Vol. 274, Issue 2, F275-F282, February 1998
Local upregulation of colonic angiotensin II receptors
enhances potassium excretion in chronic renal failure
Marguerite
Hatch,
Robert W.
Freel, and
N. D.
Vaziri
Department of Medicine, Division of Nephrology, University of
California at Irvine, Irvine, California 92697
 |
ABSTRACT |
The role of
angiotensin II (ANG II) in colonic secretion of
K+ was examined in rats with
chronic renal failure (CRF). The basal net secretory flux of
86Rb+
(as a tracer for K+) across the
CRF distal colon (
0.20 ± 0.04 µeq · cm
2 · h1)
was reversed to an absorptive flux (0.35 ± 0.05 µeq · cm2 · h1)
by injecting the rats with the AT1
receptor antagonist, losartan. A similar result was observed when
losartan was added to the CRF colonic tissue in vitro. In contrast, an
AT2 receptor antagonist, PD-123319, did not reverse the CRF-induced alterations in
Rb+ transport across the
short-circuited colonic tissue. Plasma concentrations of ANG II,
aldosterone, and K+, as well as
the ANG II content of colonic tissues from CRF and normal rats, were
similar. However, specific
125I-labeled ANG II binding sites
in rat distal colon increased twofold in CRF [maximal specific
binding (Bmax) = 28.6 ± 1.6 fmol/mg protein] compared with normal
(Bmax = 15.2 ± 0.4 fmol/mg
protein). These studies suggest that CRF-induced secretion of
K+ by the colon is mediated by an
upregulation of AT1 receptors present in CRF.
losartan; EXP-3174; PD-123319; jejunum; ileum; absorption; secretion; AT1 receptor; AT2 receptor
| |
INTRODUCTION |
POTASSIUM HOMEOSTASIS is maintained by both renal and
extrarenal mechanisms (15). Normally, the renal distal tubule secretes up to 90% of the dietary K+ load,
and the remainder is eliminated via the large intestine (15). When
renal function is compromised, such as in chronic renal failure (CRF),
plasma K+ has been shown to remain
stable (23). This plasma K+
homeostasis appears to be a consequence of an increased rate of
K+ secretion by remaining
functional nephrons in addition to an increase in colonic
K+ secretion (2, 23, 24). Both
renal and colonic K+ transport
mechanisms can be regulated by aldosterone; however, investigations of
aldosterone involvement in this CRF-adaptive response have yielded
inconsistent results in patient studies (4, 23, 24). For example,
spironolactone antagonism of colonic
K+ secretion was not demonstrated
in nephrectomized rats, and elevated aldosterone titers have not been
reported in this animal model of CRF (2).
In experimental CRF in rats, we have observed a net colonic secretion
of anions including chloride, urate, and oxalate, compared with a basal
absorptive flux in normal controls (9-11). During the course of
these previous investigations, we found that losartan, a specific
angiotensin II (ANG II) receptor antagonist, reversed the CRF-induced
anion secretion to absorption across CRF rat colon (9). These results
implicated the involvement of ANG II in mediating the generalized
secretory nature of colonic mucosa in CRF and prompted the present
study, which further investigates the possible role of angiotensin in
the regulation of colonic K+
secretion in CRF.
In this report, we provide new information regarding large intestinal
control of K+ homeostasis in CRF.
The present study suggests that ANG II has an integral role in
mediating colonic excretion of K+
in CRF by a local upregulation of ANG II binding sites.
| |
MATERIALS AND METHODS |
Reagents. We received both
AT1 receptor antagonists, losartan
and its metabolite EXP-3174, as gifts from Merck (Rahway, NJ). The
AT2 receptor antagonist,
PD-123319, was a product of Research Biochemicals International
(Natick, MA), and
[Asn1,Val5]ANG
II was obtained from Peninsula Laboratories (Belmont, CA). Bestatin, bacitracin, phosphoramidon, leupeptin,
pepstatin, and neomycin were obtained from Calbiochem (La Jolla, CA),
and all other reagents were purchased from Sigma Chemical (St. Louis, MO).
Animals. Male Sprague-Dawley rats
(285-325 g) were used in the following studies. The rats had free
access to drinking water and Purina Rat Chow 5001. Food intake was
determined, over a 48-h period, on a weekly basis in the normal and
experimental groups beginning on the second week following surgery
through the sixth week.
To produce CRF, a partial nephrectomy was performed on each animal
designated to the CRF group. General anesthesia was induced with an
intraperitoneal injection of pentobarbital, and the surgical procedure
of a right nephrectomy, followed by a left two-thirds nephrectomy 4 days later, was performed extraperitoneally under aseptic conditions.
Several series of experiments were conducted using intestinal tissues
removed from normal rats and CRF rats that were euthanized by an
intraperitoneal injection of pentobarbital sodium 6 wk after surgery.
Blood was collected from the rats at this time for the measurement of
K+ (atomic absorption
spectrometry, Perkin-Elmer, Norwalk, CT), aldosterone [solid
phase radioimmunoassay (RIA), Coat-A-Count; Diagnostic Products, Los
Angeles, CA], ANG II (RIA; Nichols Instruments, San Juan
Capistrano, CA), and creatinine (kit 555A; Sigma Chemical). The blood,
collected by cardiac puncture from unconscious rats, was divided into
several tubes for the various measurements. Within 15 s of opening the
body cavity, blood (2 ml) was collected for ANG II determination into a
syringe containing the following inhibitor cocktail (75 µl/ml blood)
to prevent generation or degradation of the peptide: 0.025 M
phenanthroline, 0.125 M EDTA, 2 mM neomycin, 103 M enalaprilat,
105 M pepstatin, and 2%
ethanol (14). An internal control provided with the RIA kit for ANG II
determination and an immunoassay tri-level control, CON6 (Diagnostic
Products), which was assayed as an unknown along with the plasma
samples in the aldosterone RIA, consistently yielded values within the
ranges specified. Creatinine was also determined in 24-h urine
specimens collected immediately before the animals were euthanized.
Creatinine clearance was calculated for each rat, according to the
standard formula, and this was used as an indicator of renal function.
In one experimental series, CRF rats were divided into two groups. Half
of the rats received intraperitoneal injections of losartan (10 mg/kg),
on a daily basis for 7 days, beginning on the fifth week after surgery.
This dosage and schedule was chosen, because it was found to achieve
the maximal antihypertensive effect in rats (30). At the same time, the
remaining half of the group received a placebo injection of saline (150 mM NaCl). The intestinal segments (primarily distal colon, but jejunum
and ileum were included in one series) were removed from both normal
and CRF rats as previously described (9), rinsed with the standard
saline solution (see below), and partially stripped of the serosal
muscularis. Flat sheets of tissue were mounted in modified Ussing
chambers with an exposed tissue area of 0.64 cm2 and bathed on both sides by 10 ml of the standard saline solution (9).
Solutions. The standard saline
contained the following solutes (in mmol/l): 139.4 Na+, 5.4 K+, 1.2 Mg2+, 123.2 Cl, 21.0 
, 1.2 Ca2+, 0.6 H2PO
,
2.4 HPO2, and 10 glucose.
Whenever
86Rb+
was used in the flux experiments,
K+ was replaced by equimolar
Rb+. ANG II at
104 M and losartan at
103 M were added to the
serosal bathing solution to ensure an effective dose at the receptor
location. In the Ussing chamber tissue preparation, these
concentrations are necessary to overcome potential time-dependent hydrolysis of the peptide and receptor antagonists (9). To inhibit
peptide degradation in the flux experiments involving ANG II addition,
the standard saline also contained 0.5 mM phenylmethylsulfonyl fluoride
(PMSF), 0.1 mM leupeptin, 0.1 mM pepstatin A, 50 µM phosphoramidon, and 0.1 µM captopril. The inclusion of these compounds in the bathing
solutions had no discernible effects on the transport characteristics
of this tissue. These solutions (pH 7.4) were maintained at 37°C
and circulated by continuous bubbling of 95% O2-5%
CO2.
Flux and electrical measurements.
Transmural fluxes of
42K+
(3.7-4.44 MBq/mg potassium chloride) or
86Rb+
(37 GBq to 1.3 TBq/g 0.5 M HCl solution) purchased from New England Nuclear (Boston, MA) were measured under short-circuit conditions with
an automatic voltage-clamping device (model VCC600; Physiologic Instruments, San Diego, CA). The use of
86Rb as a tracer for
42K+
in rat distal colon was validated by simultaneously measuring the
unidirectional fluxes of 86Rb and
42K+
in K+-absorbing (control) and
K+-secreting (dibutyryl-cAMP
stimulated) tissues. Tracer activity of
42K+
or 86Rb was determined by standard
gamma spectrometry (model 5500B; Beckman, Fullerton, CA). As shown in
Fig. 1, there is a strong correlation
(r2 = 0.94)
between the unidirectional fluxes [mucosal to serosal, Jms, and serosal
to mucosal,
Jsm]
of Rb+ and
K+; hence,
86Rb+
is a satisfactory substitute for
42K+
in rat distal colon, and
86Rb+
fluxes were subsequently measured in most of the experiments. Tissue
conductance (GT,
mS/cm2) was calculated as the
ratio of the open-circuit potential
(VT, mV) to the
short-circuit current
(Isc,
µA/cm2). The unidirectional
isotope fluxes were determined on matched tissue pairs
(GT 
20%) by
adding the isotope to one bathing solution and measuring its appearance
in the opposing solution by removing a 1 ml aliquot at 10 min intervals
over the duration of the experiment. The sampling volume was replaced
with an equal volume of unlabeled solution, and a correction was
applied for the subsequent dilution effect. Following an equilibration
period of 30 min, control fluxes were measured for a period of 30 min,
i.e., period I. This was followed by a
30-min experimental period (period
II) wherein, for example, an ANG II receptor
antagonist (losartan, EXP-3174, PD-123319) or ANG II was added to the
serosal bathing solution. A 10-min equilibration period preceded
period II, following the addition of a
drug to the tissue.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 1.
Correlation between simultaneously measured
42K+
and
86Rb+
unidirectional fluxes
(JK and
JRb,
respectively) across rat distal colon. Each point represents the mean ± SE for 3 animals. Open symbols, fluxes measured during 4 intervals in period I (control,
unstimulated); filled symbols, fluxes determined during 3 intervals in
period II subsequent to addition of
0.5 mM dibutyryl-cAMP (dB-cAMP) to the serosal compartment. Line
through the variates is the least squares regression as given by the
equation. Intercept is not significantly different from zero. Each flux
sample was counted twice by gamma spectrometry: immediately to evaluate
combined dpm, and 8-10 days later when the
42K had decayed to background.
Activity measured in the second counting was decay corrected to the
first counting and represents 86Rb
activity in the sample. Potassium activity in the sample is the
difference between total initial activity and the
86Rb activity. S-M,
serosal-to-mucosal flux; M-S, mucosal-to-serosal flux.
|
|
Determination of tissue ANG II
content. Distal colonic segments removed from both
normal and CRF rats were rinsed thoroughly with 0.9% saline containing
the same inhibitor cocktail that was used for blood collection (see
above). The tissues were stripped of the serosal muscularis in the same
way as described for the Ussing chamber preparation and dropped into
ice-cold methanol (10% wt/vol) containing 100 µl of 8 M urea and
0.1% Triton X-100 (16). The tissue was homogenized with four 10-s
pulses using a Brinkmann Polytron (Brinkmann Instruments, Westbury,
NY). The homogenate was centrifuged for 10 min at 13,000 g, and the supernatant was removed and
dried under a stream of N2 gas.
The dried extract was reconstituted into a
tris(hydroxymethyl)aminomethane (Tris) buffer, pH 7.4, and
stored at 20°C for less than 1 wk before the RIA was
conducted. In a pilot series, it was determined that this procedure
yielded 81.06 ± 2.8% (n = 5 tissues) recovery of the tracer
125I-ANG II that was added to the
tissue sample prior to homogenization. Two further extractions of the
recovered pellet in methanol, as described above, yielded tracer
recoveries of 11.74 ± 2.2% and 5.33 ± 0.8%, respectively, and
independent RIA determination of the three dried supernatant extracts
gave comparable values. On the basis of these results, the tissue
extraction procedure was standardized to one homogenization only. Both
tissue and plasma results were corrected for losses during extraction.
Preparation of membrane fragments for receptor binding
assay. Distal colonic segments were removed from both
normal and CRF rats. The tissues were handled exactly as described in
the previous section except these stripped tissues were snap frozen in
liquid nitrogen and stored at 70°C. The mucosa was
homogenized with four 10-s pulses, using the Brinkmann Polytron, in 20 vol of 250 mM sucrose (pH 7.6) containing the following: 10 mM
triethanolamine HCl, 0.1 mM PMSF, 0.1 mM bacitracin, 50 µM
phenanthroline, 10 µM phosphoramidon, 130 µM bestatin, and 1 µM
leupeptin, captopril, and pepstatin. The homogenate was centrifuged for
10 min at 50,000 g (model L5-75B,
Beckman). The pellet was recovered, rehomogenized, and centrifuged once
again. The final pellet was resuspended in an appropriate volume of 10 mM triethanolamine, containing 0.1 mM PMSF at pH 7.6 to give a solution
containing ~1 mg/ml protein. Protein was determined using the
Bradford method (Bio-Rad Protein Kit; Bio-Rad, Richmond, CA).
Receptor binding assay. The assay
buffer (pH 7.4) contained the following: 120 mM NaCl, 20 mM
Tris · HCl, 5 mM ethylene glycol-bis(
-aminoethyl ether)-N,N,N',N'-tetraacetic
acid, 0.1 mM PMSF, 0.2% bovine serum albumin (BSA, heat treated at
56°C for 30 min), 100 µM bacitracin, 50 µM phenanthroline, 10 µM phosphoramidon, 130 µM bestatin, and 1 µM leupeptin,
pepstatin, and captopril. The binding reaction was initiated by the
addition of 100 µl of membrane fraction (mean protein = 1.11 ± 0.09 mg/ml, n = 14) to a tube
containing 100 µl of assay buffer and 50 µl of varying
concentrations of labeled peptide
125I-ANG II (specific
activity = 81.4 TBq/mmol; New England Nuclear) in a 50 mM
phosphate buffer containing 0.1% BSA, pH 7.4. Nonspecific binding was
quantitated by addition of 10 µM of unlabeled ANG II to an identical
incubation mixture. The binding reaction, for both specific and
nonspecific binding, was conducted in triplicate for a period of 5 min
at 23°C (5), and it was terminated by filtering the contents of
each tube through a Millipore filter (HAWP 0.45 µm; Millipore,
Bedford, MA) under vacuum. Each filter was rinsed twice with 5 ml of
ice-cold stop solution that contained 120 mM NaCl and 20 mM
Tris · HCl, pH 7.4. Tracer activity trapped on the
filter was determined using gamma spectrometry.
Statistical methods. All results are
presented as means ± SE. Significant differences between means were
established using Student's t-test
for paired and unpaired comparisons, and differences were considered
significant if P 0.05.
| |
RESULTS |
The CRF rat model. A comparison of
plasma creatinine concentration and creatinine clearance between the
normal group of rats and the CRF group confirmed a significant
reduction in renal function in CRF rats 6 wk after five-sixths
nephrectomy. Plasma creatinine increased significantly from 0.047 ± 0.001 mM in the normal group (n = 18)
to 0.096 ± 0.008 mM in the CRF group
(n = 18). Urinary clearance of
creatinine was reduced ~50% in the CRF group compared with the
normal group (normal, 1.95 ± 0.09, n = 18; CRF, 1.0 ± 0.1 ml/min,
n = 18). Weight gain in the normal
group was greater (
= 73 ± 3 g) than in the CRF group ( = 59 ± 3 g) over the 6-wk time period. However, during this period there
was no significant difference in mean food intake between the two
groups (normal, 21.0 ± 0.5; CRF, 22.7 ± 0.7 g · rat1 · day1).
Intestinal potassium transport in CRF
rats. Unidirectional fluxes of
42K+
were measured and compared across segments of colon, ileum, and jejunum, removed from control rats and CRF rats, 6 wk after five-sixths nephrectomy. The colonic segments from the normal rats supported a net
absorptive flux of K+, which was
reversed to a significant net secretory flux in CRF (Fig.
2). This change occurred via alterations in
both unidirectional fluxes in this segment. The significant increase in
Isc across the
CRF colonic tissues confirms our previous observations of a concomitant
electrogenic chloride secretion (9-11). In the small intestine of
the normal rats, there was no significant net flux of
K+ in either direction. However,
in CRF ileum, the absorptive component of the transepithelial flux of
K+ was reduced; consequently, the
CRF ileum supported a small but significant net secretion. There was no
net secretion of K+ across the CRF
jejunum, and unidirectional K+
fluxes in this segment were not different from normal.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 2.
Basal
42K+
fluxes with associated electrical characteristics across jejunum,
ileum, and distal colon of normal and chronic renal failure (CRF) rats.
Tissue conductance
(GT) is
presented in mS/cm2, and
short-circuit current
(Isc) is given
in
µeq · cm2 · h1.
Jsm,
serosal-to-mucosal flux;
Jms,
mucosal-to-serosal flux;
JNet, net flux.
Error bars are ± SE above or below the mean;
n = 9 tissue pairs for each segment
within each group. * Significant difference
(P 0.05) between CRF and normal
animals.
|
|
Effects of ANG II receptor antagonists on
86Rb fluxes across colonic
tissues.
Seven days of losartan in vivo administration (by injection) to CRF
rats resulted in marked alterations in the direction and magnitude of
Rb+ fluxes across the colonic
tissues compared with colonic
Rb+ fluxes in CRF rats injected
with placebo (Fig. 3). It was
apparent that chronic losartan administration abolished the
characteristic net secretory flux of
Rb+ across CRF colon by way of
coordinated changes in both unidirectional fluxes. The results
presented in Fig. 3 also show that there were no significant
time-dependent changes in the fluxes or the associated electrical
parameters in either series over the duration of two flux
periods (i.e., periods I and
II). These results suggest that chronic antagonism of ANG II receptors in vivo inhibits net potassium (Rb+) secretion in this CRF
model.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 3.
Basal 86Rb transport with
associated electrical parameters of distal colonic tissues removed from
CRF rats injected with placebo (open bars) or losartan (hatched bars)
for 7 days before study.
GT is given in
mS/cm2, and
Isc is given in
µeq · cm2 · h1.
Control fluxes were determined at 10-min intervals over two 30-min
periods (Per I and
Per II). Error bars are ± SE
about the mean; n = 10 tissue pairs,
from 7 rats in each group. * Significant difference
(P 0.05) between losartan treatment
and placebo treatment within the given time period.
|
|
The possibility of acute, in vitro antagonism of CRF-induced secretion
was also evaluated. In a separate experimental series, when losartan
was added to the serosal solution bathing the CRF tissue preparation in
vitro, similar significant alterations in Rb+ transport were observed, as
depicted in Fig. 4. Furthermore, the
addition of EXP-3174, a metabolite of losartan also known to have ANG
II receptor antagonist activity (30), produced similar results. In
another experimental series (n = 6),
JRbms increased significantly
from 0.26 ± 0.05 to 0.36 ± 0.06 µeq · cm2 · h1,
and JRbsm was significantly
reduced from 0.47 ± 0.07 to 0.28 ± 0.06 µeq · cm2 · h1
following serosal EXP-3174 addition between periods
I and II. These
changes in both unidirectional fluxes resulted in a reversal of
JRbnet from 0.21 ± 0.02 to +0.08 ± 0.01 µeq · cm2 · h1.
Similar to the effects of losartan on the electrical parameters of this
tissue, EXP-3174 significantly decreased
Isc from
5.39 ± 0.27 in period I
to 2.63 ± 0.77 µeq · cm2 · h1
in period II without any alterations
in GT (8.59 ± 0.75 in period I and 8.59 ± 0.53 mS/cm2 in period
II).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 4.
Effects of losartan (103 M,
serosal) on 86Rb fluxes and
electrical parameters across CRF distal colon.
GT is given in
mS/cm2, and
Isc is given in
µeq · cm2 · h1.
Control fluxes were determined at 10-min intervals, for a period of 30 min in the first period (Per I, open
bars). In the second period (Per II,
hatched bars), fluxes were measured for an additional 30 min beginning
10 min after the addition of losartan. Error bars are ± SE above or
below the mean; n = 6 tissue
pairs. * Significant difference
(P 0.05) between Per II
and Per I.
|
|
The sensitivity of CRF tissues to exogenous ANG II was examined by
adding the peptide to the serosal solution bathing CRF colonic tissue.
The addition of ANG II at
104 M produced significant
changes in the secretory component of rubidium flux
(
JRbsm = 0.18 ± 0.01 µeq · cm2 · h1)
and Isc
( Isc = 0.85 ± 0.27 µeq · cm2 · h1)
between period I and
period II in five tissues from four
CRF rats. The time course of these responses is illustrated in Fig. 5 and compared with CRF control tissues
that were not treated with ANG II. When ANG II at
105 M was added to CRF
tissues, the increase in serosal-to-mucosal flux was not significant
( JRbsm = 0.05 ± 0.02 µeq · cm2 · h1,
n = 5). A small, transient increase in
Isc
( Isc = 0.07 ± 0.06 µeq · cm2 · h1,
n = 5) was observed following the
addition of ANG II at 106
M, which peaked over one 10-min flux period, but was not sustained through period II. Changes in
JRbsm could not be resolved
during period II with ANG II addition
at 106 M. Presumably, an
effective dose of ANG II does not reach the receptor location at these
lower concentrations because of the local degradation of the
octapeptide.1
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5.
Time course of effects of ANG II
(104 M, serosal) on
GT,
Isc, and
JRb across CRF
distal colon;  , control CRF tissues
(n = 5);  , CRF tissues treated with
ANG II (n = 5). Arrow indicates that
ANG II was added after 30 min (i.e., period
I) to serosal bathing solution, and fluxes were
measured through period II (i.e.,
40-70 min).
|
|
Losartan inhibition of CRF-induced ion secretion across the distal
colonic segment in vivo and in vitro (Fig. 3 and 4, respectively) suggests antagonism of the AT1
receptor subtype. The next question addressed was whether an
AT2 receptor antagonist would
similarly affect the CRF-induced
Rb+ secretion. As shown in Fig.
6, unidirectional and net fluxes of
Rb+ and the associated electrical
parameters across CRF colon were not affected by the addition of the
AT2 receptor antagonist,
PD-123319, to colonic segments from CRF rats.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 6.
Effects of PD-123319 (104
M) on 86Rb fluxes and electrical
parameters across CRF distal colon.
GT is given in
mS/cm2, and
Isc is given in
µeq · cm2 · h1.
Control fluxes were determined at 10-min intervals, for a period of 30 min in the first period (period I,
open bars). In the second period (period
II, hatched bars), fluxes were measured for an
additional 30 min beginning 10 min after addition of PD-123319. Error
bars are ± SE above or below the mean;
n = 5 tissue pairs.
|
|
Plasma ANG II, aldosterone, and
potassium. Since the foregoing results strongly suggest
a role for ANG II in the mediation of colonic electrolyte secretion in
CRF, the plasma ANG II concentration was compared in CRF and normal
rats. In addition, since changes in colonic
K+ transport can be mediated via
aldosterone (3, 29), plasma K+ and
aldosterone were also determined in both groups. The results of these
studies confirmed no difference in either plasma ANG II or aldosterone
concentrations in CRF rats compared with normal rats (Table
1), and despite CRF, plasma
K+ was also comparable in both
groups. The constancy of plasma ANG II levels in CRF does not preclude
a change in tissue ANG II content (1). Yet we were unable to detect a
significant difference between the ANG II content in tissues from CRF
(698 ± 148 pmol/kg tissue wet wt,
n = 7) and normal rats (907 ± 220 pmol/kg tissue wet wt, n = 7).
Since there were no differences in either the circulating or the local
tissue concentrations of ANG II, the possibility of a CRF-induced
upregulation of ANG II receptors within the large intestinal segment
was considered (1, 6). From the results of saturation binding studies,
we determined that there was an increase in the number of specific
125I-ANG II binding sites in crude
homogenates of colonic mucosa from CRF rats compared with normal rats
(Fig. 7). Specific binding of
125I-ANG II to this preparation
was saturable at ~5 nM. Maximal specific binding
(Bmax) of
125I-ANG II in colonic homogenates
from CRF rats (28.6 ± 1.6 fmol/mg protein) was significantly
greater than that from normal animals (15.22 ± 0.04 fmol/mg
protein). Ligand concentration at half-maximal binding was the same in
both preparations (3.6 ± 0.06 in normal vs. 3.3 ± 0.13 nM in
CRF).
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 7.
Specific 125I-ANG II binding capacity in crude
homogenates of distal colon removed from normal and CRF rats. Error
bars are ± SE above or below the mean;
n = 6 animals in each group.
Calculated maximal specific binding
(Bmax) of 28.6 ± 1.6 fmol/mg
protein in CRF is significantly different
(P 0.05) from normal rats, where
Bmax was 15.22 ± 0.4 fmol/mg
protein. There is no significant difference between calculated values
of ligand concentration producing half-maximal binding in the two
groups (CRF = 3.3 ± 0.06 vs. normal = 3.6 ± 0.13 nM)
|
|
| |
DISCUSSION |
Previous studies conducted in this laboratory demonstrated that
transepithelial fluxes of
Cl and organic anions
(urate and oxalate) are markedly altered in the distal colon of rats
with CRF compared with normal rats (9-11). These anions, which are
normally absorbed by the rat distal colon, are secreted in response to
CRF. On the basis of the losartan sensitivity of CRF-induced secretory
pathways, we proposed that anion secretion across this intestinal
segment in CRF involves ANG II (9). In the present study, we have more
completely addressed the possible role of ANG II in mediating enteric
elimination of K+ in CRF.
Extrarenal K+ elimination by
colonic secretion of K+ is known
to occur in patients with renal insufficiency and in animals with CRF
(2, 18, 21, 23, 24). We have confirmed this phenomenon in the present
study by directly measuring colonic fluxes of
K+ across short-circuited tissues
that were removed from both normal rats and rats with CRF (five-sixths
nephrectomized). We also confirm that the primary site for
K+ adaptation, within the
intestinal tract, is the large intestine, where coordinated alterations
in both the absorptive and the secretory components of transepithelial
K+ transport are shown to occur.
CRF had a limited affect on small intestinal
K+ transport, which was confined
to a reduction in the absorptive component of
K+ transport across the ileum; in
contrast, K+ fluxes across the
jejunal segment were not altered.
Mechanisms of
K+
absorption and secretion.
In the rat distal colon, active K+
absorption is an electroneutral, sodium-independent, and partly
chloride-independent process (26). An apical uptake mechanism involving
an
H+-K+-ATPase
has been suggested, and a basolateral exit mechanism via a
barium-sensitive conductive process has been proposed (26). The
mech-anisms explaining transcellular active
K+ secretion involve uptake at
the basolateral membrane via a ouabain-sensitive Na+-K+-ATPase
(26) and by the furosemide-sensitive
Na+-K+-2Cl
cotransporter (3). Exit across the apical membrane is conductive through barium-sensitive channels (3, 26). Modulation of colonic
K+ transport has been demonstrated
in response to changes in dietary K+ (2, 8); for example,
K+ secretion can be induced in
animals fed a high-K+ diet (22).
The mechanism of K+ adaptation in
renal insufficiency appears to be somewhat similar to the response
following an oral K+ load (22).
Since aldosterone concentrations are elevated in response to a high
dietary K+ (19) and because the
role of aldosterone in enhancing renal excretion of
K+ is well established, a central
role for aldosterone in mediating enteric elimination of
K+ has been generally assumed
(13).
Although early observations suggested that hyperaldosteronism does not
contribute to enteric K+ excretion
in CRF (12), subsequent conflicting reports did not exclude a role for
aldosterone in mediating K+
secretion by the large intestine (4, 13). Consequently, although
aldosterone clearly affects colonic
K+ transport (3), it is not
certain whether the mineralocorticoid has an integral role in the
CRF-induced enteric excretion of
K+ (4, 23, 24). We have concluded
from the present study that it is unlikely that primary
signal-initiating colonic K+
elimination in CRF is aldosterone. First, although aldosterone stimulates K+ secretion across the
rat distal colon (3), the mineralocorticoid does not stimulate
Cl secretion across this
segment (29). The distal colon, in the CRF rat model, consistently
supports a net secretion of both
Cl and other organic anions
known to have an affinity for
Cl transport systems in
large intestinal epithelia (9-11). Second, dietary
K+ intake and circulating
concentrations of aldosterone were not different in CRF rats compared
with normal rats; the latter finding is in agreement with previous
reports addressing this specific question (23, 24). Third, the addition
of ANG II to short-circuited CRF tissue preparations resulted in an
immediate alteration in electrolyte transport as judged by the
increases in Isc.
Since this response occurred within minutes of adding the octapeptide, the time frame of this effect on
Isc and on
Rb+ fluxes is not consistent with
an aldosterone-mediated effect on colonic transport. The foregoing
arguments do not, however, preclude the possibility that
hyperaldosteronism in the setting of renal insufficiency may frequently
occur and may serve as a supplementary mechanism in further enhancing
colonic K+ secretion.
ANG II mediation of colonic secretion in
CRF. Initial, convincing evidence for ANG II
involvement in CRF-induced colonic anion secretion was provided in a
recent report from our laboratory (9). The present study provides
further, substantial evidence that ANG II has an integral role in
modulating colonic K+ secretion in
CRF. ANG II is known to have a dual action on renal and intestinal
epithelia (17). At low concentrations, ANG II stimulates sodium
absorption via norepinephrine release from enteric sympathetic nerves
(17). At high concentrations, ANG II inhibits absorption by
prostaglandin production (17). Interestingly, ANG II-induced
Cl secretion by cultured
tracheal epithelial cells was found to be sensitive to both the
prostaglandin inhibitor indomethacin and to the ANG II receptor
antagonist losartan (28). In the present study, we demonstrated the
losartan sensitivity of CRF-induced net
Rb+ secretion across the rat
colon. We also showed that the basal serosal-to-mucosal flux of
Rb+ across CRF colon further
increased with the addition of ANG II. Although the latter experiments
indicated that CRF secretory tissues responded to exogenously applied
ANG II in a dose-dependent manner, quantifying the relationship between
an increase in ANG II receptor density (or ANG II sensitivity) and the
rate of ANG II-stimulated Rb+
secretion in CRF colon would be difficult given this experimental design. First, receptor agonistic activity may be coupled to one or
more signal transduction pathways, possibly involving both intercellular and intracellular mediators. Second, to what extent each
component of the secretory machinery and signaling pathways is
activated and sustained is not easily resolved. Simply stated, a
twofold increase in ANG II receptor density may not correlate with a
twofold increase in the rate of ANG II-stimulated
Rb+/K+
secretion across the isolated CRF colonic tissue that is already primed
for secretion. Although the concentrations of ANG II employed here to
further stimulate the basal CRF-induced secretion of
Rb+ are high, it was
demonstrated1 that tissue
degradation of the peptide is significant. The issue of peptide
hydrolysis within intestinal mucosa has also been addressed experimentally by Cox et al. (5) in a study of ANG II receptor binding
in the small and large intestine. These investigators examined the
susceptibility of ANG II to hydrolysis by endogenous proteases in
intestine and determined that after 5 min at 22°C, as well as in
the presence of inhibitors and nonspecific proteins, 70% of the free
hormone had been hydrolyzed (5) In the present study, the effective
dose of ANG II within the tissue is estimated to be two orders of
magnitude less than the bath concentration of
104 M. If one assumes that
this dose-degradation relationship can be linearly extrapolated, then a
local tissue concentration of 106-107
M ANG II further stimulates K+
secretion across the CRF distal colon. Although the tissues responded to a lower concentration, as judged by the response in
Isc, changes in
the secretory flux of Rb+ could not be resolved.
The apparent ANG II-mediated effects on electrolyte transport across
the CRF colon do not, however, result from either an elevation in
circulating levels of ANG II or a change in local ANG II concentrations
within colonic tissue. Normal circulating concentrations of ANG II are
maintained up to 48 h after bilateral nephrectomy, and the persistence
of these normal levels, given the short biological half-life of the
octapeptide (1), is presumably due to the remnant kidney and extrarenal
generation by other tissues. It was not, therefore, surprising to find
a comparable plasma ANG II concentration in the CRF and control groups;
however, the finding of no increase in tissue ANG II content was
unexpected. It is notable that tissue ANG II concentrations are 10-fold
higher than plasma concentrations, suggesting local production of the octapeptide. Although we found no indication that local production of
ANG II is increased in CRF tissues, local generation may also partly
explain the lack of effect of exogenous ANG II at the lower bath
concentrations.
Autoradiographic studies confirming the presence of specific ANG II
receptors in the small and large intestine (5, 7) show that the density
of ANG II receptors is greatest in the colon (7). Our saturation
binding studies confirmed the presence of specific
125I-ANG II binding sites in the
normal rat colon, and the Bmax of 15.22 ± 0.4 fmol/mg protein was comparable to that (11.31 ± 2.66 fmol/mg, n = 3) found by Cox et
al. (5) for the same tissue preparation, under similar assay
conditions. On the basis of the pharmacological designation of ANG II
receptor subtypes, the inhibition of CRF-induced electrolyte secretion
by losartan and the lack of inhibition by PD-123319 together indicate
that the predominant receptor subtype in CRF rat distal colon is
AT1. This conclusion is consistent
with previous characterizations of the ANG II receptor subtypes within
the rat colon (7, 25).
The present study clearly shows that there is a twofold increase in
specific 125I-ANG II binding in
CRF colonic tissue compared with normal tissue, and this upregulation
in ANG II receptor density may explain the ANG II mediation of
CRF-induced K+ secretion across
this tissue. Modulation of ANG II receptor density has been reported in
other aspects of K+ homeostasis.
For example, adrenal ANG II receptor density can be altered following
changes in dietary K+ intake (6)
and following bilateral nephrectomy (1). A low K+ intake was associated with a
reduction in ANG II binding capacity in the adrenal cortex, whereas
K+ loading resulted in an increase
in the number of ANG II receptors (6). Aguilera et al. (1) showed that
adrenal ANG II receptor density doubled between 24 and 48 h after
nephrectomy. Yet another observation, with relevance to the present
discussion, was reported by Modrall et al. (20), who found an
upregulation of AT1 mRNA in both
kidneys using rat model with chronic renovascular hypertension (a
two-kidney/one-clip group of rats). Similarly, ANG II receptor density
and AT1 mRNA were increased two-
and threefold, respectively, in ventricular myocardium of the same
hypertension rat model (27). The increases in specific
125I-ANG II binding by colonic
tissues, observed in the uremic hypertensive rat model used in our
studies (mean arterial blood pressure has been shown to be
significantly elevated compared with normal; = 66 mmHg, see
Ref. 9) is consistent with the notion that an upregulation of
AT1 gene expression in either
hypertensive rat model may not be confined exclusively to the kidneys
(20) or ventricular myocardium (27) or, indeed, colonic epithelium (present study).
In conclusion, ANG II has various agonistic activities in numerous
target tissues resulting in a broad range of physiological effects.
Although we cannot exclude other neuro/immune/paracrine elements or
hemodynamic influences, the results of this study suggest that ANG II
is involved in modulating transepithelial electrolyte transport in
K+ adaptation in CRF via an
upregulation of colonic ANG II receptors.
| |
ACKNOWLEDGEMENTS |
We appreciate the expert technical assistance of Fariba Oveisi, who
performed the animal surgeries, and Jennifer D. Halverson, who
determined creatinine and assisted with the animal injections and food
intake study.
| |
FOOTNOTES |
1
We evaluated the issue of peptide
hydrolysis in the CRF colonic preparation by determining the
concentration of ANG II in extracellular tissue space (ECS) of tissues
incubated in the presence of ANG II. Briefly, tissue segments, stripped
of serosal muscularis (~150 mg wet wt), were incubated in buffer
solution (see flux studies in MATERIALS AND
METHODS) containing
104 M ANG II for a period
of 30 min. For ECS measurements, paired tissues segments were incubated
in a similar buffer solution containing 3 µCi of
[14C]inulin (NEN).
Following incubation, all tissues were rinsed thoroughly in buffer
without ANG II or
[14C]inulin. The
tissues and incubation buffers that were designated for ANG II
measurements were handled and assayed as described in
MATERIALS AND METHODS. The tissues
from the
[14C]inulin-containing
buffer were further subdivided into two pieces; one set was for water
content determination, and the other pieces were each digested in 1 ml
of Beckman Tissue Solubilizer. Tracer activity was determined in the
digestate and bath samples at constant quench by liquid scintillation
spectrometry. Mean ECS in tissues removed from five CRF rats was 0.17 ± 0.04 ml extracellular water/g wet tissue wt, and the calculated
concentration of ANG II in that volume of ECS was 2.70 ± 0.18 × 106 M. Since the
measured concentration of ANG II added to the buffers was 1.02 ± 0.02 × 104 M
(n = 5), the difference in
concentration of ANG II between the bath and the tissue is
approximately two orders of magnitude.
Address for reprint requests: M. Hatch, Dept. of Medicine, Division of
Nephrology, Univ. of California at Irvine, Medical Sciences 1, Rm.
C380, Irvine, CA 92697.
Received 14 February 1997; accepted in final form 2 October 1997.
| |
REFERENCES |
1.
Aguilera, G.,
A. Scirar,
A. Baukal,
and
K. J. Catt.
Circulating angiotensin II and adrenal receptors after nephrectomy.
Nature
289:
507-509,
1981[Medline].
2.
Bastl, C.,
J. P. Hayslett,
and
H. J. Binder.
Increased large intestinal secretion of potassium in renal insufficiency.
Kidney Int.
12:
9-16,
1977[Medline].
3.
Binder, H. J.,
and
G. I. Sandle.
Electrolyte absorption and secretion in the mammalian colon.
In: Physiology of the Gastrointestinal Tract (2nd ed.), edited by L. R. Johnson. New York: Raven, 1987, p. 1389-1418.
4.
Cope, C. L.,
and
J. Pearson.
Aldosterone secretion in severe renal failure.
Clin. Sci. (Colch.)
25:
332-341,
1963.
5.
Cox, H. M.,
K. A. Munday,
and
J. A. Poat.
Identification of selective, high affinity [125I]-angiotensin and [125I]-bradykinin binding sites in rat intestinal epithelia.
Br. J. Pharmacol.
87:
201-209,
1986[Medline].
6.
Douglas, J.,
and
K. J. Catt.
Regulation of angiotensin II receptors in the rat adrenal cortex by dietary electrolytes.
J. Clin. Invest.
58:
834-843,
1976.
7.
Duggan, K. A.,
F. A. O. Mendelsohn,
and
N. R. Levens.
Angiotensin receptors and angiotensin I-converting enzyme in rat intestine.
Am. J. Physiol.
257 (Gastrointest. Liver Physiol. 20):
G504-G510,
1989[Abstract/Free Full Text].
8.
Foster, E. S.,
G. I. Sandle,
J. P. Hayslett,
and
H. J. Binder.
Dietary potassium modulates active potassium absorption and secretion in rat distal colon.
Am. J. Physiol.
251 (Gastrointest. Liver Physiol. 14):
G619-G626,
1986.
9.
Hatch, M.,
R. W. Freel,
S. Shahinfar,
and
N. D. Vaziri.
Effects of the specific angiotensin II receptor antagonist losartan on urate homeostasis and intestinal urate transport.
J. Pharmacol. Exp. Ther.
276:
187-193,
1996[Abstract/Free Full Text].
10.
Hatch, M.,
R. W. Freel,
and
N. D. Vaziri.
Intestinal excretion of oxalate in chronic renal failure.
J. Am. Soc. Nephrol.
5:
1339-1343,
1994[Abstract].
11.
Hatch, M.,
and
N. D. Vaziri.
Enhanced enteric excretion of urate in rats with chronic renal failure.
Clin. Sci. (Colch.)
86:
511-516,
1994[Medline].
12.
Hayes, C. P.,
M. E. McLeod,
and
R. R. Robinson.
An extrarenal mechanism for the maintenance of potassium balance in severe chronic renal failure.
Trans. Assoc. Am. Phys.
80:
207-216,
1967[Medline].
13.
Hayslett, J.,
and
H. J. Binder.
Mechanism of potassium adaptation.
Am. J. Physiol.
243 (Renal Fluid Electrolyte Physiol. 12):
F103-F112,
1982.
14.
Huang, H.,
T. Baussant,
R. Reade,
J. B. Michel,
and
P. Corvol.
Measurement of angiotensin II concentration in rat plasma: pathophysiological applications.
Clin. Exp. Hypertens.
11:
1535-1548,
1989.
15.
Kaufman, C. E.
Maintenance of body fluid potassium, calcium, magnesium and phosphorus.
In: Pathophysiology: Altered Regulatory Mechanisms in Disease, edited by E. D. Frolich. Philadelphia, PA: Lippencott, 1984, p. 252.
16.
Kifor, I.,
T. J. Moore,
F. Fallo,
E. Sperling,
A. Menachery,
C.-H. Chiou,
and
G. H. Williams.
The effect of sodium intake on angiotensin content of rat adrenal gland.
Endocrinology
128:
1277-1284,
1991[Abstract].
17.
Levens, N. R.
Control of intestinal absorption by the renin-angiotensin system.
Am. J. Physiol.
249 (Gastrointest. Liver Physiol. 12):
G3-G15,
1985.
18.
Martin, R. S.,
S. Panese,
M. Virginillo,
M. Gimenez,
M. Litardo,
E. Arrizurieta,
and
J. P. Hayslett.
Increased secretion of potassium in the rectum of humans with chronic renal failure.
Am. J. Kidney Dis.
8:
105-110,
1986[Medline].
19.
McCabe, R. D.,
M. J. Smith,
and
T. M. Dwyer.
Aldosterone secretion and the mechanism of potassium adaptation in rats.
Steroids
58:
305-313,
1993[Medline].
20.
Modrall, G. J.,
M. J. Quinones,
J. H. Frankhouse,
W. A. Hsueh,
F. A. Weaver,
and
L. Kedes.
Upregulation of angiotensin II type 1 receptor gene expression in chronic renovascular hypertension.
J. Surg. Res.
59:
135-140,
1995[Medline].
21.
Panese, S.,
R. S. Martin,
M. Virginillo,
M. Litardo,
E. Siga,
E. Arrizurieta,
and
J. P. Hayslett.
Mechanism of enhanced transcellular potassium secretion in man with chronic renal failure.
Kidney Int.
31:
1377-1382,
1987[Medline].
22.
Salem, M. M.,
R. M. Rosa,
and
D. C. Batlle.
Extrarenal potassium tolerance in chronic renal failure: implications for the treatment of acute hyperkalemia.
Am. J. Kidney Dis.
18:
421-440,
1991[Medline].
23.
Sandle, G. I.,
E. Gaiger,
S. Tapster,
and
T. H. J. Goodship.
Enhanced rectal potassium secretion in chronic renal insufficiency: evidence for large intestinal potassium adaptation in man.
Clin. Sci. (Colch.)
71:
393-401,
1986[Medline].
24.
Sandle, G. I.,
E. Gaiger,
S. Tapster,
and
T. H. J. Goodship.
Evidence for large intestinal control of potassium homeostasis in uraemic patients undergoing long-term dialysis.
Clin. Sci. (Colch.)
73:
247-252,
1987[Medline].
25.
Sechi, L. A.,
J.-P. Valentin,
C. A. Griffin,
and
M. Schambelan.
Autoradiographic characterization of angiotensin II receptor subtypes in rat intestine.
Am. J. Physiol.
265 (Gastrointest. Liver Physiol. 28):
G21-G27,
1993[Abstract/Free Full Text].
26.
Smith, P. L.,
S. K. Sullivan,
and
R. D. McCabe.
Potassium absorption and secretion by the intestinal epithelium.
In: Textbook of Secretory Diarrhea, edited by E. Lebenthal,
and M. Duffey. New York: Raven, 1990, p. 109-118.
27.
Suzuki, J.,
H. Matsubara,
M. Urakami,
and
M. Inada.
Rat angiotensin II (type 1A) receptor mRNA regulation and sub-type expression in myocardial growth and hypertrophy.
Circ. Res.
73:
439-447,
1993[Abstract/Free Full Text].
28.
Tamaoki, J.,
K. Isono,
A. Chiyotani,
M. Kondo,
and
K. Konno.
Angiotensin II-1 receptor-mediated Cl secretion by canine tracheal epithelium.
Am. Rev. Respir. Dis.
146:
1187-1191,
1992[Medline].
29.
Turnamian, S. G.,
and
H. J. Binder.
Regulation of active sodium and potassium transport in the distal colon of the rat.
J. Clin. Invest.
84:
1924-1929,
1989.
30.
Wong, P. C.,
W. A. Price,
A. T. Chiu,
J. V. Duncia,
R. R. Carini,
R. R. Wexler,
A. L. Johnson,
and
P. B. M. W. M. Timmermans.
Nonpeptide angiotensin II receptor antagonists. IX. Antihypertensive activity in rats of DuP 753, an orally active antihypertensive agent.
J. Pharmacol. Exp. Ther.
252:
726-732,
1990[Abstract/Free Full Text].
AJP Renal Physiol 274(2):F275-F282
0363-6127/98 $5.00
Copyright © 1998 the American Physiological Society
Top
Abstract
Introduction
