nels in Ieech integument are less sensitive to amiloride than channels known from vertebrale epithelia. With. 20 mmol Na+ ·I-1 in the mucosal solution the tissue.
}. ('omp Physiol B (1993) 163 :153-159
Joumalof
~· Comparative -En.tron· Physiology B ~~y © Springer- Verlag 1993
Ion iransport across leech integument I. Electrogenic Na+ transport and current fluctuation analysis of the apical Na+ channel W.-M. Weber, B. Dannenmaier, W. Clauss Institut für Tierphysiologie, Justus-Liebig-Universität Giessen, Wartweg 95, W-6300 Giessen, Germany Accepted: 21 October 1992
Abstract. The dorsal skin of the leech Hirudo medicinalis was used for electrophysiological measurements per(ormed in Ussing chambers. The leech skin is a tight ~>ithelium (transepithelial resistance = 10.5 ± 0.5 kO · cm - 2) with an initial short-circuit current of 29.0±2.91JA·cm- 2• Removal of Na+ from the apical bath medium reduced short-circuit current about 55%. Ouabain (50 IJmol·l- 1) added to the basolateral solution, depressed the short-circuit current completely. The Na+ current saturated at a concentration of 90 mmol Na+.J- 1 in the apical solution (KM=11.2± 1.8 mmol·l- 1). Amiloride (100 j.Jmol·l- 1) on the apical side inhibited ca. 40% of the Na+ current and indicated the presence of Na+ channels. The dependence of Na+ current on the amiloride concentration followed Michaelis-Menten kinetics (Ki=2.9±0.41Jmol·l- 1). The amiloride analogue benzamil bad a higher affinity to the Na+ channel (Ki=0.7±0.2 J.lmol·l- 1). Thus, Na+ channels in Ieech integument are less sensitive to amiloride than channels known from vertebrale epithelia. With 20 mmol Na+ ·I- 1 in the mucosal solution the tissue showed an optimum amiloride-inhibitable current, and the amiloride-sensitive current under this condition was ).8 ± 2.3% of total short-circuit current. Higher Na+ concentrations Iead to a decrease in amiloride-blockade short-circuit current. Stimulation of the tissue with cyclic adenosine monophosphate (100 IJmol·l- 1) and isobutylAbbreviations: IX, slope of the background noise component; ADH, antidiuretic hormone; cAMP, cyclic adenosine monophosphate; J, frequency; fc, comer frequency of the Lorentzian noise component; Hepes, N-hydroxyethylpiperazine-N'-ethanesulphonic acid; IBMX, isobutyl-methylxanthine; iNa• single Na+ channel current; IN• max, maximal inhibitable Na+ current; Isc• short circuit current; K;, half maximal blocker concentration; KM, Michaelis constant; MN•• channel density; RT, transepithelial resistance; SEM, standard error of the mean; S(j), power density of the Lorentzian noise component; S0 , plateau value of the Lorentzian noise component; TMA, tetramethylammonium; Trizma, TRIS-hydroxymethyl-amino-methane; vmu• maximal reaction velocity; VT, transepithelial potential. K;, half maximal blocker concentration Correspondence to: W.-M. Weber
methylxanthine (1 mmol·I- 1) nearly doubled short-circuit current and increased amiloride-sensitive Na+ currents by 50%. By current fluctuation analysis we estimated single Na+ channel current (2.7±0.9 pA) and Na+ channel density (3.6 ± 0.6 channels ·1Jm - 2) under control conditions. After cyclic adenosine monophosphate Stimulation Na+ channel density increased to 5.4 ± 1.1 channels ·1Jm - 2, whereas single Na+ channel current showed no significant change (1.9±0.2 pA). These data present a detailed investigation of an invertebrate epithelial Na+ channel, and show the similarities and differences to vertebrate Na+ channels. Whereas the channel properties are different from the classical vertebrate Na+ channel, the regulation by cyclic adenosine monophosphate seems similar. Stimulation of Na+ uptake by cyclic adenosine monophosphate is mediated by an increasing number of Na+ channels.
Key words: Na+ channels - Amiloride - Benzarnil Noise analysis - Leech, Hirudo medicinalis
Introduction The leech Hirudo medicinalis is an aquatic species of the Annelidae, occupying a freshwater habitat. With an extracellular ion concentration of about 10 mosmol · l- 1 with Na+ as the main cation, leeches are continuously challenged by the hypoosmotic environment, and have to prevent 1oss of electrolytes and uptake of water across their integument (Zerbst-Boroftlca 1984). This is achieved by homeostatic mechanisms in nephridia and integument. Whereas nephridial functions are weil characterized [review: Zerbst and Wenning (1986)], the ion transport properties of the integument are poorly investigated (Cobbold 1971; Prusch and Otter 1976) and far from being understood. From tracer flux measurements, electrical potential measurements, and the effects of the diuretic amiloride, an electrogenic mechanism of
~-~----'--'-~---·-~---- --
... . ~ - ~ .. .........
154
Na+ absorptionwas postulated for the leech integument; however, as these measurements were done either in vivo or under non-symmetrical solution conditions, the exact nature of this mechanism was not elucidated, i.e. the location of the individual transport systems in the apical and basolateral cell membranes was not probed, nor were they conclusively proven and characterized. Furthermore, nothing was known about the regulation of this putative electrogenic Na+ transport, although Zerbst et al. (1983) have argued for an antidiuretic factor in the leech. The classical mechanism of electrogenic Na+ transport in epithelia is characterized by an amiloridesensitive Na+ channel in the apical cell membrane and a basolateral Na+ /K +-ATPase. The Na+ channel of vertebrate epithelia is weil characterized by its selectivity for Na+ over K +, and for its specific blockage by amiloride (Bentley 1968; Benos 1982) and amilorideanalogues (Benos et al. 1986; Kleymann and Cragoe 1988). This amiloride-sensitive Na+ channel was investigated in various vertebrate epithelia (Benos 1982; Garty and Benos 1988). In all these vertebrate tissues the fast and reversible action of amiloride is achieved with half-maximal inhibitor concentrations (1(,5.1 J.lmol · l- 1) corresponding to an intensive blockerfchannel interaction. Recently in some vertebrate tissues amiloridesensitive Na+ channels with a Ki~ l J.lmol · 1- 1 were found and identified as a possible new dass of amiloridesensitive Na+ channels (Smith and Benos 1991). Although there have been reports of arniloride-blockage of ion transport in invertebrate epithelia (Kirschner et al. 1973; Schwarz and Graszynski 1990), the identification and analysis of invertebrate amiloride-sensitive Na+ channels has just begun (Onken et al. 1991; Dannenmaier et al. 1991 a, b; Zeiske et al. 1992). The regulation of amiloride-sensitive Na+ transport in many vertebrate epithelia is mediated by peptide hormonessuch as ADH, via the second messenger cAMP. Therefore, exposure of tissues to ADH, cAMP or IBMX (the latter a blocker of cAMP degradation) stimulates Na+ transport (Orloff and Handler 1962). Any stimulation of Na+ transport by cAMP may therefore indicate the presence of hormones in unknown epithelia which interact with the second messenger cAMP. The effect of cAMP on Na+ uptake consists of an increase in the number of conducting Na+ channels, presumably by activation of existing but non-conducting "silent" or "lazy" channels (Hetman et al. 1983; Lindemann 1984). Garty and Edelman (1983) argued that in toad bladder new Na+ channels arise by fusion of channel-containing cytoplasmic vesicles with the apical membrane, whereas Krattenmacher and Clauss ( 1988) showed for frog colon that a rapid increase in I.c after Stimulation was likely to be not produced by fusion of cytoplasmic vesicles but by activation of pre-existing Na+ channe1s in the apical membrane. The present study demonstrates, in agreement with Smith and Benos (1991), the existence of a channel that seems to belong to the dass of low amiloride affinity channels differing from the typical high amiloride affinity channels in some ofits kinetic properties. Higher specific-
-- - --- _ . . __.._____
- -- -
\,
\,
W.-M. Weber et al.: Leech epithelial Na+ channel
ity for benzamil also indicates low affinity for amiloride. Na+ movement through this channel is described, as are amiloride and benzamil binding kinetics and the effects ofcAMP and inhibitors ofcAMP degradation on macroscopic and microscopic channel characteristics. Channel/ amiloride interaction under different mucosal concentrations of Na+ is discussed. Materials and methods Animals and tissue preparation. Leeches ( Hirudo medicinalis) obtained from Altmann, Osnabrück, FRG, were kept without feeding in tap waterat room temperature (22-25 °C}. The experiments were carried out during the period March-November. Animals were hypothermically anaesthetized, the integumentwas dissected by two ventral incisions and the dorsal skin subsequently detached from the intestines. The intemal muscular layers were dissected off by gentle scraping with a scalpel until the diagonal muscle layer could be distinguished . Further dissection was not possible without damaging the tissue. A piece of tissue was then glued with tissue adhesive (Hi ~ toacryl blau, Braun, Melsungen, FRG) with its intemal side to a lucite ring. To disaggregate the collagen fibres which form a specific outer layer in this invertebrate integument and to functionally remove the intemal diagonal muscle layer, the tissue was incubated for I h in 0.1% Collagenase (Collagenase Worthington, Sigma). This treatment did not infiuence the viability or integrity of the integumental epithelium, as judged by V, and R,. The tissue was then mounted in an Ussing chamber with a 0.5-cm 2 aperture specially designed to minimize edge damage. Silicone grease was used to seal the edges on both sides of the tissue. During the whole experiment the two compartments were continuously perfused with Ringer solutions with a fiow rate ofabout 7 ml · min - 1 in the apical compartment, and 3 ml · min-I in the basolateral compartment. Solutions and chemicals. The NaCI Ringer solution contained (in mrnoi · I- 1 ) : ll5 NaCI, 4 KCI, 1.8 CaCI 2 • For Na+.free solution NaCI was replaced by TMA-CI in equimolar concentrations. The solutions were buffered with 5 mmol Hepes · J- 1 and adjusted at pH 7.4 with I mol Trizma-base · I- 1 • Amiloride was added to the mucosal solution in concentrations of 0.5, I, 2, 5, 10, 25, 50 and 100 ~mol · 1- 1 • 8-4-chlorophenylthio-cAMP ( 100 ~mol · I- 1 ) , IBMX (I mmol · l- 1) and ouabain (50 J.lmoi · J- 1 ) were all applied to the serosal solution. All these compounds were obtained from Sigma (München, FRG). Electrical measurements. Voltage- and current-measuring electrodes were Ag-AgCI-wires in a I moi·J- 1 KCI solution and the bathing compartments were connected with I mol · I- 1 KCI agar bridges. For transepithelial measurements, the tissue was voltage-clamped to zero with a low-noise voltage clamp (Van Driessche and Lindemann 1978). Ry was calculated from superimposed 10-mV pulses of 500 ms duration using Ohm's law. Short circuit current {15c) was continuously recorded by a stripchart recorder and a computer (Apple Mac Ilex) with a MacLab interface and a chart recorder program (Analog Digital Instruments, Castle Hili, Australia). In noise experiments amiloride was used as a Na+ channel blocker to induce fiuctuations in Isc. In these experiments the DC and AC components ofthe Isc were separated. The AC component containing the current fluctuations was amplified, digitized, fittered and sampled in a frequency range of0.5-l.328 Hz on a second computer (IBM PS/2). Time-domain signals were fast-Fourier transformed to obtain power density spectra ofthe current noise. Spectra calculated from 20 sequentially collected sweeps (1024 points per sweep) were averaged and stored on disk for later analysis. The required software (noise measurements and analysis) was developed by Van Driessche, Decre and Hendrickx (Leuven, Bclgium). Analy sis of macroscopic blocker kinetics. The maximal inhibitable Na+ current {IN.max) and Ki were deterrnined by the successive
W.-M . Weber et al.: Leech epithelial Na+ channel
"i
155
inhibition of the lsc by increasing amiloride or benzamil concentrations. Amiloride inhibition of lsc was tested at different Na+ concentrations. Analysis ofmicroscopic data. The blocker-induced fluctuations in lsc were analysed with a computer system as previously described (V an Driessche and Erlij 1983). The power density spectra that became visible in the presence of amiloride contained additional background or 1/f noise described by
S(f)=
sj!>
(I)
where S(l) is the spectral density at I Hz and a the slope in the double Iogarithmic plot. Random open-close switching of many Na+ channels induced Lorentzian noise. The frequency dependence of this Lorentzian noise is expressed in the following equation:
so S(f) =
(
I+
[_ fc
)2
+
korr
(3)
where kon can be determined from the slope and korr from the ordinate intercept. With these data we were able to calculate the single channel current (iNJ .
S 0 (nfc)2
INa
= /Na· k o Jarru'I]
_
/Na ·
2nfc
iNa ·
korr
Na -
"Cocktail"• Amiloride Na+-free
-6.25±0.66 -8.76±0.83•
+ 14.93 ± 1.90 -9.40± 1.64 .. -11.05 ± 1.45 ..
Difference to initial resistance (11.6± 1.8 Hl· cm 2 ) +3.13±0.39 4.50±0.8!• -0.33±0.31 +4.24±0.94 +4.50±0.53
• 100 Jlmol cAMP ·J- 1 +I mmol IBMX · ]- 1 • significantly different from amiloride •• significantly different from unactivated tissue
u L{)
t--1
5min
(5)
where /Na is the Na+ current in the presence of a distinct blocker concentration, used as the difference between maximal inhibitable Na+ current and blocker current. Statistical analysis. Results are expressed as means ± standard error of the mean (SEM), n is the number of animals. From each leech only one tissue preparation was taken. Statistical analyses were done by the Student's r-test, with a significance Ievel of Ps;0.05.
R.esults Transepithelial experiments ( macroscopic parameters)
Within 30 min incubation of the voltage-clamped tissue in NaCI Ringersolution on both sides, a stabilization of lsc and RT occurred. The initial VT was -23.I ± 1.4 mV, basolateral side as reference. During this period the initiallsc decreased and levelled at I6.I ± 1.7 ~A · cm- 2 • RT of the integument was on average I0.5±0.5 k!l · cm 2 (Table 1). Figure I demonstrates the time-course of lsc during a typical experiment. Mucosal amiloride (100 ~mol·l- 1 ) decreased Isc and the subsequent decrease in apical Na+ led to further decrease. Addition of Na+ to the mucosal bath caused a transient overshoot of lsc for about I-2 s. The fast transient inward current was followed by a slower decline to the former steady-state Ievel. The fast time-course demonstrates a rapid regulatory mechanism.
a
Amiloride Na+-free
Difference to initial current (16.1 ± 1.7 JlA · cm-2)
(4)
and the Na+ channel density (MN.)
M
Experimental condition
(2)
The plateau value (S0 ) and the corner frequency (fc) are the characteristic parameters of the Lorentzian noise. To obtain the off-rate (k 0 rc) and on-rate (k 0 J ~f the blocker-receptor interaction we have carried out a linear regression analysis of 2nfc on blocker concentra~ion [amil] 2nf=(k 0 n · [amil])
Tablc 1. Changes of lsc and RT after mucosal amiloride application and Na+ deprivation with and without activation by "cocktail"
~-_.-1. . .11---.-----llm"c"al .. Ouabain
. serosal
Fig. 1. Time-course ofa typical experiment showing the dependence of short-circuit current (/5 c) on amiloride (100 Jlmol · J- 1 ) on the mucosal side and removal of Na+ . Final addition of ouabain (50 Jlmol·I- 1 ) to the serosal compartment reduced Isc almost to zero.D NaCl-Ringer; • Amiloride; ll Na+-free Ringer
Allthese effects were reversible with NaCl Ringer solution. Basolateral ouabain (50 ~mol·l- 1 ), a specific irreversible blocker of the basolateral Na+ /K +-A TPase, inhibited the remaining current totally. Amiloridein submicromolar doses is a potent blocker of apical Na+ channels in tight epithelia (Li and Lindemann I983). The amount by which lsc is reduced by this blocker is considered tobe the maximal inhibitable Na+ current (/Namax) through amiloride-sensitive Na+ channels. Addition of I 00 ~mol amiloride · 1- 1 to the mucosal compartment caused a significant 40% reduction of lsc to 9.9± 1.5 ~A · cm- 2 , accompanied by an increase in RT to 13.4±2.1 k!l·cm 2 • Subsequently, removal of Na+ from the apical bath was followed by a further I5% decrease in Isc (Table I) to about 45% of its initial value (7.4± 1.3 ~A · cm- 2 ). The fact that removal ofNa+ after amiloride exposure caused another decrease of lsc is evidence for an amiloride-insensitive Na+ conductance that is involved in the total amount of Na+ current.
W.-M. Weber et al. : Leech epithelial Na+ channel
156
[!) NaCI-Ainger
(~mol · 1" 1)
(~mol · l" 1 )
•
100
~
~
80
c: 0
a0
f----.l.-
60
Ul
5min· ·
(])
0-----------------------------------------•••••••••••••••• ••••••••••••••••
Benzarnil
••••••••••••••••
.... •••• ••••
a:: +
ra
z
Fig. 2. Time-course of the successive decrease in lsc by addition of increasing concentrations of amiloride on the mucosal side. After washout with normal Ringer increasing concentrations of benzamil were added to the mucosal solution
40
20
0~~----~----~~----~----~-----L~
20
0
40 [Na•]
•
~
~
75
0
c:
60
80
100
(mmol-1" 1)
Fig. 4. Dependence of lsc on mucosal Na+ concentration. lsc corresponding to the Na+ concentration is shown in percent of total current at 115 mmo1 Na -1 - 1 on the mucosal side. Line was fitted using the Michaelis-Menten equation (KM = 5.9 mrno1·1- 1 , vm.. =90 mmo1 · l- 1 )
Amiloride Benzarnil
g ii .s::.
.s ö
50
c
~
::l
E
E
