys. Acta, 196 (197o) lO-19
PERMEABILITY OF CARDIOLIPIN VESICLES
13
Fig. I. Electronmicrograph of cardiolipin liquid-crystalline vesicles (unsonicated) after OsO4 fixation, acetone dehydration and lead citrate staining of Epon-embeded thin sections. Magnification J 1400 X. is not permeable to lnitochondrial m a t r i x space, b u t is permeable to outer m e m b r a n e . W h e n 14Qsucrose was t r a p p e d inside vesicles at room t e m p e r a t u r e a n d dialysed in the cold room (5°), no detectable a m o u n t of [l~C?sucrose appeared in the m~dium. This indicates t h a t cardiolipin vesicles are stable a n d p a r t i c u l a r l y impermeable to sucrose at 5 °. At room t e m p e r a t u r e (23 25°), the rate of efflux of sucrose is low. The average rate is found to be o.45 %/h (o.o8-o.75). W i t h sonicated vesicles the rate was found to be o.67 %/h (o.oo-I.4O). So even though the surface area increased with sonication, the rate did not increase. This is in agreement with earlier work with other phospholipids a n d the possible e x p l a n a t i o n s were discussed previously n. The lUCnsucrose associated with the vesicles seems to be c a p t u r e d inside the vesicles r a t h e r t h a n being adsorbed, since changing the n o n - r a d i o a c t i v e sucrose c o n c e n t r a t i o n 7° tilnes did not decrease the [laC~sucrose capture (about o.5 % per/*mole cardiolipin in sonicated vesicles). F u r t h e r m o r e , the c a p t u r e of a b o u t i ° o [14C~sucrose b y the u n s o n i c a t e d vesicles is in agreement with 2 % Na + capture p e r / , m o l e of cardiolipin. The high capture of Na + can be a t t r i b u t e d in part to the presence of Na + as a c o u n t e r ion in cardiolipin. G r a m i c i d i n up to a c o n c e n t r a t i o n of o.5 #M had no effect on sucrose permeability. As shown in Fig. 3, Ca2~ at low concentrations (up to i mM) has no significant effect on sucrose permeability. Only at higher c o n c e n t r a t i o n s (2 5 mM) does (;a a~ have an effect which results in an increase of the rate of sucrose efflux. This is in c o n t r a s t to other acidic phospholipids n, in which Ca e+ increased p e r m e a b i l i t y to
Biochim. Biophys. Acta, 196 (197o) lO-19
14
J. SAHA et a l
Fig. 2. E l e c t r o n m i c r o g r a p h of c a r d i o l i p i n l i q u i d - - c r y s t a l l i n e vesicles (sonicated) a ft e r ()sO~ f i x a t i o n as in Fig. T. M a g n i f i c a t i o n ~3 ioo × .
-i~12
lz
b_l
IlC b.l
13_ 8 LLI
nZ O
4
tl_ 0_ O' 0 ko 0~
I
,.o ;o MM
+io C a 2+
Fig. 3- Eftect of Ca 2+ on seIf-diffusion r a t e of u n i f o r m l y 14C-labeled sucrose t h r o u g h c a r d i o l i p i u l i q u i d c r y s t a l s . ©, u n s o n i c a t c d vesicles; 2x, s o n i c a t e d vesicles. All e x p e r i m e n t s were p e r f o r m e d a t room t e m p e r a t u r e , in z35 mM N a C l - i o mM Tris-HC1 buffer (pH 7-4) w i t h [a4C]sucrose as tracer.
42K~ at low concentrations (o.2 • mM). I t should be i n e n t i o n e d in this connection t h a t a fine precipitation resulting from aggregation of cardiolipin vesicles was visible at 5 mM Ca 2+ concentration. Large aggregates were observed with IO mM Ca 2+. Biockim. F3iopkys. Acta, I96 (T97 o) TO X9
t'ERMEAI3II.iTY OF CARDIOLIPIN VESICLES
15
I)(O'z~sio~ of "2X.~ The self-diffusionrate of 22Nai from unsonicated vesicles was found to be 0.23 °~)/h (3.21 9.96). [n terms of /zmoles, the rate is o.085 ~tmo]e Na- per /zmole cardiolipin per h. In a p r e l i n l i n a r y r e p o r t 1:, we have descril)ed the diffusion r a t e of "2Na ~ as IO I2 %/h. The high r a t e was p r o b a b l y due to the fact t h a t we did not use N2 during the initial phase of our experiments. All the e x p e r i m e n t s (except two e x p e r i m e n t s of t e m p e r a t u r e effect) r e p o r t e d here were carried out under N.,. The r a t e of 22Na t h r o u g h s o n i c a t e d vesicles is 4 . 6 I % / h (3.i 5 6.86), a n d in t e r m s of ,umoles the r a t e is o.o37/,mole N a ~ per/m~ole cardiolipin per h. In order to c o m p a l e the diffusion r a t e s given in this p a p e r to r e p o : t e d fluxes of ions through biological m e m b r a n e s , a t h e o r e t i c a l a r e a occupied b y I ftmole cardiolipin was calculated. Assulning an average a r e a per molecule of I2O A 2 (see ref. i8) a n d an average of ~me l a m e l l a per vesicle, an area of 36oo cm 2 per # m o l e cardiolipin was calculated. Fluxes for diffusion of Na% c a l c u l a t e d on t h e basis of the area m e n t i o n e d is approx. 0.6. Io a.5 equiv/cm.~ per sec. This value is s o m e w h a t lower t h a n the r e p o r t e d flux of K ~ t h r o u g h red cells 19 (too. zo 15 equiv/cnl'-' per see). The low r a t e ill sonicated vesMes indicates t h a t the vesicles offer considerable resistance to cation m o v e m e n t s and are v e r y stable. PAPAHA1)JOPO[TLOS AND \VATKINS 11 in their earlier studies with sonicated vesicles o b t a i n e d similar diffusion rates using the acidic phospholit)ids, p h o s p h a t i d i c acid and p h o s p h a t i d y l inositol. I t should be n o t e d t h a t the earlier studies u were carried out in the presence of air. W e have found t h a t sonication in the presence of air tends to increase the leakage r a t e of cardiolipin vesicles from three to four-fold in the I s t h a n d the rate was o b s e r v e d to increase with time, These findings suggest t h a t cardiolipin is susceptible to a u t o o x i d a t i o n . In t h e sonicated vesicles, t h e a m o u n t of ions c a p t u r e d (o.8o /tnlole N a + per /tmole cardiolipin) is a b o u t half of the u n s o n i c a t e d p r e p a r a t i o n ( I . 3 8 / , m o l e s Na~ per ,umole cardiolipin), b u t the surface a r e a increased m a n y - f o l d . Of the o . 8 / m l o l e N a + c a p t u r e per ~tnaole of cardiolipin in s o n i c a t e d vesicles, a b o u t half could be present as p h y s i c a l l y t r a p p e d NaC1 (due to C1 capture) u. F o r t h e ideal case of u n i l a m e l l a r vesicles, a t h e o r e t i c a l m i n i m u m of I equiv Na ~-p e r / , m o l e cardiolipin should be b o u n d as counter ions to the inside surface of the vesicles with a n y p h y s i c a l l y t r a p p e d NaC1 a d d i n g to the m e a s u r e d N a - content. The observed Na~ content, being significantly h)wer t h a n the t h e o r e t i c a l m i n i m u m for closed l a m e l l a r particles, i n d i c a t e s t h a t some of t h e particles m a y not have an enclosed i n t e r n a l aqueous stmce. In a n y case, sonieation, b v p r o d u c i n g smaller u n i l a m e l l a r particles would be e x p e c t e d to reduce t h e volume of the i n t e r n a l aqueous space as well as t h e n u m b e r of c o u n t e r ions inside the particles for a given a m o u n t of phospholipid. A n o t h e r e x p l a n a t i o n for the o b s e r v e d low c a p t u r e could be b a s e d on a slow r a t e of exchange of tile 2°~Na+ fronl the aqueous m e d i u m with the N a + a l r e a d y present as a counter ion in the cardiolipin samples. The e x t e n t of exchange is not definitely known. The 22Na ~ efltux r a t e is r e l a t i v e l y high in comparison to lecithin vesicles where t ) e r m e a b i l i t y to N a + a n d K ~ is v e r y low s, ~1. W e have carried out two e x p e r i m e n t s at t e m p e r a t u r e range of 5 4 °° for c a l c u l a t i n g t h e energy of a c t i v a t i o n . In one experim e n t c a r r i e d out a t p H 7.4 w i t h 135 mM NaC1 a n d IO mM Tris-HC1 buffer, t h e activ a t i o n energy was found to be i3.25 kcal/mole. A n o t h e r e x p e r i m e n t was clone at p H 5.2 w i t h I3O m M NaC1 a n d 15 mM T r i s - m a l e a t e buffer; t h e a c t i v a t i o n energy was h m n d to be I2.o 5 kcal/mole. B o t h these e x p e r i n l e n t s were done in t h e presence
Biochim. Biophys. Acta, 196 (197o) io 19
I6
j. SAHA dt gll.
of air. While these experiments were carried out in the presence o f air, the low rates of diffusion observed were indicative t h a t extensive oxidation of f a t t y acids chains of cardiolipin did n o t occur. F u r t h e r , the average value of an energy of a c t i v a t i o n was z2 14 kcal/mole which is in agreement with t h a t observed with silnilar phospholipids H. This value indicates t h a t Na ~ do not diffuse freely through aqueous channels; otherwise, the energy of a c t i v a t i o n would be very low .'° and efflux rate would be even higher. Consequently, we m a y conclude t h a t `'`'Na~ is moving across the lamellae as was found with other phospholipids s,'l
I-Z rO n.hi
5040
I.~ 5 0 n7
20
~_
mo
O (/3 ¢-~
hm o ~o
0
I.;
21.0
5 .I0
MM
Ca 2+
~4.0
51.0
Fig. 4. Effect of Ca2+ on self-diffusion rate of 2'aNa~ through cardiolipin liquid crystals. O, unsonicated vesicles; ~, sonicated vesicles. All experiments were performed at room temperature, in ] 3 5 mM NaCl-lo mM Tris-HCl buffer (pH 7-4). Tile effect of Ca 2~ o n tile diffusion rate of 2`'Na+ is shown in Fig. 4- It is clear t h a t up to a c o n c e n t r a t i o n of o. 4 mM there is no effect on 22Na+ efflux. There is some effect at z and 2 mM concentrations, a n d there is an appreciabie effect at 5 mM (;a"~ concentration. This behaviour is similar to the effect of Ca 2~ on phosphatidylinositol a n d t)hosphatidylserine p e r m e a b i l i t y to 4`'K+, b u t different froln phosphatidic acid where 42K' p e r m e a b i l i t y g r e a t l y increases at low Ca `'+ c o n c e n t r a t i o n ( < o . 4 raM) ~1. It has been suggested b y SHAH AND SCHULMAN's t h a t Ca 2+ i n t e r a c t s with two phosphate groups of the same cardiolipin molecule as shown in Fig. 5a. This type of b i n d i n g is more probable at low c o n c e n t r a t i o n s of Ca"~, when the p e r m e a b i l i t y is not affected. It has also been suggested t h a t Ca" ~ can i n t e r a c t with more t h a n one different molecules forming linear polymeric a r r a n g e m e n t 21. This idea is in agreement with the fact t h a t we o b t a i n e d large aggregates at zo mM Ca"concentration. I t has recently been proposed t h a t the increase in p e r m e a b i l i t y of acidic phospholipids in the presence of Ca 2+ results from an a s y m m e t r i c d i s t r i b u t i o n of Ca `'+ on two sides of the bilayer m e m b r a n e 2`'. X - r a y diffraction studies 1°, 2a indicate t h a t Ca 2+ brings out the water with other ions from phost)holit)id suspension, b u t still retains bimolecular leaflet structure. There is no d o u b t Ca `'+ interacts with cardiolipin vesicles at 2 5 mM concentrations, b u t there is no evidence from this s t u d y t h a t ('a`" is effective at low concentration ( < o . 2 raM), at which it s t i m u l a t e s respiration in mitoehondria. The reason for i n a c t i v i t y of Ca 2+ at low concentration in t~iochim, l~iophy.,. ,4 cla, 19¢~( i 0 7o) , o - i ()
17
PERMEABILITY OF CARDIOLIPIN VESICLES
contrast to its effect on phosphatidic acid might be due to the fact that here only one phosphate oxygen is free instead of two and consequently the polymeric arrangement involving coordination bonds between each Ca 2+ and four phospholipid molecules as proposed by PAPAHADJOPOULOS 21 is not possible.
\
/0/P%0 o#P\0\
Ca
,,o.'P% o
Co
tj \ /
Ca
o.'P,o O_#P',_U \
/
Ca
(b)
J'o- o"P' o Ca a+
(a)
~p/O-X UO-X 0~ \0 /o/P~o Co (c)
Fig. 5. a. I n t e r a c t i o n of Ca 2+ and two p h o s p h a t e groups of a single cardiolipin molecule as visualized b y SHAH AND SCHULMAN18. b. Polymeric a r r a n g e m e n t b y Ca 2+ linking different cardiolipin nlolecules, c. Dimer formation by Ca 2+ b e t w e e n two molecules of phosphatidylglycerol, p h o s p h a tidylinositol and p h o s p h a t i d i c acid at low pH. X represents either hydrogen or a glycerol or an inositol moiety.
In regard to the effect of Ca 2+ in changing tile permeability of acidic phospholipids, cardiolipin is intermediate between phospholipids as phosphatidic acid (at high pH) or phosphatidylserine, and other acidic phospholipids as phosphatidylinositol (see ref. I I ) and phosphatidylglycerol or phosphatidic acid at low pH (see ref. 24). The former group is sensitive at Ca 2~ concentrations of o.2-I.O raM. It is, therefore, tempting to speculate that cardiolipin can form polymeric arrangements with Ca ~+ linking different molecules (see Fig. 5b) because of the special molecular configuration of eardiolipin ~5which would not be possible with other acidic phospholipids such as phosphatidylglycerol, phosphatidylinositol and phosphatidic acid at low pH. The latter phospholipids would only be capable of dimer formation (see Fig. 5c). Effcct of antibiotics on diffusion of N a + and K+
The effect of gramicidin on tile self-diffusion rate of Na + efflux is shown in Fig. 6. There is an increase in the rate of Na + efflux as the concentration of gramicidin increases, with the slope of the curve (efflux vs. gramieidin concentration) rising steeply at higher concentrations. With equivalent concentrations of gramicidin, this antibiotic is less effective in increasing the 42K+ efflux in cardiolipin vesicles in comparison to its effect on 22Na+ efflux. Sonicated vesicles required a higher concentration for the same change observed with unsonicated vesicles. Valinomycin is more effective than gramicidin in accelerating the rate of K + efflux at low concentrations (Fig. 6). Furthermore, as with gramicidin, higher concentrations of valinomyein were required to induce leakage in sonieated vesicles as compared to those which were unsonicated. Thisindicates that the action of antibiotics is dependent on the surface area. Biochim. Biophys. Acta, 196 (197 o) lO-19
I8
j. SAHA et al.
At least for v a l i n o m v e i n and n o n a c t i n , it was proposed 26 t h a t the antibiotics in cyclic configuration act as carrier for K +. (;ralnieidins A and B are l i n e a r peptides 27 a n d no definitive evidence has vet been presented t h a t gramicidin can fornl a cyclic conforlnation in s(flution . r in the m e m b r a n e , although some suggestive evidence has
,./
~too
!
o,
/ z
o
50
e,,
L L
//
/[
0
•
~
0
I
0.1
~
I
1.0
ANTIBIOTIC
~
I
i
I0 CONCN.
I
I00
=
I
I000
(riM)
Fig. 6. Effect of antibiotics on self diffusion rate of 2"Na+ or ~'af(+ through cardiolipin liquid crystals. Symbols representing antibiotic-induced offlux of cations are as folIows: O, valinomycin, 42[(+., unsonicated; ~k, valiomycin, 42K~, sonicated; O, gramicidin, 22Na+, unsonicated; &, granlicidin, 22Na-, sonicated; @, gramicidin, 42K+, unsonicated. All experiments were performed at room temperature, in 135 mM NaC1-[o mM Tris-HC1 buffer or in 135 nlM KC1 i o mM Tris-HC1 buffer (pH 7.4). Each bag contains 0. 5 nil suspension and about o. 4 ~mole cardiolipin. been reported 2s. Since most of the radioactive ions came out from u n s o n i c a t e d vesicles (see Fig. 6) within a b o u t t h, one would be led to t h i n k t h a t all the lamellae i n each vesicle came in contact with antibiotics. T h a t n l e a n s each lamella received a sufficient q u a n t i t y of antibiotic molecules to increase the p e r m e a b i l i t y of the membrane. However, with sonication, the n u n l b e r of vesicles increased b y a t / e a s t IOOOfold. W h e n the c o n c e n t r a t i o n of a n t i b i o t i c was I0 nM which was the highest concent r a t i o n of a n t i b i o t i c used for u n s o n i c a t e d vesicles (ratio c a r d i o l i p i n : a n t i b i o t i c IO'5:I), the a m o u n t of antibiotics received b y each small sonicated vesicle was not enough to increase the p e r m e a b i l i t y of the m e m b r a n e . This follows the a r g u m e n t t h a t the aqueous phase per lamella in the u n s o n i c a t e d particles has a greater volume in relation to t h a t of the sonicated particles. So for one unsonicated particle, one needs fewer "holes" or "carrier" to e m p t y each bag. I n terms of existing hypothesis, this could be i n t e r p r e t e d b y assuming t h a t one needs more t h a n one a n t i b i o t i c molecules either to form a "pore ''29 or to act as a "shuttle-carrier ''a°. The present d a t a does not distinguish between these mechanisms, a n d further kinetic analysis is necessary to establish whether the efflux rate reaches a p l a t e a u as the c o n c e n t r a t i o n of a n t i b i o t i c is raised in accord with the shuttle-carrier hypothesis. Recent N M R studies ~° indicate i n a b i l i t y of the w d i n o m y c i n KCNS complex to undergo exchange in a low dielectric c o n s t a n t m e d i u m , a p r o p e r t y which was explained in support of a shuttle-carrier m e c h a n i s m for this antibiotic. However, it should be n o t e d t h a t gramicidin does n o t p a r t i t i o n in media of low dielectric c o n s t a n t (J. SAHA, D. SItEPARD, K. JACOBSON AND C. E. WENNER, u n p u b l i s h e d observation), a prerequisite property for a simple shuttle-carrier. The possibility t h a t granlicidin alters the surface properties providing favorable energy condition for ion t r a n s l o c a t i o n b y m e a n s other t h a n previously m e n t i o n e d hypothesis is n o t excluded. Biochim. Biophys. Xcta, 196 (197o) io 19
PERMEABILITY OF CARDIOLIPIN VESICLES
19
ACKNOWLEDGMENTS
We would like to thank Mr. J. Black for preparing the electronmicrographs. This project was partly supported by Western New York Health Fund, U.S. Public Health Service Grant No. CA-o5II5, and by an A.H.A. grant-in-aid No. 68698. J.S. is a National Institutes of Health post-doctoral fellow (Grant No. ToI-CAo5oI6-I2). D.P. is an Established Investigator of the A.H.A. and supported in part by the MidHudson Heart Association. REFERENCES I D. F. PARSONS, G. R. WILLIAMS AND B. CHANCE, Ann. N . Y . Aead. Sci., 137 (1966) 643. 2 D. F. PARSONS, 7th Can. Cancer Conf., Honey Harbor, Ontario, P e r g a m o n Press, Oxford, 1966, p. 193. 3 E. RACKER, Federation Proc., 26 (1967) 1335. 4 M. L. DAS, E. D. HAAK AND F. L. CRANE, Biochemistr% 4 (1965) 859. 5 S. FLEISCHER, g. FLEISCHER AND W. STOECKENIUS, J. Cell Biol., 32 (1967) 193. 6 P. MITCHELL, Nature, 191 (1961) 144. 7 P. MUELLER, D. O. RUDIN, H. T. TIEN AND W. C. ~vVESCOTT, Nature, 194 (1962) 979. 8 n . ~). BANGHAM, ~I. M. STANDISH AND J. C. WATKINS, J. Mol. Biol., 13 (1965) 238. 9 D. PAPAHADJOPOULOS AND n . D. BANGHAM, Biochim. Biophys. Acta, 126 (1966) 185. IO D. PAPAHADJOPOULOS AND N. MILLER, Biochim. Biophys. Acta, 135 (1967) 624. I I ~), PAPAHADJOPOULOS AND J. C. WATKINS, Biochim. Biophys. Acta, 135 (1967) 639. 12 K. LOHMANN AND L. JENDRASSEK, Biochemistry, 178 (1926) 419 . 13 A. D. BANGHAM AND R. ~V. HORNE, J. Mol. Biol., 8 (1964) 660. 14 D. G. DERVlCHIAN, in J. A. V. BUTLER AND H. E. HUXLEY, Progr. Biophys. Biophys. Chem., 14 (1964) 265. 15 A. D. BANGHAM, Progr. Biophys. 2,Iol. Biol., 18 (1968) 29. 16 M. I%LINGENBERG AND E. PFAFF, in J. M. TAGER, S. PAPA, E. QUAGL1ARIELLO AND E. C. SEATER, Regulation of Metabolic Processes in Mitochondria, B.B.A. Library, Vol. 7, Elsevier, Anasterdam, 1966, p. 18o. 17 J. SAHA, D. PAPAHADJOPOULOS AND C. WENNER, Federation Proc., 28 (1969) 893. I8 D. O. SHAH AND J. H. SCHULI~iAN, J. Lipid Res., 6 (1965) 341. 19 R. WHITTAM, Transport and Diffusion in Red Blood Cells, Arnold, London, 1964, p. 114. 20 L M. GLYNN, J. Physiol. London, 134 (1956) 278. 21 D. PAPAHADJOPOULOS, Biochim. Biophys. Acta, 163 (1968) 24 o. 22 D. PAPAHADJOPOULOS AND S. OKHI, Science, 164 (1969) lO75. 23 K. J. PALMER AND F. O. SCHMITT, J. Cellular Comp. Physiol., 17 (1941) 385 . 24 D. PAPAHADJOPOULOS AND S. OHKI, Syrup. Ordered Fluids Liquid Crystals, Am. Chem. Soc. 2~Ieeting, New York, ±969, P l e n u m Press, in the press. 25 G. H. DE HAAS, P. P. M. BONSEN AND L. L. M. VAN DEENEN, Biochim. Biophys. Acta, I i 6 (1966) 114. 26 B. C. PRESSMAN, Federation Pro&, 27 (1968) 1283. 27 R. SARGES AND B. WITKOP, J. Am. Chem. Soc., 87 (1965) 2027. 28 D. C. TOSTESON, T. E. ANDREOLI, M. TIEFFENBERG AND P. COOK, J. Gen. Physiol., 51 (1968) 373s. 29 P. J. F. HENDERSON, J. D. MCGIVAN AND J. B. CHAPPELL, Biochem. J., I I I (1969) 521. 3 ° D. H. HAYNES, A. KOWALSKY AND g. C. PRESSMAN, J. Biol. Chem., 244 (1969) 502.
Biochim. Biophys. Acta, 196 (I97 o) lO-19
BIOCHIMICA ET BIOPHYSICA ACTA
J~F,A 75376 STUDIES ON MODEL MEMBRANES I. EFI:ECTS OF (;a ~1 AND ANTIBIOTICS ON P E R M E A B I L I T Y OF CARI)I OLI PIN LI !~)UI1)-('RYSTALLINE VESICLES J. S:\HA, D. 1 ) A I ' A H A D J O P ( ) V L ( ) S axa) C. ]'i. \ \ : E N N E I ¢
Ne~, York State Department of Health, leoswell l)arl¢ 5[emodal Institute, Bl~jfalo, A'. Y. r42o 3 (c:.&:~ .i (Received August 4th, I969)
SUMMARY
Pure cardiolipin forms closed liquid-crystalline vesicles when suspended in aqueous salt solutions. Electronmicrographs of unsonicated suspensions indicate multilamellar vesicles and after sonication unilamellar vesicles are demonstrated by OsO~ fixation techniques. Both types of vesicles are relatively impermeable to sucrose. The self-diffusion rate of 22Na- efflux in sonicated vesicles in o.o37 /mlole Na= per ffinole cardiolipin per h and for ~"K- is about 0.o47 ,umole K~ per #mole cardiolipin per h. This rate of cation efflux by cardiolipin is typical of other vesicles prepared from acidic phospholipids. The efflux rate particularly of the sonicated particles is markedly increased in the presence of air, presumably because of the high percentage of unsaturated fatty acids which are susceptible to oxidation. Once prepared under N2, the permeability of the particles does not increase with time. Ca 2~ at low concentrations (z mM or less) is not effective in increasing permeability of cardiolipin vesicles in contrast to other acidic phospholipids. Valinomycin increases 42K rate 8-fold at o. 4 nil. Gramieidin increases a"K ~ rate 5-fold at 2o nM and also accelerates 22Na~ rate I6-fold at 2o nM.
INTRODUCe[ION
Only recently has it been possible to isolate inner and outer membranes of mitochondria in a relatively pure form 1. Analysis of the two layers indicates that cardiolipin is mainly present in inner membrane 2. It is now well established that inner membrane is the site of oxidative pbosphor3qationa. Furthermore, a role has been suggested for phospholipids in oxidative phosphorylation and electron transport in terms of maintaining electron transport activity< a. Since mitochondrial functions are intimately associated with ion movements and the recent "chemiosmotic hypothesis" of MITCHELL6 emphasizes the role of the membrane in separating H ~ and other cations, it is essential to have a better understanding of the permeability properties of inner membrane constituents. ]~'LEISCHFR et aI. 5 have shown that mitochondria can maintain their structural integrity after removal of lipids. This, of course, does not rule out the possibility that lipids might be necessary to maintain Biochim. Biophys. Acta, i96 (I970) IO--I 9
PERMEABILITY OF CARDIOLIPIN VESICLES
II
a tight membrane. In fact, in analogy with model membrane studies 7, it would appear that lipids of mitochondria might be partly responsible for separating ions and maintaining a potential difference. The discovery by BANCHA,~Iet al. s that phospholipids (amphipathic molecules) can form closed vesicles when suspended in aqueous salt solutions, has led to the study of physico-chemical properties of m a n y phospholipids except cardioliping-lL Permeability properties of cardiolipin liquid crystalline vesicles to 22Na+, aSK+ and uniformly l~C-labeled sucrose and the effects of gramicidin, valinomycin and Ca 2+ were studied in order to find the role played by cardiolipin as an inner men> brane constituent. MATERIALS
All reagents were analytical grade. The water was glass distilled. Cardiolipin was obtained from Supelco, Bellefonte, Pa., in sealed ampules under N 2 and kept at - I O ° when not in use. Microanalysis of ions present in cardiolipin was carried out by Referred Chemistry, North Hollywood, Calif., and Na + and K + contents were verified in our own laboratory. The amount of ions present in cardiolipin (expressed as /mmles per #mole of cardiolipin) are: Na + -: 2.3, K + = 0.05, Ca 2+ ~ o.I. 22Na+ was supplied by Amersham and Searle, DesPlanes, Ill., as 22NaC1 in aqueous solution (o.I mC/ml). 42K+ was obtained from Nuclear Science Division of ICN as 4eKC1 in aqueous solution (I 7 me/ml). Uniformly '4C-labeled sucrose was obtained from Nuclear Chicago in crystalline form (o. 5 mC per ampule). Gramicidin (a mixture of A and B) was obtained from Nutritional Biochemical Corp., and valinomycin was received from California Biochemical Corp. The dialysis bags (o.125 inch diameter) were a product of Union Carbide. Sephadex G-5o (coarse) was purchased from Pharmacia Fine Chemicals, Picataway, N.J. N 2 used to store cardiolipin after opening the ampule and for experiments is 99.996 % pure N~ of Liquid Carbonic Division of General Dynamics. Tris was Trizma base of Sigma. METHODS
The procedure used for the preparation of the cardiolipin vesicles and subsequent estimation of rates of diffusion was a minor modification of previous methods s,n. In most experiments, 4/~moles of cardiolipin stored in ethanol were dried under vacuum at 30-35 °. During evaporation, the tube was rotated by hand to make a thin layer at the b o t t o m of the tube. Evaporation was continued for i o - I 5 rain. Once the evaporation was complete, N 2 was allowed to flow over the dried cardiolipin layer. While N 2 was flushing over the thin shell of cardiolipin, an aqueous salt solution with radioactive solute was added to the test tube and mixed until all sides of the tube were clear. In cases where sonication was used, the tube after shaking on Vortex mixer was placed in a bath-type sonicator (Heat Systems Ultrasonic), so that the cardiolipin suspension was just below the water surface. N2, after passing through a gas-washing bottle containing distilled water, was flushed continuously during mixing and also during sonication. Sonication was carried out' for a total of 15 rain, in three 5-min intervals, each followed by a 5-min rest period to prevent elevation of temperature. The temperature during this procedure was elevated only Biochim. Biophys. Acta, I96 (I97 o) t o - ~ 9
12
j. SAHAet al.
slightly above room temperature (22 25'~). The tube with unsonicated or sonicated vesicles was closed under N 2 and left at room temperature 1-2 h for equilibration. Removal of excess radioactive material outside the vesicles was accomplished by passing through Sephadex G-5o (coarse)as described by PAPAHADJOPOULOSAND WATKINS~1. Dialysis was done at room temperature (22 25 °) in stoppered tubes with 2o inl dialysate for each bag containing o.5 ml of suspension (approx. o. 4 ~mole of cardiolipin). The salt solution contained I35 mM NaC1 or KC1 and IO mM TrisHC1 buffer adjusted to pH 7.4- For studies to deternline the activation energy of 22Na+ et'flux, bags were first dialysed at room temperature for 3o nlin, then the bags were transferred to different tubes containing 2o ml salt solution preadjusted to desired temperature. Bags were then transferred to other tubes containing 2o ml medium equilibrated at the sanle temperature for additional time intervals (iates over a period of 2-3 h were similar). Gramicidin and valinomyein in ethanol solution were added to cardiolipin suspension before pipetting into dialysis bags in io-#l quantities and controls had equivalent amount of ethanol. Ca 2~ was added to the dialysis medium. The rate of diffusion was expressed as percent of radioactivity captured per h. in this paper the term "vesicles" is used for both unsonicated and sonicated cardiolipin particles. 22Na+ and 42t( ~ were counted in Packard y-scintillation counter and rl4C]sucrose was counted in an ambient type liquid-scintillation counter (Nuclear Chicago). Cardiolipin was estimated from its phosphate content (i mole of cardiolipin per 2 moles of phosphate). Organic phosphate was estimated according to the method of LOHMANNAND JENDRASSEK TM. Electronmicrographs were prepared in the following way. Unsonicated cardiolipin vesicles were centrifuged at IOOooo x g for 9° rain. The sonicated vesicles were centrifuged at the same speed for 16 h. The pellets were then treated with OsO4, dehydrated and the photographs of lead citrate-stained thin sections were taken. RESULTS AND DISCUSSION
Physical characteristics of cardiolipin liquid-ct2vstalline vesicles Initially the electronmicrographs of cardiolipin vesicles were examined to correlate their physical characteristics to permeability properties. Fig. I (the electronmicrograph of unsonicated vesicles) shows heterogeneous nlulti-concentric, lamellar structures which vary greatly in size and shape. These vesicles are similar to other phospholipid particles as described by BANGHAM AND HORNE 13, and PAPAHAI)JOPOULOS ANt) MILLERTM. After sonieation (Fig. 2), the particles become smaller and most of them appear to be composed of one lamella of either spheroidal or crescentlike shape. Tile appearance of two lamellae is probably due to a cut through the infolded crescent-like particles. The transverse dimension of a single lamella in both sonicated and unsonicated vesicles is in agreement with bimolecular leaflet arrangement as suggested by earlier work reviewed recently ~,~5. Nevertheless, it is not possible to establish from this study involving purified cardiolipin how cardiolipin or other phospholipids are arranged in mitochondrial membrane with large amount of proteins.
Diffusio~z of ~14C]sucrose Normally sucrose is impermeable to most cell membranes and it is usually used to measure intercellular space. KI.INGENBERGAND PFAFFTM have shown that sucrose
Biochim. Biophys. Acta, 196 (197o) lO-19
PERMEABILITY OF CARDIOLIPIN VESICLES
13
Fig. I. Electronmicrograph of cardiolipin liquid-crystalline vesicles (unsonicated) after OsO4 fixation, acetone dehydration and lead citrate staining of Epon-embeded thin sections. Magnification J 1400 X. is not permeable to lnitochondrial m a t r i x space, b u t is permeable to outer m e m b r a n e . W h e n 14Qsucrose was t r a p p e d inside vesicles at room t e m p e r a t u r e a n d dialysed in the cold room (5°), no detectable a m o u n t of [l~C?sucrose appeared in the m~dium. This indicates t h a t cardiolipin vesicles are stable a n d p a r t i c u l a r l y impermeable to sucrose at 5 °. At room t e m p e r a t u r e (23 25°), the rate of efflux of sucrose is low. The average rate is found to be o.45 %/h (o.o8-o.75). W i t h sonicated vesicles the rate was found to be o.67 %/h (o.oo-I.4O). So even though the surface area increased with sonication, the rate did not increase. This is in agreement with earlier work with other phospholipids a n d the possible e x p l a n a t i o n s were discussed previously n. The lUCnsucrose associated with the vesicles seems to be c a p t u r e d inside the vesicles r a t h e r t h a n being adsorbed, since changing the n o n - r a d i o a c t i v e sucrose c o n c e n t r a t i o n 7° tilnes did not decrease the [laC~sucrose capture (about o.5 % per/*mole cardiolipin in sonicated vesicles). F u r t h e r m o r e , the c a p t u r e of a b o u t i ° o [14C~sucrose b y the u n s o n i c a t e d vesicles is in agreement with 2 % Na + capture p e r / , m o l e of cardiolipin. The high capture of Na + can be a t t r i b u t e d in part to the presence of Na + as a c o u n t e r ion in cardiolipin. G r a m i c i d i n up to a c o n c e n t r a t i o n of o.5 #M had no effect on sucrose permeability. As shown in Fig. 3, Ca2~ at low concentrations (up to i mM) has no significant effect on sucrose permeability. Only at higher c o n c e n t r a t i o n s (2 5 mM) does (;a a~ have an effect which results in an increase of the rate of sucrose efflux. This is in c o n t r a s t to other acidic phospholipids n, in which Ca e+ increased p e r m e a b i l i t y to
Biochim. Biophys. Acta, 196 (197o) lO-19
14
J. SAHA et a l
Fig. 2. E l e c t r o n m i c r o g r a p h of c a r d i o l i p i n l i q u i d - - c r y s t a l l i n e vesicles (sonicated) a ft e r ()sO~ f i x a t i o n as in Fig. T. M a g n i f i c a t i o n ~3 ioo × .
-i~12
lz
b_l
IlC b.l
13_ 8 LLI
nZ O
4
tl_ 0_ O' 0 ko 0~
I
,.o ;o MM
+io C a 2+
Fig. 3- Eftect of Ca 2+ on seIf-diffusion r a t e of u n i f o r m l y 14C-labeled sucrose t h r o u g h c a r d i o l i p i u l i q u i d c r y s t a l s . ©, u n s o n i c a t c d vesicles; 2x, s o n i c a t e d vesicles. All e x p e r i m e n t s were p e r f o r m e d a t room t e m p e r a t u r e , in z35 mM N a C l - i o mM Tris-HC1 buffer (pH 7-4) w i t h [a4C]sucrose as tracer.
42K~ at low concentrations (o.2 • mM). I t should be i n e n t i o n e d in this connection t h a t a fine precipitation resulting from aggregation of cardiolipin vesicles was visible at 5 mM Ca 2+ concentration. Large aggregates were observed with IO mM Ca 2+. Biockim. F3iopkys. Acta, I96 (T97 o) TO X9
t'ERMEAI3II.iTY OF CARDIOLIPIN VESICLES
15
I)(O'z~sio~ of "2X.~ The self-diffusionrate of 22Nai from unsonicated vesicles was found to be 0.23 °~)/h (3.21 9.96). [n terms of /zmoles, the rate is o.085 ~tmo]e Na- per /zmole cardiolipin per h. In a p r e l i n l i n a r y r e p o r t 1:, we have descril)ed the diffusion r a t e of "2Na ~ as IO I2 %/h. The high r a t e was p r o b a b l y due to the fact t h a t we did not use N2 during the initial phase of our experiments. All the e x p e r i m e n t s (except two e x p e r i m e n t s of t e m p e r a t u r e effect) r e p o r t e d here were carried out under N.,. The r a t e of 22Na t h r o u g h s o n i c a t e d vesicles is 4 . 6 I % / h (3.i 5 6.86), a n d in t e r m s of ,umoles the r a t e is o.o37/,mole N a ~ per/m~ole cardiolipin per h. In order to c o m p a l e the diffusion r a t e s given in this p a p e r to r e p o : t e d fluxes of ions through biological m e m b r a n e s , a t h e o r e t i c a l a r e a occupied b y I ftmole cardiolipin was calculated. Assulning an average a r e a per molecule of I2O A 2 (see ref. i8) a n d an average of ~me l a m e l l a per vesicle, an area of 36oo cm 2 per # m o l e cardiolipin was calculated. Fluxes for diffusion of Na% c a l c u l a t e d on t h e basis of the area m e n t i o n e d is approx. 0.6. Io a.5 equiv/cm.~ per sec. This value is s o m e w h a t lower t h a n the r e p o r t e d flux of K ~ t h r o u g h red cells 19 (too. zo 15 equiv/cnl'-' per see). The low r a t e ill sonicated vesMes indicates t h a t the vesicles offer considerable resistance to cation m o v e m e n t s and are v e r y stable. PAPAHA1)JOPO[TLOS AND \VATKINS 11 in their earlier studies with sonicated vesicles o b t a i n e d similar diffusion rates using the acidic phospholit)ids, p h o s p h a t i d i c acid and p h o s p h a t i d y l inositol. I t should be n o t e d t h a t the earlier studies u were carried out in the presence of air. W e have found t h a t sonication in the presence of air tends to increase the leakage r a t e of cardiolipin vesicles from three to four-fold in the I s t h a n d the rate was o b s e r v e d to increase with time, These findings suggest t h a t cardiolipin is susceptible to a u t o o x i d a t i o n . In t h e sonicated vesicles, t h e a m o u n t of ions c a p t u r e d (o.8o /tnlole N a + per /tmole cardiolipin) is a b o u t half of the u n s o n i c a t e d p r e p a r a t i o n ( I . 3 8 / , m o l e s Na~ per ,umole cardiolipin), b u t the surface a r e a increased m a n y - f o l d . Of the o . 8 / m l o l e N a + c a p t u r e per ~tnaole of cardiolipin in s o n i c a t e d vesicles, a b o u t half could be present as p h y s i c a l l y t r a p p e d NaC1 (due to C1 capture) u. F o r t h e ideal case of u n i l a m e l l a r vesicles, a t h e o r e t i c a l m i n i m u m of I equiv Na ~-p e r / , m o l e cardiolipin should be b o u n d as counter ions to the inside surface of the vesicles with a n y p h y s i c a l l y t r a p p e d NaC1 a d d i n g to the m e a s u r e d N a - content. The observed Na~ content, being significantly h)wer t h a n the t h e o r e t i c a l m i n i m u m for closed l a m e l l a r particles, i n d i c a t e s t h a t some of t h e particles m a y not have an enclosed i n t e r n a l aqueous stmce. In a n y case, sonieation, b v p r o d u c i n g smaller u n i l a m e l l a r particles would be e x p e c t e d to reduce t h e volume of the i n t e r n a l aqueous space as well as t h e n u m b e r of c o u n t e r ions inside the particles for a given a m o u n t of phospholipid. A n o t h e r e x p l a n a t i o n for the o b s e r v e d low c a p t u r e could be b a s e d on a slow r a t e of exchange of tile 2°~Na+ fronl the aqueous m e d i u m with the N a + a l r e a d y present as a counter ion in the cardiolipin samples. The e x t e n t of exchange is not definitely known. The 22Na ~ efltux r a t e is r e l a t i v e l y high in comparison to lecithin vesicles where t ) e r m e a b i l i t y to N a + a n d K ~ is v e r y low s, ~1. W e have carried out two e x p e r i m e n t s at t e m p e r a t u r e range of 5 4 °° for c a l c u l a t i n g t h e energy of a c t i v a t i o n . In one experim e n t c a r r i e d out a t p H 7.4 w i t h 135 mM NaC1 a n d IO mM Tris-HC1 buffer, t h e activ a t i o n energy was found to be i3.25 kcal/mole. A n o t h e r e x p e r i m e n t was clone at p H 5.2 w i t h I3O m M NaC1 a n d 15 mM T r i s - m a l e a t e buffer; t h e a c t i v a t i o n energy was h m n d to be I2.o 5 kcal/mole. B o t h these e x p e r i n l e n t s were done in t h e presence
Biochim. Biophys. Acta, 196 (197o) io 19
I6
j. SAHA dt gll.
of air. While these experiments were carried out in the presence o f air, the low rates of diffusion observed were indicative t h a t extensive oxidation of f a t t y acids chains of cardiolipin did n o t occur. F u r t h e r , the average value of an energy of a c t i v a t i o n was z2 14 kcal/mole which is in agreement with t h a t observed with silnilar phospholipids H. This value indicates t h a t Na ~ do not diffuse freely through aqueous channels; otherwise, the energy of a c t i v a t i o n would be very low .'° and efflux rate would be even higher. Consequently, we m a y conclude t h a t `'`'Na~ is moving across the lamellae as was found with other phospholipids s,'l
I-Z rO n.hi
5040
I.~ 5 0 n7
20
~_
mo
O (/3 ¢-~
hm o ~o
0
I.;
21.0
5 .I0
MM
Ca 2+
~4.0
51.0
Fig. 4. Effect of Ca2+ on self-diffusion rate of 2'aNa~ through cardiolipin liquid crystals. O, unsonicated vesicles; ~, sonicated vesicles. All experiments were performed at room temperature, in ] 3 5 mM NaCl-lo mM Tris-HCl buffer (pH 7-4). Tile effect of Ca 2~ o n tile diffusion rate of 2`'Na+ is shown in Fig. 4- It is clear t h a t up to a c o n c e n t r a t i o n of o. 4 mM there is no effect on 22Na+ efflux. There is some effect at z and 2 mM concentrations, a n d there is an appreciabie effect at 5 mM (;a"~ concentration. This behaviour is similar to the effect of Ca 2~ on phosphatidylinositol a n d t)hosphatidylserine p e r m e a b i l i t y to 4`'K+, b u t different froln phosphatidic acid where 42K' p e r m e a b i l i t y g r e a t l y increases at low Ca `'+ c o n c e n t r a t i o n ( < o . 4 raM) ~1. It has been suggested b y SHAH AND SCHULMAN's t h a t Ca 2+ i n t e r a c t s with two phosphate groups of the same cardiolipin molecule as shown in Fig. 5a. This type of b i n d i n g is more probable at low c o n c e n t r a t i o n s of Ca"~, when the p e r m e a b i l i t y is not affected. It has also been suggested t h a t Ca" ~ can i n t e r a c t with more t h a n one different molecules forming linear polymeric a r r a n g e m e n t 21. This idea is in agreement with the fact t h a t we o b t a i n e d large aggregates at zo mM Ca"concentration. I t has recently been proposed t h a t the increase in p e r m e a b i l i t y of acidic phospholipids in the presence of Ca 2+ results from an a s y m m e t r i c d i s t r i b u t i o n of Ca `'+ on two sides of the bilayer m e m b r a n e 2`'. X - r a y diffraction studies 1°, 2a indicate t h a t Ca 2+ brings out the water with other ions from phost)holit)id suspension, b u t still retains bimolecular leaflet structure. There is no d o u b t Ca `'+ interacts with cardiolipin vesicles at 2 5 mM concentrations, b u t there is no evidence from this s t u d y t h a t ('a`" is effective at low concentration ( < o . 2 raM), at which it s t i m u l a t e s respiration in mitoehondria. The reason for i n a c t i v i t y of Ca 2+ at low concentration in t~iochim, l~iophy.,. ,4 cla, 19¢~( i 0 7o) , o - i ()
17
PERMEABILITY OF CARDIOLIPIN VESICLES
contrast to its effect on phosphatidic acid might be due to the fact that here only one phosphate oxygen is free instead of two and consequently the polymeric arrangement involving coordination bonds between each Ca 2+ and four phospholipid molecules as proposed by PAPAHADJOPOULOS 21 is not possible.
\
/0/P%0 o#P\0\
Ca
,,o.'P% o
Co
tj \ /
Ca
o.'P,o O_#P',_U \
/
Ca
(b)
J'o- o"P' o Ca a+
(a)
~p/O-X UO-X 0~ \0 /o/P~o Co (c)
Fig. 5. a. I n t e r a c t i o n of Ca 2+ and two p h o s p h a t e groups of a single cardiolipin molecule as visualized b y SHAH AND SCHULMAN18. b. Polymeric a r r a n g e m e n t b y Ca 2+ linking different cardiolipin nlolecules, c. Dimer formation by Ca 2+ b e t w e e n two molecules of phosphatidylglycerol, p h o s p h a tidylinositol and p h o s p h a t i d i c acid at low pH. X represents either hydrogen or a glycerol or an inositol moiety.
In regard to the effect of Ca 2+ in changing tile permeability of acidic phospholipids, cardiolipin is intermediate between phospholipids as phosphatidic acid (at high pH) or phosphatidylserine, and other acidic phospholipids as phosphatidylinositol (see ref. I I ) and phosphatidylglycerol or phosphatidic acid at low pH (see ref. 24). The former group is sensitive at Ca 2~ concentrations of o.2-I.O raM. It is, therefore, tempting to speculate that cardiolipin can form polymeric arrangements with Ca ~+ linking different molecules (see Fig. 5b) because of the special molecular configuration of eardiolipin ~5which would not be possible with other acidic phospholipids such as phosphatidylglycerol, phosphatidylinositol and phosphatidic acid at low pH. The latter phospholipids would only be capable of dimer formation (see Fig. 5c). Effcct of antibiotics on diffusion of N a + and K+
The effect of gramicidin on tile self-diffusion rate of Na + efflux is shown in Fig. 6. There is an increase in the rate of Na + efflux as the concentration of gramicidin increases, with the slope of the curve (efflux vs. gramieidin concentration) rising steeply at higher concentrations. With equivalent concentrations of gramicidin, this antibiotic is less effective in increasing the 42K+ efflux in cardiolipin vesicles in comparison to its effect on 22Na+ efflux. Sonicated vesicles required a higher concentration for the same change observed with unsonicated vesicles. Valinomycin is more effective than gramicidin in accelerating the rate of K + efflux at low concentrations (Fig. 6). Furthermore, as with gramicidin, higher concentrations of valinomyein were required to induce leakage in sonieated vesicles as compared to those which were unsonicated. Thisindicates that the action of antibiotics is dependent on the surface area. Biochim. Biophys. Acta, 196 (197 o) lO-19
I8
j. SAHA et al.
At least for v a l i n o m v e i n and n o n a c t i n , it was proposed 26 t h a t the antibiotics in cyclic configuration act as carrier for K +. (;ralnieidins A and B are l i n e a r peptides 27 a n d no definitive evidence has vet been presented t h a t gramicidin can fornl a cyclic conforlnation in s(flution . r in the m e m b r a n e , although some suggestive evidence has
,./
~too
!
o,
/ z
o
50
e,,
L L
//
/[
0
•
~
0
I
0.1
~
I
1.0
ANTIBIOTIC
~
I
i
I0 CONCN.
I
I00
=
I
I000
(riM)
Fig. 6. Effect of antibiotics on self diffusion rate of 2"Na+ or ~'af(+ through cardiolipin liquid crystals. Symbols representing antibiotic-induced offlux of cations are as folIows: O, valinomycin, 42[(+., unsonicated; ~k, valiomycin, 42K~, sonicated; O, gramicidin, 22Na+, unsonicated; &, granlicidin, 22Na-, sonicated; @, gramicidin, 42K+, unsonicated. All experiments were performed at room temperature, in 135 mM NaC1-[o mM Tris-HC1 buffer or in 135 nlM KC1 i o mM Tris-HC1 buffer (pH 7.4). Each bag contains 0. 5 nil suspension and about o. 4 ~mole cardiolipin. been reported 2s. Since most of the radioactive ions came out from u n s o n i c a t e d vesicles (see Fig. 6) within a b o u t t h, one would be led to t h i n k t h a t all the lamellae i n each vesicle came in contact with antibiotics. T h a t n l e a n s each lamella received a sufficient q u a n t i t y of antibiotic molecules to increase the p e r m e a b i l i t y of the membrane. However, with sonication, the n u n l b e r of vesicles increased b y a t / e a s t IOOOfold. W h e n the c o n c e n t r a t i o n of a n t i b i o t i c was I0 nM which was the highest concent r a t i o n of a n t i b i o t i c used for u n s o n i c a t e d vesicles (ratio c a r d i o l i p i n : a n t i b i o t i c IO'5:I), the a m o u n t of antibiotics received b y each small sonicated vesicle was not enough to increase the p e r m e a b i l i t y of the m e m b r a n e . This follows the a r g u m e n t t h a t the aqueous phase per lamella in the u n s o n i c a t e d particles has a greater volume in relation to t h a t of the sonicated particles. So for one unsonicated particle, one needs fewer "holes" or "carrier" to e m p t y each bag. I n terms of existing hypothesis, this could be i n t e r p r e t e d b y assuming t h a t one needs more t h a n one a n t i b i o t i c molecules either to form a "pore ''29 or to act as a "shuttle-carrier ''a°. The present d a t a does not distinguish between these mechanisms, a n d further kinetic analysis is necessary to establish whether the efflux rate reaches a p l a t e a u as the c o n c e n t r a t i o n of a n t i b i o t i c is raised in accord with the shuttle-carrier hypothesis. Recent N M R studies ~° indicate i n a b i l i t y of the w d i n o m y c i n KCNS complex to undergo exchange in a low dielectric c o n s t a n t m e d i u m , a p r o p e r t y which was explained in support of a shuttle-carrier m e c h a n i s m for this antibiotic. However, it should be n o t e d t h a t gramicidin does n o t p a r t i t i o n in media of low dielectric c o n s t a n t (J. SAHA, D. SItEPARD, K. JACOBSON AND C. E. WENNER, u n p u b l i s h e d observation), a prerequisite property for a simple shuttle-carrier. The possibility t h a t granlicidin alters the surface properties providing favorable energy condition for ion t r a n s l o c a t i o n b y m e a n s other t h a n previously m e n t i o n e d hypothesis is n o t excluded. Biochim. Biophys. Xcta, 196 (197o) io 19
PERMEABILITY OF CARDIOLIPIN VESICLES
19
ACKNOWLEDGMENTS
We would like to thank Mr. J. Black for preparing the electronmicrographs. This project was partly supported by Western New York Health Fund, U.S. Public Health Service Grant No. CA-o5II5, and by an A.H.A. grant-in-aid No. 68698. J.S. is a National Institutes of Health post-doctoral fellow (Grant No. ToI-CAo5oI6-I2). D.P. is an Established Investigator of the A.H.A. and supported in part by the MidHudson Heart Association. REFERENCES I D. F. PARSONS, G. R. WILLIAMS AND B. CHANCE, Ann. N . Y . Aead. Sci., 137 (1966) 643. 2 D. F. PARSONS, 7th Can. Cancer Conf., Honey Harbor, Ontario, P e r g a m o n Press, Oxford, 1966, p. 193. 3 E. RACKER, Federation Proc., 26 (1967) 1335. 4 M. L. DAS, E. D. HAAK AND F. L. CRANE, Biochemistr% 4 (1965) 859. 5 S. FLEISCHER, g. FLEISCHER AND W. STOECKENIUS, J. Cell Biol., 32 (1967) 193. 6 P. MITCHELL, Nature, 191 (1961) 144. 7 P. MUELLER, D. O. RUDIN, H. T. TIEN AND W. C. ~vVESCOTT, Nature, 194 (1962) 979. 8 n . ~). BANGHAM, ~I. M. STANDISH AND J. C. WATKINS, J. Mol. Biol., 13 (1965) 238. 9 D. PAPAHADJOPOULOS AND n . D. BANGHAM, Biochim. Biophys. Acta, 126 (1966) 185. IO D. PAPAHADJOPOULOS AND N. MILLER, Biochim. Biophys. Acta, 135 (1967) 624. I I ~), PAPAHADJOPOULOS AND J. C. WATKINS, Biochim. Biophys. Acta, 135 (1967) 639. 12 K. LOHMANN AND L. JENDRASSEK, Biochemistry, 178 (1926) 419 . 13 A. D. BANGHAM AND R. ~V. HORNE, J. Mol. Biol., 8 (1964) 660. 14 D. G. DERVlCHIAN, in J. A. V. BUTLER AND H. E. HUXLEY, Progr. Biophys. Biophys. Chem., 14 (1964) 265. 15 A. D. BANGHAM, Progr. Biophys. 2,Iol. Biol., 18 (1968) 29. 16 M. I%LINGENBERG AND E. PFAFF, in J. M. TAGER, S. PAPA, E. QUAGL1ARIELLO AND E. C. SEATER, Regulation of Metabolic Processes in Mitochondria, B.B.A. Library, Vol. 7, Elsevier, Anasterdam, 1966, p. 18o. 17 J. SAHA, D. PAPAHADJOPOULOS AND C. WENNER, Federation Proc., 28 (1969) 893. I8 D. O. SHAH AND J. H. SCHULI~iAN, J. Lipid Res., 6 (1965) 341. 19 R. WHITTAM, Transport and Diffusion in Red Blood Cells, Arnold, London, 1964, p. 114. 20 L M. GLYNN, J. Physiol. London, 134 (1956) 278. 21 D. PAPAHADJOPOULOS, Biochim. Biophys. Acta, 163 (1968) 24 o. 22 D. PAPAHADJOPOULOS AND S. OKHI, Science, 164 (1969) lO75. 23 K. J. PALMER AND F. O. SCHMITT, J. Cellular Comp. Physiol., 17 (1941) 385 . 24 D. PAPAHADJOPOULOS AND S. OHKI, Syrup. Ordered Fluids Liquid Crystals, Am. Chem. Soc. 2~Ieeting, New York, ±969, P l e n u m Press, in the press. 25 G. H. DE HAAS, P. P. M. BONSEN AND L. L. M. VAN DEENEN, Biochim. Biophys. Acta, I i 6 (1966) 114. 26 B. C. PRESSMAN, Federation Pro&, 27 (1968) 1283. 27 R. SARGES AND B. WITKOP, J. Am. Chem. Soc., 87 (1965) 2027. 28 D. C. TOSTESON, T. E. ANDREOLI, M. TIEFFENBERG AND P. COOK, J. Gen. Physiol., 51 (1968) 373s. 29 P. J. F. HENDERSON, J. D. MCGIVAN AND J. B. CHAPPELL, Biochem. J., I I I (1969) 521. 3 ° D. H. HAYNES, A. KOWALSKY AND g. C. PRESSMAN, J. Biol. Chem., 244 (1969) 502.
Biochim. Biophys. Acta, 196 (I97 o) lO-19