Current Clamp Experiments - rupress.org

Calcium-Activated Conductance in Skate Electroreceptors

Current Clamp Experiments

W. T. CLUSIN and M. V. L. BENNETT

From the Division of Cellular Neurobiology, Department of Neuroscience, Albert Einstein College of Medicine, Bronx, New York 10461, and the Marine Biological Laboratory, Woods Hole, Massachusetts 02543. Dr. Clusin's present address is the Department of Medicine, Stanford University Hospital, Stanford, California 94305.

A B S T R A C T When current c lamped, skate electroreceptor epithelium produces large action potentials in response to stimuli that depolarize the lumenal faces of the receptor cells. With increasing stimulus strength these action potentials become prolonged. When the peak voltage exceeds about 140 mV the repolarizing phase is blocked until the end of the stimulus. Perfusion exper iments show that the rising phase of the action potential results f rom an increase in calcium permeabili ty in the lumenal membranes . Perfusion of the lumen with cobalt or with a zero calcium solution containing EGTA blocks the action potential. Perfusion of the lumen with a solution containing 10 mM Ca and 20 mM EGTA initially slows the repolar izing process at all voltages and lowers the potential at which it is blocked. With prolonged perfusion, repolarization is blocked at all voltages. When excitability is abolished by perfusion with cobalt, or with a zero calcium solution containing EGTA, no delayed rectification occurs. We suggest that repolarization dur ing the action potential depends on an influx of calcium into the cytoplasm, and that the rate of repolarization depends on the magni tude of the inward calcium current . Increasingly large stimuli reduce the rate of repolarization by reducing the driving force for calcium, and then block repolarization by causing the lumenal membrane potential to exceed Eca. Changes in extracellular calcium affect repolarization in a manner consistent with the resulting change in Eca.

I N T R O D U C T I O N

A m p u l l a e o f L o r e n z i n i a r e exqu i s i t e ly sens i t ive e l e c t r o r e c e p t o r s f o u n d b e n e a t h the skin o f e l a s m o b r a n c h f ishes . T h e s e r e c e p t o r s a r e c a p a b l e o f d e t e c t i n g gill a n d musc le po t en t i a l s o f p r e y , a n d a r e mos t n u m e r o u s a r o u n d the m o u t h . K a l m i j n (1971) s h o w e d tha t e lec t r ica l s ignals e m i t t e d by p r e y can elici t f e e d i n g b e h a v i o r . D i j k g r a a f a n d K a l m i j n (1963) d e m o n s t r a t e d b e h a v i o r a l r e s p o n s e s to p o t e n t i a l g r a d i e n t s as smal l as 0.01 /zV/cm 2. M u r r a y (1967) f o u n d tha t i n d i v i d u a l a m p u l l a e r e s p o n d to vo l t age s t imul i o f a few mic rovo l t s , b u t can a c c o m m o d a t e w i thou t loss o f sens i t iv i ty to m a i n t a i n e d vo l tages tha t a r e m u c h l a r g e r . Exc i sed a m p u l l a r y e l e c t r o r e c e p t o r s give t r a n s i e n t r e s p o n s e s to m e c h a n i c a l s t imul i (Low- ens t e in , 1960), c h a n g e s in sa l in i ty ( M u r r a y , 1960; L o w e n s t e i n a n d I s h i k o , 1962),

THE JOURNAL OF GENERAL PHYSIOLOGY " VOLUME 6 9 , 1977 • pages 121-143 121

Dow

nloaded from http://rupress.org/jgp/article-pdf/69/2/121/1246366/121.pdf by guest on 26 April 2022

122 T H E J O U R N A L O F G E N E R A L P H Y S I O L O G Y " V O L U M E 69 • 1 9 7 7

and changes in temperature (Murray, 1967; Sand, 1938). However, Kalmijn (1975) showed that in free-swimming animals, the responses of the ampullary nerves to thermal and mechanical stimuli are insignificant compared to the responses evoked by electrical stimuli.

The resistivity of the skin and body tissues of the skate are low, and a field in the environment is not significantly distorted by the presence of the fish (Mur- ray, 1967). On the other hand, the canals of the ampullae have a very long space constant (Waltman, 1966). Thus the receptor epithelium in the ampulla is acted upon by the voltage difference between the canal opening and body interior adjacent to the ampulla. The longest canals give the greatest sensitivity to uniform fields while a spectrum of canal lengths and orientations provides information about local variations.

An important property of these receptors is their electrical excitability, which. was first suggested by Murray's (1965) report that low voltage oscillations can be recorded when a wire is thrust into the pore of the ampullary canal of a freshly killed skate. Waltman (1966) subsequently reported that excitatory (lumen negative) voltage stimuli applied between the pore of the canal and the ampulla evoke graded oscillatory responses across the ampullary epithelium. Obara and Ben- nett (1968, 1972) evoked similar responses by passing constant current into the lumen of the canal with a microelectrode. They showed that these responses are regenerative. Moreover, 30-mV action potentials with a well-defined threshold sometimes occur. Waltman (1968) used electronic feedback to prevent current from flowing around the edges of the epithelium, so that the epithelium was current clamped. Under these conditions, long trains of "all-or-none" action potentials are generated. The action potentials reach 60 mV in amplitude and have a duration of 100 ms. Under voltage clamp conditions, when the epithelium is effectively short-circuited by the feedback amplifier, an N-shaped current- voltage relation is obtained (Waltman, 1968). From Waltman's experiments it appears that, as the conductance shunting the epithelium is reduced from the voltage clamped condition, one obtains graded regenerative responses, and finally, action potentials with a well-defined threshold.

By penetrating the afferent fibers, Obara and Bennett (1972) recorded postsynaptic potentials that followed the regenerative responses of the ampullary epithelium. They concluded that the ampullary potentials are generated by the receptor cells. They suggested that the response of the electroreceptors to physiological stimuli results from a depolarizing excitability of the lumenal faces of the receptor cells which depolarizes the passive presynaptic basal faces. Experiments with large stimuli supported this proposed mechanism. Obara and Bennett also discussed the possibility that excitability of the lumenal membranes of the receptor cells might be due to an active inward calcium current. The evidence was primarily comparative: that is, the electrically excitable responses of teleost phasic receptors are insensitive to TTX (Zipser and Bennett, 1973), while the receptor potential in ampullary electroreceptors of Plotosus varies as the Nernst potential for calcium (Akutsu and Obara, 1974).

In this and the following paper (Clusin and Bennett, 1977) the permeability changes responsible for action potentials in skate electroreceptors are analyzed. The unusual anatomic features of the electroreceptors have permitted detailed

Dow


CLUSIN AND BENNETr Calcium-Activated Conductance in Electroreceptors. I 123

analysis of m e m b r a n e proper t ies without impalement of the receptor cells, which are very small and invested with connective tissue. This paper deals with cur ren t clamp and the following paper with voltage clamp exper iments . We find that excitability o f the ampulla involves a permeabili ty process that has not been found in axons, but has been described in several nerve cell bodies (Meech and Standen, 1975). T h e rising phase of the ampul lary action potential results f rom an increase in calcium permeabili ty in the lumenal membranes o f the receptor cells. T h e conductance increase responsible for repolarizat ion is caused not by the change in voltage but by the calcium influx across the lumenal membranes . Large stimuli that prevent this calcium influx by exceeding the calcium equilib- r ium potential also prevent repolarizat ion.

Prel iminary communicat ions have appeared (Clusin and Bennet t , 1973; Clusin et al., 1974; Clusin et al., 1975).

Anatomical Considerations

Fig. 1 is a d iagram of a typical ampul lary e lect roreceptor . It consists of a lobulated bag, or ampulla , connected to an external orifice by a long canal. T h e canal wall is comprised of two layers o f f lat tened cells. T h e cells of the basal layer are separated f rom each o ther by a un i fo rm 100-~ extracellular gap. However , the cells abutt ing on the lumen are jo ined together by apical tight junct ions which completely occlude the extracellular space (Waltman, 1966), and presumably fo rm a barr ier to ionic movement . T h e canal wall is electrically l inear and has a very large resistivity (10 ~ l~/cm 2) in parallel with a capacity of about 0.5 pF/ cm 2. To account for the large mural resistivity, Waltman proposed that the inner and outer membranes of cells abutt ing on the lumen have an extremely high

FIGURE 1. Skate ampullary electroreceptor and canal (after Waltman, 1966). The ampulla consists of a cluster of alveoli, one of which is shown in cross section. The epithelium of each alveolus is a single cell layer in which receptor cells are inter- spersed among supporting cells. Ampullae used in these experiments were 1.2 mm in diameter. The basal surfaces of the receptor cells are innervated by five to seven nerve fibers which ramify profusely over the surface of the alveoli. The neck of the ampulla is called the marginal zone (MZ).

Dow


124 T H E J O U R N A L O F 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

resistance. He fu r t he r suggested that the mura l capacity is the capacity o f the two m e m b r a n e faces in series.

T h e epi thel ium of the ampul la is d i a g r a m m e d in Fig. 2A. I t consists of a single sheet o f cells that is cont inuous with the high resistance layer of the canal. As in the canal, adjacent cells are jo ined by occ ludingjunct ions that f o rm a collar a round each cell and part i t ion its m e m b r a n e into lumenal and basal faces. However , two types o f cells are p resen t in the ampul la . Spherical r ecep tor cells are in te rspersed a m o n g co lumnar suppor t ing cells. These recep tor cells have

A LUMEN

B

~ARGINA~ C__~ z o N e ~

FIGURE 2. Diagram of ampullary cell types. A, Details of two receptor cells in the ampullary epithelium. The apical tight junctions (zonulae occludentes, zo) partition the cell membranes into lumenal and basal faces. The basal membranes of the receptor cells form characteristic ribbon synapses with the afferent nerve fibers (Waltman, 1966). A number of synaptic ribbons are present in each receptor cell. Each receptor cell is about 13 ktm in diameter and bears an apical cilium (not shown). B, Details of interdigitating marginal zone cells. These are modified supporting cells whose basal membrane area greatly exceeds their lumenal surface area owing to the extensive interdigitation.

cytoplasmic features that are quite distinct f rom those of the suppor t ing cells and canal wall cells. Each recep tor cell has a small lumenal face and a much larger basal face . T h e lumenal face bears an apical cilium (not shown) and the basal face forms several synaptic contacts with branches of a f fe ren t fibers f rom the eighth cranial nerve (Wal tman, 1966). Af fe ren t fibers ramify profusely and five to seven of them innervate the several thousand receptor cells in each ampul la . T h e e lec t roreceptor synapses resemble o ther sensory synapses of the acoustical lateralis system in that they have presynapt ic r ibbons.

Both the ampul la and the canal are filled with a gelatinous substance whose conductivity is slightly grea ter than that o f seawater (Murray and Putts, 1961). This jelly contains fine prote in fibers that are anchored to the lumenal surface o f the epi thel ium at hemidesmosomes . T h e neck o f the ampul la is called the

Dow


CLUSIN AND BENNETT Calcium-Activated Conductance in Electroreceptors. I 125

m a r g i n a l zone by W a l t m a n (1966) a n d is a zone o f t r a n s i t i o n b e t w e e n the e p i t h e l i u m o f t h e a m p u l l a a n d t h e e p i t h e l i u m o f t h e cana l . T h e r e a r e n o r e c e p t o r cells in t he m a r g i n a l zone . I ts e p i t he l i a l cells a r e m o d i f i e d s u p p o r t i n g cells which i n t e r d i g i t a t e ex t ens ive ly be low the t igh t j u n c t i o n s (Fig. 2 B). As a r e su l t , t he basa l faces o f cells in t he m a r g i n a l zone have m a n y t imes m o r e a r e a t h a n the l u m e n a l faces , a n d m a y c o n t a i n m u c h o f the s u r f a c e m e m b r a n e in an a m p u l l a ( S z a m i e r , p e r s o n a l c o m m u n i c a t i o n ) .

M A T E R I A L S A N D M E T H O D S

The Preparation

In o rder to fit the recording appara tus and simplify numerical t reatment of data, electroreceptors of a uniform size were used in these experiments . Skates (Raja oscellata and R. erinacea) 45-55 cm in length were obtained by trawling near Woods Hole, Mass. The canals that run posteriorly from the mandibular capsule along the dorsal surface (Murray, 1966, Fig. 1 A) were exposed and transected within 4 cm of their terminal pores. Other canals and the afferent nerve were transected close to the capsule. The capsule, with attached poster ior canals, was then removed from the fish and placed in a t ransparent, ice-cooled dissecting dish containing either cerebrospinal fluid or Ringer's solution (composition below). Individual electroreceptors were detached under a dissecting mi- croscope, along with several millimeters of the ampul lary nerve branch. The average length of excised canal was 7 cm and the average diameter near the ampulla was 1 ram.

The recording appara tus in these exper iments is d iagrammed in Fig. 3. The canal was

/

AIR

i0 e

FIGURE 3. Experimental appara tus for passing constant current across the ampullary epi thel ium. Current was measured with a current-voltage t ransducer made from an operat ional amplif ier with output voltage propor t ional to I. The ampul lary potential was measured with a microelectrode thrust through the canal wall. Postsynaptic activity was recorded by drawing the ampul lary nerve into oil on chlorided silver hooks at tached to the input terminals of a differential amplifier .

Dow


126 T H E J O U R N A L O F G E N E R A L P H Y S I O L O G Y • V O L U M E 69 • 1 9 7 7

suspended on thin pins across a n a i r gap between two saline pools. The transected end of the canal was in one saline pool and the ampulla in another. The portion of canal in the air gap was washed with a solution of 526 mM sucrose and 350 mM urea to increase the resistance of the air gap.

Electrical Stimulation and Recording

The voltage across the ampulla was measured with a Bioelectric NF1 preamplifier and a microelectrode thrust through the canal wall at the neck of the ampulla. Conventional glass pipettes filled with 2 M KC1 and having a resistance of 1 MI'I or higher were sufficient to impale the canal without measurably reducing its resistance. The reference electrode, which also served as a current re turn , was a chlorided silver plate with a surface area of 1 cm 2 and a resistance of less than 100 gI. In some experiments, the silver plate was embedded in an agar-filled glass tube, the resistance of which was less than 500 ~. The reference electrode was held at "virtual ground" by a current-voltage transducer made from an Analog Devices 40J operational amplifier (Analog Devices, Inc., Norwood, Mass.).

Zero voltage was established by short-circuiting the two saline pools through a low resistance (2 kl~) salt bridge made with physiological saline and 2% agar. The potential across the ampulla is within 1 mV of zero when the distal end of the canal is short- circuited to the basal surface of the ampulla, as occurs under physiological conditions (Obara and Bennett, 1972).

Pulses of constant current were produced by using a voltage source in series with a 100 MI) resistor. Current was applied through a chlorided silver wire to the pool containing the open end of the canal. Because the portion of the canal in the air gap had been washed with an ion-free solution, current passed between the saline pools flowed down the lumen of the canal and across the ampullary epithelium. The resistance along the external surface of the sucrose-treated canal was estimated as 7.5 Ml'~/cm by determining the distance from the ampulla at which a cut produced about 10% reduction in input resistance. From Waltman's figure for wall resistivity one can calculate that the mural resistance of each centimeter of the canals used in these experiments was 15 MI). Since the resistance along the lumen was only 2.3 kli/cm, it is unlikely that significant current would have crossed the canal wall and flowed along its external surface.

Postsynaptic activity was recorded extracellularly by draping the nerve over a chlorided silver hook that was then retracted into an oil-filled glass capillary. Differential recordings were sometimes obtained with a Metametrix (DC-coupled) amplifier and a second silver wire placed near the opening of the capillary. Postsynaptic potentials (PSPs) as large as 1 mV were often recorded with superimposed action potentials that were even larger. Sometimes 10 -7 M TTX was added to the saline to abolish the postsynaptic action potentials, leaving a smooth PSP (Steinbach, 1974; Steinbach and Bennett, 1971).

Solutions

Stable recordings could be obtained for several hours if the ampulla was bathed in cerebrospinal fluid (CSF). For experiments involving ionic substitutions, a modified F/ihner's (1908) elasmobranch Ringer was nearly as satisfactory as CSF. This solution contained 415 mM urea, 340 mM NaCI, 6 mM KCI, 2.5 mM MgCI~, 1.8 mM CaCI2, 2.5 mM NaHCO3, and 5 mM HEPES buffer (pK a = 7.55) adjusted to pH 7.4. The temperature of solutions bathing the ampulla was regulated by using a Peltier device activated by a small thermistor. Experiments were performed at 10°C unless otherwise indicated.

Solutions used in perfusing the lumen of the ampulla were made from a control solution containing 428 mM NaC1, 13 mM KCI, 50 mM MgCI2, 10 mM CaC12, 75 mM urea, and 5 mM HEPES at pH 8.1. In a few preliminary experiments, a control solution was made by dissolving 75 mM urea in seawater.

Dow



The 100 mM CaCI2 solution was made by replacement of 150 mM NaCI in the control solution. The 100 mM CaCI~ solution was made by substitution of 90 mM CaCI~ for 135 mM NaCI.

The EGTA solutions were prepared by substitution of 20 mM EGTA and 55 mM HEPES for NaC1 in the control solution. The usual 10 mM CaCI2 was present in the Ca- EGTA solution, but was replaced by MgC12 in the Ca-free EGTA solution. Both solutions were neutralized with NaOH so that the final pH was 8.1 and the final osmolarity was equal to that of the control solution. The total of 60 mM HEPES in both solutions stabilized them against changes in pH caused by chelation of additional calcium. Addition of 10 mM CaCi to either solution caused a pH reduction of only 0.4.

Internal Perfusion of the Ampulla

Internal perfusion of the ampulla was difficult because the lumenal surface of the high resistance cells in the canal wall is not protected by connective tissue. Moreover, the canal is tortuous and must be "straightened to prevent damage to these high resistance cells dur ing cannulation.

Cannulat ion was accomplished by using the chamber shown in Fig. 4. This chamber had two saline pools separated by an air gap but, in addition, the portion of the canal nearest the ampulla was forced into a Vaseline-smeared groove in a Sylgard resin block. The Vaseline-filled groove served to straighten the canal and to immobilize it dur ing cannulation. In the initial experiments, it was found that the saline solution bathing the ampulla tended to seep into the groove along the Vaseline-coated surface of the canal,

A M P U L L A

kI R GAP

ZASELINE

~ILLED

_~ROOVE

,'ANAL

FIGURE 4. A diagram of the apparatus used for cannulation and internal perfusion of ampullary electroreceptors. The ampulla lay in one saline pool, while the canal was draped across an air gap which ran perpendicular to the Vaseline groove into a second saline pool (not shown). Perfusate was forced out of the cannula under pressure and flowed back along its outer surface to the nick in the canal wall where it was aspirated. The stimulating and recording apparatus was the same as in Fig. 3, except that the perfusion cannula was used in place of a microelectrode to measure the ampullary voltage.

Dow


128 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 6 9 " 1 9 7 7

reducing its sensitivity. This problem was solved by cutting a l - ram wedge in the Sylgard block perpendicular to the groove.

After the canal was straightened, its wall was nicked at the edge of the Sylgard block furthest from the ampulla. The perfusion cannula was in t roduced through this nick in the canal wall. I f the outer aspect of the canal was adequately deionized, this nick had no effect on the input resistance.

The perfusion cannula was a glass capillary several centimeters long and about 100/~m in d iameter . This cannula was also used to record voltage. The perfusate was forced out of the cannula under modest pressure produced by compressed air. The maximum rate of flow was 100/~l/min. The cannula was mounted on an XYZ micromanipula tor and was advanced down the straightened port ion of the canal. Fluid forced into the lumen o f the canal flowed back along the cannula to the nick in the canal wall where it was aspirated. Because the ampul lary je l ly was viscous and the cannula was somewhat flexible, the tip of the cannula could be steered as it advanced. Since the Sylgard block was t ransparent , the position of the tip could be viewed directly from above and also f rom the side by using a 45 ° front-silvered mirror . By viewing the tip in three dimensions, it was possible to prevent the cannula from touching the inner aspect of the canal wall as it advanced. Jelly in the canal was removed by perfusing with protease as the cannula was advanced (1 mg/ ml of Sigma Type VI protease [Sigma Chemical Co., St. Louis, Mo.] in the control solution). The input resistance of the prepara t ion was unaffected. Perfusion with a protease-free solution was begun when the tip of the cannula was within 2 mm of the neck of the ampulla.

With the tip of the cannula at the neck of the ampulla, ionic exchange was limited by diffusion into individual ampul lary lobules, a distance of up to 1 mm. When the lumen was perfused with fluorescein or phenol red, about 20 rain was required before the intensity of dye within the ampui lary lobules was indistinguishable from that of the perfusate flowing out of the cannula. This is approximately the amount of t ime one would estimate for free diffusion of a substance across 1 mm of gelatin. Because of this prolonged diffusion time, it was not possible to adjust ionic concentrations to specified levels or to el iminate completely a part icular ion from the ampul lary lumen. However, it was quite feasible to introduce a new ion or chelating agent into the ampulla by perfusing with a fairly high concentration.

R E S U L T S

Passive Properties

W h e n the a m p u l l a is e lec t r ica l ly i so l a t ed by the a i r g a p a n d no c u r r e n t is p a s s e d across it , a l u m e n - p o s i t i v e t r a n s e p i t h e l i a l p o t e n t i a l is r e c o r d e d . Th i s r e s t i n g p o t e n t i a l is 10-30 m V in f r e sh ly d i s s ec t ed p r e p a r a t i o n s . Li t t le r e s t i n g p o t e n t i a l is p r e s e n t u n d e r phys io log i ca l c o n d i t i o n s ( O b a r a a n d B e n n e t t , 1972) w h e n the a m p u l l a is s h u n t e d by the r e s i s t ance o f t he cana l (20-30 kf~ in the p r e s e n t e x p e r i m e n t s ) . W h e n phys io log ica l c o n d i t i o n s a r e m i m i c k e d by s h o r t - c i r c u i t i n g the two sa l ine poo l s wi th a salt b r i d g e , t he t r a n s e p i t h e l i a l p o t e n t i a l is c lose to ze ro . ( T h e s o m e w h a t g r e a t e r cana l l e n g t h in vivo s h o u l d m a k e n o s ign i f i can t d i f f e r e n c e . ) T h e r e s t i n g p o t e n t i a l in t he i so l a t ed p r e p a r a t i o n p r e s u m a b l y c o r r e - s p o n d s to a d i f f e r e n c e in t he p o t e n t i a l s ac ross t he two faces o f the r e c e p t o r cells. U n d e r phys io log i ca l c o n d i t i o n s , t he m e m b r a n e po t e n t i a l s o f t he two faces m u s t be n e a r l y e q u a l b e c a u s e t he faces a r e c o n n e c t e d t h r o u g h the low re s i s t ance o f the cana l .

T h e ef fec ts o f c o n s t a n t c u r r e n t a r e s h o w n in Fig. 5. M o d e s t c u r r e n t s o f e i t h e r

Dow


CLUSXN ANn BENsE'r'r Calcium-Activated Conduaanee in Electroreceptors. I 129

polarity cause the epithelial voltage to approach a value that is propor t ional to applied cur ren t (Figs. 5 A, 7 B). The ampulla is typically linear between 0 and 130 mV lumen positive. In these experiments intact ampullae had input impedances

A C

Bs° I 35

~ 2 0 ~ g

1(3

5

0

PsP

D

v ~

PSP . ,~.. . . ~ ,~,,w~,~,u.~,.,.,.~ t I

lOO 20o TIME (ms)

FIGURE 5. Effects of constant current on the ampullary epithelium. Applied current is shown in the top trace, lumen-negative stimuli upward. The transepithelial voltage appears below with the lumen negative responses shown upward. In A, hyperpolarizing and subthreshold depolarizing stimuli cause the epithelium to exponentially approach a voltage that is proportional to applied current. There is a 30-mV lumen-positive resting potential, the threshold for the action potential being about 0 mV. In B, a semilog plot of voltage minus final voltage vs. time after stimulus onset confirms the exponential time course of the passive response. The time constant of the inactive epithelium is 126 ms (average value from plots 1 and 2 in B). The inactive or leakage resistance of the epithelium is 372 kfl and the capacity is therefore 0.34/~F. An ampullary action potential and corresponding postsynaptic potential (PSP) are shown in C. The trace passing through the base of the action potential parallel to the base line indicates 0 mV. The extracellular recording of the PSP (bottom trace) shows superimposed action potentials arising in the nerve fibers. There is no change in postsynaptic activity from the spontaneous level until the threshold is reached. In D, transmitter is released by a strong lumen-positive stimulus which directly depolarizes the basal, secretory membranes of the receptor cells. The current calibration (vertical bar) is 0.4 p,A in A, 1.0 t~A in C, and 2.0/~A in D. The transepithelial voltage calibration (vertical bar) is 70 mV in A, 40 mV in C, and 200 mV in D. The postsynaptic voltage calibration (vertical bar) is 2.0 mV in C and 0.8 mV in D. The time calibration (horizontal bar) is 0.4 s in A and C and 0.2 in D.

ranging f rom 0.2 to 0.4 MI~. Waltman (1968) reports a passive input impedance o f 0.3 MI~ in ampullae of similar dimensions isolated by electronic feedback. In Fig. 5 B the logari thm of voltage minus final voltage is plotted against time for two traces. The plot shows that voltage rises nearly exponentially when a step of

Dow


130 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 • 1 9 7 7

current is applied. This result suggests that the resistance in series with the capacity of the ampullary epithelium is small, although voltage clamp experiments show that part of the capacity is in series with a measurable resistance (Clusin and Bennett, 1977).

Lumen-positive stimuli applied to the opening of the canal can directly depolarize the secretory membranes of the receptor cells and cause transmitter secretion (Obara and Bennett, 1972). However, lumen-positive stimulation of the isolated ampulla produces no postsynaptic potentials until the epithelial voltage exceeds 120 mV, lumen positive (Fig. 5 D). This finding implies that most of the voltage drop across inactive receptor cells occurs in the uninnervated lumenal membranes, and that the resistance of these membranes is high.

Action Potentials Evoked by Weak Stimuli

Lumen-negative stimuli greater than 50-100 nA evoke action potentials across the epithelium as in Fig. 5 C. The action potential has a well-defined threshold at about 0 mV (20 mV lumen negative from the resting potential) as shown. The active character of the response is indicated by the fact that it occurs at the end of the stimulus and that the voltage continues to rise after the stimulus is terminated. The notch on the rising phase of the action potential corresponds to the termination of the stimulus. After the action potential there is a lumen-positive afterpotential lasting several hundred milliseconds. In deteriorating preparations, the resting transepithelial potential declines, and constant lumen-positive current has to be applied to keep the ampullary epithelium from becoming spontaneously active. In such preparations, lumen-negative rather than lumen- positive afterpotentials follow the action potential.

Recordings from the afferent nerve of an electrically isolated ampulla show postsynaptic potentials whose onset coincides with the rising phase of the ampullary action potential (Fig. 5 C). Under physiological conditions when the opening of the canal is short-circuited to the basal surface of the epithelium, graded responses of the epithelium and graded postsynaptic responses are usually obtained (Murray, 1967; Obara and Bennett, 1972). Lumen-positive stimuli produce graded hyperpolarizing PSPs presumably resulting from a reduction in spontaneous activity of the receptor cells and reduced transmitter release (Obara and Bennett, 1972), while lumen-negative stimuli produce graded depolarizing PSPs. Graded responses to voltage stimuli of a few microvolts can be obtained in vitro if the two saline pools shown in Fig. 3 are short-circuited by a salt bridge (Clusin and Bennett, 1974). In the electrically isolated preparation, however, no hyperpolarizing postsynaptic potentials are seen and no depolarizing PSPs are seen until the epithelium reaches threshold for the production of an all-or-none response. Presumably, the lumenal faces of the individual receptor cells of an isolated ampulla are not spontaneously active and become synchronously excited when they are depolarized by applied current.

Maintained stimuli evoke trains of action potentials (Fig. 6 A). In an intact, freshly dissected preparation which has not been stimulated for at least 10 s the amplitude of the first action potential is between 60 and 100 mV and the rise time is about 60 ms. Subsequent action potentials of a train are uniform, but are 50% smaller in amplitude and duration. When brief stimuli are repeated at 2-s

Dow



intervals the ampl i tude and dura t ion o f the action potential is progressively reduced (Fig. 6 B), reaching a steady state af ter several stimuli. T h e basis for this refractoriness is discussed in the following paper (Clusin and Bennet t , 1977).

Effects of Strong Stimuli

As in o ther excitable cells (Fitzhugh, 1961), large long-lasting stimuli evoke only a single action potential , which occurs at the beginning o f the stimulus. Action potentials for a family o f stimuli are shown in Fig. 7 A. In Fig. 7 B, voltage is plotted as a funct ion of cur ren t . T h e peak voltage and the min imum voltage af ter the action potential but before terminat ion of the stimulus are both linearly related to cu r ren t for modera te stimulus strengths. T h e resistance of the inactive epi thel ium is 331 k n and the resistance immediately after an action potential is

A B

FIGURE 6. Effects of repetitive activity. In A a prolonged stimulus (lower trace) evokes a train of action potentials (upper trace). The first acdon potential is greater in amplitude and slightly longer in duration than the subsequent action potentials. In B, two brief identical stimuli are applied 2 s apart and the voltage traces are superimposed. The action potential evoked by the first stimulus (trace 1) is of greater amplitude and duration than that evoked by the second stimulus (trace 2). The vertical bar represents 0.4 /xA and 45 mV in A, 0.2/xA and 40 mV in B. The horizontal bar represents 0.4 s in A and 0.1 s in B.

84 kl]. When br ie f excitatory stimuli are super imposed on a large long-lasting stimulus just af ter the action potential no additional active response occurs and the cur ren t voltage relation is identical to that obtained by using a single stimulus and plotting the min imum voltage af ter the action potential . Thus , for large stimuli the slope conductance of the epi thel ium is increased four fo ld af ter the action potential. Obara and Bennet t (1972) r epor ted a less than twofold conductance increase associated with the ampul lary action potential. However , since their results were obtained in ampullae that were shunted by the resistance o f the canal, there is no disagreement .

I f the increase in slope conductance associated with the action potential results entirely f rom increased conductance in the lumenal membranes of the receptor cells (see below), then the resting resistance of the lumenal membranes must be at least three times grea ter than the resting resistance of the basal faces. Calcula- tions based on voltage clamp exper iments show that the ratio is actually much larger (Clusin and Bennet t , 1977). This asymmetry of m e m b r a n e resistance would be consistent with the observations on evoked t ransmit ter release described above.

A striking fea ture of the ampul lary action potentials is that they become greatly p ro longed when large stimuli are applied as shown in Fig. 7A. With

Dow


132 T H E J O U R N A L O F 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

increasing st imulus s t rength , the onset o f repolar izat ion occurs later and the rate of repolar izat ion is slowed. Above 158 mV, no repolar izat ion occurs du r ing the stimulus (Fig. 7A, Fig. 10A), even when the stimulus lasts more than 1 s. However , af ter t e rmina t ion of the st imulus, there is an inflection in the voltage trace (arrow) beyond which the trace is nearly super imposab le on the falling phase of an action potential evoked by a weak stimulus.

A

"AI

I J

B 035s

mV 150

lOg

5O j l

0 . 7 ~ 1 I 0 '~ 0.5 1.0,~A

. / t -,5o ,~ v ~ Z I • v,,,,

/ k,o o FXGURE 7. Ef fec t s o f c o n s t a n t c u r r e n t s t imul i a n d the r e s u l t i n g c u r r e n t vo l t age

relation. In A, the voltage responses (upper traces) to a family of current stimuli (lower traces) are shown. The resting transepithelial potential is 30 mV, lumen positive. A current voltage relation is plotted in B. The peak voltage of the action potential trilled circles) and the minimum voltage between the peak of the action potential and the termination of the stimulus (open circles) are linearly related to current over a broad range. With large stimuli repolarization is incomplete and the voltage at the end of the stimulus is plotted.

Thus , with large stimuli repolar izat ion is progressively slowed and delayed until a "suppress ion potential" is reached , above which repolar izat ion is indefi- nitely blocked. T h e t ime course of the voltage trace af ter t e rmina t ion of a large stimulus suggests that the process which normal ly repolar izes the action potential is delayed until the st imulus ends.

Ionic Substitution outside the Basal Faces

In o rde r to investigate the ionic basis for the ampul la ry action potential , it is necessary to vary the composi t ion o f the saline pe r fus ing the two surfaces of the

Dow



epithelium. It is assumed that the ampullary action potential results from changes in the permeability of membranes to ions, and that the total transepithelial potential represents the sum of ionic potentials generated in the lumenal and basal membranes. If the action potential results from significant permeability changes in both the lumenal and basal membranes, then it should be possible to change the form of the action potential by varying one or more ionic gradients across either membrane. If the permeability changes are confined to one face, only substitutions at that face will affect the action potential.

The basal membranes are most readily studied because they are more accessi- ble to perfusion. Perfusion experiments show that a variety of changes in the basal membrane's ionic milieu do not significantly affect the form of the ampullary action potential. Replacement of 90% of extracellular Na with K or with choline does not substantially alter the rising phase or the repolarizing phase of the action potential within 5 min. Steinbach (1974) previously reported that application of T T X to the basal membrane blocks action potentials in the postsynaptic nerve but does not affect the ampullary potential. In the present study total replacement of Na has no immediate effect on the ampul!ary action potential, although there is a gradual increase in epithelial conductance over several minutes. The significance of this observation is discussed below.

When calcium is completely removed from the Ringer and 1 mM EGTA is added, there is no effect on the ampullary action potential but synaptic transmission is abolished in less than 1 min. The rapid action of the calcium-free solution suggests that ionic exchange across the gelatinous material adhering to the basal surface of the epithelium is rapid. The dependence of synaptic transmission on extraceUular calcium suggests that there is a voltage-dependent calcium conductance in the basal membranes of the receptor cells (Steinbach, 1974). This observation fur ther suggests that any calcium crossing the lumenal membranes during the action potential does not diffuse across the receptor cells in sufficient quantity to mediate transmitter release. Since the transepithelial action potential is unaffected by removal of calcium from the basal face, we can conclude that, as in squid giant synapse (Katz and Miledi, 1969), the active calcium permeability in the secretory membrane has little effect on its potential under physiological conditions. Moreover, we conclude that calcium entering across the basal membranes does not affect the permeability of the lumenal membranes during the action potential (Clusin and Bennett, 1973). Rose and Lowenstein (1975) inferred from aequorin studies that at low concentrations calcium diffuses only a few micrometers in epithelial cells of an insect salivary gland before it is sequestered.

Even more drastic ionic substitutions in the saline bathing the basal membranes do not affect the form of the action potential. Fig. 8 B shows an action potential recorded in a calcium-free saline containing 1 mM EGTA in which 80% of the NaCl has been replaced by potassium acetate. Exposure to potassium acetate saline causes the epithelium to become spontaneously active, so that a steady hyperpolarizing current has to be applied to maintain the resting potential. As in other experiments where a DC holding current is used to prolong viability (Steinbach, 1974), a lumen negative afterpotential is recorded. How- ever, the shape of the action potential is normal, and the regenerative character of the response (not shown) is preserved. Moreover, there is little change when

Dow


134 T H E J O U R N A L O F G E N E R A L P H Y S I O L O G Y ' V O L U M E 6 9 • 1 9 7 7

the epi the l ium is r e t u rned to the control saline (Fig. 8 A). Since nei ther the rising nor the falling phase o f the action potential is significantly affected by drastic pe r tu rba t ion o f ionic gradients across the basal m e m b r a n e s , the action potential and the associated increase in conductance must result f rom permeabi l i ty changes in the lumenal m e m b r a n e s .

Ionic Substitution outside the Lumenal Faces

Because o f the long diffusion times involved, no a t t empt was made to r emove sodium f rom the lumen of the canal. However , when the lumen is pe r fused with 1 /zM tetrodotoxin (TTX) , the action potential is unchanged as shown in Fig. 9 D. T T X selectively abolishes the vo l t age -dependen t sodium permeabi l i ty o f axons

A. CONTROL B. K ACETATE

045 s

FIGURE 8. Effects of multiple ionic substitutions in the fluid bathing the basal surface of the ampullary epithelium. In B, calcium has been removed, 1 mM EGTA was present, and 80% of the NaCI has been replaced by potassium acetate. In A the epithelium has been returned to the control saline for several minutes. There is little change in the form of the action potential. A well-defined threshold (not shown) can be demonstrated under both conditions. In both records a sustained 0.12 ~A hyperpolarizing current has been applied to prevent spontaneous activity (Steinbach, 1974). Lumen-positive afterpotentials are common under these conditions. The holding potential is 28 mV lumen positive in A, and 41 mV lumen positive in B. This difference in potential with the same holding current is consistent with a potassium- or chloride-sensitive resting potential in the basal membranes of the receptor or supporting cells.

and of many o ther excitable tissues, t hough TTX-res i s t an t sod ium conductances have been described (Kao, 1966).

When the lumen is per fused with cobalt, excitability is abolished. Cobalt is known to be a selective blocker o f vo l t age -dependen t calcium permeabil i t ies . 30 mM cobalt nearly abolishes the inward calcium cur ren t in barnacle muscle fibers (Hagiwara et al., 1969), but has little effect on the ou tward potass ium cur ren t . Fig. 9 B shows the effects of cons tan t -cur ren t pulses appl ied across an ampul l a ry epi thel ium af ter the lumen has been pe r fused with a 100 mM cobalt solution for 25 min. Not only is the action potential abolished, but there is very little delayed rectification. T h e concentra t ion of cobalt requi red to p roduce these effects is undoubted ly lower than 100 mM, but because of diffusion delays, the concentration within the alveoli is not known. Fig. 9 C shows the same epi thel ium af ter 35 min of per fus ion with the control solution. T h e normal fo rm o f the action potential at d i f fe ren t peak voltages is largely res tored . Comple te reversibility is

Dow



not achieved, p re sumab ly because the p repa ra t ion deter iora tes before comple te washout of cobalt occurs.

In most excitable m e m b r a n e s , de layed rectification causes repolar izat ion during large exci tatory stimuli, even if the active inward cu r r en t has been blocked. Reversible loss o f excitability du r ing per fus ion o f the lumen with cobalt suggests that the action potent ial in skate e lect roreceptors is ascribable to a calcium permeabi l i ty increase in the lumenal m e m b r a n e s . However , the absence of delayed rectification du r ing per fus ion with cobalt suggests that cobalt somehow interferes with the repolar iz ing process as well.

A. CONTROL B. COBALT

C. WASHOUT D TTX

P

FIGURE 9. Effects of perfusing the lumen with cobalt and TTX. In A the lumen is perfused with the control solution. In B the ampulla has been perfused for 25 min with a 100-mM cobalt solution. The cobalt concentration at the alveolar epithelium is not known because of the long diffusion time. Perfusion with cobalt renders the ampulla inexcitable. Considerable recovery occurs when the ampulla is perfused with the control solution for 35 min, as shown in C. There is, to be sure, a progressive irreversible decline in leakage resistance, which is invariably seen in preparations more than 3 h old and need not be attributed to cobalt. D shows a normal action potential recorded after perfusing the lumen for 25 min with a solution containing 1/~M tetrodotoxin. The vertical bar represents 1 #A and 88 mV in A-C and 0.2 /zA and 20 mV in D. The horizontal bar represents 0.35 s in A-C and 0.2 s in D. The resting potential was 20 mV in all four records.

T h e role of calcium can be clarified by per fus ion o f the lumen of the e lec t roreceptor with a solution containing 20 mM E G T A , 10 mM calcium, and 50 mM magnes ium. T h e ionized calcium concentra t ion in this per fusa te is 0 .2/xM at p H 8, according to the m e thod of Portzehl et al. (1964). However , because the m o v e m e n t o f E G T A into the lumen is l imited by diffusion, it is unlikely that the ionized calcium concent ra t ion at the recep to r cells reaches this level. After 15 min of per fus ion , the repolar iz ing phase o f the action potent ial in Fig. 10 C is slowed at all voltages, and the suppress ion potential is r educed f rom 145 to 91 mV. T h e t ime course of delayed repolar izat ion is also slowed in low calcium.

I f per fus ion is con t inued for ano the r 10 min (Fig. 10D) repolar izat ion no longer occurs at any voltage. T h e action potential evoked by a jus t - threshold stimulus is a sustained plateau, which can last at least 10 s and has to be t e rmina ted by appl ied cur ren t . In several exper iments , per fus ion of the lumen

Dow


1 3 6 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 • 1 9 7 7

with a calcium-free solution containing EGTA was found completely to abolish electrical excitability and to render the epithelium electrically linear within 30 min.

The relationship between the extracellular calcium concentrat ion and the size of the action potential is complex. A comparison of cur rent and voltage clamp data (Clusin and Bennett , 1977) indicates that the repolarizing process begins before the peak of the action potential, so that the latency of the repolarizing process affects spike amplitude. However , the rate of repolarization varies

A. CONTROL B HIGH Ca

D. LOWER Ca

FIGURE 10. Effects of varying the ionized calcium concentration in the lumen. The records in A have been obtained during perfusion with the control solution. In C the lumen has been perfused for 15 rain with a solution containing 20 mM EGTA and 10 mM Ca. The Ca concentration at this time cannot be determined. In D the lumen has been perfused for another 10 min with the EGTA solution. Spontaneous repolarization no longer occurs and the resulting plateau-shaped action potential is terminated with applied current. A well-defined threshold for termination (not shown) can be demonstrated. B has been obtained from the same preparation after 30 min of perfusion with a solution containing 100 mM Ca and no EGTA. The resting potential is 30 mV lumen positive in A and C. The epithelium is held at 44 mV lumen positive in B and 51 mV lumen positive in D. The holding current is 0.19 /zA in each case. The vertical calibration represents 2/zA in A, B, and C and 0.8/zA in D. The horizontal calibration represents 0.35 s in A, B, and C and 0.9 s in D.

inversely with extracellular calcium (Fig. 10). Thus a decrease in extracellular calcium would have antagonistic effects: it would tend to decrease spike amplitude by decreasing inward calcium current , but it would tend to increase spike ampli tude by slowing the onset of repolarization.

Fig. 10 B is f rom the same preparat ion as above after 30 min of perfusion with a solution containing 100 mM Ca and no EGTA. The effect of calcium chelation on the action potential has been clearly reversed. The durat ion of the action potential is shorter at all voltages than it is in normal calcium and repolarization is no longer suppressed at 145 mV. The suppression potential in high calcium was not de te rmined for fear of damaging the epithelium.

The effects o f changing the calcium concentrat ion in the lumen of the ampulla

Dow



indicate that electrical excitability results f rom a vol tage-dependent calcium conductance in the lumenal membranes of the receptor cells. Moreover , it appears that calcium influx is necessary for the repolar izing process to occur. T h e rate of repolar izat ion for a given stimulus and the suppression potential for repolarizat ion both depend on the gradient of ionized calcium across the lumenal membranes .

Effects of Dinitrophenol

Another way of al ter ing the calcium concentrat ion gradient across cell membranes is t rea tment with 2,4-dini t rophenol (DNP), which causes eff lux of mito- chondrial calcium into the cytoplasm (Drahota et al., 1965). When the electroreceptor epi thel ium is t reated with 10 -4 M DNP, a fourfo ld transepithelial conductance increase develops over a per iod of 10 min. Dur ing this per iod there is progressive loss of electrical excitability. Excitability and a normal transepithelial resistance cannot be res tored by passing large inhibi tory currents which hyper- polarize the lumenal membranes . This suggests that the transepithelial conductance p roduced by DNP is insensitive to voltage. In te rpre ta t ion o f these results is complicated because d in i t rophenol could act directly, or by some metabolic effect o ther than liberation of calcium. However , the result is consistent with the suggestion that ionized intracellular calcium produces a conductance increase in the normally high-resistance lumenal membranes of the receptor cells.

T h e effects of DNP are similar to those observed when the basal surface of the epithel ium is ba thed in a solution in which all o f the sodium has been replaced by choline. Over a per iod of 10 min the transepithelial conductance increases and excitability is lost. A normal resting resistance cannot be res tored by passing large lumen-posit ive currents . In squid axons, Blaustein and Hodgkin (1969) showed that extrusion o f calcium is r educed in sodium-free saline, while Baker et al. (1971) used aequor in to show that the cytoplasmic concentra t ion of ionized calcium in squid axons rises when the extracellular sodium is replaced. In skate electroreceptors , calculations based on Fig. 32 o f Waltman (1966) show that the basal faces comprise more than 98% of the receptor cell surface membrane . Removal of extracellular sodium f rom the basal faces could t h e r e fo r e lead to a rise in the cytoplasmic calcium concentra t ion. Katz and Miledi (1969) suggested that a similar mechanism may account for the disappearance of the calcium- dependen t action potential when presynaptic fibers of squid stellate synapse are bathed in a sodium-free saline.

D I S C U S S I O N

Origin of the Epithelial Potentials

Voltage clamp exper iments (Waltman, 1968; Clusin and Bennet t , 1977) show that excitability of the ampulla results f rom an active cur ren t which flows inward across the lumenal membranes and outward across the basal membranes of cells in the epithel ium. Since the action potential is associated with t ransmit ter release, at least some of this cur ren t must flow outward across the basal membranes of the receptor cells. T h e present data do not exclude the possibility that both the receptor cells and the suppor t ing cells are electrically excitable. How-

Dow


138 THE JOURNAL OF GENERAL PHYSIOLOGY " VOLUME 69 " 1977

ever, on morphologica l and funct ional g rounds we suspect that the suppor t ing cells are inexcitable high-resistance cells like those o f the canal wall, and that only the recep tor cells p roduce active cur ren t . Voltage c lamp data (Clusin and Bennet t , 1977) s u p p o r t this inference .

T h e per fus ion expe r imen t s clearly suggest that it is the lumenal m e m b r a n e s of the receptor cells that are excitable. Various al ternatives, such as a hyperpolar iz - ing po ta s s ium-dependen t response in the basal m e m b r a n e s , are excluded by the fact that the action potentials persist when every conceivable ionic gradient across the basal m e m b r a n e is al tered. T h e effects o f pe r fus ing the lumen with high Ca, E G T A , and cobalt suggest that both the depolar iz ing and repolar iz ing phases of the action potential are at t r ibutable to permeabi l i ty changes in the lumenal m e m b r a n e s of the receptor cells.

The Synchronization of Receptor Cells

U n d e r cur ren t c lamp condit ions, the ampul l a ry epi the l ium behaves like a single sheet o f excitable m e m b r a n e , genera t ing large action potentials with a well- def ined threshold . Activity of individual patches of excitable m e m b r a n e in the lumenal faces of the recep tor cells in an electrically isolated ampul la must the re fo re be closely synchronized. T h e basis for this synchronizat ion is illus- t ra ted in Fig. 11. T h e lumenal faces are r ep re sen t ed by a fixed resistance, rLUM, in parallel with an active calcium conductance , gca, while the basal m e m b r a n e is r ep resen ted by a f ixed resistance, rBAS. TheYe are two reasons to believe that the resistance o f the basal m e m b r a n e s is low c o m p a r e d with the rest ing resistance of the lumenal membranes :

SALT BRIDGE

o ~ r l " • gc u~ /,'~ -.~

\T 4 I '~ RSH

~Rcanal

:J2 / ' - ~ - - r - T \

I t

\ , . ~>' BAS ,,,

7- FIGURE 11. Equivalent circuit illustrating spread of electrical excitation between two receptor cells in the ampullary epithelium. The lumenal faces are represented as an active calcium conductance, gca, in parallel with a fixed resistance rLUM. (The corresponding batteries Eca and ELUM are drawn but not labeled.) The basal faces are represented as fixed resistors, rBAS. A single resistance RsH represents shunt pathways across inexcitable cells in the epithelium and through the intercellular clefts. The numbered arrows indicate possible return pathways for inward current generated by an active response in the lumenal membrane of the receptor cell on the right.

Dow


CLusIN AND BENNETT Calcium-Activated Conductance in Electroreceptors. I 139

(a) There is a large increase in the slope conductance across the epithelium during the action potential. In the following paper, (Clusin and Bennett, 1977) it will be shown that this increase in slope conductance is due to an increase in the instantaneous conductance of the epithelium, and that the conductance increase across the receptor cells is considerably larger than the fourfold transepithelial conductance increase. Since the transepithelial conductance increase during the action potential is due to an increased conductance of the lumenal membranes, it follows that the resistance of the basal membranes is low.

(b) The transepithelial potential must exceed 100 mV lumen positive in order to directly excite the presynaptic calcium conductance in the basal faces and cause release of transmitter, while lumen-negative potentials of less than 20 mV directly excite the voltage-dependent calcium permeability of the lumenal membranes.

The basal membrane resistance need not be lower than the active resistance of the lumenal membranes. In fact, under voltage clamp conditions the basal membranes appear to constitute a significant series resistance as discussed in the following paper.

The low resting resistance of the basal membranes has two consequences: (a) most of the voltage drop across an inactive receptor cell will occur in the lumenal membrane; (b) before onset of the late outward current most of the early inward current through the lumenal membrane of an excited receptor cell will flow outward across the basal membrane.

In Fig. 11, some of the current generated by a response in the receptor cell on the right flows outward across the lumenal membrane of the adjacent receptor cell, thereby depolarizing it (current path 1). Because of the low resistance of the basal membrane, active current is not greatly shunted by the leakage resistance of the membrane in which it arose (current path 3).

Leakage of current through the intercellular clefts (included in current path 2) is minimized by the zonulae occludentes. The supporting cells must also be of high resistance as evidenced by the large input impedance of the isolated ampulla. However, introduction of an external low resistance pathway across the ampullary epithelium loads down active receptor cells and shunts current away from adjacent inactive receptor cells so that interaction among them is reduced (current path 4). This shunting can be done reversibly by cutting open the canal wall in the air gap and short-circuiting the incision to the saline pool containing the ampulla with a salt bridge. When an isolated ampulla with an input resistance of 300 k12 is shunted by the remaining canal resistance of 15 k12, the response becomes graded and no threshold can be demonstrated.

The disappearance of the well-defined threshold when the epithelium is shunted excludes the possibility that the receptor cells are synchronized mainly by way of electrotonic junctions between cells in the epithelium. Voltage clamp experiments in the following paper indicate that the lumenal faces probably remain excitable when the canal is short-circuited because of the series resistance of the basal faces. I f synchronization of the receptor cells were mediated primarily by electrotonic junctions between them, the threshold characteristic would be little affected by reducing the input resistance of the epithelium. The

Dow


1 4 0 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 - 1 9 7 7

above experiments therefore demonstrate that the synchronization is mediated mainly by extracellular pathways.

Since there have been no freeze-fracture studies of skate electroreceptor epithelium, the possibility of small gap junctions within the zonulae occludentes cannot be excluded. However the existence of such gap junctions, with a small degree of coupling between receptor cells, would not affect the conclusions in this or the following paper.

Calcium Dependence of Repolarization

Sufficient reduction in the extracellular free calcium concentration in the lumen of the ampulla causes the action potential to become plateau shaped with no signs of repolarization more than 20 s after excitation. The inhibition of the repolarizing process by low calcium and by cobalt could be interpreted in several ways. For example, the reduction in extracellular calcium could led to inactiva- tion of the repolarizing process by a nonspecific effect on the membrane not involving intracellular calcium. The effect of cobalt could be explained by a direct action on the repolarizing process. However, when the similar effects of cobalt and low calcium are considered together, the simplest explanation is that the repolarizing process is initiated by an influx of calcium into the cytoplasm. Since the calcium influx associated with the plateau-shaped action potential evoked in the low calcium solution was not sufficient to initiate repolarization, it appears that a certain minimum level of intracellular calcium is required.

There is a marked similarity between the effect of lowering the extracellular calcium concentration in the lumen and that of passing large excitatory current pulses across the epithelium. In both cases, the rate of repolarization becomes progressively slowed until repolarization is completely blocked. The effect of reduced calcium and the effect of large excitatory currents are additive, as shown by the fact that the suppression potential for repolarization is reduced when the lumen is bathed in low calcium saline and increased in high calcium saline.

In squid giant synapse, passage of large outward currents was found to block another calcium-dependent process, release of synaptic transmitter. Katz and Miledi (1967) and Kusano et al. (1967) injected TEA into the presynaptic terminal of squid giant synapse so that large depolarizations could be achieved. They found that depolarization of the terminal beyond + 130 mV blocks transmitter release until the end of the stimulus. The lowest potential at which transmission is blocked was termed the suppression potential, and the postsynaptic potential occurring at the end of the stimulus was attributed to "delayed release." The suppression potential was presumed to be the calcium equilibrium potential. Stimuli causing the presynaptic membrane to exceed this voltage block the influx of calcium into the cytoplasm. When the stimulus is terminated, a significant calcium influx occurs before the membrane returns to its resting potential and the calcium channels close. This calcium influx produces the delayed release. Using aequorin, Llin~s and Nicholson (1975) demonstrated that the suppression potential corresponds to the voltage at which the calcium influx is blocked. An influx of calcium after termination of the stimulus accompanies delayed release. Blockage of aequorin luminescence during large excitatory stimuli, with delayed

Dow



luminescence on repo la r i za t ion , had previously been d e m o n s t r a t e d in Aplysia neurons by S t innakre and Tauc (1973).

In skate e l ec t ro recep to r s with a no rma l concen t ra t ion o f calcium in the l umen , exci ta tory st imuli which cause the epi thel ia l voltage to exceed 150 mV block repola r iza t ion . Since most of the t ransepi the l ia l vol tage is deve loped across the excitable lumena l m e m b r a n e , the effect of large st imuli can be exp la ined by suppos ing that they cause the l umena l m e m b r a n e to exceed the calcium equil ib- r ium potent ia l . Repola r iza t ion is de layed unti l the end o f the s t imulus because the repo la r i z ing process is in i t ia ted by a calcium inf lux which does not occur unt i l the lumena l m e m b r a n e falls below the calc ium equ i l ib r ium potent ia l . T h e ra te at which the r epo la r i z ing conduc tance develops should d e p e n d on the ra te of calcium inf lux. T h u s the fal l ing phase of an action poten t ia l evoked by a weak st imulus and the t ime course of de layed repo la r iza t ion af te r a s t rong s t imulus are faster in h ighe r calcium concent ra t ions (Fig. 10). Moreover , in both high and low calcium solut ions repo la r iza t ion is slowed as the suppress ion potent ia l is a p p r o a c h e d . Voltage c lamp data to be p r e sen t ed in the fol lowing p a p e r (Clusin a n d Bennet t , 1977) s u p p o r t this i n t e rp re t a t i on .

We are grateful to Drs. C. M. Armstrong, F. Bezanilla, L. B. Cohen, and B. Hille for valuable discussions and useful technical suggestions. Critical review of the manuscript by Drs. A. Finkelstein, E. R. Kandel, D. P. Purpura, J. M. Ritchie, and C. F. Stevens was also very beneficial. The assistance of Dr. D. C. Spray was most helpful in developing the techniques for perfusion of the ampulla. We thank Ms. R. Sekhar for her secretarial assistance. W. T. Clusin was supported by the Nadonal Institutes of Health Medical Scientist Training Grant No. 5T5 GM 1674-12 to the Albert Einstein College of Medicine and by the Epilepsy Foundation of America. The work was also supported in part by grants HD-04248 and NS-07512 from the National Institutes of Health.

Received for publication 7 June 1976.

REFERENCES

AKUTSU, Y., and OBARA, S. 1974. Calcium dependent receptor potential of electroreceptor of marine catfish. Proc. Jpn. Acad. 50:247-251.

BAKER, P. F., A. L. HODGKIN, and E. P. RIDGWAY. 1971. Depolarization and calcium entry in squid giant axons. J. Physiol. (Lond.). 218:709-755.

BLAUSTEIN, M. P., and A. L. HODGKIN. 1961. Effect of cyanide on the efflux of calcium from squid giant axons.J. Physiol. (Lond.). 200:497-527.

CLtJSIN, W. T., and M. V. L. BENNETT. 1973. Calcium electrogenesis in skate electroreceptors. Biol. Bull. (Woods Hole). 145:429.

CLUSIN, W. T., and M. V. L. BENNETT. 1974. Negative slope resistance in skate electroreceptors. Fed. Proe. 33:1471.

CLUSlN, W. T., and M. V. L. BENNETT. 1977. Calcium activated conductance in skate electroreceptors: Voltage clamp experiments. J. Gen. Physiol. 69:145-182.

CLUSIN, W. T., D. C. SPRAY, and M. V. L. BENNETT. 1974. Activation of a voltage insensitive conductance by inward calcium current. Biol. Bull. (Woods Hole). 147:472.

CLUSlN, W. T., D. C. SPRAY, and M. V. L. BENNZTT. 1975. Activation of a voltage- insensitive conductance by inward calcium current. Nature (Lond.). 256:425-427.

DIJKGRAFF, S., and A. J. KAL~aIJN. 1063. Verhaltensversuche zur Funktion der Lorenzin- ischen Ampulla. Naturwissenschaften. 49:400-401.

Dow


142 THE JOURNAL OF GENERAL PHYSIOLOGY " VOLUME 69 • 1977

DRAHOTA, Z., E. CARAFOLI, C. S. R o s s I , R. L. GAMBLE, a n d A. L. LEHNINGER. 1965. T h e

steady state maintenance of accumulated calcium in rat liver mitochondria . J. Biol. Chem. 240:2712.

FXTZHUOH, R. 1961. Impulses and physiological states in theoretical models of nerve membrane. Biophys. J. 1:445-466.

Fi)HNER, H. 1908. fQber eine Speisunsflussigkeit ffir Salachierherzen. Z. Allg. Physiol. 8:485-491.

HAGIWARA, S., H . HAYASHI, and K. TAKAHASHI. 1969. Calcium and potassium currents of the membrane of a barnacle muscle fiber in relation to the calcium spike.J . Physiol. 205:115-129.

KALMIJN, A . J . 1971. The electric sense of sharks and rays .J . Exp. Biol. 55:371. KALMZJN, A . J . 1975. The detection of electric fields from inanimate and animate sources

other than electric organs. In Handbook of Sensory Physiology. Vol. III/3. Electro. receptors and Other Specialized Receptors in Lower Vertebrates. A. Fessard, editor. Springer , New York. 147-200.

KAO, C. Y. 1966. Te t rodotoxin , saxitoxin and their significance in the study of excitation phenomena. Pharmacol. Rev. 18:997-1099.

KATZ, B., and R. MILEDI. 1967. A study of synaptic transmission without impulses. J. Physiol. (Lond. ). 192:407-436.

KATZ, B., and R. MILEr1. 1969. Tet rodotoxin resistant electric activity in presynaptic terminals. J. Physiol. ( Lond. ). 205:459-487.

KUSANO, K., G. LIVENOOOD, and R. WERMAN. 1967. Correlat ion of t ransmitter release with membrane propert ies of the presynaptic fiber of squid giant synapse. J. Gen. Physiol. 50:2579-2601.

LHN/,S, R., and C. NXCHOLSON. 1975. Calcium role in depolarization-secretion coupling: an aequorin study. Proc. Natl. Acad. Sci. U.S.A. 72:187-190.

LOWENSTEIN, W. R. 1960. Mechanisms of nerve impulse initiation in a pressure receptor (Lorenzinian ampulla) . Nature (Lond.). 188:1034.

LOWENSTEIN, W. R., and N. ISnIKO. 1962. Sodium chloride sensitivity and electrochemi- cal effects in a Lorenzinian ampulla. Nature (Lond.). 194:292-294.

MEECH, R. W., and N. B. STANDEN. 1975. Potassium activation in Helix aspersa neurones under voltage clamp: a component mediated by calcium influx. J. Physiol. (Lond.). 249:211-239.

MURRAY, R. W. 1960. The response of ampullae of Lorenzini of elasmobranchs to mechanical stimulation. J. Exp. Biol. $7:417-424.

MURRAY, R. W. 1965. Electroreceptor mechanisms: the relation of impulse frequency to stimulus strength and responses to pulsed stimuli in the ampullae of Lorenzini of elasmobranchs. J. Physiol. (Lond.). 180:592-606.

MURRAY, R. W. 1966. Receptor mechanisms in the ampullae of Lorenzini of elasmobranch fishes. Cold Spring Harbor Symp. Quant. Biol. $0:233-243.

MURRAY, R. W. 1967. The function of the ampullae of Lorenzini of elasmobranchs. In Lateral Line Detectors. P. Cahn, editor. Indiana University Press. 277-293.

MURRAY, R. W., and W. T. W. POTTS. 1961. The composition of the endolymph, per i lymph and other body fluids of elasmobranchs. Comp. Biochem. Physiol. 2:65-75.

OBARA, S., and M. V. L. BENNETT. 1968. Receptor and generator potentials of ampulla of Lorenzini in the skate, Raja. Biol. Bull. (Woods Hole). 155"430.

OBARA, S., and M. V. L. BENNETT. 1972. Mode of operat ion of ampullae of Lorenzini of the skate Raja. J. Gen, Physiol. 60:534-557.

Dow



PORTZEHL, H. , P. C. CALDWELL, and J. C. RUEGG. 1964. The dependence of contraction and relaxation of muscle fibers from the crab Maia squinado on the internal concentration of free calcium ions. Biochim. Biophys. Acta. 79:581-591.

RosE, B., and W. R. LOWENSTEIN. 1975. Calcium ion distribution in cytoplasm visualized by aequorin: Diffusion in cytosol restricted by energized sequestering. Science (Wash. D. C.). 190:1204-1206.

SAND, A. 1938. T h e function of the ampullae of Lorenzini with some observations on the effect of t empera tu re on sensory rhythms. Proc. R. Soc. Lond. Ser. B. Biol. Sci. 125:524- 553.

STINNAKRE, J. , and L. TAUC. 1973. Calcium influx in active Aplysia neurones detected by injected aequorin. Nat. New Biol. 242:113-115.

STEXNBACH, A. B. 1974. Transmission from receptor cells to afferent nerve fibers. In Synaptic Transmission and Neuronal Interact ion. M. V. L. Bennett , editor. Raven Press, New York. 105-140.

STEINBACH, A. B., and M. V. L. BENNETT. 1971. Presynaptic actions of calcium and magnesium and post-synaptic actions of glutamate at a sensory synapse. Biol. Bull. (Woods Hole). 141:403.

WALTMAN, B. 1966. Electrical propert ies and fine structure of the ampullary canals of Lorenzini. Acta Physiol. Scand. Suppl. 264:1-60.

WALTMAN, B. 1968. Electrical excitability of the ampullae of Lorenzini in the ray. Acta Physiol. Scand. 74:29a (Abstr.).

ZXPSER, B., and M. V. L. BENNETT. 1973. Tet rodotoxin resistant electrically excitable responses of receptor cells. Brain Res. 62:253-259.

Dow


Current Clamp Experiments - rupress.org

Documents

Transcript of Current Clamp Experiments - rupress.org