Effect of acid-base balance on biliary bicarbonate secretion in the isolated perfused guinea pig liver M. BLOT-CHABAUD, M. DUMONT, M. CORBIC, AND S. ERLINGER Unit6 de Recherches de Physiopathologie Hkpatique, Institut National de la Santk et de la Recherche Mgdicale, H6pital Beaujon, 92118 Clichy Cedex, France
BLOT-CHABAUD, M.,M. DUMONT, M. CORBIC,AND S.ERLINGER. Effect of acid-base balance on biliary bicarbonate secretion in the isolated perfused guinea pig liver. Am. J. Physiol. 258 (Gastrointest. Liver Physiol. 21): G863-G872, 1990.-Secretin-induced choleresis is of ductal origin and involves bicarbonate transport. Its mechanism is unknown. To determine the relative effects of systemic pH, Pco~, and bicarbonate concentration on secretin-stimulated bicarbonate transport, states of acute metabolic and respiratory acidosis or alkalosis were created in isolated perfused guinea pig livers with or without secretin infusion. During spontaneous secretion conditions, biliary bicarbonate secretion was not correlated with perfusate pH (7.19-7.62) or perfusate PCO~ (23.9-59.7) but was significantly correlated with perfusate bicarbonate concentration (17.5-37.9 mM). Under secretin infusion (25 mU/min), bile flow and biliary bicarbonate concentration increased significantly (109 and 51%, respectively). Biliary bicarbonate secretion was not correlated with perfusate pH (7.19-7.60) but was significantly correlated both with perfusate bicarbonate concentration (14.6-36.8 mM) and PCO~ (25.8-54.3 mmHg). Spontaneous and secretin-induced bile flow were correlated with biliary bicarbonate concentration. The correlation between biliary bicarbonate secretion and PCO~ during secretin-induced choleresis supports the hypothesis that secretin-induced biliary bicarbonate secretion could, at least in part, involve a transport of H’ (or OH-) rather than HCO: itself and that intracellular pH could play a role in the regulation of this secretion. Amiloride (5 x low4 M) did not influence secretin-induced biliary bicarbonate secretion. This result suggests that the Na’-H’ exchange is not involved in bicarbonate secretion by ductular cells. acidosis; amiloride
alkalosis;
bile flow;
secretin;
bicarbonate
transport;
MANY SPECIES (including guinea pigs, pigs, and humans), infusion of secretin induces an increase in bile flow and biliary bicarbonate concentration. Different pieces of evidence support the view that secretin choleresis is mostly of ductular or ductal origin (9, 29). The mechanism of this secretion is not known, but might involve stimulation of bicarbonate transport. Bicarbonate transport has been well characterized in other epithelia, and in most cases such transport is not primarily a bicarbonate movement but is the result of H+ transport in the other direction (4, 6, 24, 25, 27). Such transport might be influenced by systemic or luminal pH: reabsorption of bicarbonate in the isolated proximal rabbit tubule depends on luminal and peritubular pH IN
0193-1857/90 $1.50 Copyright
(22); secretin-induced pancreatic secretion of bicarbonate in the pig in vivo depends on systemic pH (20). Nevertheless, other experiments have shown that systemic bicarbonate concentration may also influence bicarbonate secretion; rat ileal bicarbonate secretion in vivo does not depend on plasma pH but depends on systemic bicarbonate concentration (7). The aim of this study was to examine the regulation of bicarbonate biliary secretion in guinea pigs under basal conditions and during secretin infusion. Isolated perfused livers were used to disturb perfusate acid-base balance and induce respiratory and metabolic acidosis and alkalosis. Thus we have examined separately the effects of perfusate pH, perfusate carbon dioxide tension (Pco~), and bicarbonate concentration on bile flow and biliary bicarbonate secretion. In addition, the effect of amiloride, an inhibitor of the Na+-H+ exchanger, on the secretin-induced biliary bicarbonate secretion was tested. MATERIALS
AND
METHODS
Isolated perfused liver. Hartley albino male guinea pigs (Charles River, Saint-Aubin-les-Elbeuf, France) weighing 280-400 g were used as donors. They were fed with UAR 106 biscuits (UAR, Villemoisson-sur-Orge, France) and supplemented with ascorbic acid. To prepare the isolated liver perfusion, a technique derived from that of Miller et al. (19) was used. Animals were anesthetized intraperitoneally with urethan (25% solution; 0.6 ml/l00 g body wt). They were tracheotomized, and ventilation was started with a rodent respirator (Harvard Apparatus, Millis, MA) according to ventilation standards (15). The cystic duct was tied. The common bile duct was cannulated near its entrance into the duodenum with a no. 8 catheter (1.19 mm ID, Biotrol, Paris, France), which was inserted to a point just short of the confluence of the hepatic ducts. The gastroduodenal vein was tied, and the animal was heparinized (40 IU). The liver was perfused in situ through the portal vein with a Krebs-Henseleit solution, previously warmed at 37°C and oxygenated with a humidified mixture of 5% C02-95% 02. After the catheterization of the portal vein (Biotrol no. 10,1.57 mm ID) and before the beginning of the perfusion, the inferior vena cava was cut distal to the renal vein. The thoracic inferior vena cava was catheterized with a no. IO catheter. The liver was dissected from the animal, transferred to a prewarmed (37°C) Plexiglas cabinet (Medical Re-
0 1990 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G863
G864
ACID-BASE Reqlrrtofy ReqimJry
0 A
Lquilibr8tion 30 min
0B
Lquilibrrtion 30 min
30 min
or mmb,llc or notobalic
“Exparimantrl period” 30 -40 nin
Equilibrltion 5-10 min
“Experiment81 period” 30 - 40 min
5-10min
1
AND
“Experiment81 period” 30 - 10 min
,
BILIARY
HCQ:
SECRETION
8CilOJS rlllrI@s
Equilibrrtion v 5-10 min
“88x81 period” 30 min
I
BALANCE
4
FIG.
1. Experimental
design.
0D Compenrrted Equilibrrtion 30 min I
0E 4[,/ : '
/'
l cretw
Equilibrrtion 30 min
t
Equilibrrtion 5- 10 min t
I
II won
“88~81 period” 30 min
I
rrapir8tory
l Ik8ksis
Boul
rrd rcilotic
“lnitirl brul period” 30 min
“Exprrim8ntrl period” 30 - 40 min
b8s81
L
“Fin81 period” 30 min 4
,,,,,, //,,,/ ,,/,, ..j /
Equilibrltion 20 min t
Equilibr8tion 5-10 min
1
5 mU/mba
conditions
“Experiment81 period” 30 - 40 min I
4
search Apparatus, Boston, MA), and connected to a recirculating perfusion system. The time interval between interruption of the portal circulation and the onset of artificial perfusion was ~60 s, and the time between interruption of the portal circulation to installation into the cabinet was 8, bicarbonate concentration was determined with a total CO, analyzer (Corning 960), PCO~ was measured with the blood gas analyzer, and pH was calculated. Sodium and potassium concentrations were measured by flame photometry (Eppendorf flame photometer, Hamburg, FRG). Bile acid concentration was measured by an enzymatic technique using as-hydroxysteroid dehydrogenase (Worthington Biochemical, Freehold, NJ), according to a method previously described (3). Statistical analysis. Results were expressed as means
HCO,
G865
SECRETION
t SE. To determine statistical significance between paired data, the nonparametric Wilcoxon test was employed where absolute data were compared (basal vs. experimental period in Tables l-5. In Table 4 basal period is the average of the initial and final basal period). The experimental value of the control group was compared with the experimental value of each other group in Tables l-3 and 5 by the nonparametric Mann-Whitney test. Because the test was used four times in Tables 1 and 3 (multiplying 4 times the probability of erroneously discovering “significant” differences), only differences at 0.01 significance level were indicated. Regression lines were calculated by the least-squares method, from the individual data. The calculation of the partial correlation coefficient was done according to Snedecor and Cochran (26). RESULTS
Effect of perfusate acid-base balance on spontaneous bile flow and biliary bicarbonate concentration. Metabolic acidosis was characterized by a decrease in perfusate pH and bicarbonate concentration compared with control values. During simulated metabolic alkalosis, perfusate pH and bicarbonate concentration was higher than in controls. During respiratory alkalosis, perfusate pH increased because of a decrease in perfusate Pco~. Finally, respiratory acidosis was characterized by a decrease in perfusate pH due to an increase in perfusate Pco~. The values obtained are indicated in Table 1. The effects of perfusate acid-base disturbances on spontaneous bile flow, bile acid secretion, and biliary electrolyte concentrations are summarized in Table 1. During control experiments, bile acid secretion rate, bile flow, and biliary bicarbonate concentration were significantly lower during the experimental period than during the basal period. For each metabolic or respiratory condition, we compare the values obtained during the experimental period with those of the basal period. In addition, we compare the values of the experimental periods obtained in con-
1. Effect of acid-base balance disturbances on bile flow, bile acid secretion rate, and biliary electrolyte concentration in isolated perfused guinea pig liver TABLE
Bile Perfusate Group
Respiratory alkalosis (n = 5) Metabolic acidosis (n = 5) Control (n = 5) Metabolic alkalosis (n = 4) Respiratory acidosis (n = 4)
Period
Basal Exptl Basal Exptl Basal Exptl Basal Exptl Basal Exptl
Flow, & mine1 . g liver-l
2.9eo.42 2.33+0.29$ 2.77t0.30 2.04+0.29-f 2.5ozto.09 2.24+0.08"f 2.88t0.61 3.62kO.73"f 2.63t0.12 3.03+0.18-f§
Acid secretion rate, nmol . min-l . g liver-l 1.7kO.3 0.6+0.2-f 1.7t0.1 0.7+0.2j1.4t0.2 0.7+0.1"f 1.2t0.4 1.2+0.3$ 1.2t0.2 0.9+0.1$
Biliary Na+ 159t8 158+3$ P57t5 150+4$ 156t5 152+4$ 143-r-8 146+5$ 157t3 151+3$
concentration, K’
6.8t0.5 7.1+0.4$ 9.0t0.9 9.8+0.9$ 7.OkO.7 7.4+0.9$ 6.7-t-2.7 6.1+2.8x 6.1t0.5 5.8kO.55
mM Cl-
91t2 100+4$ 87t2 92+2$ 102t5 100+4$ 82-r-12 69_+9$ 9lt2 84+4$
HCO, 63.8t4.7 53.0*4.3-f 60.3k2.8 52.2&4.3j52.9t3.5 49.9rt4.0-f 58.154.5 70.8k3.47 50.922.0 54.3+1.5?
PH 7.46t0.01 7.62+0.02-j-§ 7.45t0.01 7.19+0.01-t-~ 7.44kO.01 7.41-+0.01* 7.40t0.02 7.6140.02"f§ 7.39t0.02 7.20+0.01-/-§
Values are means t SE. Statistical values refer to comparison, for each variable, of basal period (first 30 min) 30-40 min when acid-base balance was modified) (Wilcoxon test: * P < 0.05; “f P < 0.01; $ P > 0.05, not significant). periods of control livers with alterations of acid-base balance were performed (Mann-Whitney test: § P < 0.01).
PC% mmHg
HCO,, mM
36.lkl.O 23.9+1.4t§ 37.5tl.5 45.4+2.0? 38.4t1.0 39.7+1.1$ 38.6t0.8 38.5+1.1$ 40.3t1.8 59.7+2.0-i-§
25.2t0.5 24.9+1.7$ 25.6tl.3 17.5+0.9-Q 25.9+0.6 24.9IO.8$ 23.7t0.6 37.9+1.5§ 24.0t0.3 22.6+0.9$
with experimental period (last Comparisons of experimental
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G866
ACID-BASE
BALANCE
AND
trol animals with animals having a disturbed acid-base balance. During metabolic alkalosis, bile flow and biliary bicarbonate concentration were significantly higher during the experimental period than during the basal period. Biliary bicarbonate concentration during the experimental period was significantly higher than during the experimental period obtained in control animals. Bile flow during the experimental period was higher than during the experimental period obtained in control animals, but the difference did not reach the statistical significance. During metabolic acidosis, bile flow and biliary bicarbonate concentration were significantly lower during the experimental period than during the basal period. Bile flow and biliary bicarbonate concentration during the experimental period were lower than during the experimental period obtained in control animals, but the difference did not reach the statistical significance. During respiratory alkalosis, biliary bicarbonate concentration was significantly lower during the experimental period than during the basal period (bile flow was lower, but the difference did not reach statistical significance). Comparison of bile flow and biliary bicarbonate concentration during the experimental period with values obtained during the experimental period in control animals shows that differences were not statistically significant. During respiratory acidosis, bile flow and biliary bicarbonate concentration were significantly higher during the experimental period than during the basal period. Bile flow during the experimental period was significantly higher than during the experimental period obtained in control animals (difference in biliary bicarbonate concentration was not statistically significant). Figure 2 shows variations of bicarbonate biliary secretion, when compared with the basal period, as a function of the perfusate pH. During metabolic disturbances, variations of biliary bicarbonate secretion were positively VARIATIONS HCO;SECl?ETION (nmol/min.g
4
OF
BILIARY
HCO,?
SECRETION
correlated with the perfusate pH (r = 0.84; n = 14; P < 0.001). In contrast, during respiratory disturbances, variation of biliary bicarbonate secretion was negatively correlated with the perfusate pH (r = 80; n = 14; P < 0.001). Therefore, when all conditions were considered together, there was no relation between perfusate pH and biliary bicarbonate secretion. Figure 3 shows that there was a significant correlation between biliary bicarbonate concentration and perfusate bicarbonate concentration. Figure 4 shows that there was no correlation between biliary bicarbonate concentration and perfusate PCO~ during spontaneous secretion. Effect of secretin on bile flow and biliary bicarbonate concentration. Table 2 summarizes the effects of secretin infusion on bile flow, bile acid secretion rate, and biliary electrolyte concentrations. Bile acid secretion rate was slightly increased by secretin infusion. There was an increase of 109% in bile flow and an increase of 51% in biliary bicarbonate concentration under secretin infusion compared with the basal period. Effect of acid-base disturbances on bile flow and biliary bicarbonate concentration under secretin infusion. Table 3 summarizes the effects of acid-base disturbances on secretin-stimulated bile flow, bile acid secretion rate, and biliary electrolyte concentrations. Control experiments showed that there was a significant decrease in bile flow and in biliary bicarbonate concentration with time, whereas bile acid secretion rate and biliary sodium, potassium, and chloride concentration were not modified. During respiratory acidosis and during metabolic alkalosis, bile flow and biliary bicarbonate concentration were significantly higher during the experimental period than during the basal period. Bile flow and biliary bicarbonate concentration during the experimental period were significantly higher than during the experimental
BILIARY METABOLIC ALKALOSIS (n = 4)
liver)
RESPIRATORY ACIDOSIS (n=4)
FIG. 2. Effect of perfusate pH on biliary HCO; secretion in isolated perfused guinea pig liver. Biliary HCO: secretion was expressed as variation of biliary HCO; output during acid-base balance disturbances when compared with basal period (this variation corresponds to difference between biliary HCO: secretion during experimental period and biliary HCO; secretion during basal period). Results are expressed as means t SE.
CONTROL (n =5)
-60
I
ME’TABOLIC ACIDOSIS (n=5)
RESPIRATORY ALKALOSIS (n=5)
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
ACID-BASE
AND
HCO j
Biliary
8a
BALANCE
- concentration
(mM)
6a
40
20
I
0
1
20
10 Perfusate
FIG. 3. Relation HCO; concentration taneous choleresis.
HCOj
1
I
1
30
40
50
concentration
(m M)
between perfusate HCO, concentration and biliary in isolated perfused guinea pig liver during sponEach point represents result of 1 experiment.
period obtained in control animals. During respiratory alkalosis and during metabolic acidosis, bile flow and biliary bicarbonate concentration were significantly lower during the experimental period than during the basal period. Biliary bicarbonate concentration during the experimental periods was significantly lower than during the experimental period obtained in control animals (difference in bile flow did not reach statistical significance). Figure 5 shows variations of biliary bicarbonate secretion, when compared with the basal period, as a function of the perfusate pH. There was a significant positive correlation between perfusate pH and variation of biliary bicarbonate secretion during metabolic disturbances (r = 0.93; n = 12; P < 0.001) and a negative correlation Biliary HCOi concentration
BILIARY
HCO;
G867
SECRETION
between perfusate pH and variation in the biliary bicarbonate secretion during respiratory disturbances (r = 0.88; n = 12; P < 0.01). Figure 6 shows that there was a significant correlation between perfusate bicarbonate concentration and biliary bicarbonate concentration under secretin infusion, as already observed during spontaneous secretion. The slope observed with secretin choleresis (1.70) was twice that obtained during spontaneous secretion (0.83). There was a significant correlation between biliary bicarbonate concentration and perfusate PCO~under secretin infusion (Fig. 7). In Table 4, the effects of compensated respiratory alkalosis and compensated respiratory acidosis on bile flow, bile acid secretion rate, and biliary electrolyte concentrations under secretin infusion are shown. During compensated respiratory alkalosis, bile flow and biliary bicarbonate concentration were significantly lower during the experimental period than during the basal period (mean value between a preexperimental and a postexperimental basal period), whereas during compensated respiratory acidosis, bile flow was significantly higher. In this last condition, biliary bicarbonate concentration was higher during the experimental period than during the basal period, but the difference did not reach statistical significance. Taken together with those of Table 3, these data clearly show that, under secretin infusion, the biliary bicarbonate secretion does not depend on the perfusate pH but does depend on perfusate PCO~and bicarbonate concentration. The calculation of the partial correlation coefficients showed that, under secretin infusion, there was a relation between biliary bicarbonate concentration and perfusate bicarbonate concentration, independent of PCO~ (r: biliary bicarbonate concentration, perfusate bicarbonate concentration. Perfusate PCO~ = 0.41; n = 61; P < 0.01) and that there was a significant correlation between the biliary bicarbonate concentration and the
(mM)
70
60
4
4 4
50
4 between perfusate HCO, concentration in isolated perfused guinea pig liver during spontaneous choleresis. Each point represents result of 1 experiment. FIG.
40
30 20
4. Relation
PCO~ and biliary
t
t
10 1
01
I
1
I
I
I
I
I
10
20
30
40
50
60
70
Perfusate
PCO, (mmHg)
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G868
ACID-BASE
BALANCE
AND
BILIARY
HCO,
SECRETION
TABLE 2. Effect of secretin (25 mU/min) on bile acid secretion rate and biliary electrolyte concentrations in isolated perfused guinea pig liver Bile Perfusate Flow, ~1. min-l . g liver-l
Period
Basal Experimental (under secretin infusion) with
2.68t0.08 5.59+0.25-p
Results of 5 experiments experimental period
are expressed (under secretin
Acid secretion rate nmol . min-l . g liver-l
Biliary Na’
143t2 146&l*
1.ltO.l
1.2kO.l.j.
concentration,
mM
K’
Cl-
HCO,
6.1t0.2 8.9+0.8-j-
87t5 65+5-j-
59.9k4.0 90.2+8.9-j-
PH
7.42t0.01 7.42+0.01$
Pco~, mmHg
40.9t1.8 39.8+1.3$
as means t SE. Statistical values refer to comparison, for each variable, of basal period infusion; 25 mU/min). Wilcoxon test: * P < 0.05; j- P < 0.01; $ P < 0.05, not significant.
HCO;,
mM
26.321.2 25.1kl.Q (first
30 min)
3. Effect of acid-base balance disturbances on bile flow, bile acid secretion rate, and biliary electrolyte concentration in isolated guinea pig liver under secretin infusion TABLE
Bile Perfusate Group
Period
Respiratory alkalosis (n = 4) Metabolic acidosis (n = 5) Control (n = 4) Metabolic alkalosis (n = 3) Respiratory acidosis (n = 4)
Flow, pl. min-l . g liver-l
Acid secretion rate, nmol . min-l . g liver-l
Biliary
concentration, K’
Na’
mM Cl-
HCO,
mM
4.62t0.50 2.53t0.40*
1.1t0.4 0.6kO.l”r
143t2 150+7t
6.1t0.4 5.7kO.23
77tl2 74-+ll”r
75.6t5.6 57.1+3.8*$
7.36t0.02 7.57+0.02*$
41.7t0.7 25.8+1.7*$
23.6t0.9
Basal Exptl
5.61t0.52 4.15t0.43’”
0.9kO.2 0.6tO.l-f
147t3 143t2t
5.4t1.1 6.6tl.l”
72t13 61tllt
70.6rir3.3 60.1t3.3*
7.38t0.01 7.19+0.02*$
38.5k0.9 38.9kO.6t
23.0t0.5 14.6+0.4*$
Basal Exptl Basal Exptl
3.64k0.22 3.18-t-0.24* 5.16t0.47 6.87zkO.04”
0.9kO.l 0.8+O.l”r 1.0tO.l 0.8+0.lt
154t9 153+8j146&6 159+5t
7.3t0.9 7.3+1.0t 7.2rt1.2 8.0+0.5-j-
72t4 78+7t 70t12 67tl.f
82.1t1.9 71.8t3.1* 84.6-1-5.0 93.4+7.8*$
7.4lkO.02 7.40+0.01t 7.40t0.01 7.60+0.02*$
39.1t1.4 37.5+0.4-f 38.621.2 38.0tl.O.f
24.8t0.3 23.1kO.8” 24.0t0.5 36.8+1.6*$
Basal Exptl
4.68t0.34 5.36+0.38*$
1.1kO.l
14o-t3 137+4-f
7.Ot1.1 6.4-t-1.0-f
71t3 79+2”f
76.2tl.l 88.8+2.6*$
7.42kO.02 7.22*0.02*$
38.3t0.6 54.3+0.9*$
24.3tl.O 21.9kO.8.f
0.7t0.2*
VARIATIONS OF HCO; SECRETION (nmol/min.g liver)
I
RESPIRATORY ACIDOSIS (n= 4)
a @ .
7.2
30 min) with Comparisons
22.8+1.3t
experimental period (last of experimental periods
BILIARY
z .
HCO;,
Basal Exptl
Values are means t SE. Statistical values refer to comparison, for each variable, of basal period (first 30-40 min during acid-base balance disturbances) (Wilcoxon test: * P < 0.01; “f P < 0.05, not significant). of control livers with alterations of acid-base balance were performed (Mann-Whitney test: $ P < 0.01).
-
PC%?, mmHg
PH
CON1 (n=4)
FIG. 5. Effect of perfusate pH on biliary HCO; secretion rate in isolated perfused guinea pig liver under secretin infusion (25 mU/min). Biliary HCO; secretion was expressed as variation of biliary HCO; output during acid-base balance disturbances when compared with basal period (this variation corresponds to the difference between biliary HCO, secretion during experimental period and biliary HCO: secretion during basal period). Results are expressed as means t SE.
ROL
METABOLIC ACIDOSIS (n=S 1
7.3
METABOLIC ALKALOSIS (n=3)
RESPIRATORY ALKALdSlS (n=4)
7.4
7.5 PERFUSATE
7.7
7.8
pH
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
ACID-BASE
BALANCE
AND
BILIARY
HCO;
SECRETION
G869
40
Effect of 5 x 10B4 M amiloride on biliary bicarbonate secretion under secretin infusion. Table 5 shows the effects of 5 x 10s4 M amiloride on bile flow, bile acid secretion rate, biliary bicarbonate concentration, and biliary bicarbonate secretion. Values obtained during the experimental period in both the absence (control livers) and presence of 5 x 10B4M amiloride are given. Results show that 5 X 10M4M amiloride did not modify biliary bicarbonate concentration and biliary bicarbonate secretion, compared with control livers. However, under amiloride infusion, bile acid secretion rate was significantly higher than in control conditions.
20
DISCUSSION
Biliary concent
140
HC03 rat ion ( mM)
120 100
80
0
60
spontaneous
LO.50 po.01
0
1
I
I
10
20
30
Perfusate
HCO;
1
40
concentration
I
50 (mM)
FIG. 6. Relation between perfusate HCO; concentration and biliary HCO, concentration in isolated perfused guinea pig liver under secretin infusion (25 mU/min) and during spontaneous choleresis. Each point represents result of 1 experiment. Biliary HCOi concentration(mM) 0
0
I
I
10
20 Perfusate
7. Relation between tration in isolated perfused Each point represents result FIG.
1
I
30
40
PC02
I
50
I
60
(mmHg)
perfusate PCO~ and biliary HCO, concenguinea pig liver under secretin infusion. of 1 experiment.
perfusate Pco~, independent of the perfusate bicarbonate concentration (r: biliary bicarbonate concentration, perfusate Pco~. Perfusate bicarbonate concentration = 0.50; n = 61; P c 0.01). Relation between bile flow and biliary bicarbonate concentration. Figure 8 shows that during spontaneous secretion and under secretin infusion there was a significant correlation between bile flow and biliary bicarbonate concentration. There was no correlation between bile flow and biliary chloride, sodium, or potassium concentration. There was an increase in bile flow of 56 ~1. min-l l g liver-’ for an increase of 1 pmol/l in the biliary bicarbonate concentration during spontaneous secretion. Under secretin infusion, the increase in bile flow was 60 ~1. min-l g liver-l for the same increase in the biliary bicarbonate concentration. These values were not significantly different. l
The aim of this study was to investigate separately the role of pH, Pco~, and systemic bicarbonate concentration on the biliary bicarbonate secretion in the guinea pig liver during spontaneous secretion and secretin-induced choleresis. The pH, Pco~, and bicarbonate concentration of the perfusate were systematically varied, and bicarbonate secretion was measured under these various situations. Experiments were done with isolated livers so that acute metabolic or respiratory acidosis and alkalosis were easily created, and no respiratory or metabolic compensation occurred. Tavoloni and Schaffner (28) showed that the isolated perfused liver was a good experimental model to study secretin-induced choleresis in the guinea pig; in particular, suppression of hepatic arterial flow did not alter bile secretion. Our results show that, during spontaneous secretion, biliary bicarbonate secretion was not correlated with systemic pH (at least in the range of pH 7.19-7.62) but was significantly correlated with perfusate bicarbonate concentration (17.5-37.9 mM). Under secretin infusion (25 mU/min), bile flow and biliary bicarbonate concentration increased significantly (109 and 5l%, respectively). Clearly, in this condition, biliary bicarbonate secretion was not correlated with perfusate pH (7.197.60) but was significantly correlated with perfusate bicarbonate concentration (14.6-36.8 mM) and PCO~ (25.8-54.3 mmHg). Amiloride, a potent inhibitor of the Na+-H+ exchanger, did not affect the secretin-induced biliary bicarbonate secretion. During spontaneous biliary secretion, we have shown that there was a significant correlation between biliary bicarbonate concentration and perfusate bicarbonate concentration. The correlation coefficient indicates a weak correlation. This suggeststhat variables other than perfusate bicarbonate concentration influence biliary bicarbonate concentration. Garcia-Marin et al. (11) have also studied the effects of acid-base balance disturbances on biliary bicarbonate secretion during spontaneous and ursodeoxycholate-induced choleresis in a perfused rat liver model. Their results during spontaneous choleresis are similar to those of this study: biliary bicarbonate secretion was not correlated with perfusate pH, whereas bile flow and biliary bicarbonate concentration were significantly correlated with perfusate bicarbonate concentration. However, in guinea pigs, biliary bicarbonate concentration was always greater than perfusate bicarbonate concentration. This suggests that an active trans-
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G870
ACID-BASE BALANCE AND BILIARY
HCO; SECRETION
TABLE 4. Effect of compensated respiratory alkalosis and acidosis on bile flow, bile acid secretion and biliary electrolyte concentration in isolated guinea pig liver under secretin infusion
rate,
Bile Perfusate Group
Flow, & mine1 . g liver-’
Period
Compensated respiratory alkalosis (n = 4) Compensated respiratory acidosis (n = 3)
Acid secretion rate, nmol . min-’ - g liver-l
Biliary
concentration,
Na+
K+
7.8t0.2 7.7+0.1$
Basal Exptl
2.25kO.15 1.87t0.13”r
0.59t0.28 0.26+0.07$
145t2 141+7$
Basal Exptl
2.86t0.43 3.57&0.65-j-
0.41kO.21 0.46+0.40$
150tl 151+2$
mM
96.8k5.0 102.0t4.4”
11.2kO.8 12.6t1.3*
HCO,
Cl-
74.5t4.0 74.7*3.8$
Bile flow (pl/min.g liver)
mmHg
HCO,,
mM
45.00t2.80 36.40_+3.71j-
7.43t0.01 7.43+0.01$
34.3620.91 27.15&1.20-I
21.6020.49 17.3*0.56t
76.73t3.91 83.47+5.81$
7.44t0.02 7.42+0.03$
34.39t1.87 59.11+7.221-
22.33t0.21 36.57t1.56t
Values are means t SE. Statistical values refer to comparison, for each variable, of the basal period (mean initial and postexperimental final period of 30 min) with experimental period (compensated respiratory alkalosis < 0.05; t P < 0.01; $ P > 0.05, not significant.
10
Pco~,
PH
value between preexperimental or acidosis). Wilcoxon test: * P
r= 0.72 p
0.05, not significant.
port not regulated by systemic pH could be involved. In rats, in contrast, during spontaneous biliary secretion, there seemsto be a passive equilibration between plasma bicarbonate concentration and biliary bicarbonate concentration (11). The mechanism and site of bicarbonate secretion in guinea pigs are not known. Under secretin infusion, we have shown that there was no correlation between perfusate pH and biliary bicarbonate secretion but that there was a significant corre-
Bile HCO, Secretion, nmol . mine1 liver-l
of 5 x low4 M. Data
48.88k3.75 51.30t1.15t were
compared
between
control
and
lation between biliary bicarbonate secretion and systemic bicarbonate concentration; there was also a correlation between biliary bicarbonate secretion and perfusate Pco~. These results are at variance with those of Mathisen and Raeder (17) in pigs. These authors have found a significant correlation between biliary bicarbonate secretion and perfusate pH in pigs in vivo under secretin infusion. The discordance between the results of our study and those of Mathisen and Raeder are difficult to
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
ACID-BASE
BALANCE
AND
explain. This difference might be due to a difference in species or in experimental protocols. In our study, we have found a good correlation between biliary bicarbonate secretion and perfusate bicarbonate concentration. This observation suggests that a basolateral bicarbonate influx mechanism could be present in the secretin-responsive cells; in particular, one can speculate that a Na+-HCOY symporter mediates the bicarbonate transport. Such a mechanism has been described in the basolateral membrane of the rat hepatocyte (21) where it could act in conjunction with a canalicular Cl--HCOg exchanger (18). Under secretin infusion, we have also found a correlation between bicarbonate secretion and perfusate Pco~. A bicarbonate influx mechanism in the ductular cells can not account alone for this effect. This suggests that, in addition or instead of this mechanism, a transport of H+ (or OH-) could occur. It had been suggested by Garcia-Marin et al. (10) that the transport of bicarbonate ions could be due to transport of H+ rather than transport of bicarbonate itself. The most likely hypothesis would be that the mechanism responsible for bicarbonate secretion is a H+ extrusion at the basolateral membrane of the secretin sensitive cell. The mechanism of this H+ extrusion is generally thought to be either a Na+-H+ exchanger or a H+ pump (23). In this hypothesis, plasma bicarbonate could be associated with a proton (transported from the cell via the H+ transport system) to produce carbon dioxide. This carbon dioxide would diffuse across the ductular cell basolateral membrane and then be hydrated inside the cell upon the action of carbonic anhydrase to form H+ and bicarbonate. The H+ would then supply the H+ transport system, whereas HCO, would be secreted into bile. In different epithelia, such as kidney tubule and particularly proximal tubule (4, 5) or pancreas (24), the transport of bicarbonate occurs secondary to a Na+-H+ exchange localized at the contraluminal side of the cell. This Na+-H+ exchange is responsible for a large or the whole part of the bicarbonate transport. In other epithelia, such as urinary bladder, a proton pump is thought to have a role in the bicarbonate secretion (1). Using isolated perfused rat liver, Lake et al. (16) have shown that amiloride, a well-known inhibitor of the Na’H+ exchange, decreased the ursodeoxycholic acid-stimulated bile flow and biliary HCO; output. Their findings were in agreement with the hypothesis of Garcia-Marin et al. (lo), who had proposed that such an exchange could be responsible for the biliary bicarbonate secretion in the rat. In a recent paper, Grotmol et al. (13) have shown that ursodeoxycholic acid-dependent canalicular HCO, secretion was inhibited by amiloride. They suggested that the Na’-H’ exchange could play an important role in mediating canalicular choleresis. Arias and Forgac (2) have shown that such a Na+-H+ exchanger was present in sinusoidal membrane vesicles. In the same paper, Grotmol et al. (13) showed that amiloride did not alter secretin-dependent biliary bicarbonate secretion, but it was inhibited by dicyclohexylcarbodiimide, an inhibitor of all H+ translocating ATPases (12). They concluded that the secretin-dependent ductular biliary bicarbonate secretion in the pig was most likely mediated
BILIARY
HCO.?
G871
SECRETION
by a proton pump. Using a computer simulation, they confirmed their findings (14). In our study, the effect of 5 X lo-* M amiloride was tested. Under secretin infusion, it appeared that amiloride did not modify either biliary bicarbonate concentration or bile flow; hence, amiloride did not modify biliary bicarbonate secretion. These findings suggest that the Na+-H’ exchanger is not involved in the secretininduced bicarbonate secretion in guinea pigs. Another H+ transport system is certainly involved, maybe an H+ ATPase, as described in the pig (12, 14). Anyway, whatever the mechanism of H+ transport, the role of carbonic anhydrase would be essential, since this enzyme is responsible for the equilibrium between COZ, H+, and HCO;. In addition, the influence of PCO~ on biliary bicarbonate secretion under secretin infusion supports the view that intracellular pH may be involved in the regulation of biliary bicarbonate secretion by ductular cells. It is known that an increase in PCO~ induces an acidification of the hepatocyte cytosol and, likewise, a decrease in PCO~ induces an alkalinization of the cytosol (8). In the hypothesis of a H+ transport mechanism at the basal side of the cell, the regulation of bicarbonate secretion would be mainly due to the activity of this system. Finally, we have observed that the regression lines between bile flow and biliary HCO; concentration in basal (spontaneous) and secretin-stimulated conditions were parallel (there was no relation between bile flow and biliary chloride, sodium, or potassium concentration). This suggests that the choleretic effect of the bicarbonate ion is the same in the two conditions. In conclusion, this study shows that spontaneous and secretin-stimulated biliary bicarbonate secretion in the isolated perfused guinea pig liver was not regulated by the perfusate pH but was strongly correlated with perfusate bicarbonate concentration. Under secretin infusion, there was also a relation between PCO~ and biliary bicarbonate secretion. The results are consistent with the view that the cellular mechanism responsible for secretin-stimulated bicarbonate secretion could, at least in part, involve transport of H+ (or OH-) rather than transport of bicarbonate itself and that intracellular pH could be involved in the regulation of this secretion. The absence of effect of 5 x lo-* M amiloride on biliary bicarbonate secretion supports the hypothesis that the Na’-H’ exchanger is probably not involved in secretininduced biliary bicarbonate secretion. The authors the preparation Address for Department of France. Received
7 March
thank M. H. Badoureaux for secretarial assistance in of this manuscript. reprint requests: M. Blot-Chabaud, INSERM U 246, Biology, SBCE, CEN Saclay, 91191 Gif-sur-Yvette, 1988; accepted
in final
form
26 January
1990.
REFERENCES 1. AL-AWQATI, A. MUELLER, AND P. R. STEINMETZ. Transport of H+ against electrochemical gradient in turtle urinary bladder. Am. J. Physiol. 233 (Renal Fluid Electrolyte Physiol. 2): F502-F508, 1987. 2. ARIAS, I. M., AND M. FORGAC. The sinusoidal domain of the plasma membrane of rat hepatocytes contains an amiloride-sensitive Na’H+ antiport. J. Biol. Chem. 259: 5406-5408, 1984.
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G872
ACID-BASE
BALANCE
AND
3. BERTHELOT, P., S. ERLINGER, D. DHUMEAUX, AND A. M. PREAUX. Mechanism of phenobarbital-induced hypercholeresis in the rat. Am. J. Physiol. 219: 809-813, 1970. 4. BICHARA, M., M. PAILLARD, F. LEVIEL, AND J. P. GARDIN. Hydrogen transport in rabbit kidney proximal tubules. Na+-H+ exchange. Am. J. Physiol. 238 (Renal Fluid Electrolyte Physiol. 7): F445--F451, 1980. 5. BORON, W. F., AND E. L. BOULPAEP. Intracellular pH regulation in the renal proximal tubule of the salamander. Basolateral HCO; transport. J. Gen. Physiol. 81: 53-94, 1983. 6. CASE, R. M., A. D. CONIGRAVE, E. J. FAVALORO, I. NOVAK, C. H. THOMPSON, AND J. A. YOUNG. The role of buffer anions and protons in secretion by the rabbit mandibular salivary gland. J. Physiol. Lond. 322: 273-286, 1982. 7. CHARNEY, A. N., AND L. P. HASKELL. Relative effects of systemic pH, PCO~ and bicarbonate concentration on ileal ion transport. Am. J. Physiol. 245 (Gastrointest. Liver Physiol. 8): G230-G235, 1983. 8. COHEN, R. D., R. M. HENDERSON, R. A. ILES, AND J. A. SMITH. Metabolic interrelationships of intracellular pH measured by double-barrelled microelectrodes in perfused rat liver. J. Physiol. Lond. 330: 69-80, 1982. 9. FORKER, E. L. Two sites of bile formation as determined by mannitol and erythritol clearance in the guinea pig. J. CZin. Invest. 46: 1189-1195,1967. 10. GARCIA MARIN, J. J., M. CORBIC, M. DUMONT, G. DE COUET, AND S. ERLINGER. Role of H’ transport in ursodeoxycholate-induced biliary HCO,? secretion in the rat. Am. J. Physiol. 249 (Gastrointest. Liver Physiol. 12): G335-G341, 1985. 11. GARCIA MARIN, J. J., M. DUMONT, M. GORBIC, G. DE COUET, AND S. ERLINGER. Effect of acid-base balance and acetazolamide on ursodeoxycholate-induced biliary bicarbonate secretion. Am. J. Physiol. 248 (Gastrointest. Liver Physiol. 11): GZO-G27, 1985. 12. GROTMOL, T., T. BUANES, AND M. G. RAEDER. DCCD (N,N’Dicyclohexylcarbodiimide) inhibits biliary secretion of HCO;. Stand. J. Gastroenterol. 22: 207-213, 1987. 13. GROTMOL, T., T. BUANES, AND M. G. RAEDER. The effect of amiloride on biliary bicarbonate secretion in the anesthetized pig. Acta Physiol. Stand. 130: 447-455, 1987. 14. GROTMOL, T., T. BUANES, AND M. G. RAEDER. Effect of arterial pH and pC0, on biliary HCO, secretion in the pig. Acta Physiol. Stand. 131: 183-193,1987. 15. KLEINMAN, L. I., AND E. P. RADFORD. Ventilation standards for small animals. J. Appl. Physiol. 19: 360-362, 1964.
BILIARY
HCO;
SECRETION
16. LAKE, J. R., R. W. VAN DYKE, AND B. F. SCHARSCHMIDT. Effects of Na’ replacement and amiloride on ursodeoxycholic acid-stimulated choleresis and biliary bicarbonate secretion. Am. J. Physiol. 252 (Gastrointest. Liver Physiol. 15): Gl63-G169, 1987. 17. MATHISEN, O., AND M. RAEDER. Mechanism of hepatic bicarbonate secretion and bile acid independent bile secretion. Eur. J. Invest. 13: 193-200, 1983. 18. MEIER, P. J., R. KNICKELBEIN, R. H. MOSELEY, J. W. DOBBINO, AND J. L. BOYER. Evidence for carrier-mediated chloride/bicarbonate exchange in canalicular rat liver plasma membrane vesicles. J. Clin. Invest. 75: 1256-1263, 1985. 19. MILLER, L. L., C. G. BLY, M. L. WATSON, AND W. F. BALE. The dominant role of the liver in plasma protein synthesis. J. Exp. Med. 94: 431-445, 1951. 20. RAEDER, M., A. MO, S. AUNE, AND 0. MATHISEN. Relationship between plasma pH and pancreatic HCO, secretion at different intravenous secretin infusion rates. Acta Physiol. Stand. 109: 187191, 1980. 21. RENNER, E. L., J. R. LAKE, B. F. SCHARSCHMIDT, B. ZIMMERLI, AND P. J. MEIER. Rat hepatocytes exhibit basolateral Na’/HCO; cotransport. J. Clin. Invest. 83: 1225-1235, 1989. 22. SASAKI, S., C. A. BERRY, AND F. C. RECTOR. Effect of luminal and peritubular HCO; concentrations and pC0, on HCO: reabsorption in rabbit proximal convoluted tubules perfused in vitro. J. Clin. Invest. 70: 639-649, 1982. 23. SCHARSCHMIDT, B. F., AND R. W. VAN DYKE. Mechanisms of hepatic electrolyte transport. Gastroenterology 85: 1199-1214,1983. 24. SCHULZ, I. Bicarbonate transport in the exocrine pancreas. Ann. NY Acad Sci. 341: 191-209, 1980. 25. SCHWARTZ, G. J. Na+-dependent H’ efflux from proximal tubule: evidence for reversible Na+-H+ exchange. Am. J. Physiol. 241 (Renal Fluid Electrolyte Physiol. 10): F380-F385, 1981. 26. SNEDECOR, G. W., AND W. G. COCHRAN. StatisticaL Methods (6th ed.). Ames: Iowa State Univ., 1967, p. 381-418. 27. STEINMETZ, P. R., AND L. R. LAWSON. Defect in urinary acidification induced in vitro by amphotericin B. J. CZin. Invest. 49: 596601,197O. 28. TAVOLONI, N., AND F. SCHAFFNER. The intrahepatic biliary epithelium in the guinea-pig: is hepatic artery blood flow essential in maintaining its function and structure. Hepatology Baltimore 5: 666-672,1985. 29. WHEELER, H. O., AND P. L. MANCUSI-UNGARO. Role of bile ducts during secretin choleresis in dogs. Am. J. Physiol. 210: 1153-1159, 1966.
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
BLOT-CHABAUD, M.,M. DUMONT, M. CORBIC,AND S.ERLINGER. Effect of acid-base balance on biliary bicarbonate secretion in the isolated perfused guinea pig liver. Am. J. Physiol. 258 (Gastrointest. Liver Physiol. 21): G863-G872, 1990.-Secretin-induced choleresis is of ductal origin and involves bicarbonate transport. Its mechanism is unknown. To determine the relative effects of systemic pH, Pco~, and bicarbonate concentration on secretin-stimulated bicarbonate transport, states of acute metabolic and respiratory acidosis or alkalosis were created in isolated perfused guinea pig livers with or without secretin infusion. During spontaneous secretion conditions, biliary bicarbonate secretion was not correlated with perfusate pH (7.19-7.62) or perfusate PCO~ (23.9-59.7) but was significantly correlated with perfusate bicarbonate concentration (17.5-37.9 mM). Under secretin infusion (25 mU/min), bile flow and biliary bicarbonate concentration increased significantly (109 and 51%, respectively). Biliary bicarbonate secretion was not correlated with perfusate pH (7.19-7.60) but was significantly correlated both with perfusate bicarbonate concentration (14.6-36.8 mM) and PCO~ (25.8-54.3 mmHg). Spontaneous and secretin-induced bile flow were correlated with biliary bicarbonate concentration. The correlation between biliary bicarbonate secretion and PCO~ during secretin-induced choleresis supports the hypothesis that secretin-induced biliary bicarbonate secretion could, at least in part, involve a transport of H’ (or OH-) rather than HCO: itself and that intracellular pH could play a role in the regulation of this secretion. Amiloride (5 x low4 M) did not influence secretin-induced biliary bicarbonate secretion. This result suggests that the Na’-H’ exchange is not involved in bicarbonate secretion by ductular cells. acidosis; amiloride
alkalosis;
bile flow;
secretin;
bicarbonate
transport;
MANY SPECIES (including guinea pigs, pigs, and humans), infusion of secretin induces an increase in bile flow and biliary bicarbonate concentration. Different pieces of evidence support the view that secretin choleresis is mostly of ductular or ductal origin (9, 29). The mechanism of this secretion is not known, but might involve stimulation of bicarbonate transport. Bicarbonate transport has been well characterized in other epithelia, and in most cases such transport is not primarily a bicarbonate movement but is the result of H+ transport in the other direction (4, 6, 24, 25, 27). Such transport might be influenced by systemic or luminal pH: reabsorption of bicarbonate in the isolated proximal rabbit tubule depends on luminal and peritubular pH IN
0193-1857/90 $1.50 Copyright
(22); secretin-induced pancreatic secretion of bicarbonate in the pig in vivo depends on systemic pH (20). Nevertheless, other experiments have shown that systemic bicarbonate concentration may also influence bicarbonate secretion; rat ileal bicarbonate secretion in vivo does not depend on plasma pH but depends on systemic bicarbonate concentration (7). The aim of this study was to examine the regulation of bicarbonate biliary secretion in guinea pigs under basal conditions and during secretin infusion. Isolated perfused livers were used to disturb perfusate acid-base balance and induce respiratory and metabolic acidosis and alkalosis. Thus we have examined separately the effects of perfusate pH, perfusate carbon dioxide tension (Pco~), and bicarbonate concentration on bile flow and biliary bicarbonate secretion. In addition, the effect of amiloride, an inhibitor of the Na+-H+ exchanger, on the secretin-induced biliary bicarbonate secretion was tested. MATERIALS
AND
METHODS
Isolated perfused liver. Hartley albino male guinea pigs (Charles River, Saint-Aubin-les-Elbeuf, France) weighing 280-400 g were used as donors. They were fed with UAR 106 biscuits (UAR, Villemoisson-sur-Orge, France) and supplemented with ascorbic acid. To prepare the isolated liver perfusion, a technique derived from that of Miller et al. (19) was used. Animals were anesthetized intraperitoneally with urethan (25% solution; 0.6 ml/l00 g body wt). They were tracheotomized, and ventilation was started with a rodent respirator (Harvard Apparatus, Millis, MA) according to ventilation standards (15). The cystic duct was tied. The common bile duct was cannulated near its entrance into the duodenum with a no. 8 catheter (1.19 mm ID, Biotrol, Paris, France), which was inserted to a point just short of the confluence of the hepatic ducts. The gastroduodenal vein was tied, and the animal was heparinized (40 IU). The liver was perfused in situ through the portal vein with a Krebs-Henseleit solution, previously warmed at 37°C and oxygenated with a humidified mixture of 5% C02-95% 02. After the catheterization of the portal vein (Biotrol no. 10,1.57 mm ID) and before the beginning of the perfusion, the inferior vena cava was cut distal to the renal vein. The thoracic inferior vena cava was catheterized with a no. IO catheter. The liver was dissected from the animal, transferred to a prewarmed (37°C) Plexiglas cabinet (Medical Re-
0 1990 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G863
G864
ACID-BASE Reqlrrtofy ReqimJry
0 A
Lquilibr8tion 30 min
0B
Lquilibrrtion 30 min
30 min
or mmb,llc or notobalic
“Exparimantrl period” 30 -40 nin
Equilibrltion 5-10 min
“Experiment81 period” 30 - 40 min
5-10min
1
AND
“Experiment81 period” 30 - 10 min
,
BILIARY
HCQ:
SECRETION
8CilOJS rlllrI@s
Equilibrrtion v 5-10 min
“88x81 period” 30 min
I
BALANCE
4
FIG.
1. Experimental
design.
0D Compenrrted Equilibrrtion 30 min I
0E 4[,/ : '
/'
l cretw
Equilibrrtion 30 min
t
Equilibrrtion 5- 10 min t
I
II won
“88~81 period” 30 min
I
rrapir8tory
l Ik8ksis
Boul
rrd rcilotic
“lnitirl brul period” 30 min
“Exprrim8ntrl period” 30 - 40 min
b8s81
L
“Fin81 period” 30 min 4
,,,,,, //,,,/ ,,/,, ..j /
Equilibrltion 20 min t
Equilibr8tion 5-10 min
1
5 mU/mba
conditions
“Experiment81 period” 30 - 40 min I
4
search Apparatus, Boston, MA), and connected to a recirculating perfusion system. The time interval between interruption of the portal circulation and the onset of artificial perfusion was ~60 s, and the time between interruption of the portal circulation to installation into the cabinet was 8, bicarbonate concentration was determined with a total CO, analyzer (Corning 960), PCO~ was measured with the blood gas analyzer, and pH was calculated. Sodium and potassium concentrations were measured by flame photometry (Eppendorf flame photometer, Hamburg, FRG). Bile acid concentration was measured by an enzymatic technique using as-hydroxysteroid dehydrogenase (Worthington Biochemical, Freehold, NJ), according to a method previously described (3). Statistical analysis. Results were expressed as means
HCO,
G865
SECRETION
t SE. To determine statistical significance between paired data, the nonparametric Wilcoxon test was employed where absolute data were compared (basal vs. experimental period in Tables l-5. In Table 4 basal period is the average of the initial and final basal period). The experimental value of the control group was compared with the experimental value of each other group in Tables l-3 and 5 by the nonparametric Mann-Whitney test. Because the test was used four times in Tables 1 and 3 (multiplying 4 times the probability of erroneously discovering “significant” differences), only differences at 0.01 significance level were indicated. Regression lines were calculated by the least-squares method, from the individual data. The calculation of the partial correlation coefficient was done according to Snedecor and Cochran (26). RESULTS
Effect of perfusate acid-base balance on spontaneous bile flow and biliary bicarbonate concentration. Metabolic acidosis was characterized by a decrease in perfusate pH and bicarbonate concentration compared with control values. During simulated metabolic alkalosis, perfusate pH and bicarbonate concentration was higher than in controls. During respiratory alkalosis, perfusate pH increased because of a decrease in perfusate Pco~. Finally, respiratory acidosis was characterized by a decrease in perfusate pH due to an increase in perfusate Pco~. The values obtained are indicated in Table 1. The effects of perfusate acid-base disturbances on spontaneous bile flow, bile acid secretion, and biliary electrolyte concentrations are summarized in Table 1. During control experiments, bile acid secretion rate, bile flow, and biliary bicarbonate concentration were significantly lower during the experimental period than during the basal period. For each metabolic or respiratory condition, we compare the values obtained during the experimental period with those of the basal period. In addition, we compare the values of the experimental periods obtained in con-
1. Effect of acid-base balance disturbances on bile flow, bile acid secretion rate, and biliary electrolyte concentration in isolated perfused guinea pig liver TABLE
Bile Perfusate Group
Respiratory alkalosis (n = 5) Metabolic acidosis (n = 5) Control (n = 5) Metabolic alkalosis (n = 4) Respiratory acidosis (n = 4)
Period
Basal Exptl Basal Exptl Basal Exptl Basal Exptl Basal Exptl
Flow, & mine1 . g liver-l
2.9eo.42 2.33+0.29$ 2.77t0.30 2.04+0.29-f 2.5ozto.09 2.24+0.08"f 2.88t0.61 3.62kO.73"f 2.63t0.12 3.03+0.18-f§
Acid secretion rate, nmol . min-l . g liver-l 1.7kO.3 0.6+0.2-f 1.7t0.1 0.7+0.2j1.4t0.2 0.7+0.1"f 1.2t0.4 1.2+0.3$ 1.2t0.2 0.9+0.1$
Biliary Na+ 159t8 158+3$ P57t5 150+4$ 156t5 152+4$ 143-r-8 146+5$ 157t3 151+3$
concentration, K’
6.8t0.5 7.1+0.4$ 9.0t0.9 9.8+0.9$ 7.OkO.7 7.4+0.9$ 6.7-t-2.7 6.1+2.8x 6.1t0.5 5.8kO.55
mM Cl-
91t2 100+4$ 87t2 92+2$ 102t5 100+4$ 82-r-12 69_+9$ 9lt2 84+4$
HCO, 63.8t4.7 53.0*4.3-f 60.3k2.8 52.2&4.3j52.9t3.5 49.9rt4.0-f 58.154.5 70.8k3.47 50.922.0 54.3+1.5?
PH 7.46t0.01 7.62+0.02-j-§ 7.45t0.01 7.19+0.01-t-~ 7.44kO.01 7.41-+0.01* 7.40t0.02 7.6140.02"f§ 7.39t0.02 7.20+0.01-/-§
Values are means t SE. Statistical values refer to comparison, for each variable, of basal period (first 30 min) 30-40 min when acid-base balance was modified) (Wilcoxon test: * P < 0.05; “f P < 0.01; $ P > 0.05, not significant). periods of control livers with alterations of acid-base balance were performed (Mann-Whitney test: § P < 0.01).
PC% mmHg
HCO,, mM
36.lkl.O 23.9+1.4t§ 37.5tl.5 45.4+2.0? 38.4t1.0 39.7+1.1$ 38.6t0.8 38.5+1.1$ 40.3t1.8 59.7+2.0-i-§
25.2t0.5 24.9+1.7$ 25.6tl.3 17.5+0.9-Q 25.9+0.6 24.9IO.8$ 23.7t0.6 37.9+1.5§ 24.0t0.3 22.6+0.9$
with experimental period (last Comparisons of experimental
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G866
ACID-BASE
BALANCE
AND
trol animals with animals having a disturbed acid-base balance. During metabolic alkalosis, bile flow and biliary bicarbonate concentration were significantly higher during the experimental period than during the basal period. Biliary bicarbonate concentration during the experimental period was significantly higher than during the experimental period obtained in control animals. Bile flow during the experimental period was higher than during the experimental period obtained in control animals, but the difference did not reach the statistical significance. During metabolic acidosis, bile flow and biliary bicarbonate concentration were significantly lower during the experimental period than during the basal period. Bile flow and biliary bicarbonate concentration during the experimental period were lower than during the experimental period obtained in control animals, but the difference did not reach the statistical significance. During respiratory alkalosis, biliary bicarbonate concentration was significantly lower during the experimental period than during the basal period (bile flow was lower, but the difference did not reach statistical significance). Comparison of bile flow and biliary bicarbonate concentration during the experimental period with values obtained during the experimental period in control animals shows that differences were not statistically significant. During respiratory acidosis, bile flow and biliary bicarbonate concentration were significantly higher during the experimental period than during the basal period. Bile flow during the experimental period was significantly higher than during the experimental period obtained in control animals (difference in biliary bicarbonate concentration was not statistically significant). Figure 2 shows variations of bicarbonate biliary secretion, when compared with the basal period, as a function of the perfusate pH. During metabolic disturbances, variations of biliary bicarbonate secretion were positively VARIATIONS HCO;SECl?ETION (nmol/min.g
4
OF
BILIARY
HCO,?
SECRETION
correlated with the perfusate pH (r = 0.84; n = 14; P < 0.001). In contrast, during respiratory disturbances, variation of biliary bicarbonate secretion was negatively correlated with the perfusate pH (r = 80; n = 14; P < 0.001). Therefore, when all conditions were considered together, there was no relation between perfusate pH and biliary bicarbonate secretion. Figure 3 shows that there was a significant correlation between biliary bicarbonate concentration and perfusate bicarbonate concentration. Figure 4 shows that there was no correlation between biliary bicarbonate concentration and perfusate PCO~ during spontaneous secretion. Effect of secretin on bile flow and biliary bicarbonate concentration. Table 2 summarizes the effects of secretin infusion on bile flow, bile acid secretion rate, and biliary electrolyte concentrations. Bile acid secretion rate was slightly increased by secretin infusion. There was an increase of 109% in bile flow and an increase of 51% in biliary bicarbonate concentration under secretin infusion compared with the basal period. Effect of acid-base disturbances on bile flow and biliary bicarbonate concentration under secretin infusion. Table 3 summarizes the effects of acid-base disturbances on secretin-stimulated bile flow, bile acid secretion rate, and biliary electrolyte concentrations. Control experiments showed that there was a significant decrease in bile flow and in biliary bicarbonate concentration with time, whereas bile acid secretion rate and biliary sodium, potassium, and chloride concentration were not modified. During respiratory acidosis and during metabolic alkalosis, bile flow and biliary bicarbonate concentration were significantly higher during the experimental period than during the basal period. Bile flow and biliary bicarbonate concentration during the experimental period were significantly higher than during the experimental
BILIARY METABOLIC ALKALOSIS (n = 4)
liver)
RESPIRATORY ACIDOSIS (n=4)
FIG. 2. Effect of perfusate pH on biliary HCO; secretion in isolated perfused guinea pig liver. Biliary HCO: secretion was expressed as variation of biliary HCO; output during acid-base balance disturbances when compared with basal period (this variation corresponds to difference between biliary HCO: secretion during experimental period and biliary HCO; secretion during basal period). Results are expressed as means t SE.
CONTROL (n =5)
-60
I
ME’TABOLIC ACIDOSIS (n=5)
RESPIRATORY ALKALOSIS (n=5)
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
ACID-BASE
AND
HCO j
Biliary
8a
BALANCE
- concentration
(mM)
6a
40
20
I
0
1
20
10 Perfusate
FIG. 3. Relation HCO; concentration taneous choleresis.
HCOj
1
I
1
30
40
50
concentration
(m M)
between perfusate HCO, concentration and biliary in isolated perfused guinea pig liver during sponEach point represents result of 1 experiment.
period obtained in control animals. During respiratory alkalosis and during metabolic acidosis, bile flow and biliary bicarbonate concentration were significantly lower during the experimental period than during the basal period. Biliary bicarbonate concentration during the experimental periods was significantly lower than during the experimental period obtained in control animals (difference in bile flow did not reach statistical significance). Figure 5 shows variations of biliary bicarbonate secretion, when compared with the basal period, as a function of the perfusate pH. There was a significant positive correlation between perfusate pH and variation of biliary bicarbonate secretion during metabolic disturbances (r = 0.93; n = 12; P < 0.001) and a negative correlation Biliary HCOi concentration
BILIARY
HCO;
G867
SECRETION
between perfusate pH and variation in the biliary bicarbonate secretion during respiratory disturbances (r = 0.88; n = 12; P < 0.01). Figure 6 shows that there was a significant correlation between perfusate bicarbonate concentration and biliary bicarbonate concentration under secretin infusion, as already observed during spontaneous secretion. The slope observed with secretin choleresis (1.70) was twice that obtained during spontaneous secretion (0.83). There was a significant correlation between biliary bicarbonate concentration and perfusate PCO~under secretin infusion (Fig. 7). In Table 4, the effects of compensated respiratory alkalosis and compensated respiratory acidosis on bile flow, bile acid secretion rate, and biliary electrolyte concentrations under secretin infusion are shown. During compensated respiratory alkalosis, bile flow and biliary bicarbonate concentration were significantly lower during the experimental period than during the basal period (mean value between a preexperimental and a postexperimental basal period), whereas during compensated respiratory acidosis, bile flow was significantly higher. In this last condition, biliary bicarbonate concentration was higher during the experimental period than during the basal period, but the difference did not reach statistical significance. Taken together with those of Table 3, these data clearly show that, under secretin infusion, the biliary bicarbonate secretion does not depend on the perfusate pH but does depend on perfusate PCO~and bicarbonate concentration. The calculation of the partial correlation coefficients showed that, under secretin infusion, there was a relation between biliary bicarbonate concentration and perfusate bicarbonate concentration, independent of PCO~ (r: biliary bicarbonate concentration, perfusate bicarbonate concentration. Perfusate PCO~ = 0.41; n = 61; P < 0.01) and that there was a significant correlation between the biliary bicarbonate concentration and the
(mM)
70
60
4
4 4
50
4 between perfusate HCO, concentration in isolated perfused guinea pig liver during spontaneous choleresis. Each point represents result of 1 experiment. FIG.
40
30 20
4. Relation
PCO~ and biliary
t
t
10 1
01
I
1
I
I
I
I
I
10
20
30
40
50
60
70
Perfusate
PCO, (mmHg)
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G868
ACID-BASE
BALANCE
AND
BILIARY
HCO,
SECRETION
TABLE 2. Effect of secretin (25 mU/min) on bile acid secretion rate and biliary electrolyte concentrations in isolated perfused guinea pig liver Bile Perfusate Flow, ~1. min-l . g liver-l
Period
Basal Experimental (under secretin infusion) with
2.68t0.08 5.59+0.25-p
Results of 5 experiments experimental period
are expressed (under secretin
Acid secretion rate nmol . min-l . g liver-l
Biliary Na’
143t2 146&l*
1.ltO.l
1.2kO.l.j.
concentration,
mM
K’
Cl-
HCO,
6.1t0.2 8.9+0.8-j-
87t5 65+5-j-
59.9k4.0 90.2+8.9-j-
PH
7.42t0.01 7.42+0.01$
Pco~, mmHg
40.9t1.8 39.8+1.3$
as means t SE. Statistical values refer to comparison, for each variable, of basal period infusion; 25 mU/min). Wilcoxon test: * P < 0.05; j- P < 0.01; $ P < 0.05, not significant.
HCO;,
mM
26.321.2 25.1kl.Q (first
30 min)
3. Effect of acid-base balance disturbances on bile flow, bile acid secretion rate, and biliary electrolyte concentration in isolated guinea pig liver under secretin infusion TABLE
Bile Perfusate Group
Period
Respiratory alkalosis (n = 4) Metabolic acidosis (n = 5) Control (n = 4) Metabolic alkalosis (n = 3) Respiratory acidosis (n = 4)
Flow, pl. min-l . g liver-l
Acid secretion rate, nmol . min-l . g liver-l
Biliary
concentration, K’
Na’
mM Cl-
HCO,
mM
4.62t0.50 2.53t0.40*
1.1t0.4 0.6kO.l”r
143t2 150+7t
6.1t0.4 5.7kO.23
77tl2 74-+ll”r
75.6t5.6 57.1+3.8*$
7.36t0.02 7.57+0.02*$
41.7t0.7 25.8+1.7*$
23.6t0.9
Basal Exptl
5.61t0.52 4.15t0.43’”
0.9kO.2 0.6tO.l-f
147t3 143t2t
5.4t1.1 6.6tl.l”
72t13 61tllt
70.6rir3.3 60.1t3.3*
7.38t0.01 7.19+0.02*$
38.5k0.9 38.9kO.6t
23.0t0.5 14.6+0.4*$
Basal Exptl Basal Exptl
3.64k0.22 3.18-t-0.24* 5.16t0.47 6.87zkO.04”
0.9kO.l 0.8+O.l”r 1.0tO.l 0.8+0.lt
154t9 153+8j146&6 159+5t
7.3t0.9 7.3+1.0t 7.2rt1.2 8.0+0.5-j-
72t4 78+7t 70t12 67tl.f
82.1t1.9 71.8t3.1* 84.6-1-5.0 93.4+7.8*$
7.4lkO.02 7.40+0.01t 7.40t0.01 7.60+0.02*$
39.1t1.4 37.5+0.4-f 38.621.2 38.0tl.O.f
24.8t0.3 23.1kO.8” 24.0t0.5 36.8+1.6*$
Basal Exptl
4.68t0.34 5.36+0.38*$
1.1kO.l
14o-t3 137+4-f
7.Ot1.1 6.4-t-1.0-f
71t3 79+2”f
76.2tl.l 88.8+2.6*$
7.42kO.02 7.22*0.02*$
38.3t0.6 54.3+0.9*$
24.3tl.O 21.9kO.8.f
0.7t0.2*
VARIATIONS OF HCO; SECRETION (nmol/min.g liver)
I
RESPIRATORY ACIDOSIS (n= 4)
a @ .
7.2
30 min) with Comparisons
22.8+1.3t
experimental period (last of experimental periods
BILIARY
z .
HCO;,
Basal Exptl
Values are means t SE. Statistical values refer to comparison, for each variable, of basal period (first 30-40 min during acid-base balance disturbances) (Wilcoxon test: * P < 0.01; “f P < 0.05, not significant). of control livers with alterations of acid-base balance were performed (Mann-Whitney test: $ P < 0.01).
-
PC%?, mmHg
PH
CON1 (n=4)
FIG. 5. Effect of perfusate pH on biliary HCO; secretion rate in isolated perfused guinea pig liver under secretin infusion (25 mU/min). Biliary HCO; secretion was expressed as variation of biliary HCO; output during acid-base balance disturbances when compared with basal period (this variation corresponds to the difference between biliary HCO, secretion during experimental period and biliary HCO: secretion during basal period). Results are expressed as means t SE.
ROL
METABOLIC ACIDOSIS (n=S 1
7.3
METABOLIC ALKALOSIS (n=3)
RESPIRATORY ALKALdSlS (n=4)
7.4
7.5 PERFUSATE
7.7
7.8
pH
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
ACID-BASE
BALANCE
AND
BILIARY
HCO;
SECRETION
G869
40
Effect of 5 x 10B4 M amiloride on biliary bicarbonate secretion under secretin infusion. Table 5 shows the effects of 5 x 10s4 M amiloride on bile flow, bile acid secretion rate, biliary bicarbonate concentration, and biliary bicarbonate secretion. Values obtained during the experimental period in both the absence (control livers) and presence of 5 x 10B4M amiloride are given. Results show that 5 X 10M4M amiloride did not modify biliary bicarbonate concentration and biliary bicarbonate secretion, compared with control livers. However, under amiloride infusion, bile acid secretion rate was significantly higher than in control conditions.
20
DISCUSSION
Biliary concent
140
HC03 rat ion ( mM)
120 100
80
0
60
spontaneous
LO.50 po.01
0
1
I
I
10
20
30
Perfusate
HCO;
1
40
concentration
I
50 (mM)
FIG. 6. Relation between perfusate HCO; concentration and biliary HCO, concentration in isolated perfused guinea pig liver under secretin infusion (25 mU/min) and during spontaneous choleresis. Each point represents result of 1 experiment. Biliary HCOi concentration(mM) 0
0
I
I
10
20 Perfusate
7. Relation between tration in isolated perfused Each point represents result FIG.
1
I
30
40
PC02
I
50
I
60
(mmHg)
perfusate PCO~ and biliary HCO, concenguinea pig liver under secretin infusion. of 1 experiment.
perfusate Pco~, independent of the perfusate bicarbonate concentration (r: biliary bicarbonate concentration, perfusate Pco~. Perfusate bicarbonate concentration = 0.50; n = 61; P c 0.01). Relation between bile flow and biliary bicarbonate concentration. Figure 8 shows that during spontaneous secretion and under secretin infusion there was a significant correlation between bile flow and biliary bicarbonate concentration. There was no correlation between bile flow and biliary chloride, sodium, or potassium concentration. There was an increase in bile flow of 56 ~1. min-l l g liver-’ for an increase of 1 pmol/l in the biliary bicarbonate concentration during spontaneous secretion. Under secretin infusion, the increase in bile flow was 60 ~1. min-l g liver-l for the same increase in the biliary bicarbonate concentration. These values were not significantly different. l
The aim of this study was to investigate separately the role of pH, Pco~, and systemic bicarbonate concentration on the biliary bicarbonate secretion in the guinea pig liver during spontaneous secretion and secretin-induced choleresis. The pH, Pco~, and bicarbonate concentration of the perfusate were systematically varied, and bicarbonate secretion was measured under these various situations. Experiments were done with isolated livers so that acute metabolic or respiratory acidosis and alkalosis were easily created, and no respiratory or metabolic compensation occurred. Tavoloni and Schaffner (28) showed that the isolated perfused liver was a good experimental model to study secretin-induced choleresis in the guinea pig; in particular, suppression of hepatic arterial flow did not alter bile secretion. Our results show that, during spontaneous secretion, biliary bicarbonate secretion was not correlated with systemic pH (at least in the range of pH 7.19-7.62) but was significantly correlated with perfusate bicarbonate concentration (17.5-37.9 mM). Under secretin infusion (25 mU/min), bile flow and biliary bicarbonate concentration increased significantly (109 and 5l%, respectively). Clearly, in this condition, biliary bicarbonate secretion was not correlated with perfusate pH (7.197.60) but was significantly correlated with perfusate bicarbonate concentration (14.6-36.8 mM) and PCO~ (25.8-54.3 mmHg). Amiloride, a potent inhibitor of the Na+-H+ exchanger, did not affect the secretin-induced biliary bicarbonate secretion. During spontaneous biliary secretion, we have shown that there was a significant correlation between biliary bicarbonate concentration and perfusate bicarbonate concentration. The correlation coefficient indicates a weak correlation. This suggeststhat variables other than perfusate bicarbonate concentration influence biliary bicarbonate concentration. Garcia-Marin et al. (11) have also studied the effects of acid-base balance disturbances on biliary bicarbonate secretion during spontaneous and ursodeoxycholate-induced choleresis in a perfused rat liver model. Their results during spontaneous choleresis are similar to those of this study: biliary bicarbonate secretion was not correlated with perfusate pH, whereas bile flow and biliary bicarbonate concentration were significantly correlated with perfusate bicarbonate concentration. However, in guinea pigs, biliary bicarbonate concentration was always greater than perfusate bicarbonate concentration. This suggests that an active trans-
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G870
ACID-BASE BALANCE AND BILIARY
HCO; SECRETION
TABLE 4. Effect of compensated respiratory alkalosis and acidosis on bile flow, bile acid secretion and biliary electrolyte concentration in isolated guinea pig liver under secretin infusion
rate,
Bile Perfusate Group
Flow, & mine1 . g liver-’
Period
Compensated respiratory alkalosis (n = 4) Compensated respiratory acidosis (n = 3)
Acid secretion rate, nmol . min-’ - g liver-l
Biliary
concentration,
Na+
K+
7.8t0.2 7.7+0.1$
Basal Exptl
2.25kO.15 1.87t0.13”r
0.59t0.28 0.26+0.07$
145t2 141+7$
Basal Exptl
2.86t0.43 3.57&0.65-j-
0.41kO.21 0.46+0.40$
150tl 151+2$
mM
96.8k5.0 102.0t4.4”
11.2kO.8 12.6t1.3*
HCO,
Cl-
74.5t4.0 74.7*3.8$
Bile flow (pl/min.g liver)
mmHg
HCO,,
mM
45.00t2.80 36.40_+3.71j-
7.43t0.01 7.43+0.01$
34.3620.91 27.15&1.20-I
21.6020.49 17.3*0.56t
76.73t3.91 83.47+5.81$
7.44t0.02 7.42+0.03$
34.39t1.87 59.11+7.221-
22.33t0.21 36.57t1.56t
Values are means t SE. Statistical values refer to comparison, for each variable, of the basal period (mean initial and postexperimental final period of 30 min) with experimental period (compensated respiratory alkalosis < 0.05; t P < 0.01; $ P > 0.05, not significant.
10
Pco~,
PH
value between preexperimental or acidosis). Wilcoxon test: * P
r= 0.72 p
0.05, not significant.
port not regulated by systemic pH could be involved. In rats, in contrast, during spontaneous biliary secretion, there seemsto be a passive equilibration between plasma bicarbonate concentration and biliary bicarbonate concentration (11). The mechanism and site of bicarbonate secretion in guinea pigs are not known. Under secretin infusion, we have shown that there was no correlation between perfusate pH and biliary bicarbonate secretion but that there was a significant corre-
Bile HCO, Secretion, nmol . mine1 liver-l
of 5 x low4 M. Data
48.88k3.75 51.30t1.15t were
compared
between
control
and
lation between biliary bicarbonate secretion and systemic bicarbonate concentration; there was also a correlation between biliary bicarbonate secretion and perfusate Pco~. These results are at variance with those of Mathisen and Raeder (17) in pigs. These authors have found a significant correlation between biliary bicarbonate secretion and perfusate pH in pigs in vivo under secretin infusion. The discordance between the results of our study and those of Mathisen and Raeder are difficult to
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
ACID-BASE
BALANCE
AND
explain. This difference might be due to a difference in species or in experimental protocols. In our study, we have found a good correlation between biliary bicarbonate secretion and perfusate bicarbonate concentration. This observation suggests that a basolateral bicarbonate influx mechanism could be present in the secretin-responsive cells; in particular, one can speculate that a Na+-HCOY symporter mediates the bicarbonate transport. Such a mechanism has been described in the basolateral membrane of the rat hepatocyte (21) where it could act in conjunction with a canalicular Cl--HCOg exchanger (18). Under secretin infusion, we have also found a correlation between bicarbonate secretion and perfusate Pco~. A bicarbonate influx mechanism in the ductular cells can not account alone for this effect. This suggests that, in addition or instead of this mechanism, a transport of H+ (or OH-) could occur. It had been suggested by Garcia-Marin et al. (10) that the transport of bicarbonate ions could be due to transport of H+ rather than transport of bicarbonate itself. The most likely hypothesis would be that the mechanism responsible for bicarbonate secretion is a H+ extrusion at the basolateral membrane of the secretin sensitive cell. The mechanism of this H+ extrusion is generally thought to be either a Na+-H+ exchanger or a H+ pump (23). In this hypothesis, plasma bicarbonate could be associated with a proton (transported from the cell via the H+ transport system) to produce carbon dioxide. This carbon dioxide would diffuse across the ductular cell basolateral membrane and then be hydrated inside the cell upon the action of carbonic anhydrase to form H+ and bicarbonate. The H+ would then supply the H+ transport system, whereas HCO, would be secreted into bile. In different epithelia, such as kidney tubule and particularly proximal tubule (4, 5) or pancreas (24), the transport of bicarbonate occurs secondary to a Na+-H+ exchange localized at the contraluminal side of the cell. This Na+-H+ exchange is responsible for a large or the whole part of the bicarbonate transport. In other epithelia, such as urinary bladder, a proton pump is thought to have a role in the bicarbonate secretion (1). Using isolated perfused rat liver, Lake et al. (16) have shown that amiloride, a well-known inhibitor of the Na’H+ exchange, decreased the ursodeoxycholic acid-stimulated bile flow and biliary HCO; output. Their findings were in agreement with the hypothesis of Garcia-Marin et al. (lo), who had proposed that such an exchange could be responsible for the biliary bicarbonate secretion in the rat. In a recent paper, Grotmol et al. (13) have shown that ursodeoxycholic acid-dependent canalicular HCO, secretion was inhibited by amiloride. They suggested that the Na’-H’ exchange could play an important role in mediating canalicular choleresis. Arias and Forgac (2) have shown that such a Na+-H+ exchanger was present in sinusoidal membrane vesicles. In the same paper, Grotmol et al. (13) showed that amiloride did not alter secretin-dependent biliary bicarbonate secretion, but it was inhibited by dicyclohexylcarbodiimide, an inhibitor of all H+ translocating ATPases (12). They concluded that the secretin-dependent ductular biliary bicarbonate secretion in the pig was most likely mediated
BILIARY
HCO.?
G871
SECRETION
by a proton pump. Using a computer simulation, they confirmed their findings (14). In our study, the effect of 5 X lo-* M amiloride was tested. Under secretin infusion, it appeared that amiloride did not modify either biliary bicarbonate concentration or bile flow; hence, amiloride did not modify biliary bicarbonate secretion. These findings suggest that the Na+-H’ exchanger is not involved in the secretininduced bicarbonate secretion in guinea pigs. Another H+ transport system is certainly involved, maybe an H+ ATPase, as described in the pig (12, 14). Anyway, whatever the mechanism of H+ transport, the role of carbonic anhydrase would be essential, since this enzyme is responsible for the equilibrium between COZ, H+, and HCO;. In addition, the influence of PCO~ on biliary bicarbonate secretion under secretin infusion supports the view that intracellular pH may be involved in the regulation of biliary bicarbonate secretion by ductular cells. It is known that an increase in PCO~ induces an acidification of the hepatocyte cytosol and, likewise, a decrease in PCO~ induces an alkalinization of the cytosol (8). In the hypothesis of a H+ transport mechanism at the basal side of the cell, the regulation of bicarbonate secretion would be mainly due to the activity of this system. Finally, we have observed that the regression lines between bile flow and biliary HCO; concentration in basal (spontaneous) and secretin-stimulated conditions were parallel (there was no relation between bile flow and biliary chloride, sodium, or potassium concentration). This suggests that the choleretic effect of the bicarbonate ion is the same in the two conditions. In conclusion, this study shows that spontaneous and secretin-stimulated biliary bicarbonate secretion in the isolated perfused guinea pig liver was not regulated by the perfusate pH but was strongly correlated with perfusate bicarbonate concentration. Under secretin infusion, there was also a relation between PCO~ and biliary bicarbonate secretion. The results are consistent with the view that the cellular mechanism responsible for secretin-stimulated bicarbonate secretion could, at least in part, involve transport of H+ (or OH-) rather than transport of bicarbonate itself and that intracellular pH could be involved in the regulation of this secretion. The absence of effect of 5 x lo-* M amiloride on biliary bicarbonate secretion supports the hypothesis that the Na’-H’ exchanger is probably not involved in secretininduced biliary bicarbonate secretion. The authors the preparation Address for Department of France. Received
7 March
thank M. H. Badoureaux for secretarial assistance in of this manuscript. reprint requests: M. Blot-Chabaud, INSERM U 246, Biology, SBCE, CEN Saclay, 91191 Gif-sur-Yvette, 1988; accepted
in final
form
26 January
1990.
REFERENCES 1. AL-AWQATI, A. MUELLER, AND P. R. STEINMETZ. Transport of H+ against electrochemical gradient in turtle urinary bladder. Am. J. Physiol. 233 (Renal Fluid Electrolyte Physiol. 2): F502-F508, 1987. 2. ARIAS, I. M., AND M. FORGAC. The sinusoidal domain of the plasma membrane of rat hepatocytes contains an amiloride-sensitive Na’H+ antiport. J. Biol. Chem. 259: 5406-5408, 1984.
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
G872
ACID-BASE
BALANCE
AND
3. BERTHELOT, P., S. ERLINGER, D. DHUMEAUX, AND A. M. PREAUX. Mechanism of phenobarbital-induced hypercholeresis in the rat. Am. J. Physiol. 219: 809-813, 1970. 4. BICHARA, M., M. PAILLARD, F. LEVIEL, AND J. P. GARDIN. Hydrogen transport in rabbit kidney proximal tubules. Na+-H+ exchange. Am. J. Physiol. 238 (Renal Fluid Electrolyte Physiol. 7): F445--F451, 1980. 5. BORON, W. F., AND E. L. BOULPAEP. Intracellular pH regulation in the renal proximal tubule of the salamander. Basolateral HCO; transport. J. Gen. Physiol. 81: 53-94, 1983. 6. CASE, R. M., A. D. CONIGRAVE, E. J. FAVALORO, I. NOVAK, C. H. THOMPSON, AND J. A. YOUNG. The role of buffer anions and protons in secretion by the rabbit mandibular salivary gland. J. Physiol. Lond. 322: 273-286, 1982. 7. CHARNEY, A. N., AND L. P. HASKELL. Relative effects of systemic pH, PCO~ and bicarbonate concentration on ileal ion transport. Am. J. Physiol. 245 (Gastrointest. Liver Physiol. 8): G230-G235, 1983. 8. COHEN, R. D., R. M. HENDERSON, R. A. ILES, AND J. A. SMITH. Metabolic interrelationships of intracellular pH measured by double-barrelled microelectrodes in perfused rat liver. J. Physiol. Lond. 330: 69-80, 1982. 9. FORKER, E. L. Two sites of bile formation as determined by mannitol and erythritol clearance in the guinea pig. J. CZin. Invest. 46: 1189-1195,1967. 10. GARCIA MARIN, J. J., M. CORBIC, M. DUMONT, G. DE COUET, AND S. ERLINGER. Role of H’ transport in ursodeoxycholate-induced biliary HCO,? secretion in the rat. Am. J. Physiol. 249 (Gastrointest. Liver Physiol. 12): G335-G341, 1985. 11. GARCIA MARIN, J. J., M. DUMONT, M. GORBIC, G. DE COUET, AND S. ERLINGER. Effect of acid-base balance and acetazolamide on ursodeoxycholate-induced biliary bicarbonate secretion. Am. J. Physiol. 248 (Gastrointest. Liver Physiol. 11): GZO-G27, 1985. 12. GROTMOL, T., T. BUANES, AND M. G. RAEDER. DCCD (N,N’Dicyclohexylcarbodiimide) inhibits biliary secretion of HCO;. Stand. J. Gastroenterol. 22: 207-213, 1987. 13. GROTMOL, T., T. BUANES, AND M. G. RAEDER. The effect of amiloride on biliary bicarbonate secretion in the anesthetized pig. Acta Physiol. Stand. 130: 447-455, 1987. 14. GROTMOL, T., T. BUANES, AND M. G. RAEDER. Effect of arterial pH and pC0, on biliary HCO, secretion in the pig. Acta Physiol. Stand. 131: 183-193,1987. 15. KLEINMAN, L. I., AND E. P. RADFORD. Ventilation standards for small animals. J. Appl. Physiol. 19: 360-362, 1964.
BILIARY
HCO;
SECRETION
16. LAKE, J. R., R. W. VAN DYKE, AND B. F. SCHARSCHMIDT. Effects of Na’ replacement and amiloride on ursodeoxycholic acid-stimulated choleresis and biliary bicarbonate secretion. Am. J. Physiol. 252 (Gastrointest. Liver Physiol. 15): Gl63-G169, 1987. 17. MATHISEN, O., AND M. RAEDER. Mechanism of hepatic bicarbonate secretion and bile acid independent bile secretion. Eur. J. Invest. 13: 193-200, 1983. 18. MEIER, P. J., R. KNICKELBEIN, R. H. MOSELEY, J. W. DOBBINO, AND J. L. BOYER. Evidence for carrier-mediated chloride/bicarbonate exchange in canalicular rat liver plasma membrane vesicles. J. Clin. Invest. 75: 1256-1263, 1985. 19. MILLER, L. L., C. G. BLY, M. L. WATSON, AND W. F. BALE. The dominant role of the liver in plasma protein synthesis. J. Exp. Med. 94: 431-445, 1951. 20. RAEDER, M., A. MO, S. AUNE, AND 0. MATHISEN. Relationship between plasma pH and pancreatic HCO, secretion at different intravenous secretin infusion rates. Acta Physiol. Stand. 109: 187191, 1980. 21. RENNER, E. L., J. R. LAKE, B. F. SCHARSCHMIDT, B. ZIMMERLI, AND P. J. MEIER. Rat hepatocytes exhibit basolateral Na’/HCO; cotransport. J. Clin. Invest. 83: 1225-1235, 1989. 22. SASAKI, S., C. A. BERRY, AND F. C. RECTOR. Effect of luminal and peritubular HCO; concentrations and pC0, on HCO: reabsorption in rabbit proximal convoluted tubules perfused in vitro. J. Clin. Invest. 70: 639-649, 1982. 23. SCHARSCHMIDT, B. F., AND R. W. VAN DYKE. Mechanisms of hepatic electrolyte transport. Gastroenterology 85: 1199-1214,1983. 24. SCHULZ, I. Bicarbonate transport in the exocrine pancreas. Ann. NY Acad Sci. 341: 191-209, 1980. 25. SCHWARTZ, G. J. Na+-dependent H’ efflux from proximal tubule: evidence for reversible Na+-H+ exchange. Am. J. Physiol. 241 (Renal Fluid Electrolyte Physiol. 10): F380-F385, 1981. 26. SNEDECOR, G. W., AND W. G. COCHRAN. StatisticaL Methods (6th ed.). Ames: Iowa State Univ., 1967, p. 381-418. 27. STEINMETZ, P. R., AND L. R. LAWSON. Defect in urinary acidification induced in vitro by amphotericin B. J. CZin. Invest. 49: 596601,197O. 28. TAVOLONI, N., AND F. SCHAFFNER. The intrahepatic biliary epithelium in the guinea-pig: is hepatic artery blood flow essential in maintaining its function and structure. Hepatology Baltimore 5: 666-672,1985. 29. WHEELER, H. O., AND P. L. MANCUSI-UNGARO. Role of bile ducts during secretin choleresis in dogs. Am. J. Physiol. 210: 1153-1159, 1966.
Downloaded from www.physiology.org/journal/ajpgi at Karolinska Institutet University Library (130.237.122.245) on February 12, 2019.
