On the Inhibition of Muscle Membrane Chloride Conductance by Aromatic Carboxylic Acids P. T . P A L A D E
and R. L. B A R C H I
From the Departments of Neurology and of Biochemistry and Biophysics, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19174. Dr. Palade's present address is the Department of Physiology and Biophysics, University of Washington School of Medicine, Seattle, Washington 98195.
A B S T R A C T 25 aromatic carboxylic acids which are analogs o f benzoic acid were tested in the rat d i a p h r a g m preparation for effects on chloride conductance (Gc~). O f the 25, 19 were shown to reduce m e m b r a n e Gca with little effect on other m e m b r a n e parameters, although their a p p a r e n t Kl varied widely. This inhibition was reversible if exposure times were not prolonged. T h e most effective analog studied was anthracene-9-COOH (9-AC; Kt = 1.1 x 10-s M). Active analogs p r o d u c e d concentration-dependent inhibition o f a type consistent with interaction at a single site or g r o u p o f sites having similar binding affinities, although a correlation could also be shown between lipophilicity and Kl. Structure-activity analysis indicated that hydrophobic ring substitution usually increased inhibitory activity while para polar substitutions reduced effectiveness. These compounds do not a p p e a r to inhibit Gcl by altering m e m b r a n e surface charge and the inhibition p r o d u c e d is not voltage d e p e n d e n t . Qualitative characteristics o f the I-V relationship for CI- current are not altered. Conductance to all anions is not uniformly altered by these acids as would be expected from steric occlusion o f a common chafinel. Concentrations o f 9-AC reducing Gci by >90% resulted in slight augmentation of GI. T h e complete conductance sequence obtained at high levels o f 9-AC was the reverse o f that obtained u n d e r control conditions. Permeability sequences underwent progressive changes with increasing 9-AC concentration and ultimately inverted at high levels o f the analog. Aromatic carboxylic acids a p p e a r to inhibit Gc~ by binding to a specific i n t r a m e m b r a n e site and altering the selectivity sequence o f the m e m b r a n e anion channel. INTRODUCTION I n s k e l e t a l m u s c l e o f m a n y s p e c i e s t h e r e s t i n g m e m b r a n e i o n c o n d u c t a n c e is p r e d o m i n a n t l y t h a t a t t r i b u t a b l e to c h l o r i d e (25, 37, 1). A l t h o u g h n o t d i r e c t l y i n v o l v e d in t h e g e n e s i s o f a n a c t i o n p o t e n t i a l , c h l o r i d e c o n d u c t a n c e p l a y s a n i m p o r t a n t r o l e in t h e m a i n t e n a n c e o f m e m b r a n e p o t e n t i a l stability. R e c e n t s t u d i e s in p a t h o l o g i c a l l y u n s t a b l e m u s c l e m e m b r a n e s h a v e s u p p o r t e d this c o n c e p t (2, 38). M y o t o n i a , a n e x a m p l e o f s u c h a n u n s t a b l e p a t h o l o g i c s t a t e , is a p h e n o m e n o n c h a r a c t e r i z e d b y r e p e t i t i v e a c t i o n p o t e n t i a l g e n e r a t i o n in t h e s a r c o l e m m a in THE JOURNAL OF GENERALPHYSIOLOGY° VOLUME 69, 1977 " pages 879-896
879
880
T H E JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 6 9 • 1 9 7 7
response to a b r i e f stimulus (27). This p h e n o m e n o n has been described in several naturally o c c u r r i n g diseases in m a n a n d animals (11,27). In those that have been studied e x p e r i m e n t a l l y the repetitive activity is associated with an absolute reduction in m e m b r a n e c o n d u c t a n c e to chloride (10, 30). Myotonia can be p r o d u c e d in otherwise n o r m a l animals by various chemical agents, including inhibitors o f cholesterol biosynthesis (40), certain herbicides (38), and in some species iodide (35). Many agents which reversibly p r o d u c e a myotonic s y n d r o m e in n o r m a l animals are m e m b e r s o f a class o f substituted benzoic acids (15). In several cases these agents have b e e n e x p e r i m e n t a l l y shown to reduce muscle m e m b r a n e Ge~ u n d e r in vitro conditions (12). We have previously r e p o r t e d the general characteristics of chloride conductance in the rat d i a p h r a g m (37), a n d evidence has b e e n p r e s e n t e d f r o m this laboratory a n d others that a r e d u c t i o n in m e m b r a n e Gel in muscle is sufficient g r o u n d s for repetitive electrical activity (6, 3, 8). In o r d e r to e x t e n d o u r u n d e r standing o f chloride c o n d u c t a n c e m e c h a n i s m s in n o r m a l a n d in pathologic states, we have studied in the rat d i a p h r a g m the m e c h a n i s m o f action o f a series o f aromatic acids similar to those k n o w n to p r o d u c e m y o t o n i a in o t h e r species. We have d o c u m e n t e d that these c o m p o u n d s do i n d e e d reversibly r e d u c e m e m b r a n e Gel a n d that direct correlations can be d e m o n s t r a t e d between the physical p r o p e r t i e s o f these analogs a n d their effectiveness in m o d i f y i n g chloride conductance. Earlier r e p o r t s have postulated that the o b s e r v e d r e d u c t i o n in m e m b r a n e Ge~ p r o d u c e d by the benzoic acid derivatives is d u e to steric blocking o f the anion p e r m e a t i o n pathway (12). We show here, however, that this inhibition most p r o b a b l y involves a direct interaction with the anion p e r m e a t i o n site in a m a n n e r affecting its ion selectivity r a t h e r t h a n by steric blocking o f these channels or by altering m e m b r a n e surface charge. MATERIALS
AND
METHODS
The majority of the 25 aromatic carboxylic acids studied were obtained from Aldrich Chemical Co., Milwaukee, Wis., from K & K Laboratories, Plainview, N. Y., or from Eastman Kodak Co., Rochester, N. Y. Several were the generous gift of Dr. R. B. Moffett of the Upjohn Pharmaceutical Co., Kalamazoo, Mich. Neuraminidase and phosphatidyl choline were obtained from Sigma Chemical Co., St. Louis, Mo. Male Wistar rats of 200-300 g were used for all experiments. Ringer's solution with or without various concentrations of aromatic carboxylic acids was continuously perfused over sections of diaphragm which were obtained and handled as previously described (37). Temperature control, recording methods, and solution compositions were all as described in an earlier paper ($7). For routine determination of half-maximal blocking concentration (K0 several determinations of the percent of inhibition of Gc~ were made at three or more concentrations of each aromatic acid. For each determination the percent inhibition of Gel was calculated after three-point cable analysis of five or more fibers in each of the following solutions: (a) normal Ringer's; (b) normal Ringer's + carboxylic acid; and (c) chloride-free Ringer's + carboxylic acid. Control values for Gel were obtained by subtraction of an average GE = 0.34 mmho/cm 2 from the average Gm obtained in the control solution. Subtraction of the average Gm in solution (c) from that in solution (b) yielded a value for Gel in the presence of the carboxylic acid, allowing a percent inhibition to be calculated. Double reciprocal
PALADEAND BARCHI Aromatic Carboxylic Acids and Chloride Conductance
881
plots of inhibition vs. analog concentration were then constructed and an observed Kl determined graphically. Partition coefficients for the benzoic acids were calculated by summation of values for the benzoic acid nucleus and factors specific for the effect on partitioning of a substitution at a given location on the parent compound based on the experimental results and theoretical methods of Hansch and co-workers (24, 23, 19). Hammett ¢r values and values for steric hindrance were also obtained from the chemical literature (26, 32). (See Results for a discussion of these parameters.) Statistical analysis was performed by standard computer methods (36). Electrophoretic mobilities of erythrocytes and liposomes were determined in a cylindrical microelectrophoresis cell 10 cm long with large volume electrode vessels at either end. The design was that of Bangham et al. (4). Constant temperature was maintained by immersion in a regulated water bath. Measurements were made with frequent reversals of field direction. Erythrocytes were prepared from human blood by dilution with 0.145 M NaCI and repeated centrifugation. Liposomes were prepared after the method of Bangham et al. (5). Phosphatidyl choline was dissolved in hexane-chloroform 1:10 and deposited on the sides of a round-bottom flask by flash evaporation. Liposomes were formed by the addition of 100 p~l of 0.145 M NaCI (pH 7.35) to the dried residue and sonication until vesicular formation was optimized. RESULTS
A n t h r a c e n e - 9 - C O O H (9-AC) may be considered representative o f the aromatic carboxylic acids studied in its effects on the rat d i a p h r a g m . T h e addition o f 10-5 M 9-AC to the Ringer's solution s u r r o u n d i n g a section o f d i a p h r a g m being maintained in vitro results in a rapid increase in the m e m b r a n e resistance. This increase begins almost immediately and reaches a new steady value within 10 min. W h e n n o r m a l Ringer's solution is r e p e r f u s e d over the d i a p h r a g m , memb r a n e resistance gradually r e t u r n s to near control levels over a 30-60-min period. W h e n similar e x p e r i m e n t s are p e r f o r m e d in methylsulfate-substituted chloride-free Ringer's, m e m b r a n e resistance is u n a f f e c t e d by this aromatic acid at the same concentration, suggesting that the c o m p o u n d acts specifically to r e d u c e the m e m b r a n e c o n d u c t a n c e to chloride (Gel-). W h e n a complete series o f steady-state c o n d u c t a n c e m e a s u r e m e n t s was m a d e on d i a p h r a g m fibers equilibrated in various concentrations (1-100/~M) o f 9-AC in n o r m a l Ringer's and C1--free Ringer's, a sigmoid relationship could be d e m o n s t r a t e d between the m e a s u r e d m e m b r a n e Gem a n d the logarithm o f the concentration o f 9-AC (Fig. 1). W h e n the average Gem at each 9-AC concentration is calculated a n d replotted a c c o r d i n g to the Hill equation, a diagnostic plot for binding characteristics, a linear relationship is obtained (Fig. 2). (9, see r e f e r e n c e 31 for recent discussion). T h e Hill coefficient calculated f r o m these data is not significantly different f r o m one, suggesting that the inhibitory action o f the a n t h r a c e n e derivative is mediated via the binding at a single class o f noninteracting sites. A dissociation constant o f 1.1 × 10-5 M can be calculated for the a p p a r e n t ligand-site interaction. This constant may be m o r e conveniently considered to be an inhibition constant for 9-AC with respect to m e a s u r e d Gel and for this and subsequent analogs will be r e f e r r e d to as Ki. As would be anticipated f r o m the n a t u r e o f the Hill plot for the above data, the relationship between the percent inhibition o f Gem and the concentration o f 9-
882
THE
JOURNAL
OF
GENERAL
PHYSIOLOGY"
VOLUME
69'
1977
AC is hyperbolic in nature, and a double reciprocal plot o f data in this f o r m is linear. A value for Kl identical to that obtained f r o m the Hill plot can be d e t e r m i n e d m o r e conveniently f r o m this g r a p h , a n d this m e t h o d was used subsequently to evaluate the a p p a r e n t inhibition constants with respect to Gc~ o f 24 o t h e r related derivatives o f the benzoic acid nucleus. For each analog determinations o f percent inhibition were m a d e at at least three analog concentrations (see Materials and Methods) a n d values o f Kl d e t e r m i n e d graphically• Significant reduction o f m e m b r a n e Gcl was p r o d u c e d by 19 o f these compounds. In all cases in which inhibition o f chloride c o n d u c t a n c e was observed, a
•
II
2.0
-O
E E i o
O
? 1.0
•
10-6
.
,
.
,,,,
.
.
.
.
10-5 ANTHRACENE-9-COOH (M)
,
,,,.
,i
10-4
10-6
10-5
10-4
ANTHRACENE-9-COOH (M)
FZGURE 1. Membrane chloride conductance (Gcl) observed after incubation of rat diaphragm preparation in various concentrations of 9-AC for 30 min at 35°C. Each point at a given 9-AC concentration represents the average of fibers measured in a different diaphragm. Measurements were made in C1--containing and C1--free Ringer's solution for conductance calculations (see Materials and Methods). FIGURE 2. The average values of GCl at each concentration of 9-AC from Fig. 1 are plotted according to the Hill equation. The average Gcl of all preparations before addition of 9-AC was taken as Gcl~max). The Hill coefficient of the regression line to this data is not significantly different from one. The dissociation constant for the apparent 9-AC site complex calculated from this data is 1.1 x 10-5 M. hyperbolic relationship between analog concentration and percent inhibition could be d e m o n s t r a t e d and the plots o f (1/percent inhibition) vs. (1/[aromatic acid]) were linear. Inhibition constants d e t e r m i n e d f r o m the double-reciprocal plots are listed in Table I. It can be seen that the concentrations required to p r o d u c e 50% inhibition o f Gcl varied a m o n g these c o m p o u n d s over a r a n g e o f three o r d e r s o f m a g n i t u d e . Resting m e m b r a n e potential (RMP) c h a n g e d little after exposure to 9-AC. T h u s the average RMP o f all fibers before e x p o s u r e to 9-AC was 70.3 -+ 5.2 (SD) mV while fibers exposed to 2 × 10-s M, 1 × 10-5 M, and 5 × 10-s M 9-AC had average RMP o f 68.3 + 3.8, 67.8 -+ 4.9, and 67.7 -+ 4.8 mV, respectively. T h e small d e c r e m e n t observed may represent the i n h e r e n t " r u n d o w n " o f the dia-
PALADE AND BARCHI
Aromatic Carboxylic Acids and Chloride Conductance
883
p h r a g m f i b e r s a t 35°C s i n c e all m e a s u r e m e n t s on control fibers were carried out w i t h i n t h e f i r s t 30 m i n o f a n e x p e r i m e n t while determinations in 9-AC were m a d e f r o m 30 m i n t o 2 h a f t e r d i s s e c t i o n o f t h e d i a p h r a g m . Overall, no correlation was noted between RMP and concentration of 9-AC in the perfusion m e d i u m o v e r t h e r a n g e i n d i c a t e d i n F i g . 1. For each compound,
change in residual conductance TABLE
INHIBITION
Antbracene-9-COOH 2,4,5-Trimethyl benzoic acid Pentachloro BA 3,~-Dichloro-2-amino BA 2,3,5,6-Tetrachloro BA 3,5-Dichloro BA 3,4-Dichloro BA 2,5-Dichloro BA 2,5-Dimethyl BA 2,3,5,6-Tetramethyl BA 2,3,6-Trichloro BA 3,5-Dimethyl BA 2,4-Dimethyl BA 3,6-Dichloro-2-nitro BA 3,4-Dimethyl BA 2,4-Dichloro phenoxyacetic acid 2,6-Dichloro BA 3,5-Dimethyl-4-nitro BA 2,4,6-Trimethyl BA Benzoic acid (BA) 4,5-Dichioro-2-COOH BA 3,5-Dimethyl-4-amino BA 2,5-Dimethyl benzene SOsH 2,3,5,6-Tetramethyl-4-COOH-BA 3,5-Dimethyl-4-hydroxy BA
solution
I
O F Gct I N R A T D I A P H R A G M
Compound
in chloride-free
BY A R O M A T I C
ACIDS
Kt(M)
Relative potency
Calc. log P
GK
1.1 x 10-s 6.6x 10-s 8,0x 10-5 1.3x 10-4 2.1 x 10-4 2.5x 10-4 2.6x 10-4 3.0x 10-4 3.1 x 10-4 6.0x 10-4 9.5 x 10-4 1 .Sx 10-s 1.6x 10-a 1.7x 10-3 2.0x 10-s 2.0 x 10-a 2.0x 10-s 2.8 x 10-a 4.0x 10-8 > l . 0 x 10-2 1.6x 10-2 >3.0x 10-~ >3.0x 10-2 > l . 0 x l 0 -1 > l . 0 x 10-1
909 152 125 76 47 40 38 33.3 32 16.7 10.5 6.7 6.2 6.0 5.0 5.0 5.0 3.6 2.5 B r - > I - . After exposure to 5 x 10-5 M 9-AC, the potential shift on c h a n g i n g to I - - c o n t a i n i n g solution reversed sign, indicating that u n d e r these conditions the m e m b r a n e is now more permeable to I - than to e l - (Fig. 6). Increasing [9-AC] to 2.5 x 10-4 M resulted in both I - and B r - solutions p r o d u c i n g a h y p e r p o l a r i z i n g response, the overall permeability sequence now being inverted to I - > B r - > e l - in a m a n n e r analogous to that observed for the c o n d u c t a n c e sequence. Such sequence inversions could also be d e m o n s t r a t e d in K+-free m e d i u m with Rb + substitution, CONTROL
8 -78mV cv.~.--~=O--cv~--z-
I
cv
1-s-~--cr-.---{
ANTHRACENE-9-COOH
5 x 10"SM
z mV
-75 mV - - - ~
c,--.--I-z--I
o 0Q.
/ ~
cl-
.,.
I- e~--I--c, -..-1- z--I--c,---I
FIGURE 6. A, Continuous traces of membrane potential demonstrating depolarizing shifts when external el- is rapidly replaced by Br- or I-. In each case return to normal el- solution results in membrane repolarization. B, After exposure to 9-AC (5 x 10-5 M) for 30 min, similar recordings in the same diaphragm demonstrate a hyperpolarization when I- is rapidly substituted for el- and depolarization when Br- is substituted for el-. Figures represent ink trackings of original records. eliminating changes in K + permeability as a d e t e r m i n a n t factor in the observed responses. Similar results in the presence o f $203 (42) (5 mM) rule out contribution f r o m an I-s carrier system. 13 experiments were d o n e c o m p a r i n g control permeability sequences to sequences obtained after equilibration in Ringer's solution containing 9-AC in concentrations between 1 a n d 100 × 10 -5 M. T h e permeability sequences obtained are shown graphically in Fig. 7. A progressive and reproducible shift in sequence is seen with increasing concentrations o f the aromatic acid. T h e same sequence inversions were observed with 2,4,5-trimethyl benzoic acid. Analogs with lower potencies p r o d u c e d partial sequence inversions in similar experiments, but concentrations in solution proportionately h i g h e r than their Kl could not be obtained due to their limited solubility. Again, permeability sequences d e t e r m i n e d at r e d u c e d p H showed a pattern identical to that obtained at p H 7.
PALADEAND BARCHI
891
Aromatic Carboxylic Acids and Chloride Conductance
DISCUSSION
In rat d i a p h r a g m , halogenated and methylated derivatives o f benzoic acid increase m e m b r a n e resistance by specifically r e d u c i n g m e m b r a n e conductance to CI-. Such a reduction in Gcl has previously been suggested as the primary physiologic effect o f several o f these c o m p o u n d s (12) o f which 2,4-dichlorophenoxyacetic acid (2,4-D) is the best known example (38). This behavior is now shown to be characteristic o f this class o f c o m p o u n d s in general. In vivo studies have d e m o n s t r a t e d a transient myotonic s y n d r o m e in mice associated with a single injection o f analogs similar to those tested h e r e (34). We confirm that the effects o f these c o m p o u n d s on Gcl in vitro is indeed reversible if the concentraI-> BF->Cl-
~
~
2'~
~o
0 w u} >"
Cl-ffi T - > B r -
.J m hi
w n
N
CF>I->Br-
el > B r
->I
/
~
l
[Anthracene-9-COOH]
,b M
,80
x I0 -5
FmURE 7. Summary of average permeability sequences determined from experiments similar to that shown in Fig. 6 after exposure of diaphragm preparations to 9-AC at various concentrations for 30 min at 35°C and pH 7.5. tion o f the analog is kept below cytotoxic levels. By extrapolation f r o m o t h e r work; it would a p p e a r likely that these c o m p o u n d s induce a myotonic s y n d r o m e in vivo by virtue o f the reduction in GcI which they p r o d u c e (6, 38). T h e benzoic acid derivatives r e p o r t e d h e r e reversibly depress Cl- conductance at extracellular concentrations which can be directly related to their relative solubility in a n o n p o l a r e n v i r o n m e n t . This observation suggests that the comp o u n d s must partition into the lipid phase o f the muscle m e m b r a n e b e f o r e p r o d u c i n g their effect. T h e hyperbolic relationship observed between extracellular analog concentration and percent reduction in Gcl, however, suggests that the inhibition o f Gcl is not simply related to i n t r a m e m b r a n e analog concentration. T h e n a t u r e o f ' t h e observed relationship implies that binding at a specific site or class o f sites having similar affinities is a critical" step leading to inhibition. T h e d e p e n d e n c e on partition coefficient would suggest that the specific site o f action is located either within or across the surface m e m b r a n e .
892
T H E J O U R N A L OF G E N E R A L P H Y S I O L O G Y • V O L U M E
69
• 1977
When calculated inhibition constants are corrected for intramembrane concentration based on partition coefficients for each analog they are still found to vary considerably among the analogs tested. This suggests that structural features related to specific binding also play a role in determining the effectiveness of a given analog. Examination of these factors should yield information about the local environment of the benzoic acid binding site. In general this site appears to have two regions, o n e relatively polar and one nonpolar. T h e unsaturated hydrophobic ring structure common to all effective analogs must associate with a rather loosely defined nonpolar region since this region can accommodate simple methyl groups as ring substituents as well as complete additional 6-carbon rings as in 9-AC without significantly affecting the ability of the compound to inhibit Gc~. Attachment of a polar residue to the ring at sites removed from the carboxyl residue, on the other hand, drastically reduces activity although steric considerations should be minimal. This is particularly true of substitutions at the para position. Steric effects alone appear to be most significant in the positions ortho to the carboxyl residue, suggesting that conformational requirements for binding in the vicinity of this residue are more restrictive than elsewhere on the ring. T h e data presented in this paper allow us to differentiate between several potential mechanisms by which these carboxylic acids might have produced the observed inhibition of Gc~. Initially, the amphipathic nature of the derivatives suggested that they might intercalate their hydrophobic ring structure into the membrane lipid phase while the polar carboxyl residues remained at the membrane-water interface. T h e net effect of such partitioning wo~Ad be to increase membrane fixed negative surface charge density; such an increase would be expected to retard transmembrane anion movement. Failure to find any effect of these analogs within their physiologically active concentration range on the electrophoretic mobilities of several test systems, as well as the lack o f competition by Ca 2+, would seem to preclude the surface charge mechanism in this case. A second hypothesis which had been proposed earlier for several of these compounds suggested that they interact in the membrane in such a way that their charged carboxyl residue sterically blocks access to the anion channel. This would produce inhibition of Gel in a manner similar to that observed in the sodium system with tetrodotoxin. A testable consequence of this hypothesis is that conductance to all ions passing through this channel ought to be equally affected. Evidence to date suggests that in normal muscle all the halide ions pass through the membrane predominantly along the same permeation pathway. In light of the trace amounts of I- outside the thyroid and of Br- normally present in the rat, it would seem teleologically unlikely that separate pathways of relatively high capacity would have developed for each anion in rat sarcolemma, and several points of experimental evidence support the concept that indeed a single channel serves for all these anions. We have shown that changes in pH affect the movement of all three halides in the same reversible manner and that the decrease in conductance observed with decreasing pH shows the same apparent inflection point in each case (37). Further, significant interaction is observed between CI- and I- and to a lesser extent between CI- and Br- when
PALADEAND BARCHI Aromatic Carboxylic Acids and Chloride Conductance
893
conductance m e a s u r e m e n t s are m a d e with varying mole fractions o f these ions. Low concentrations o f I - markedly inhibit m o v e m e n t o f C1- t h r o u g h the membrane. T h u s at least a portion o f these ions must move t h r o u g h the same pathway. Although absolute p r o o f is lacking, it seems unlikely that multiple noninteracting channels specific for each anion are present in the m e m b r a n e . I f the premise o f a single p r e d o m i n a n t anion channel is accepted, t h e n the observation that 5 × 10 -s M 9-AC reduces Gcl by 80% while GBr remains nearly constant and G1 a p p a r e n t l y increases cannot be reconciled with a mechanism which implies simple steric blocking o f the channel by the analog. Most o f the evidence p r e s e n t e d h e r e would suggest that the benzoic acid derivatives exert their effect by interacting with the anion channel in such a way as to alter its ion selectivity sequence. Similar changes in ion selectivity have been r e p o r t e d as a function o f p r o t o n concentration in the gall bladder epithelium (41) and were attributed by those authors to an alteration in the field strength o f a polar site or sites controlling selectivity in a m a n n e r analogous to that described by Eisenman in his classical studies with glass electrodes (16). Levitan and B a r k e r have studied the effects o f a series o f salicylates and o t h e r benzoic acid derivatives on the ion permeability characteristics o f molluscan n e u r o n s . T h e y find that these c o m p o u n d s alter the cation selectivity series in this p r e p a r a t i o n in a m a n n e r similar to that observed for anion permeability and conductance sequences in the present r e p o r t (29). Simultaneous effects were also seen in anion permeability. T h e y r e p o r t a very high correlation between effective aromatic acid concentration and octanol-water partition coefficient and conclude that in their system the effects on ion permeability are due to a nonspecific increase in m e m b r a n e anion field strength p r o d u c e d by partitioning o f these c o m p o u n d s into the m e m b r a n e . T h e y conclude that in their system a change in surface charge alone could not p r o d u c e the observed effects (28, 7). We also present evidence that m e m b r a n e surface charge is not significantly altered by concentrations o f o u r analogs which affect Gcl. H o w e v e r , while partitioning into the m e m b r a n e appears to be o f i m p o r t a n c e in the effects which we r e p o r t , there is evidence suggesting f u r t h e r specific interaction with intram e m b r a n e sites. T h e observed linearity o f the Hill plot as well as the nature o f the alterations in anion selectivity sequence suggest that in the rat d i a p h r a g m direct interaction with the anion conductance channel is probably involved in the inhibitory effect o f these aromatic carboxylic acids. Such interaction might well act to alter the field strength o f a channel region involved in p r o d u c i n g ion selectivity and result in alterations in ion selectivity sequence o f a type predicted by Eisenman on the basis o f his field strength theory (17). T h e progressive alteration in permeability sequence d e m o n s t r a t e d with increasing concentrations o f 9-AC may be i n t e r p r e t e d in several ways. I f one assumes the interaction o f a single molecule o f 9-AC with a single channel to convert the selectivity of that channel from Cl-> B r - > I- to I - > B r - > CI- then the observed intermediate permeability sequences f o u n d in Fig. 7 may r e p r e s e n t the average permeability o f a m e m b r a n e containing various ratios o f the two channel types. T h e sequence obtained at any one concentration o f 9-AC would reflect the percent o f anion channels whose sites are occupied by the analog and would
894
THE JOURNAL OF GENERAL PHYSIOLOGY " VOLUME 69 " 1977
relate directly to the affinity o f this site for the analog. A l t h o u g h the intermediate sequences r e p o r t e d in Fig. 7 are not a m o n g those predicted by Eisenman on the basis o f his field strength theory o f ion selectivity (16), they are precisely those which would be anticipated f r o m a m e m b r a n e having varying ratios of the two channel types described. Alternatively, the permeability sequence o f a single channel might u n d e r g o a progressive c h a n g e due to the b i n d i n g o f a n u m b e r o f molecules o f the benzoic acid derivative. T h e m e a s u r e d m e m b r a n e permeability sequence would then represent a more complex ensemble average of channels having multiple intermediate sequences. T h e a g r e e m e n t between the shape of the saturation curves for the analogs studied and that expected f r o m b i n d i n g to a single active site would tend to s u p p o r t the first hypothesis but c a n n o t be considered conclusive at this point. T h e major question raised by these data is how the interaction o f an amphipathic molecule such as 9-AC changes the selectivity of a single channel. Mechanisms based on alterations in field strength o f an ion-exchange site within the channel or on alterations in critical channel g e o m e t r y can be postulated, but there clearly is at present insufficient information on the basic selectivity mechanism in the anion channel to make such hypotheses more than speculation. T h i s w o r k w a s s u p p o r t e d in p a r t by N a t i o n a l I n s t i t u t e s o f H e a l t h g r a n t N S 08075 a n d b y a g r a n t from the Muscular Dystrophy Society.
Received for publication 11 January 1977. REFERENCES
1. ADRIAN, R. H. 1961. Internal chloride concentration and chloride efflux of frog muscle. J. Physiol. (Lond.). 156:632. 2. ADRIAN, R. H., and S. H. BRYANT. 1974. On the repetitive discharges in myotonic muscle fibers. J. Physiol. (Lond.). 240:505-515. 3. ADRIAN, R. H., and M. W. MARSHALL. 1976. Action potentials reconstructed in normal and myotonic muscle fibers. J. Physiol. (Lond.). 258:125-143. 4. BANGHAM, A. D., R. FLEMENS, A. D. HEARD, and G. V. F. SEAMAN. 1958. An apparatus for the microelectrophoresis of small particles. Nature (Lond.). 182:642644. 5. BANGHAM, A. D., M. W. HILL, and N. G. MILLER. 1974. Preparation and use of liposomes as models of biological membranes. In Methods in Membrane Biology. E. D. Korn, editor. Plenum Publishing Corp., New York. 6. BARCHI,R. L. 1975. Myotonia: an evaluation of the chloride hypothesis. Arch. Neurol. 32:175-180. 7. BARKER,J. L., and H. LEVITAN. 1975. Mitochondriai uncoupling agents. Effects on membrane permeability of molluscan neurons. J. Membr. Biol. 25:361-380. 8. BRETAG, A. H. 1973. Mathematical modelling of the myotonic action potential. In New Developments in Electromyography and Clinical Neurophysiology. Vol. 1. J. E. Desmedt, editor. S. Karger, Basel. 464-482 9. BROWN,W. E., and A. V. HILL. 1922. The oxygen dissociation curve of blood and its thermodynamic basis. Proc. R. Soc. Lond. Ser. B Biol. Sci. B94:297-333. 10. BRYANT, S. H. 1969. Cable properties of external intercostal muscle fibers from myotonic and non-myotonic goats. J. Physiol. (Lond.). 204:539-550.
PALADE AND BARCHI
AromaticCarboxylic Acids and Chloride Conductance
895
11. BRYANT,S. H. 1973. T h e electrophysiology of myotonia with a review o f congenital myotonia o f goats. In New Developments in Electromyography and Clinical Neurophysiology. Vol. 1. J. E. Desmedt, editor. S. Karger, Basel. 420-450. 12. BRYANT, S., and A. MORALES-AGUILERA.1971. Chloride conductance in normal and myotonic muscle fibers and the action of monocarboxylic aromatic acids.J. Physiol. (Lond.). 219:367-383. 13. D'ARRIGO, J. S. 1973. Possible screening o f surface charges on crayfish axons by polyvalent metal ions. J. Physiol. (Lond.). 231:117-128. 14. D'ARRIGO, J. S. 1974. Axonal surface charges: binding or screening by divalent cations governed by external p H . J. Physiol. (Lond.). 243:757-764. 15. EDSON, E. F., and D. M. SANDZRSON. 1964. Toxicities and myotonic activities o f certain polychlorobenzoic acids. Biochem. Pharmacol. 15:1538-1539. 16. EISENMAN, G. 1962. Cation selective glass electrodes and their mode of operation. Biophys. J. 2:259-323. 17. EISENMAN, G. 1965. Proceedings o f the 23rd International Congress o f Physiological Sciences, Tokyo. 489-506. Exerpta Medica Foundation, A m s t e r d a m . 18. FRANKENHAEUSER,J. A., and A. L. HODGKIN. 1957. T h e action of calcium on the electrical properties o f squid axons. J. Physiol. (Lond.). 137:218-244. 19. FUJITA,T., J. IWASA,and C. HANSCH. 1964. A new substituent constant, *r, derived from partition coefficients.J. Am. Chem. Soc. 86:5175-5180. 20. FuKuI, K., G. NAGATA, and T. YONEZAWA. 1958. Electronic structure and auxin activity of benzoic acid derivatives. J. Am. Chem. Soe. 80:2267-2270. 21. Hammett, L. P. 1970. Physical Organic Chemistry. 2nd edition. McGraw-Hill Book Company, New York. 22. HANSCH,C., and T. FUJITA. 1964. P-or-zr analysis. A method for the correlation of biological activity and chemical structure. J. Am. Chem. Soc. 86:1616-1626. 23. HANSCH,C., R. M. MUIR, T. FUJITA, P. D. MALONEY,F. GEIGER,and M. STREICH. 1963. T h e correlation of biological activity o f plant growth regulators and chloromycetin derivatives with H a m m e t t constants and partition coefficients. J. Am. Chem. Soc. 85:2817-2824. 24. HANSCH,C., and W. R. GLAVE. 1971. Structure-activity relationships in m e m b r a n e perturbing agents. Mol. Pharmacol. 7:337-354. 25. HUTTER, O. F., and D. NOBLE. 1960. The chloride conductance o f frog skeletal muscleJ. Physiol. (Lond.). 151:89-102. 26. JAFFE, H. H. 1953. A reexamination of the Hammett equation. Chem. Rev. 53:191261. 27. LANDAU, W. M. 1952. T h e essential mechanism in myotonia. An electromyographic study. Neurology. 2:369-388. 28. LrVITAN, H., and J. L. BARRrR. 1972a. Salicylate: a structure-activity study o f its effects on m e m b r a n e permeability. Science (Wash. D. C.). 176:1423-1425. 29. LEVtTAN, H., and J. L. BAR~ER. 1972b. Membrane permeability: cation selectivity reversibly altered by salicylate. Science (Wash. D. C.). 178:63-64. 30. Ln, icKY, R. J., S. H. BRYANT, and J. H. SALMOn. 1971. Cable parameters, sodium, potassium, chloride and water content, and potassium efflux in isolated intercostal muscles o f normal volunteers and patients with myotonia congenita.J. Clin. Invest. 50:2091-2103. 31. MAHLER, H. R., and E. H. CORDES. 1971. Biological Chemistry. 2nd edition. H a r p e r & Row, Publishers, Inc., New York.
896
THE JOURNAL OF GENERAL PHYSIOLOGY" VOLUME 6 9 .
1977
32. MCDANIEL, D., and H. BROWN. 1958. An extended table of Hammet substituent constants based on the ionization of substituted benzoic acids.J. Org. Chem. 23:420427. 33. MCLAUGHLIN,S. G. A., G. SZABO, and G. EISENMAN. 1971. Divalent cations and the surface potential of charged phospholipid m e m b r a n e s . J . Gen. Physiol. 58:667-687. 34. MOFFETT, R. B., and A. H. TANG. 1968. Skeletal muscle stimulants. Substituted benzoic acids. J. Med. Pharm. Chem. 11:1020-1022. 35. MORGAN, K. G., R. K. ENTRIKIN, and S. H. BRYANT. 1975. Myotonia and block of chloride conductance by iodide in avian muscle. Am. J. Physiol. 229:1155-1158. 36. NIE, N., C. HULL, J. JENKINS, K. STEINBRENNER, and O. BENT. 1975. Statistical package for the social sciences. McGraw-Hill Book Company, New York. 37. PALADE, P. T., and R. L. BARCHI. 1977. Characteristics of chloride conductance in muscle fibers of the rat diaphragm. J. Gen. Physiol. 69:325-342. 38. RUt)EL, R., and J. SENGES. 1972. Mammalian skeletal muscle: Reduced chloride conductance in drug-induced myotonia and induction of myotonia by low chloride solution. Naunyn-Schmiedeberg's Arch. Pharmakol. 274:337-347. 39. SCHAUF, C. L. 1975. The interactions of Ca ++ with Myxicola giant axons and a description in terms of a simple surface charge model.J. Physiol. (Lond.). 248:613624. 40. WINER, N., D. M. KLACHKO, R. D. BAER, P. L. LANGLEY, and T. W. BURNS. 1966. Myotonic response induced by inhibitors of cholesterol biosynthesis. Science (Wash. D. C.). 153:312-313. 41. WRIGHT, E. N., and J. M. DIAMOND. 1968. Effects of pH and polyvalent cations on the selective permeability of the gallbladder epithelium to monovalent ions. Biochim. Biophys. Acta. 165:57-74. 42. YERMISHKIN,L. N., Y. A. LmERMAN, and V. V. SMOLYANINOV. 1968. Permeability to ions of a m e m b r a n e with narrow pores. Biofizika. 13:205-207.
and R. L. B A R C H I
From the Departments of Neurology and of Biochemistry and Biophysics, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19174. Dr. Palade's present address is the Department of Physiology and Biophysics, University of Washington School of Medicine, Seattle, Washington 98195.
A B S T R A C T 25 aromatic carboxylic acids which are analogs o f benzoic acid were tested in the rat d i a p h r a g m preparation for effects on chloride conductance (Gc~). O f the 25, 19 were shown to reduce m e m b r a n e Gca with little effect on other m e m b r a n e parameters, although their a p p a r e n t Kl varied widely. This inhibition was reversible if exposure times were not prolonged. T h e most effective analog studied was anthracene-9-COOH (9-AC; Kt = 1.1 x 10-s M). Active analogs p r o d u c e d concentration-dependent inhibition o f a type consistent with interaction at a single site or g r o u p o f sites having similar binding affinities, although a correlation could also be shown between lipophilicity and Kl. Structure-activity analysis indicated that hydrophobic ring substitution usually increased inhibitory activity while para polar substitutions reduced effectiveness. These compounds do not a p p e a r to inhibit Gcl by altering m e m b r a n e surface charge and the inhibition p r o d u c e d is not voltage d e p e n d e n t . Qualitative characteristics o f the I-V relationship for CI- current are not altered. Conductance to all anions is not uniformly altered by these acids as would be expected from steric occlusion o f a common chafinel. Concentrations o f 9-AC reducing Gci by >90% resulted in slight augmentation of GI. T h e complete conductance sequence obtained at high levels o f 9-AC was the reverse o f that obtained u n d e r control conditions. Permeability sequences underwent progressive changes with increasing 9-AC concentration and ultimately inverted at high levels o f the analog. Aromatic carboxylic acids a p p e a r to inhibit Gc~ by binding to a specific i n t r a m e m b r a n e site and altering the selectivity sequence o f the m e m b r a n e anion channel. INTRODUCTION I n s k e l e t a l m u s c l e o f m a n y s p e c i e s t h e r e s t i n g m e m b r a n e i o n c o n d u c t a n c e is p r e d o m i n a n t l y t h a t a t t r i b u t a b l e to c h l o r i d e (25, 37, 1). A l t h o u g h n o t d i r e c t l y i n v o l v e d in t h e g e n e s i s o f a n a c t i o n p o t e n t i a l , c h l o r i d e c o n d u c t a n c e p l a y s a n i m p o r t a n t r o l e in t h e m a i n t e n a n c e o f m e m b r a n e p o t e n t i a l stability. R e c e n t s t u d i e s in p a t h o l o g i c a l l y u n s t a b l e m u s c l e m e m b r a n e s h a v e s u p p o r t e d this c o n c e p t (2, 38). M y o t o n i a , a n e x a m p l e o f s u c h a n u n s t a b l e p a t h o l o g i c s t a t e , is a p h e n o m e n o n c h a r a c t e r i z e d b y r e p e t i t i v e a c t i o n p o t e n t i a l g e n e r a t i o n in t h e s a r c o l e m m a in THE JOURNAL OF GENERALPHYSIOLOGY° VOLUME 69, 1977 " pages 879-896
879
880
T H E JOURNAL OF GENERAL PHYSIOLOGY • VOLUME 6 9 • 1 9 7 7
response to a b r i e f stimulus (27). This p h e n o m e n o n has been described in several naturally o c c u r r i n g diseases in m a n a n d animals (11,27). In those that have been studied e x p e r i m e n t a l l y the repetitive activity is associated with an absolute reduction in m e m b r a n e c o n d u c t a n c e to chloride (10, 30). Myotonia can be p r o d u c e d in otherwise n o r m a l animals by various chemical agents, including inhibitors o f cholesterol biosynthesis (40), certain herbicides (38), and in some species iodide (35). Many agents which reversibly p r o d u c e a myotonic s y n d r o m e in n o r m a l animals are m e m b e r s o f a class o f substituted benzoic acids (15). In several cases these agents have b e e n e x p e r i m e n t a l l y shown to reduce muscle m e m b r a n e Ge~ u n d e r in vitro conditions (12). We have previously r e p o r t e d the general characteristics of chloride conductance in the rat d i a p h r a g m (37), a n d evidence has b e e n p r e s e n t e d f r o m this laboratory a n d others that a r e d u c t i o n in m e m b r a n e Gel in muscle is sufficient g r o u n d s for repetitive electrical activity (6, 3, 8). In o r d e r to e x t e n d o u r u n d e r standing o f chloride c o n d u c t a n c e m e c h a n i s m s in n o r m a l a n d in pathologic states, we have studied in the rat d i a p h r a g m the m e c h a n i s m o f action o f a series o f aromatic acids similar to those k n o w n to p r o d u c e m y o t o n i a in o t h e r species. We have d o c u m e n t e d that these c o m p o u n d s do i n d e e d reversibly r e d u c e m e m b r a n e Gel a n d that direct correlations can be d e m o n s t r a t e d between the physical p r o p e r t i e s o f these analogs a n d their effectiveness in m o d i f y i n g chloride conductance. Earlier r e p o r t s have postulated that the o b s e r v e d r e d u c t i o n in m e m b r a n e Ge~ p r o d u c e d by the benzoic acid derivatives is d u e to steric blocking o f the anion p e r m e a t i o n pathway (12). We show here, however, that this inhibition most p r o b a b l y involves a direct interaction with the anion p e r m e a t i o n site in a m a n n e r affecting its ion selectivity r a t h e r t h a n by steric blocking o f these channels or by altering m e m b r a n e surface charge. MATERIALS
AND
METHODS
The majority of the 25 aromatic carboxylic acids studied were obtained from Aldrich Chemical Co., Milwaukee, Wis., from K & K Laboratories, Plainview, N. Y., or from Eastman Kodak Co., Rochester, N. Y. Several were the generous gift of Dr. R. B. Moffett of the Upjohn Pharmaceutical Co., Kalamazoo, Mich. Neuraminidase and phosphatidyl choline were obtained from Sigma Chemical Co., St. Louis, Mo. Male Wistar rats of 200-300 g were used for all experiments. Ringer's solution with or without various concentrations of aromatic carboxylic acids was continuously perfused over sections of diaphragm which were obtained and handled as previously described (37). Temperature control, recording methods, and solution compositions were all as described in an earlier paper ($7). For routine determination of half-maximal blocking concentration (K0 several determinations of the percent of inhibition of Gc~ were made at three or more concentrations of each aromatic acid. For each determination the percent inhibition of Gel was calculated after three-point cable analysis of five or more fibers in each of the following solutions: (a) normal Ringer's; (b) normal Ringer's + carboxylic acid; and (c) chloride-free Ringer's + carboxylic acid. Control values for Gel were obtained by subtraction of an average GE = 0.34 mmho/cm 2 from the average Gm obtained in the control solution. Subtraction of the average Gm in solution (c) from that in solution (b) yielded a value for Gel in the presence of the carboxylic acid, allowing a percent inhibition to be calculated. Double reciprocal
PALADEAND BARCHI Aromatic Carboxylic Acids and Chloride Conductance
881
plots of inhibition vs. analog concentration were then constructed and an observed Kl determined graphically. Partition coefficients for the benzoic acids were calculated by summation of values for the benzoic acid nucleus and factors specific for the effect on partitioning of a substitution at a given location on the parent compound based on the experimental results and theoretical methods of Hansch and co-workers (24, 23, 19). Hammett ¢r values and values for steric hindrance were also obtained from the chemical literature (26, 32). (See Results for a discussion of these parameters.) Statistical analysis was performed by standard computer methods (36). Electrophoretic mobilities of erythrocytes and liposomes were determined in a cylindrical microelectrophoresis cell 10 cm long with large volume electrode vessels at either end. The design was that of Bangham et al. (4). Constant temperature was maintained by immersion in a regulated water bath. Measurements were made with frequent reversals of field direction. Erythrocytes were prepared from human blood by dilution with 0.145 M NaCI and repeated centrifugation. Liposomes were prepared after the method of Bangham et al. (5). Phosphatidyl choline was dissolved in hexane-chloroform 1:10 and deposited on the sides of a round-bottom flask by flash evaporation. Liposomes were formed by the addition of 100 p~l of 0.145 M NaCI (pH 7.35) to the dried residue and sonication until vesicular formation was optimized. RESULTS
A n t h r a c e n e - 9 - C O O H (9-AC) may be considered representative o f the aromatic carboxylic acids studied in its effects on the rat d i a p h r a g m . T h e addition o f 10-5 M 9-AC to the Ringer's solution s u r r o u n d i n g a section o f d i a p h r a g m being maintained in vitro results in a rapid increase in the m e m b r a n e resistance. This increase begins almost immediately and reaches a new steady value within 10 min. W h e n n o r m a l Ringer's solution is r e p e r f u s e d over the d i a p h r a g m , memb r a n e resistance gradually r e t u r n s to near control levels over a 30-60-min period. W h e n similar e x p e r i m e n t s are p e r f o r m e d in methylsulfate-substituted chloride-free Ringer's, m e m b r a n e resistance is u n a f f e c t e d by this aromatic acid at the same concentration, suggesting that the c o m p o u n d acts specifically to r e d u c e the m e m b r a n e c o n d u c t a n c e to chloride (Gel-). W h e n a complete series o f steady-state c o n d u c t a n c e m e a s u r e m e n t s was m a d e on d i a p h r a g m fibers equilibrated in various concentrations (1-100/~M) o f 9-AC in n o r m a l Ringer's and C1--free Ringer's, a sigmoid relationship could be d e m o n s t r a t e d between the m e a s u r e d m e m b r a n e Gem a n d the logarithm o f the concentration o f 9-AC (Fig. 1). W h e n the average Gem at each 9-AC concentration is calculated a n d replotted a c c o r d i n g to the Hill equation, a diagnostic plot for binding characteristics, a linear relationship is obtained (Fig. 2). (9, see r e f e r e n c e 31 for recent discussion). T h e Hill coefficient calculated f r o m these data is not significantly different f r o m one, suggesting that the inhibitory action o f the a n t h r a c e n e derivative is mediated via the binding at a single class o f noninteracting sites. A dissociation constant o f 1.1 × 10-5 M can be calculated for the a p p a r e n t ligand-site interaction. This constant may be m o r e conveniently considered to be an inhibition constant for 9-AC with respect to m e a s u r e d Gel and for this and subsequent analogs will be r e f e r r e d to as Ki. As would be anticipated f r o m the n a t u r e o f the Hill plot for the above data, the relationship between the percent inhibition o f Gem and the concentration o f 9-
882
THE
JOURNAL
OF
GENERAL
PHYSIOLOGY"
VOLUME
69'
1977
AC is hyperbolic in nature, and a double reciprocal plot o f data in this f o r m is linear. A value for Kl identical to that obtained f r o m the Hill plot can be d e t e r m i n e d m o r e conveniently f r o m this g r a p h , a n d this m e t h o d was used subsequently to evaluate the a p p a r e n t inhibition constants with respect to Gc~ o f 24 o t h e r related derivatives o f the benzoic acid nucleus. For each analog determinations o f percent inhibition were m a d e at at least three analog concentrations (see Materials and Methods) a n d values o f Kl d e t e r m i n e d graphically• Significant reduction o f m e m b r a n e Gcl was p r o d u c e d by 19 o f these compounds. In all cases in which inhibition o f chloride c o n d u c t a n c e was observed, a
•
II
2.0
-O
E E i o
O
? 1.0
•
10-6
.
,
.
,,,,
.
.
.
.
10-5 ANTHRACENE-9-COOH (M)
,
,,,.
,i
10-4
10-6
10-5
10-4
ANTHRACENE-9-COOH (M)
FZGURE 1. Membrane chloride conductance (Gcl) observed after incubation of rat diaphragm preparation in various concentrations of 9-AC for 30 min at 35°C. Each point at a given 9-AC concentration represents the average of fibers measured in a different diaphragm. Measurements were made in C1--containing and C1--free Ringer's solution for conductance calculations (see Materials and Methods). FIGURE 2. The average values of GCl at each concentration of 9-AC from Fig. 1 are plotted according to the Hill equation. The average Gcl of all preparations before addition of 9-AC was taken as Gcl~max). The Hill coefficient of the regression line to this data is not significantly different from one. The dissociation constant for the apparent 9-AC site complex calculated from this data is 1.1 x 10-5 M. hyperbolic relationship between analog concentration and percent inhibition could be d e m o n s t r a t e d and the plots o f (1/percent inhibition) vs. (1/[aromatic acid]) were linear. Inhibition constants d e t e r m i n e d f r o m the double-reciprocal plots are listed in Table I. It can be seen that the concentrations required to p r o d u c e 50% inhibition o f Gcl varied a m o n g these c o m p o u n d s over a r a n g e o f three o r d e r s o f m a g n i t u d e . Resting m e m b r a n e potential (RMP) c h a n g e d little after exposure to 9-AC. T h u s the average RMP o f all fibers before e x p o s u r e to 9-AC was 70.3 -+ 5.2 (SD) mV while fibers exposed to 2 × 10-s M, 1 × 10-5 M, and 5 × 10-s M 9-AC had average RMP o f 68.3 + 3.8, 67.8 -+ 4.9, and 67.7 -+ 4.8 mV, respectively. T h e small d e c r e m e n t observed may represent the i n h e r e n t " r u n d o w n " o f the dia-
PALADE AND BARCHI
Aromatic Carboxylic Acids and Chloride Conductance
883
p h r a g m f i b e r s a t 35°C s i n c e all m e a s u r e m e n t s on control fibers were carried out w i t h i n t h e f i r s t 30 m i n o f a n e x p e r i m e n t while determinations in 9-AC were m a d e f r o m 30 m i n t o 2 h a f t e r d i s s e c t i o n o f t h e d i a p h r a g m . Overall, no correlation was noted between RMP and concentration of 9-AC in the perfusion m e d i u m o v e r t h e r a n g e i n d i c a t e d i n F i g . 1. For each compound,
change in residual conductance TABLE
INHIBITION
Antbracene-9-COOH 2,4,5-Trimethyl benzoic acid Pentachloro BA 3,~-Dichloro-2-amino BA 2,3,5,6-Tetrachloro BA 3,5-Dichloro BA 3,4-Dichloro BA 2,5-Dichloro BA 2,5-Dimethyl BA 2,3,5,6-Tetramethyl BA 2,3,6-Trichloro BA 3,5-Dimethyl BA 2,4-Dimethyl BA 3,6-Dichloro-2-nitro BA 3,4-Dimethyl BA 2,4-Dichloro phenoxyacetic acid 2,6-Dichloro BA 3,5-Dimethyl-4-nitro BA 2,4,6-Trimethyl BA Benzoic acid (BA) 4,5-Dichioro-2-COOH BA 3,5-Dimethyl-4-amino BA 2,5-Dimethyl benzene SOsH 2,3,5,6-Tetramethyl-4-COOH-BA 3,5-Dimethyl-4-hydroxy BA
solution
I
O F Gct I N R A T D I A P H R A G M
Compound
in chloride-free
BY A R O M A T I C
ACIDS
Kt(M)
Relative potency
Calc. log P
GK
1.1 x 10-s 6.6x 10-s 8,0x 10-5 1.3x 10-4 2.1 x 10-4 2.5x 10-4 2.6x 10-4 3.0x 10-4 3.1 x 10-4 6.0x 10-4 9.5 x 10-4 1 .Sx 10-s 1.6x 10-a 1.7x 10-3 2.0x 10-s 2.0 x 10-a 2.0x 10-s 2.8 x 10-a 4.0x 10-8 > l . 0 x 10-2 1.6x 10-2 >3.0x 10-~ >3.0x 10-2 > l . 0 x l 0 -1 > l . 0 x 10-1
909 152 125 76 47 40 38 33.3 32 16.7 10.5 6.7 6.2 6.0 5.0 5.0 5.0 3.6 2.5 B r - > I - . After exposure to 5 x 10-5 M 9-AC, the potential shift on c h a n g i n g to I - - c o n t a i n i n g solution reversed sign, indicating that u n d e r these conditions the m e m b r a n e is now more permeable to I - than to e l - (Fig. 6). Increasing [9-AC] to 2.5 x 10-4 M resulted in both I - and B r - solutions p r o d u c i n g a h y p e r p o l a r i z i n g response, the overall permeability sequence now being inverted to I - > B r - > e l - in a m a n n e r analogous to that observed for the c o n d u c t a n c e sequence. Such sequence inversions could also be d e m o n s t r a t e d in K+-free m e d i u m with Rb + substitution, CONTROL
8 -78mV cv.~.--~=O--cv~--z-
I
cv
1-s-~--cr-.---{
ANTHRACENE-9-COOH
5 x 10"SM
z mV
-75 mV - - - ~
c,--.--I-z--I
o 0Q.
/ ~
cl-
.,.
I- e~--I--c, -..-1- z--I--c,---I
FIGURE 6. A, Continuous traces of membrane potential demonstrating depolarizing shifts when external el- is rapidly replaced by Br- or I-. In each case return to normal el- solution results in membrane repolarization. B, After exposure to 9-AC (5 x 10-5 M) for 30 min, similar recordings in the same diaphragm demonstrate a hyperpolarization when I- is rapidly substituted for el- and depolarization when Br- is substituted for el-. Figures represent ink trackings of original records. eliminating changes in K + permeability as a d e t e r m i n a n t factor in the observed responses. Similar results in the presence o f $203 (42) (5 mM) rule out contribution f r o m an I-s carrier system. 13 experiments were d o n e c o m p a r i n g control permeability sequences to sequences obtained after equilibration in Ringer's solution containing 9-AC in concentrations between 1 a n d 100 × 10 -5 M. T h e permeability sequences obtained are shown graphically in Fig. 7. A progressive and reproducible shift in sequence is seen with increasing concentrations o f the aromatic acid. T h e same sequence inversions were observed with 2,4,5-trimethyl benzoic acid. Analogs with lower potencies p r o d u c e d partial sequence inversions in similar experiments, but concentrations in solution proportionately h i g h e r than their Kl could not be obtained due to their limited solubility. Again, permeability sequences d e t e r m i n e d at r e d u c e d p H showed a pattern identical to that obtained at p H 7.
PALADEAND BARCHI
891
Aromatic Carboxylic Acids and Chloride Conductance
DISCUSSION
In rat d i a p h r a g m , halogenated and methylated derivatives o f benzoic acid increase m e m b r a n e resistance by specifically r e d u c i n g m e m b r a n e conductance to CI-. Such a reduction in Gcl has previously been suggested as the primary physiologic effect o f several o f these c o m p o u n d s (12) o f which 2,4-dichlorophenoxyacetic acid (2,4-D) is the best known example (38). This behavior is now shown to be characteristic o f this class o f c o m p o u n d s in general. In vivo studies have d e m o n s t r a t e d a transient myotonic s y n d r o m e in mice associated with a single injection o f analogs similar to those tested h e r e (34). We confirm that the effects o f these c o m p o u n d s on Gcl in vitro is indeed reversible if the concentraI-> BF->Cl-
~
~
2'~
~o
0 w u} >"
Cl-ffi T - > B r -
.J m hi
w n
N
CF>I->Br-
el > B r
->I
/
~
l
[Anthracene-9-COOH]
,b M
,80
x I0 -5
FmURE 7. Summary of average permeability sequences determined from experiments similar to that shown in Fig. 6 after exposure of diaphragm preparations to 9-AC at various concentrations for 30 min at 35°C and pH 7.5. tion o f the analog is kept below cytotoxic levels. By extrapolation f r o m o t h e r work; it would a p p e a r likely that these c o m p o u n d s induce a myotonic s y n d r o m e in vivo by virtue o f the reduction in GcI which they p r o d u c e (6, 38). T h e benzoic acid derivatives r e p o r t e d h e r e reversibly depress Cl- conductance at extracellular concentrations which can be directly related to their relative solubility in a n o n p o l a r e n v i r o n m e n t . This observation suggests that the comp o u n d s must partition into the lipid phase o f the muscle m e m b r a n e b e f o r e p r o d u c i n g their effect. T h e hyperbolic relationship observed between extracellular analog concentration and percent reduction in Gcl, however, suggests that the inhibition o f Gcl is not simply related to i n t r a m e m b r a n e analog concentration. T h e n a t u r e o f ' t h e observed relationship implies that binding at a specific site or class o f sites having similar affinities is a critical" step leading to inhibition. T h e d e p e n d e n c e on partition coefficient would suggest that the specific site o f action is located either within or across the surface m e m b r a n e .
892
T H E J O U R N A L OF G E N E R A L P H Y S I O L O G Y • V O L U M E
69
• 1977
When calculated inhibition constants are corrected for intramembrane concentration based on partition coefficients for each analog they are still found to vary considerably among the analogs tested. This suggests that structural features related to specific binding also play a role in determining the effectiveness of a given analog. Examination of these factors should yield information about the local environment of the benzoic acid binding site. In general this site appears to have two regions, o n e relatively polar and one nonpolar. T h e unsaturated hydrophobic ring structure common to all effective analogs must associate with a rather loosely defined nonpolar region since this region can accommodate simple methyl groups as ring substituents as well as complete additional 6-carbon rings as in 9-AC without significantly affecting the ability of the compound to inhibit Gc~. Attachment of a polar residue to the ring at sites removed from the carboxyl residue, on the other hand, drastically reduces activity although steric considerations should be minimal. This is particularly true of substitutions at the para position. Steric effects alone appear to be most significant in the positions ortho to the carboxyl residue, suggesting that conformational requirements for binding in the vicinity of this residue are more restrictive than elsewhere on the ring. T h e data presented in this paper allow us to differentiate between several potential mechanisms by which these carboxylic acids might have produced the observed inhibition of Gc~. Initially, the amphipathic nature of the derivatives suggested that they might intercalate their hydrophobic ring structure into the membrane lipid phase while the polar carboxyl residues remained at the membrane-water interface. T h e net effect of such partitioning wo~Ad be to increase membrane fixed negative surface charge density; such an increase would be expected to retard transmembrane anion movement. Failure to find any effect of these analogs within their physiologically active concentration range on the electrophoretic mobilities of several test systems, as well as the lack o f competition by Ca 2+, would seem to preclude the surface charge mechanism in this case. A second hypothesis which had been proposed earlier for several of these compounds suggested that they interact in the membrane in such a way that their charged carboxyl residue sterically blocks access to the anion channel. This would produce inhibition of Gel in a manner similar to that observed in the sodium system with tetrodotoxin. A testable consequence of this hypothesis is that conductance to all ions passing through this channel ought to be equally affected. Evidence to date suggests that in normal muscle all the halide ions pass through the membrane predominantly along the same permeation pathway. In light of the trace amounts of I- outside the thyroid and of Br- normally present in the rat, it would seem teleologically unlikely that separate pathways of relatively high capacity would have developed for each anion in rat sarcolemma, and several points of experimental evidence support the concept that indeed a single channel serves for all these anions. We have shown that changes in pH affect the movement of all three halides in the same reversible manner and that the decrease in conductance observed with decreasing pH shows the same apparent inflection point in each case (37). Further, significant interaction is observed between CI- and I- and to a lesser extent between CI- and Br- when
PALADEAND BARCHI Aromatic Carboxylic Acids and Chloride Conductance
893
conductance m e a s u r e m e n t s are m a d e with varying mole fractions o f these ions. Low concentrations o f I - markedly inhibit m o v e m e n t o f C1- t h r o u g h the membrane. T h u s at least a portion o f these ions must move t h r o u g h the same pathway. Although absolute p r o o f is lacking, it seems unlikely that multiple noninteracting channels specific for each anion are present in the m e m b r a n e . I f the premise o f a single p r e d o m i n a n t anion channel is accepted, t h e n the observation that 5 × 10 -s M 9-AC reduces Gcl by 80% while GBr remains nearly constant and G1 a p p a r e n t l y increases cannot be reconciled with a mechanism which implies simple steric blocking o f the channel by the analog. Most o f the evidence p r e s e n t e d h e r e would suggest that the benzoic acid derivatives exert their effect by interacting with the anion channel in such a way as to alter its ion selectivity sequence. Similar changes in ion selectivity have been r e p o r t e d as a function o f p r o t o n concentration in the gall bladder epithelium (41) and were attributed by those authors to an alteration in the field strength o f a polar site or sites controlling selectivity in a m a n n e r analogous to that described by Eisenman in his classical studies with glass electrodes (16). Levitan and B a r k e r have studied the effects o f a series o f salicylates and o t h e r benzoic acid derivatives on the ion permeability characteristics o f molluscan n e u r o n s . T h e y find that these c o m p o u n d s alter the cation selectivity series in this p r e p a r a t i o n in a m a n n e r similar to that observed for anion permeability and conductance sequences in the present r e p o r t (29). Simultaneous effects were also seen in anion permeability. T h e y r e p o r t a very high correlation between effective aromatic acid concentration and octanol-water partition coefficient and conclude that in their system the effects on ion permeability are due to a nonspecific increase in m e m b r a n e anion field strength p r o d u c e d by partitioning o f these c o m p o u n d s into the m e m b r a n e . T h e y conclude that in their system a change in surface charge alone could not p r o d u c e the observed effects (28, 7). We also present evidence that m e m b r a n e surface charge is not significantly altered by concentrations o f o u r analogs which affect Gcl. H o w e v e r , while partitioning into the m e m b r a n e appears to be o f i m p o r t a n c e in the effects which we r e p o r t , there is evidence suggesting f u r t h e r specific interaction with intram e m b r a n e sites. T h e observed linearity o f the Hill plot as well as the nature o f the alterations in anion selectivity sequence suggest that in the rat d i a p h r a g m direct interaction with the anion conductance channel is probably involved in the inhibitory effect o f these aromatic carboxylic acids. Such interaction might well act to alter the field strength o f a channel region involved in p r o d u c i n g ion selectivity and result in alterations in ion selectivity sequence o f a type predicted by Eisenman on the basis o f his field strength theory (17). T h e progressive alteration in permeability sequence d e m o n s t r a t e d with increasing concentrations o f 9-AC may be i n t e r p r e t e d in several ways. I f one assumes the interaction o f a single molecule o f 9-AC with a single channel to convert the selectivity of that channel from Cl-> B r - > I- to I - > B r - > CI- then the observed intermediate permeability sequences f o u n d in Fig. 7 may r e p r e s e n t the average permeability o f a m e m b r a n e containing various ratios o f the two channel types. T h e sequence obtained at any one concentration o f 9-AC would reflect the percent o f anion channels whose sites are occupied by the analog and would
894
THE JOURNAL OF GENERAL PHYSIOLOGY " VOLUME 69 " 1977
relate directly to the affinity o f this site for the analog. A l t h o u g h the intermediate sequences r e p o r t e d in Fig. 7 are not a m o n g those predicted by Eisenman on the basis o f his field strength theory o f ion selectivity (16), they are precisely those which would be anticipated f r o m a m e m b r a n e having varying ratios of the two channel types described. Alternatively, the permeability sequence o f a single channel might u n d e r g o a progressive c h a n g e due to the b i n d i n g o f a n u m b e r o f molecules o f the benzoic acid derivative. T h e m e a s u r e d m e m b r a n e permeability sequence would then represent a more complex ensemble average of channels having multiple intermediate sequences. T h e a g r e e m e n t between the shape of the saturation curves for the analogs studied and that expected f r o m b i n d i n g to a single active site would tend to s u p p o r t the first hypothesis but c a n n o t be considered conclusive at this point. T h e major question raised by these data is how the interaction o f an amphipathic molecule such as 9-AC changes the selectivity of a single channel. Mechanisms based on alterations in field strength o f an ion-exchange site within the channel or on alterations in critical channel g e o m e t r y can be postulated, but there clearly is at present insufficient information on the basic selectivity mechanism in the anion channel to make such hypotheses more than speculation. T h i s w o r k w a s s u p p o r t e d in p a r t by N a t i o n a l I n s t i t u t e s o f H e a l t h g r a n t N S 08075 a n d b y a g r a n t from the Muscular Dystrophy Society.
Received for publication 11 January 1977. REFERENCES
1. ADRIAN, R. H. 1961. Internal chloride concentration and chloride efflux of frog muscle. J. Physiol. (Lond.). 156:632. 2. ADRIAN, R. H., and S. H. BRYANT. 1974. On the repetitive discharges in myotonic muscle fibers. J. Physiol. (Lond.). 240:505-515. 3. ADRIAN, R. H., and M. W. MARSHALL. 1976. Action potentials reconstructed in normal and myotonic muscle fibers. J. Physiol. (Lond.). 258:125-143. 4. BANGHAM, A. D., R. FLEMENS, A. D. HEARD, and G. V. F. SEAMAN. 1958. An apparatus for the microelectrophoresis of small particles. Nature (Lond.). 182:642644. 5. BANGHAM, A. D., M. W. HILL, and N. G. MILLER. 1974. Preparation and use of liposomes as models of biological membranes. In Methods in Membrane Biology. E. D. Korn, editor. Plenum Publishing Corp., New York. 6. BARCHI,R. L. 1975. Myotonia: an evaluation of the chloride hypothesis. Arch. Neurol. 32:175-180. 7. BARKER,J. L., and H. LEVITAN. 1975. Mitochondriai uncoupling agents. Effects on membrane permeability of molluscan neurons. J. Membr. Biol. 25:361-380. 8. BRETAG, A. H. 1973. Mathematical modelling of the myotonic action potential. In New Developments in Electromyography and Clinical Neurophysiology. Vol. 1. J. E. Desmedt, editor. S. Karger, Basel. 464-482 9. BROWN,W. E., and A. V. HILL. 1922. The oxygen dissociation curve of blood and its thermodynamic basis. Proc. R. Soc. Lond. Ser. B Biol. Sci. B94:297-333. 10. BRYANT, S. H. 1969. Cable properties of external intercostal muscle fibers from myotonic and non-myotonic goats. J. Physiol. (Lond.). 204:539-550.
PALADE AND BARCHI
AromaticCarboxylic Acids and Chloride Conductance
895
11. BRYANT,S. H. 1973. T h e electrophysiology of myotonia with a review o f congenital myotonia o f goats. In New Developments in Electromyography and Clinical Neurophysiology. Vol. 1. J. E. Desmedt, editor. S. Karger, Basel. 420-450. 12. BRYANT, S., and A. MORALES-AGUILERA.1971. Chloride conductance in normal and myotonic muscle fibers and the action of monocarboxylic aromatic acids.J. Physiol. (Lond.). 219:367-383. 13. D'ARRIGO, J. S. 1973. Possible screening o f surface charges on crayfish axons by polyvalent metal ions. J. Physiol. (Lond.). 231:117-128. 14. D'ARRIGO, J. S. 1974. Axonal surface charges: binding or screening by divalent cations governed by external p H . J. Physiol. (Lond.). 243:757-764. 15. EDSON, E. F., and D. M. SANDZRSON. 1964. Toxicities and myotonic activities o f certain polychlorobenzoic acids. Biochem. Pharmacol. 15:1538-1539. 16. EISENMAN, G. 1962. Cation selective glass electrodes and their mode of operation. Biophys. J. 2:259-323. 17. EISENMAN, G. 1965. Proceedings o f the 23rd International Congress o f Physiological Sciences, Tokyo. 489-506. Exerpta Medica Foundation, A m s t e r d a m . 18. FRANKENHAEUSER,J. A., and A. L. HODGKIN. 1957. T h e action of calcium on the electrical properties o f squid axons. J. Physiol. (Lond.). 137:218-244. 19. FUJITA,T., J. IWASA,and C. HANSCH. 1964. A new substituent constant, *r, derived from partition coefficients.J. Am. Chem. Soc. 86:5175-5180. 20. FuKuI, K., G. NAGATA, and T. YONEZAWA. 1958. Electronic structure and auxin activity of benzoic acid derivatives. J. Am. Chem. Soe. 80:2267-2270. 21. Hammett, L. P. 1970. Physical Organic Chemistry. 2nd edition. McGraw-Hill Book Company, New York. 22. HANSCH,C., and T. FUJITA. 1964. P-or-zr analysis. A method for the correlation of biological activity and chemical structure. J. Am. Chem. Soc. 86:1616-1626. 23. HANSCH,C., R. M. MUIR, T. FUJITA, P. D. MALONEY,F. GEIGER,and M. STREICH. 1963. T h e correlation of biological activity o f plant growth regulators and chloromycetin derivatives with H a m m e t t constants and partition coefficients. J. Am. Chem. Soc. 85:2817-2824. 24. HANSCH,C., and W. R. GLAVE. 1971. Structure-activity relationships in m e m b r a n e perturbing agents. Mol. Pharmacol. 7:337-354. 25. HUTTER, O. F., and D. NOBLE. 1960. The chloride conductance o f frog skeletal muscleJ. Physiol. (Lond.). 151:89-102. 26. JAFFE, H. H. 1953. A reexamination of the Hammett equation. Chem. Rev. 53:191261. 27. LANDAU, W. M. 1952. T h e essential mechanism in myotonia. An electromyographic study. Neurology. 2:369-388. 28. LrVITAN, H., and J. L. BARRrR. 1972a. Salicylate: a structure-activity study o f its effects on m e m b r a n e permeability. Science (Wash. D. C.). 176:1423-1425. 29. LEVtTAN, H., and J. L. BAR~ER. 1972b. Membrane permeability: cation selectivity reversibly altered by salicylate. Science (Wash. D. C.). 178:63-64. 30. Ln, icKY, R. J., S. H. BRYANT, and J. H. SALMOn. 1971. Cable parameters, sodium, potassium, chloride and water content, and potassium efflux in isolated intercostal muscles o f normal volunteers and patients with myotonia congenita.J. Clin. Invest. 50:2091-2103. 31. MAHLER, H. R., and E. H. CORDES. 1971. Biological Chemistry. 2nd edition. H a r p e r & Row, Publishers, Inc., New York.
896
THE JOURNAL OF GENERAL PHYSIOLOGY" VOLUME 6 9 .
1977
32. MCDANIEL, D., and H. BROWN. 1958. An extended table of Hammet substituent constants based on the ionization of substituted benzoic acids.J. Org. Chem. 23:420427. 33. MCLAUGHLIN,S. G. A., G. SZABO, and G. EISENMAN. 1971. Divalent cations and the surface potential of charged phospholipid m e m b r a n e s . J . Gen. Physiol. 58:667-687. 34. MOFFETT, R. B., and A. H. TANG. 1968. Skeletal muscle stimulants. Substituted benzoic acids. J. Med. Pharm. Chem. 11:1020-1022. 35. MORGAN, K. G., R. K. ENTRIKIN, and S. H. BRYANT. 1975. Myotonia and block of chloride conductance by iodide in avian muscle. Am. J. Physiol. 229:1155-1158. 36. NIE, N., C. HULL, J. JENKINS, K. STEINBRENNER, and O. BENT. 1975. Statistical package for the social sciences. McGraw-Hill Book Company, New York. 37. PALADE, P. T., and R. L. BARCHI. 1977. Characteristics of chloride conductance in muscle fibers of the rat diaphragm. J. Gen. Physiol. 69:325-342. 38. RUt)EL, R., and J. SENGES. 1972. Mammalian skeletal muscle: Reduced chloride conductance in drug-induced myotonia and induction of myotonia by low chloride solution. Naunyn-Schmiedeberg's Arch. Pharmakol. 274:337-347. 39. SCHAUF, C. L. 1975. The interactions of Ca ++ with Myxicola giant axons and a description in terms of a simple surface charge model.J. Physiol. (Lond.). 248:613624. 40. WINER, N., D. M. KLACHKO, R. D. BAER, P. L. LANGLEY, and T. W. BURNS. 1966. Myotonic response induced by inhibitors of cholesterol biosynthesis. Science (Wash. D. C.). 153:312-313. 41. WRIGHT, E. N., and J. M. DIAMOND. 1968. Effects of pH and polyvalent cations on the selective permeability of the gallbladder epithelium to monovalent ions. Biochim. Biophys. Acta. 165:57-74. 42. YERMISHKIN,L. N., Y. A. LmERMAN, and V. V. SMOLYANINOV. 1968. Permeability to ions of a m e m b r a n e with narrow pores. Biofizika. 13:205-207.
