Mechanism of active chloride secretion by shark
rectal gland: role of Na-K-ATPase
in chloride transport
PATRICIO SILVA, JEFFREY STOFF, MICHAEL
FIELD,
JOHN N. FORREST, AND FRANKLIN
H. EPSTEIN
LEON FINE,
SILVA,
PATRICIO,
JEFFREY
STOFF, MICHAEL
FIELD, LEON
FINE, JOHN N. FORREST, AND FRANKLIN
H. EPSTEIN. Mecha-
secretion by shark rectal gland:
role of
nism of active chloride
Na-K-ATPase
in chloride
transport.
Am. J. Physiol.
233(4):
F298-F306,
1977 or Am. J. Physiol.:
Renal Fluid Electrolyte
Physiol.
2(4): F298-F306,
1977. -The
isolated
rectal gland of
Squab
acanthias
was stimulated
to secrete chloride
against
an electrical
and a chemical
gradient
when perfused in vitro
by theophylline
and/or dibutyryl
cyclic AMP.
Chloride
secretion was depressed by ouabain
which inhibits
Na-K-ATPase.
Thiocyanate
and furosemide
also inhibited
chloride
secretion
but ethoxzolamide,
a carbonic
anhydrase
inhibitor,
did not.
Chloride
transport
was highly
dependent
on sodium concentration
in the perfusate.
The intracellular
concentration
of
chloride
averaged 70-80 meq/liter
in intact glands, exceeding
the level expected at electrochemical
equilibrium
and suggesting active transport
of chloride
into the cell. These features
suggest a tentative
hypothesis
for chloride
secretion
by the
rectal gland in which the uphill
transport
of chloride
into the
cytoplasm
is coupled
through
a membrane
carrier
to the
downhill
movement
of sodium along its electrochemical
gradient. The latter is maintained
by the Na-K-ATPase
pump
while chloride is extruded
into the duct by electrical
forces.
electrochemical
carrier
equilibrium;
Squalus
acanthias;
membrane
GLAND OF THE spiny dogfish, Squalus acanthias, secretes fluid with a high concentration of sodium
THE RECTAL
chloride, thus providing an efficient mechanism for the
excretion of salt in the interest of homeostasis (4, 5, 6).
The gland carries out active secretion when perfused in
vitro and under these circumstances chloride appears
to be transported against both an electrical and a
chemical gradient (18, 42, 43). The perfused rectal
gland, therefore, provides a model for study of the
mechanisms of active chloride transport. These mechanisms were examined in the present series of experiments.
We have recently determined that secretion by the
perfused rectal gland is modulated by the adenylate
cyclase-cyclic AMP system, since the addition of theophylline or dibutyryl cyclic AMP to the perfusate immediately increases the basal level of chloride secretion
F298
School
and
several times (45). With the addition of these agents,
the response of rectal gland secretion to several different inhibitors
of ion transport can be more easily
studied. In addition, the intracellular
composition of
rectal glands in vivo and in vitro has been determined.
The results permit the formulation of a general hypothesis of chloride transport linked to the operation of the
Na-K-ATPase pump.
METHODS
Spiny dogfish, Squalus acanthias,
of either sex,
weighing 2-6 kg, were caught by hook and line in
Frenchman’s Bay, Maine. The animals were kept in
marine livecars without food until sacrificed, usually
within 4 days of capture. After segmental transsection
of the cord the rectal gland was removed via a lower
abdominal incision and its artery immediately cannulated with a PE-90 polyethylene catheter. After the
arterial cannula was tied in place the perfusion was
started. Cannulation of the rectal gland duct and vein
with PE-90 polyethylene catheters was then performed,
after which the gland was transferred into an aluminum and Plexiglas perfusion chamber kept at 16 t l°C
by running seawater. The gland was perfused from an
oxygenated reservoir by gravity flow at a pressure of
approximately
4 mmHg and a flow rate of 3.5-9 ml/
min. The perfusion solution (shark-Ringer) contained
(in millimoles per liter): Na, 280; K, 5; Cl, 270; bicarbonate, 8; Ca, 2.5; Mg, 1.2; phosphate, 1; sulfate, 0.5; urea,
350. The pH was 7.6 when gassed with 99% O2 and 1%
CO,. Glucose (5 mM) was used as the sole exogenous
substrate. The rectal gland secretion was collected at
timed intervals in 1.5ml conical centrifuge tubes or,
when volume was small, in loo-p1 disposable pipettes.
All rectal gland vein effluent was collected for determination of flow rate and electrolyte concentration. Arterial perfusate samples were obtained through a selfsealing rubber tube placed just proximal to the arterial
cannula. Transglandular
potential differences were
measured with 1 M KC1 agar bridges previously equilibrated with perfusate solution and an electronic voltmeter (Hewlett-Packard
410C) equipped with two calome1 electrodes. The tip of one bridge, connected to the
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
Department
of Medicine and Thorndike
Laboratory,
Harvard
Medical
Beth Israel Hospital,
Boston, Massachusetts;
and Mount Desert
Island Biological
Laboratory,
Salsbury
Cove, Maine
CHLORIDE
SECRETION
BY SHARK
RECTAL
F299
GLAND
completeness
and stability of the labeling of the extracellular
space. The sections were then lightly blotted
on filter paper, weighed, placed in a scintillation
vial,
and digested with 0.4 ml of Nuclear Chicago solubilizer
for 24 h. Separate samples were digested in hot concenmeasuretrated nitric acid for sodium and potassium
ments or boiled in distilled water for chloride determinations.
Ouabain
(K & K Laboratories),
furosemide
(Lasix,
Hoechst
Pharmaceuticals),
sodium thiocyanate,
and
ethoxzolamide
were dissolved in perfusate solution and
added directly
to the perfusion
reservoir.
Dibutyryl
cyclic AMP (Sigma Chemical
Co. or Calbiochem)
and
theophylline
were also previously
dissolved in perfusate
solution and added directly to the perfusion
reservoir.
Sodium and potassium
were measured in an Instrumentation
Laboratory
343 flame photometer.
Chloride
was measured in a Buchler Cotlove chloridometer.
Results are expressed as means t SE of the mean.
Statistical
significance
was determined
by Student t
test or pairedt
test wherever
applicable.
RESULTS
Perfusion
of Unstimulated
Rectal
Gland
The initial secretion rate of 10 isolated rectal glands
perfused with shark-Ringer
solution was 568 t 96 &h
per g wet weight. The secreted fluid contained 449 t 18
meq/liter
of sodium, 11.7 +_ 0.7 of potassium, and 446 t
11 of chloride. The concentration
of these electrolytes
in the perfusing
solution was: Na, 280; K, 5; and Cl,
270 meq/liter.
The osmolalities
of the perfusion solution
and the fluid secreted by the gland were equal, the
difference
in electrolyte
con .centrati .on being balanced
by the high concentration
of urea in th .e perfusate. The
potential-difference
across the gland, measured
in 30
perfused
rectal glands, ranged from 0.5 to 19 mV,
lumen negative,
averaging
6.2 t 4.7. After the initial
15 min of perfusion,
rectal gland secretion
rapidly
declined to 35-40% of its initial value, and diminished
more slowly over the course of the first hour without
change in the concentration
of electrolytes.
The average
rate of decline observed in 10 glands is shown in Fig. 1.
Chloride
excretion fell from 254.7 t 45.0 to 41.3 t 7.7
peq/h per wet wt. These results resemble the results
reported previously
for this preparation
(17, 42, 43).
Perfusion
Dibutyryl
with Theophylline
Cyclic AMP
and
Both theophylline
(0.01 and 0.5 mM> and dibutyryl
cyclic AMP (0.05-O. 2 mM) evoked a rapid rise in the
volume of rectal gland secretion with little change in
its electrolyte
composition
and induced an increase in
the negative voltage recorded from the duct. Perfusion
of the gland with a combination
of theophylline
and
dibutyryl
cyclic AMP enhanced their stimulatory
effect
and resulted
in a reasonably
stable and sustained
stimulation
of the gland (Fig. 2).
Figure 3 shows the transglandular
potential
differencebefore
and after stimulation
of the gland with 0.25
mM theophylline
and 0.05 mM dibutyryl
cyclic AMP in
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
exploring
electrode,
was immersed
in the receptacle
collecting secreted fluid. The tip of the bridge connected
to the reference electrode was either submerged
in the
solution contained in the perfusion
reservoir
or placed
in the fluid surrounding
the perfused rectal gland in
close proximity
to its surface. Placement of this bridge
in either of these positions did not show difference
in
the PD measurements
across the gland.
Sodium-potassium-ATPase
activity was measured in
whole homogenates
of fresh and perfused rectal glands.
The glands were dissected free of connective
tissue,
cut, weighed, and homogenized
with a Teflon pestle in
a glass homogenizer
in a 20/l vollwt
homogenizing
solution containing:
0.25 M sucrose; 20 mM imidazole; 6
mM EDTA; and 0.1% wt/vol sodium deoxycholate.
The
assay was done in Erlenmeyer
flasks in 5 ml of a
reaction mixture
containing:
NaCl, 100 KCl, 20 mM;
imidazole, 10 mM; MgCl*, 6 mM; ATP, 6 mM; and 50 ~1
of enzyme suspension;
at pH 7.8. The reaction
was
started by the addition of MgCl* and ATP and incubated
for 15 min at 37OC in a shaking metabolic
incubator.
The reaction was stopped by the addition of 1 ml of icecold 35% trichloroacetic
acid. After centrifugation
the
supernatant
was assayed for inorganic
phosphate.
NaK-ATPase
was defined as the difference
in inorganic
phosphate liberated in the presence or absence of potassium. Results are expressed as micromoles
of inorganic
phosphate liberated
per milligram
of protein per hour.
Intracellular
electrolyte
content was determined
in
rectal glands in vivo and in vitro by labeling
the
extracellular
space with [l-‘C]inulin.
For the in vivo
measurements
11 dogfish were each injected with a
single intravenous
bolus of 20 &i of [lC]inulin
as a
marker of extracellular
space. Eight hours later blood
was collected from the dorsal aorta and the animals
were sacrificed. The rectal gland was then extirpated,
dissected free of connective
tissue, weighed,
divided
into lOO- to 200-mg pieces, and placed in tared glass
vials. Wet and dry weights were determined
and the
tissues were either digested in hot concentrated
nitric
acid or boiled in distilled
water and homogenized
in a
glass homogenizer
with a Teflon pestle. The [l-‘C]inulin
concentration
was determined
in aliquots of tissue and
plasma after digestion in Nuclear Chicago solubilizer,
and the extracellular
space was calculated
from these
values. For the in vitro measurements,
the rectal glands
were removed and perfused with a medium containing
0.1 &i/ml
of [lC]inulin
for at least 30 min. Thereafter
the glands were processed as described above. In separate in vitro measurements,
the rectal glands were
removed from five dogfish and sectioned with a StadieRiggs microtome.
Sections approximately
200 pm in
thickness
and 4-5 mm in diameter
were examined
under a dissecting microscope to ensure that the individual glandular
tubules were cut in cross section, the
open central lumen surrounded
by cells giving a doughnut-like appearance.
The sections were then transferred
to Erlenmeyer
flasks containing
4 ml of shark-Ringer
solution
and 50 &i/ml
of [lC]inulin,
at 15°C and
bubbled with air. After 15 min of incubation
three
sections were removed
while another
section was allowed to incubate
for a total of 30 min to ensure
F300
SILVA
700
1
and 0.05 mM dibutyryl
cyclic AMP were used in subs_equent studies of the effect of inhibitors.
I-L
Effect
600
ab
I
200
1
too
i
15
30
60
45
MINUTES
2000
I
0
fluid secretion in 10 rectal glands perfused
at 15-min intervals.
Initial
secretory
rate
1st h to values of less than 20% of initial
t SE.
1
1500
1000
G
500
1
04
1
I5
r
30
I
45
I
60
MINUTES
2. Secretion of rectal glands when stimulated
with theophylline and DBcAMP.
When rectal glands were perfused in vitro with
theophylline,
0.25 mM, and dibutyryl
cyclic AMP, 0.05 mM, rate of
secretion of chloride was not only greatly stimulated
above initial
control value but was maintained
constant over a period of at least
1 h. Values are means * SE. Number of experiments
was 12.
FIG.
Control
Theophyl
Dibutyryl
line
cyclic AMP
3. Effect of secretory
stimulation
on potential
difference
across rectal gland. Potential
difference
(mV, duct negative)
is
shown before and after addition
of theophylline,
0.25 mM, and
dibutyryl
cyclic AMP, 0.05 mM. Stimulation
of perfused
rectal
glands was followed in every case by an increase in negative PD.
FIG.
13 glands. The PD in the basal state was 6.8 t 0.9,
duct negative,
and rose after stimulation
to 15.0 t 1.6
mV. Glands exposed for the duration of the experiment
to 0.25 mM theophylline
or to 0.25 mM theophylline
of Ouabain
The addition
of ouabain
to homogenates
of rectal
gland inhibited
Na-K-ATPase
activity by an average of
50% at a concentration
of lOweM, while complete inhibition was found at a concentration
of 10m4M. When
10e4M ouabain was added to solutions perfusing rectal
glands in vitro, Na-K-ATPase
activity in gland homogenates, prepared
and incubated
in the absence of the
glycoside, was reduced to 20% of control levels (50.5 t
4.1 PM Pi/mg protein per h in four control glands vs.
9.1 t 10.5 PM Pi/mg protein
per h in four glands
perfused with 10e4M ouabain).
The addition of 10m4M
ouabain
to the perfusing
medium
sharply
inhibited
rectal gland secretion,
producing
a fall in secretory
volume as well as a decrease in the concentration
of
sodium and chloride (Table 1 and Fig. 4). The transglandular potential
difference
also decreased, falling from
17.9 t 2.7 to 4.5 t 1.4 mV (n, 8, P < 0.01). Even after
inhibition
with ouabain a small amount of duct fluid
continued
to be formed at about the same level seen in
resting glands unstimulated
by theophylline
or dibutyryl cyclic AMP. The inhibitory
effect of ouabain was
not reversed by perfusion
with ouabain-free
solution,
probably reflecting
the tight binding of ouabain to cell
membranes.
Effects of Thiocyanate,
and Ethoxzolamide
Furosemide,
Because both thiocyanate
and furosemide
inhibit
chloride transport
in different
tissues it was of interest
to study their effect on rectal gland secretion. Thiocyanate added to the perfusate as 10 mM NaSCN inhibited
sodium chloride secretion by 60%, the chloride output
falling from 702 t 106 to 293 t 55 peq/h per g wet wt
(Fig. 5). This inhibitory
effect was manifested
mainly
by a change in secretory fluid volume, with little or no
change in its electrolyte
composition
(Table 2). The
effect of thiocyanate
was only partially
reversible,
secretion returning
toward but not quite to control rates
when thiocyanate
was removed from the perfusate.
Furosemide,
10v4M, reversibly inhibited rectal gland
secretion, reducing fluid secretory rate to 40% of control,
without
change in the electrolyte
composition
of the
fluid. Chloride
secretion (Fig. 6) fell from 691 t 94 to
250 t 41 peq/h per g wet wt after the addition of the
drug. Removal
of furosemide
from the perfusate
returned chloride secretory rate to 760 t 91 peqlh per g
wet wt.
The rectal gland of the spiny dogfish contains substantial amounts of carbonic anhydrase (28). Nevertheless, when ethoxzolamide,
a potent inhibitor
of carbonic
anhydrase,
was added in a supramaximal
dose (J-O-4 M)
to the perfusate of three theophylline-stimulated
rectal
glands, their secretory rate was unaltered.
Dependence
of Secretion
on Sodium
in Perfusate
The importance
of sodium for glandular
secretion is
illustrated
by Fig. 7, which shows the result of substi-
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
1. Basal rate of
in vitro is shown here
declined over course of
rate. Values are means
FIG.
3
3
P
<
sa
ET AL.
CHLORIDE
TABLE
SECRETION
1. Effect
BY SHARK
RECTAL
of ouabain on secretory
---_
rate and electrolyte
Secretory
Time,
______-
Control
Ouabain,
min
-I__----~
Vol,
O-30
lo+
M
30-60
ml/h
per
g wet
peq/h
per
+
\
600
8
a
I
0
400
of Secreted
K , meq/li
meq/liter
Fluid
ter
Cl,
meq/liter
2.15 2 0.37
(17)
909.3 + 156.1
(17)
419.7 2 7.8
(17)
10.2 2 0.7
(17)
0.42 + 0.07*
(171
163.0 + 30.3*
(17)
391.3 + 12.9t
(17)
12.2 + 1.0
(17)
391.5 2 12.u
(17)
in parentheses.
*P < 0.0005.
0
CONTROL
OUABAIN
IO-4 M
4. Effect of ouabain on chloride secretion
by rectal gland.
Ouabain,
10eJ M, sharply
reduced rate of chloride secretion
in 17
rectal glands previously
stimulated
with theophylline,
0.25 mM,
and dibutyryl
cyclic AMP, 0.05 mM. Values are means + SE.
FIG.
1000,
8001
I
Control
Glands
tP
were
< 0.025.
stimulated
throughout
the experiment
with
0.25 mM
$ P < 0.01 .
glands was directly proportional
to the concentration
of
sodium in the perfusate.
As sodium concentration
varied from 0 to 70, 140, and 280 mM, chloride secretion
increased progressively
from close to 0 to 735.2 t 154.8
peq/h per g wet wt (r = 0.77; P < 0.001).
Additional
evidence of sodium dependence
was obtained
by ubstituting
Tris for sodium.
When Tris
replaced sodium in the medium perfusing
the stimulated rectal gland, chloride
secretion also decreased,
from 1,629 t 224 to 134 t 36 peqlh per g wet wt (n, 4).
Restoring
sodium concentration
to 280 mM increased
secretion to 1,404 t 123 peq/h per g wet wt, as in the
experiments
with choline chloride.
_-
2 600
.
\
I
w” 400’
3
1
& 200’
i
0i
Na,
wt
426.1 + 8.9
(17)
200
$
g wet
fluid
Post-SCNSCNIO mM
FIG.
5. Effect of thiocyanate
on chloride secretion by rectal gland.
Thiocyanate,
lo--” M, added to perfusion
medium as NaSCN inhibited by about 60% rate of chloride secretion
in stimulated
perfused rectal glands. Removal of SCN from perfusion
medium returned secretion toward but not quite to control values. Values are
means 4 SE of 7 experiments.
tuting
choline for sodium in the perfusion
medium.
Secretion of chloride averaged
877 t 272 peq/h per g
wet wt when perfused with a normal (280 mM> sodium
concentration.
In the absence of sodium (osmolality
maintained
constant with choline chloride),
chloride
secretion
dropped to 66 t 29 peq/h per g wet wt,
representing
essentially
complete inhibition.
Chloride
secretion returned to the previous level (885 t 387 peq/
hr per g wet wt) when sodium concentration
in the
perfusate was restored to 280 mM. Dependence
of chloride secretion on the sodium concentration
of the perfusate is further demonstrated
by the experiment
shown
in Fig. 8. Secretion of chloride by five perfused rectal
Intracellular
Composition
of Rectal
Gland
In intact rectal glands from 11 live dogfish,
the
intracellular
concentration
of potassium was about 150
meq/liter
and that of sodium about 20 meq/liter
(Table
3). The calculated intracellular
concentration
of chloride
greatly exceeded that of sodium.
b)78.9 t 7.1 meq/liter)
Inulin space was 26.7 t 0.5%.
In perfused rectal glands stimulated
with 0.25 mM
theophylline
and 0.05 mM dibutyryl
cyclic AMP, calculated intracellular
sodium concentration
was 26.6 t 5.3
meq/liter
and potassium 127.1 t 6.2 meq/liter.
Intracellular chloride concentration
was 69.7 t 5.6 meq/liter,
a
value not significantly
different from that in intact fish.
Because inulin is excluded from the tubular
lumina
of the intact rectal glands, it was thought this might
interfere
with the estimation
of intracellular
ionic content. Accordingly,
freshly harvested
rectal glands of
five dogfish were sectioned with a microtome
so that
the individual
glandular
tubules were cut in cross
section, giving
a doughnut-like
appearance
when inspected through
a dissecting
microscope.
Sections of
200 pm thickness were then incubated with radioactive
inulin in oxygenated
shark-Ringer
solution for 15 or 30
min as detailed in METHODS.
There was no difference
between the inulin space at 15 and 30 min, suggesting
that isotopic labeling of both extracellular
and luminal
fluid was achieved.
The inulin
space averaged
29 t
0.8% of wet wt, slightly higher than in intact glands.
Calculated
intracellular
concentration
of sodium was
42.1 +- 3.8, potassium
161.6 t 8.5, and chloride 116 t
10.6 meq/liter.
DISCUSSION
The rectal gland appears to have evolved in elasmobranch fishes living
in seawater as a way to excrete
excess salt on behalf of the constancy of the internal
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
800
of rectal gland
Compoaition
Cl,
IO00
3
3?
composition
Rate
wt
Values are means 2 SE. Number
of observations
theophylline
and 0.05 mM dibutyryl
cyclic AMP.
I200
F301
GLAND
F302
TABLE
SILVA
2. Effect
and electrolyte
of furosemide,
thiocyanate,
and ethoxdamide
composition
of rectal gland fluid
Secretory
Time,
Control
min
O-30
Thiocyanate,
10d2 M
30-60
Recovery
60-90
Control
O-30
Furosemide,
lOed M
Recovery
low4 M
per
g wet
Composition
Rate
wt
Cl,
peqlh
Na,
wt
Cl,
meq/liter
1.39 + 0.18
690.6 + 93.9
490.2 + 23.7
9.9 2 1.1
497.9 2 10.3
2 0.17t
30-60
0.54 t 0.09
249.8 + 41.4*
(8)
(8)
60-90
1.58 + 0.18*
(5)
760.0 iz 90.6*
(5)
484.7 f 17.6
(5)
8.7 + 0.8
(5)
477.3 + 6.7
(5)
3.14 2 0.88
(3)
3.33 2 0.89
(3)
1672.4 + 497.7
(3)
1699.2 + 473.0
(3)
529.8 +_ 44.8
(3)
540.5 + 39.6
(3)
15.6 + 2.5
(3)
15.9 IL 1.6
(3)
527.5 2 12.7
(3)
508.5 + 8.0
(3)
is the number
cyclic AMP.
475.0 + 17.3
10.9 2 1.5
(8)
461.4 + 6.6
(8)
(8)
in vitro and stimulated
by 0.25
t P < 0.05 when compared
The conclusion that chloride is actively transported
by the rectal gland is strengthened
by the increase in
transglandular
potential,
duct lumen negative,
that
occurred in every case when secretion was stimulated
by theophylline.
(It should be appreciated
that the
actual potential
difference
across the glandular
epithelium in the tubule at the site where secretion takes
place may not be precisely indicated
by the present
measurements
in which voltage was recorded in the
main rectal gland duct.) An increase in the transport of
Post-Furosemide
V
by rectal gland.
chloride secretion
of drug reduced
its removal from
to control values.
environment. The glandular
secretion of live dogfish
contains sodium and chloride at approximately
the
concentration of seawater. The organ is said to regress
in elasmobranchs living in fresh water (33). Extirpation
of the rectal gland of the seawater spiny dogfish,
Squalus acanthias, produces a progressive rise in
plasma sodium (14), and it is possible to evoke secretion
by the gland in live dogfish by injections of hypertonic
sodium chloride into the bloodstream (4).
The isolated rectal gland can be perfused easily in
vitro, and it has been established that a small basal
secretion can be elicited, dwindling with time, in which
chloride moves against both an electrical and a chemical
gradient (18,42-44). In previous experiments this basal
secretion did not appear to be inhibited by either
ouabain or furosemide (18, 44), although it is sensitive
sodium
(8)
of observations.
All glands were perfused
* P < 0.01 when compared with controls.
v
Control
(8)
(8)
(8)
30-60
of
Fluid
(7)
0.93
FIG.
6. Effect of furosemide
on chloride secretion
Furosemide,
10m4 M, reversibly
inhibited
rate of
in stimulated
perfused
rectal glands. Addition
secretory
rate of chloride to 40% of control, and
perfusate was followed by restoration
of secretion
Values are means + SE of 8 experiments.
concentration
ter
478.7 k 106.1
(7)
426.8 +_ 22.9*
(7)
427.7 AI 27.9
(7)
U4M
the
of Secreted
K, meq/li
9.4 iI 1.4
(7)
8.6 + 0.8
(7)
8.5 + 2.0
(71
Furosemide
to
meq/liter
443.3 2 22.9
(7)
414.6 + 33.3
(7)
404.6 + 33.5
(7)
looo-
O*
g wet
702.6 + 106.1
(7)
293.4 * 55.4*
(7)
412.1 + 78.3
(7)
1.38 t 0.19
(7)
0.66 + 0.13*
(7)
Values are means + SE. Number in parentheses
mM theophylline
alone or with 0.05 mM dibutyryl
with preceding period.
7L600
<
po
:-‘#
b 200
per
(8)
O-30
Ethoxzolamide,
ml/h
rate
and
chloride
in
the
perfusate (42). The discovery that glandular activity is
greatly stimulated by theophylline and cyclic AMP has
permitted a closer look at the mechanisms of active
secretion.
1200
3 IO00
3
\
+
$=
800
600
200
Control
Choline
Chloride
Post-Choline
Chloride
FIG.
7. Dependence
of rectal gland secretion on presence of sodium in perk&e.
Substitution
of choline for sodium in perfusate of
rectal glands stimulated
with theophylline,
0.25 mM, and dibutyryl
cyclic AMP, 0.05 mM, dropped chloride secretory
rate to less than
10% of control. When sodium concentration
was restored to normal,
chloride secretion returned
to control level. Values are means + SE
of 5 experiments.
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
Control
Vol,
on secretory
ET AL.
CHLORIDE
SECRETION
BY SHARK
RECTAL
F303
GLAND
limb of Henle’s loop (38) and in the cornea (52), while
thiocyanate
interferes
with chloride transfer
in gastric
mucosa (19), cornea (52), and teleost gill (12). Although
the rectal gland contains considerable
carbonic anhydrase, the secretion
of chloride was unaffected
by a
carbonic anhydrase
inhibitor,
in confirmation
of previous work using the unstimulated
gland (18). This
contrasts
with the effect of acetazolamide
to inhibit
secretion
of salt by the avian salt gland (29), the
pancreas,
gastric mucosa, ciliary
body, and choroid
plexus (28), and to block inward transport
(absorption)
of chloride across the ileal mucosa (24, 32), amphibian
skin (l), and teleost gill (12). Carbonic anhydrase inhibitors do not, however,
inhibit
outward
secretion
of
chloride by the gills of marine teleosts (12) or theophylline-stimulated
chloride secretion by intestinal
mucosa
(32)
The striking inhibition
of secretion produced by ouabain was of special interest,
since ouabain had previously failed to affect low levels of secretion by resting
glands. The rectal gland is particularly
rich in Na-KATPase (3), and since inhibition
by cardiac glycosides
blocks CAMP-stimulated
secretion, it may be supposed
that the enzyme plays a key role in the secretory
process. The mechanism
by which this occurs poses a
dilemma,
in part because of the anatomical
location of
Na-K-ATPase
on the surface of rectal gland cells. These
cells have extensive basal and lateral infoldings
facing
the extracellular
fluid and blood, rather than the duct
lumen. Autoradiographic
studies with radioactive
ouabain indicate that Na-K-ATPase
is localized to these
basolateral
infoldings
(20); sodium would therefore
be
pumped out of the cell into the blood, in a direction
opposite to that in which glandular
secretion actually
takes place. In this respect the rectal gland resembles
the chloride cells of the gills of seawater teleosts (21),
the salt gland of birds (13), or the mammalian
salivary
gland (35).
A further point of interest in the present experiments
is the dependence of glandular
secretion on the concentration
of sodium in the perfusate.
Chloride
was not
secreted in the absence of sodium and the rate of
secretion produced by theophylline
increased with increasing sodium concentration.
This was also true of
CAMP-stimulated
secretion by intestinal
mucosa (30,
1000
3
3
\
800
600
c
0
PERFUSATE
70
140
280
No’ CONCENTRATION
mEq/L
8. Chloride secretion by perfused rectal gland is proportional
to concentration
of sodium in perfusate.
As sodium concentration
rose from 0 to 70, 140, and 280 mM the rate of chloride secretion
increased. Osmolality
was maintained
constant with choline cloride.
Values are means + SE of 5 perfused rectal glands.
FIG.
3. Intracellular
TABLE
electrolyte
composition
of rectal gland
In Vivo
In Vitro
Perfused
Nitric
acid
digest
24.8
K, meqlliter
+ 6.0
(5)
156.3 + 1.6
(5)
Cl, meqlliter
17.9 + 2.5
acid
26.6
(6)
147.7
f 3.9
digest
iz 5.3
127.1
(6)
26.7
+ 0.5
Values
are means
+ SE. Number
of observations
digest
+ 3.8
161.6
t 8.5
(5)
116 + 10.6
(5)
(7)
in 0.9
Leaching
(5)
+ 6.2
(7)
in parentheses.
acid
69.7 zt 5.6
27.1
(5)
Nitric
42.1
(7)
78.9 + 7.1
space, % wet wt
Slices
Leaching
(5)
(6)
Inulin
gland
Leaching
Nitric
Na, meq/liter
stimulated
29.0 + 0.8
(5)
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
chloride
was thus evoked by theophylline
or CAMP
against a steeper electrical
(and sometimes
a larger
chemical)
gradient
than had previously
existed. The
electrical gradient
for sodium, on the other hand, was
always changed during
stimulated
secretion so as to
favor the passive movement
of Na+ from capillary
to
duct lumen. The average chemical gradient
opposing
passive diffusion
of Na+ from capillary
to duct lumen
(duct Na+/perfusate
Na+ = 1.7) would be counterbalanced by a potential
difference
of -lOmV,
duct lumen
negative.
Potential differences
more negative than this
strongly
suggest passive movement
of Na+ into the
lumen.
Secretion
of salt against
a mucosa-negative
potential difference,
implying
anion secretion, has also
been reported for the small intestine
(37), colon (16),
the gill of marine teleosts (25), pancreatic
intra- and
extratubular
ducts (47, 48), biliary ducts (9), and salivary acini (49). In these organs, as in the rectal gland,
mucosal negativity
increases with stimulation
of electrolyte secretion.
Chloride secretion by the rectal gland was inhibited
by furosemide
and thiocyanate.
Both substances also
inhibit active chloride transport
in other tissues. Furosemide blocks :hloride transport
in the thick ascending
F304
ET AL.
the contraluminal
cell border, together with the transport of Na+ by the Na-K-ATPase
pump into the basolatera1 spaces constitute
the operational
equivalent
of a
chloride
pump that actively
transports
Cl- into the
cell. Chloride could be extruded across the luminal cell
border by electrical
forces. Recycled Na+, returned
to
the lateral spaces by the sodium extrusion pump, would
diffuse down its electrical
gradient
into the lumen.
This model does not account for the low level of basal
secretion of chloride present in the unstimulated
gland,
which is not inhibited
by ouabain.
The attraction
of this hypothesis (schematized in Fig.
9) is that it accounts for the transport
of chloride by
rectal gland cells against an electrochemical
gradient
in a way that is linked to and indirectly
energized by
Na-K-ATPase.
The function
of the enzyme is to maintain a low intracellular
concentration
of sodium that
facilitates
the downhill
entry of sodium into the cell,
and a high intracellular
concentration
of potassium.
The latter is responsible
in large part for the negative
intracellular
electrical potential
that extrudes chloride
into the gland lumen and also serves as a force favoring
the passive entry of sodium into the cell. The location
of Na-K-ATPase
on the contraluminal
side of the cell
does not pose a logical dilemma
in this system. The
Extracel Mar
/vu+
Intracellular
Duct Lumen
450
280
20
K+
5
150
10
cl-
270
70
460
ml/
0
-70
-15
Nu - K-ATiPase
K+
Na+
Lhked /Vu - Cl Cuffief -clNa+
Electrochemical potential
across peritubular membrane
Electrochemical potential
across
luminal iem brane
-~
cl-
(Opposing)
36.5
mV
8.3 mV (Favoring)
No+
(Favoring)
135.5 mV
132.2 mV (Opposing)
I
I
FIG. 9. Schematic
model for movement
of chloride across rectal
gland epithelium.
Passive
ion movements
are shown by dotted
lines; active transport
by solid arrows.
A neutral sodium chloride
carrier located in basolateral
cell membrane effects active movement
of chloride into cell, coupled to downhill
movement of sodium. Low
intracellular
sodium concentration
and large downhill
electrochemical gradient for sodium is maintained
by activity of Na-K-ATPase.
Chloride
diffuses passively
from cell into tubular
lumen down an
electrical gradient. Sodium moves down its electrochemical
gradient
into tubules through paracellular
pathways,
though an Na-K-ATPase pump on luminal cell border is not excluded. Lower two columns
represent
the electrochemical
potentials
(EC) for chloride and sodium across peritubular
and luminal
membranes,
respectively.
Calculations
are based on Nernst
equation
where EC -= PD +
chemical potential,
and chemical
potential
= (RT/zF)
In (C/C).
Values for PD and electrolyte
concentrations
(mM) in extracellular,
intracellular.
and ductal fluid are shown in unner columns.
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
36), as well as of pancreatic
(7), and salivary
gland
secretion (34). Chloride
transport
by the cornea (50),
gastric mucosa (lo), and frog skin (51) was sodium
dependent as well.
If chloride
is actively
transported
by rectal gland
cells, a critical
question
in localizing
the site of the
active process is the concentration
of chloride
within
the cell. The present data indicated
that intracellular
chloride concentration
probably equaled or exceeded 70
meq/liter
in the intact gland and was even higher in
slices incubated
in vitro. The intracellular
concentration of chloride was substantially
higher than that of
sodium. Intracellular
potassium
approximated
levels
seen in mammalian
tissues and in mammalian
and
elasmobranch
muscle with a ratio of intracellular-toextracellular
K+ of approximately
30/l. The ratio of
extracellular-to-intracellular
chloride,
on the other
hand, was between 3/l and 4/l. These values suggested
that the chloride
content of the cell exceeded that
predicted
at electrochemical
equilibrium.
Assuming
that no intracellular
Cl- was bound (measurements
with Cl-sensitive
microelectrodes
would be necessary
to ascertain
this), the intracellular
potential
of rectal
gland cells would have to be -31 mV for chloride to be
in electrochemical
equilibrium
across the basal surface
of the cell. Preliminary
measurements
of intracellular
potential
in cells of rectal gland slices indicated a value
of about -60 to - 70 mV, consonant with the intracellular electrical
potential
of a variety
of other secretory
tissues (2). Chloride
thus appears to be present in
rectal gland cells at a concentration
2-4 times higher
than that expected for passive distribution,
suggesting
that chloride
is transported
uphill
into the cell. In
contrast to the low concentration
of chloride in skeletal
muscle, this high chloride concentration
is reminiscent
of the concentration
seen in the distal renal tubule
(23), salivary
gland (49), avian salt gland (39), mammary gland (27), and intestinal
mucosa (17).
The outstanding
characteristics
of chloride transport
by the rectal gland can be summarized
as follows. 1)
The transepithelial
transfer of chloride proceeds against
an electrochemical
gradient and is, therefore,
an active
process. 2) Intracellular
chloride concentration
exceeds
that expected for electrochemical
equilibrium
with extracellular
fluid, implying
that chloride must be transported uphill into cells at their basolateral
margins. 3)
Chloride transport
is highly dependent on sodium concentration
in the perfusate.
4) The process is blocked
by inhibition
of Na-K-ATPase,
which is located chiefly
on the contraluminal
membrane.
These features
suggest a tentative
hypothesis
for
chloride secretion by the rectal gland that is analogous
to one already developed for the sodium-linked
absorption of glucose and amino acids (40) and of chloride
(31). A membrane
carrier that coupled the movement
of chloride tightly with the inward movement
of sodium
would accomplish
the uphill transport
of chloride into
the cytoplasm. The energy for this process would derive
from the movement
of sodium downhill along its electrochemical gradient,
and ultimately
from the hydrolysis
of ATP by Na-K-ATPase,
which maintains
this gradient. The inward
movement
of neutral
NaCl across
SJLVA
CHLORIDE
SECRETION
BY SHARK
RECTAL
F305
GLAND
was invoked by Keynes (22) to account for an unexpectedly high concentration
of chloride in the giant axon of
the squid, thought to be due to uphill inward transport
of chloride. Chloride
pumps have also been postulated
in crayfish axons (46>, cardiac (26) and smooth muscle
(8) fibers, and toad bladders (11). It seems possible that
coupled sodium-chloride
transport,
the effect of which
is to act as an inwardly
directed chloride pump and
elevate intracellular
chloride, is a general phenomenon,
characteristic
of many kinds of cells, and specialized to
provide directionality
for transepithelial
chloride transport in certain secretory tissues.
The authors
are grateful
to Ms. Katherine
Spokes and Mr.
Arthur
Stevens for invaluable
assistance.
This study was supported
by Public Health Service Grant AM18078 (to F. H. Epstein),
and Grant BG7581 from the National
Science Foundation
(to Mount Desert Island Biological Laboratory).
Received
for publication
5 January
1977.
REFERENCES
R. H., T. H. DIETZ,
AND T. L. MULLEN.
Chloride
transport
across isolated skin of Rana pipiens. Am. J. Physiol.
229: 869-876, 1975.
ARMSTRONG,
W. M. Intracellular
potential in shark rectal gland.
BUZZ. Mount Desert ZsZand Biol. Lab. In press.
BONTING,
S. L. Studies on sodium-potassium-activated
adenosine triphosphatase.
XV. The rectal gland of the elasmobranchs.
Comp. Biochem. PhysioZ. 17: 953-966, 1966.
Burger,
J. W. Further
studies on the function
of the rectal
gland in the spiny dogfish. PhysioZ. 2002. 35: 205-217, 1962.
BURGER,
J. W. Roles of the rectal gland and the kidneys
in salt
and water excretion in the spiny dogfish. Physiol. ZooZ. 38: 191196, 1965.
BURGER,
J. W., and W. N. Hess. Function of the rectal gland in
the spiny dogfish. Science 131: 670-671, 1960.
CASE, R. M., AND T. SCRATCHERD.
The secretion of alkali metal
ions by the perfused cat pancreas as influenced by the composition and osmolality
of the external environment
and by inhibitors of metabolism
and Na+-K+-ATPase
activity.
J. Ph.ysioZ.,
London 242: 415-428, 1974.
CASTEELS,
R. The distribution
of chloride ions in smooth muscle
cells of the guinea pig’s taenia coli. J. PhysioZ., London 214:
225-243, 1971.
CHENDEROVITCH,
J. Secretory
function
of the rabbit common
bile duct. Am. J. PhysioZ. 223: 695-706, 1972.
CUMMINS,
J. T., AND VAUGHN,
B. E. Ionic relations
of the
bioelectric
mechanism
in isolated rat stomach.
Biochim.
Biophys. Acta 94: 280-292, 1965.
DAVIES,
H. E. F., D. G. MARTIN,
AND G. W. D. SHARP.
Differences in the physiological
characteristics
of bladders of toads
from different
geographical
sources. Biochim.
Biophys.
Acta
150: 315-318, 1968.
EPSTEIN,
F. H., J. MAETZ,
AND G. DE RENZIS.
Active transport
of chloride by the teleost gill: inhibition
by thiocyanate.
Am. J.
Physiol. 224: 1295-1299, 1975.
ERNST, S. A. Transport
adenosine triphosphatase
cytochemistry.
II. Cytochemical
localization
of ouabain-sensitive,
potassiumdependent phosphatase
activity
in the secretory
epithelium
of
the avian salt gland. J. Histochem.
Cytochem. 20: 23-38, 1972.
FORREST,
J. N., P. SILVA,
A. EPSTEIN,
AND F. H. EPSTEIN.
Effect
of rectal gland extirpation
on plasma
sodium in the spiny
dogfish. Bull. Mount Desert Island Biol. Lab. 13: 41-42, 1973.
FRIZZELL,
R. A., M. C. DUGAS,
AND S. G. SCHULTZ.
Sodium
chloride transport
by rabbit gall bladder. J. Gen. PhysioZ. 65:
769-795, 1975.
FRIZZELL,
R. A., M. J. KOCH, AND S. G. SCHULTZ.
Ion transport
by rabbit colon. J. Membrane BioZ. 27: 297-316, 1976.
FRIZZELL,
R. A., H. N. NELLANS,
R. C. ROSE, I. MARKSCHEIDKASPI,
AND S. G. SCHULTZ.
Intracellular
Cl concentrations
and
1. ALVARADO,
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
influxes across the brush border of rabbit ileum. Am. J. PhysioZ.
224: 328-337, 1973.
HAYSLETT,
J. P., D. A. SCHON,
M. EPSTEIN,
AND C. A. M.
HOGBEN.
In vitro perfusion
of the dog&h rectal gland. Am. J.
Physiol. 226: 1188-1192, 1974.
HOGBEN,
C. A. M. Active transport
of chloride by isolated frog
gastric epithelium - origin of the gastric mucosal potential. Am.
J. Physiol.
108: 641-649, 1955.
KARNAKY,
K. J., JR., H. CHURCH,
W. B. KINTER,
AND P. SILVA.
Autoradiographic
localization
of “H-ouabain
binding by Na,KATPase in rectal gland of dogfish (SquaZus acanthias).
BUZZ.
Mount Desert Island Biol. Lab. 16: 64-65, 1976.
KARNAKY,
K. J. JR., L. B. KINTER,
W. B. KINTER,
AND C. E.
STIRLING.
Teleost chloride cell. II. Autoradiographic
localization
of gill Na,K-ATPase
in killifish
(FunduZus
heterocZitus) adapted
to low and high salinity environments.
J. CeZZ BioZ. 70: 157-177,
1976.
KEYNES,
R. D. Chloride
in the squid giant axon. J. Physiol.,
London 169: 690-705, 1963.
KHURI,
R. N., S. K. AGULIAN,
AND K. BOGHARIAN.
Electrochemical potentials
of chloride in distal renal tubule of the rat. Am.
J. Physiol. 227: 1352-1355, 1974.
KINNEY,
V. R., AND C. F. CODE. Canine ileal chloride adsorption:
effect of carbonic
anhydrase
inhibitor
on transport.
Am. J.
PhysioZ. 207: 998-1004, 1964.
KIRSCHNER,
L. B., L. GREENWALD,
AND M. SANDERS.
On the
mechanism
of sodium
extrusion
across the irrigated
gill of
seawater-adapted
rainbow
trout
(SaZmo gairdneri).
J. Gen.
PhysioZ. 64: 148-165, 1974.
LAMB,
J. F. The chloride content of rat auricle. J. PhysioZ.,
London 157: 415-425, 1961.
LINZELL,
J. L., AND M. PEAKER.
Intracellular
concentration
of
sodium, potassium
and chloride in lactating
mammary
gland
and their relation
to the secretory
mechanism.
J. Physiol.,
London 216: 683-700, 1971.
MAREN,
T. H. Carbonic
anhydrase:
chemistry,
physiology
and
inhibition.
Physiol. Rev. 47: 595-781, 1967.
NECHAY,
B. R., J. L. LARIMER,
AND T. H. MAREN.
Effects of
drugs and physiologic
alterations
on nasal salt secretion in sea
gulls. J. PharmacoZ Exptl. Therap. 130: 401-410, 1960.
NELLANS,
H. N., R. A. FRIZZELL,
AND S. G. SCHULTZ.
Brush
border processes and transepithelial
Na and Cl transport
by
rabbit ileum. Am. J. Physiol. 226: 1131-1141, 1974.
NELLANS,
H. N., R. A. FRIZZELL,
AND S. G. SCHULTZ.
Coupled
sodium-chloride
influx across the brush border of rabbit ileum.
Am. J. Physiol. 225: 467-475, 1973.
NELLANS,
H. N., R. A. FRIZZELL,
AND S. G. SCHULTZ.
Effect of
acetazolamide
on sodium and chloride transport
by in vitro
rabbit ileum. Am. J. Phvsiol. 228: 1808-1814, 1975.
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
direction of net chloride movement across the cell would
be determined
by the predomin .ant anatomical
location
of the coupled sodium-chloride
carrier and by the relative chloride permeabilities
of luminal
and contraluminal cell borders.
The chief features of chloride secretion in the rectal
gland are shared by a number of other biological
systems in which active chloride transport
has been demonstrated
or is suspected. Examples
include cornea1
epithelium,
the thick ascending limb of Henle’s loop in
the kidney, “chloride
cells” of the gills of teleost fish,
mammalian
salivary
glands, chloride-secreting
cells of
intestinal
mucosa, and, possibly, the avian salt gland
and the mammary
gland. The present hypothesis seems
equally applicable
to these tissues.
Co-transport
of sodium and chloride has been demonstrated in the gallbladder
(15) and across the mucosal
border of intestinal
cells (31). Linked transport
of NaCl
as an ion pair was first suggested by Shanes (41) and
F306
44.
45.
46.
47.
48.
49.
50.
51.
52.
ET AL.
HAYSLETT. Functional
correlates
of the dogfish rectal gland
during in vitro perfusion.
Comp. Biochem. Physiol.
5lA: 593597, 1975.
SOLOMON, R., L. HAYS, S. EPSTEIN, J. FORREST, P. SILVA, AND F.
H. EPSTEIN. Studies of rectal gland function in SquaZus acanthias in vivo and in vitro. BUZZ. Mount Desert IsZand BioZ. Lab.
14: 122-126, 1974.
STOFF, J. S., P. SILVA, M. FIELD, J. FORREST, A. STEVENS, AND
F. H. EPSTEIN. Cyclic AMP regulation
of active chloride transport in the rectal gland of marine elasmobranchs.
J. Exptl.
ZooZ. 199: 443-448, 1977.
STRICKHOLM, A., AND B. G. WALLIN.
Intracellular
chloride
activity
of crayfish
giant axons. Nature 208: 790-791, 1965.
SWANSON, C. H., AND A. K. SOLOMON. A micropuncture
investigation of the whole tissue mechanism
of electrolyte
secretion by
the in vitro rabbit pancreas. J. Gen. Physiol. 62: 407-429, 1973.
WAY, L. W., AND J. M. DIAMOND. The effect of secretion on
electrical potential differences
in the pancreatic
duct. Biochim.
Biophys. Acta 203: 298-307, 1970.
YOUNG, J. A. Electrolyte
transport
by salivary
epithelia. Proc.
AustraZian
PhysioZ. PharmacoZ. Sot. 4: 101-121, 1973.
ZADUNAISKY,
J. A. Active
transport
of chloride
in the frog
cornea. Am. J. Physiol.
211: 506-510, 1966.
ZADUNAISKY, J. A., 0. A. CANDIA, AND D. J. CHIARANDINI.
The
origin of the short-circuit
current
in the isolated skin of the
South American
frog (Leptodactylus
oceZZatus). J. Gen. PhysioZ.
47: 393-402, 1963.
ZADUNAISKY,
J. A., M. A. LANDE, AND J. HAFNER. Further
studies on chloride transport
in the frog cornea. Am. J. Physiol.
221: 1832-1836, 1971.
Downloaded from http://ajprenal.physiology.org/ by 10.220.33.1 on June 17, 2017
33. OGURI, M. Rectal glands of marine and fresh-water
sharks:
comparative
histology.
Science 144: 1151-1152, 1964.
34. PETERSEN, 0. H. The importance
of extracellular
sodium and
potassium for acetylcholine-evoked
salivary
secretion. Experientia 26: 1103-1104, 1970.
35. POULSEN, J. H., M. BUNDGUARD, AND M. MOLLER. Localization
of (Na+-K+)-activated
ATPase
in “forward”
and “backward”
epithelial in salivary
glands. Physiologist
18: 356, 1975.
36. POWELL, D. W. Intestinal
conductance
and permselectivity
changes with theophylline
and choleragen. Am. J. Physiol. 227:
1436-1443, 1974.
37. POWELL, D. W., R. K. FARRIS, AND S. T. CARBONELTA. Theophylline, cyclic AMP, choleragen,
and electrolyte
transport
by rabbit
ileum. Am. J. Physiol. 227: 1428-1435, 1974.
38. ROCHA, A. S., AND J. P. KOKKO. Sodium chloride and water
transport
in the medullary
thick
ascending
limb of Henle.
Evidence for active chloride transport.
J. Clin. Invest. 52: 612623, 1973.
39. SCHMIDT-NIELSON,
B. Intracellular
concentration
of the salt
gland of the herring gull Laries argentatus.
Am. J. Physiol. 230:
514-521, 1976.
40. SCHULTZ, S. G., AND P. F. CURRAN. Coupled transport
of sodium
and organic solutes. PhysioZ. Rev. 50: 637-718, 1970.
41. SHANES, A. M. Electrochemical
aspects of physiological
and
pharmacological
action in excitable cells. Part I. The resting
cell and its alteration
by extrinsic
factors. PharmacoZ.
Rev. 10:
59-164, 1958.
42. SIEGEL, N. J., D. A. SCHON, AND J. P. HAYSLETT. Evidence for
active chloride transport
in dogfish rectal gland. Am. J. Physiol.
230: 1250-1254, 1976.
43. SIEGEL, N. J., P. SILVA, F. H. EPSTEIN, T. H. MAREN, AND J. P.
SILVA