Reversal of Ethinyl Estradiol-induced Bile Secretory Failure with Triton WR-1339

The effects of Triton WR-1339 and phenobarbital on ethinyl estradiol bile secretory failure were examined to determine the mechanism responsible for decreased bile salt excretion. When administered to ethinyl estradiol-treated rats, Triton WR-1339 restored bile salt independent bile flow and maximum taurocholate transport, whereas phenobarbital corrected bile flow only. Ethinyl estradiol decreased the activities of Na(+)-K(+)-ATPase, 5'-nucleotidase, while increasing the activities of Mg(++)-ATPase and alkaline phosphatase. In contrast to these heterogeneous changes in surface membrane enzyme activities, the number and affinity of [(14)C]cholic acid carriers were not altered. When administered in vivo or added directly to surface membrane fractions Triton WR-1339 restored the activities of Na(+)-K(+)-ATPase and Mg(++)-ATPase of rats treated with ethinyl estradiol through a process that did not require protein synthesis (unaffected by cycloheximide). Phenobarbital also restored the activity of Na(+)-K(+)-ATPase to control levels, but, unlike Triton WR-1339 it did not correct the defect responsible for reduced bile salt secretion. Ethinyl estradiol increased the concentration of cholesterol esters in surface membrane fractions. When administered to ethinyl estradiol-treated rats, Triton WR-1339 restored cholesterol ester concentrations to normal, whereas phenobarbital did not. These combined data suggest that decreased or altered bile salt carriers or reduced sodium driving forces resulting from impaired activity of Na(+)-K(+)-ATPase are not responsible for decreased bile salt excretion in ethinyl estradiol-treated rats. It is proposed that the diverse changes in surface membrane function, which are associated with ethinyl estradiol bile secretory failure, may be the result of a generalized alteration in membrane lipid structure.     

Role of liver plasma membrane fluidity in the pathogenesis of estrogen-induced cholestasis

The role of liver plasma membrane (LPM) fluidity in the pathogenesis of intrahepatic cholestasis in rats was assessed by comparing the effects of ethinyl estradiol, a cholestatic agent, and spironolactone on membrane fluidity and bile flow. Spironolactone is a steroid that has some feminizing actions but that lacks the phenolic A ring necessary for estrogens to cause cholestasis. Bile flow was reduced 42% (p less than 0.01) by ethinyl estradiol and increased 22% (p less than 0.05) by spironolactone; however, both agents produced a significant reduction of membrane Na+, K+-ATPase activity (p less than 0.01) and fluidity (p less than 0.01). The decreased fluidity persisted in liposomes prepared from the total lipid extract as well as the phospholipid extract of these membranes. Both agents produced similar significant increases in the cholesterol ester content and cholesterol-to-phospholipid molar ratio of the membranes. In addition, ethinyl estradiol and spironolactone increased the membrane sphingomyelin content (15% and 11%, respectively); however, neither agent altered the fatty acid composition of the phospholipids. Because the decreased fluidity persisted in liposomes prepared from phospholipids extracted from the LPMs of treated rats, changes in membrane cholesterol are not the sole cause of the altered membrane fluidity. Rather, the increased sphingomyelin is at least partially responsible for these changes. Also, because ethinyl estradiol and spironolactone produce similar changes in LPM lipid composition and fluidity but disparate effects on bile flow, membrane fluidity as assessed by fluorescence polarization does not appear to be the rate-limiting determinant of bile flow in estrogen-induced cholestasis.     

The effect of complete biliary obstruction on bile secretion. Studies on the mechanisms of postcholestatic choleresis in the rat

It has been previously shown that rats subjected to obstructive cholestasis demonstrate in the postcholestatic period, after release of common duct obstruction, a marked increase in canalicular bile flow relative to bile acid excretion. To characterize this phenomenon further, we investigated whether changes in canalicular permeability and in the activity of (Na+-K+)-ATPase in isolated liver surface membranes are associated with postcholestatic choleresis. With this purpose, the clearances of 14C-erythritol and 3H-insulin were simultaneously measured in rats subjected to a 3-day obstructive cholestasis and in controls, during spontaneous choleresis as well as during the intravenous infusion of sodium taurocholate at both submaximal and saturating rates. In additional groups of bile duct-ligated rats and controls, liver surface membrane fractions were isolated, and the activity of appropriate marker enzymes and (Na+-K+)-ATPase was determined. In both groups 14C-erythritol clearance closely approximated total bile flow, suggesting that bile flow was of canalicular origin. However, cholestatic rats showed a sixfold increase in 3H-inulin clearance compared to controls. These results suggest that canalicular permeability to inulin is markedly increased in cholestatic rats. On the other hand, (Na+-K+)-ATPase activity was significantly higher in cholestatic rats than in controls in both homogenate (1.27 +/- 0.07 and 0.89 +/- 0.07 U/mg of protein, respectively, p less than 0.001) and liver surface membranes (22.6 +/- 1.2 and 17.5 +/- 1.2 U/mg of protein, respectively, p less than 0.001). Thus enhanced choleretic response to bile acids in the postcholestatic period is associated with an increased permeability of canalicular structure to inulin and with a significant increase in both homogenate and surface membrane (Na+-K+)-ATPase activity. In addition, this study points out some important differences between bile secretory function of rats subjected to obstructive cholestasis and that described in models of bile secretory failure induced by drugs or monohydroxy-bile acids.     

Class III P-glycoproteins mediate the formation of lipoprotein X in the mouse.

Cholestasis is associated with hypercholesterolemia and appearance of the abnormal lipoprotein X (LpX) in plasma. Using mice with a disrupted Mdr2 gene, we tested the hypothesis that LpX originates as a biliary lipid vesicle. Mdr2-deficient mice lack Mdr2 P-glycoprotein, the canalicular translocator for phosphatidylcholine, and secrete virtually no phospholipid and cholesterol in bile. Bile duct ligation of Mdr2(+)/+ mice induced a dramatic increase in the plasma cholesterol and phospholipid concentration. Agarose electrophoresis, density gradient ultracentrifugation, gel permeation, and electron microscopy revealed that the majority of phospholipid and cholesterol was present as LpX, a 40-100 nm vesicle with an aqueous lumen. In contrast, the plasma cholesterol and phospholipid concentration in Mdr2(-)/- mice decreased upon bile duct ligation, and plasma fractionation revealed a complete absence of LpX. In mice with various expression levels of Mdr2 or MDR3, the human homolog of Mdr2, we observed that the plasma level of cholesterol and phospholipid during cholestasis correlated very closely with the expression level of these canalicular P-glycoproteins. These data demonstrate that during cholestasis there is a quantitative shift of lipid secretion from bile to the plasma compartment in the form of LpX. The concentration of this lipoprotein is determined by the activity of the canalicular phospholipid translocator.     

Effect of alpha-asarone and a derivative on lipids, bile flow and Na+/K+-ATPase in ethinyl estradiol-induced cholestasis in the rat

Administration of ethinyl estradiol (EE), a widely used component of oral contraceptives, has been associated with impairment of bile flow and the capacity to excrete organic anions in man and experimental animals. alpha-Asarone (2,4,5-trimethoxypropenylbenzene) and 2-methoxy-4-(2-propenyl) phenoxyacetic acid (MPPA) have shown hypolipidemic effects. In addition to these effects, we decided to evaluate the properties of these compounds on EE-induced cholestasis. Wistar male rats were injected subcutaneously with 10 mg/kg of EE for 5 days; simultaneously, alpha-asarone or MPPA were also administered and appropriate controls were performed. alpha-asarone and MPPA decreased plasma and bile cholesterol. EE diminished triglycerides total, low-density lipoprotein, high-density lipoprotein and bile cholesterol. MPPA further decreased these lipid parameters. Alkaline phosphatase (an enzyme marker of cholestasis) was increased after administration of EE, but this effect was prevented significantly by alpha-asarone or MPPA administration. Bile flow was importantly decreased by EE and increased by alpha-asarone alone. Furthermore, alpha-asarone or MPPA preserved the normal bile flow in EE-treated rats. EE inhibited the activity of the Na(+)/K(+)-ATPase, while both alpha-asarone and MPPA preserved this enzyme activity. Na(+)/K(+)-ATPase is involved in Na(+)-coupled uptake of bile acids into hepatocytes and, therefore, ultimately is the driving force for the generation of bile flow. Therefore, the anticholestatic effects of alpha-asarone and MPPA, described herein by the first time, may be due to its ability to preserve ATPase activity. This enzyme is negatively regulated by membrane cholesterol, thus the hypolipidemic effects of the compounds tested may be responsible for Na(+)/K(+)-ATPase activity and bile flow maintenance.     

Relationship between bile flow and Na+, K+-adenosinetriphosphatase in liver plasma membranes enriched in bile canaliculi.

The relationship between bile flow and Na+,K+-ATPase activity in liver plasma membranes enriched in bile canaliculi was studied in rats treated with ethinyl estradiol, phenobarbital, or 20-methyl cholanthrene. In comparison with controls (1.49+/-0.12 microliter/min per g liver), bile flow was significantly diminished by ethinyl estradiol, increased by phenobarbital, and unchanged by 20-methyl cholanthrene or the solvent, propanediol (0.92+/-0.31, 2.50+/-0.21, 1.62+/-0.18, and 1.64+/-0.30 microliter/min per g liver, respectively). The corresponding values for canalicular Na+,K+-ATPase activity were 80.7+/-19.2, 50.0+/-18.4, 231.7+/-42.6, 82.7+/-30.7, and 143.6+/-55.3 micronmol Pi/h per g liver. Canalicular Na+,K+-ATPase activity was significantly correlated (r=0.785, n=31) with bile flow. These findings support the hypothesis that a fraction of bile flow is related to Na+,K+-ATPase activity and canalicular Na+ transport.     

Effect of cholestasis on hepatic transport of 99mtechnetium p-isopropyl-iminodiacetic acid

Hepatobiliary imaging with 99mTc-p-isopropyl-iminodiacetic acid (PIPIDA) and other acetanilidoiminodiacetic acid derivatives is a frequently used clinical tool in evaluating patients with jaundice. However, there has been little objective assessment of the effects of cholestasis on hepatic transport of acetanilioiminodiacetic acid derivatives. In our studies, transport of 99mTc-PIPIDA by isolated rat hepatocytes obtained from animals with extrahepatic obstruction secondary to bile duct ligation or intrahepatic cholestasis induced by ethinyl estradiol therapy was determined. Uptake constants for 99mTc-PIPIDA by hepatocytes isolated from livers of animals with ligated bile ducts were significantly decreased compared with uptake by liver cells from sham-operated controls (0.0030 +/- 0.0003 vs. 0.0089 +/- 0.0010 femtomole X 10(6) cells-1 X min-1 X pmol/L-1; p less than 0.001). Hepatocytes isolated from livers of animals given ethinyl estradiol also demonstrated significantly reduced 99mTc-PIPIDA uptake compared with controls given propylene glycol (0.0034 +/- 0.0002 vs. 0.0060 +/- 0.0004 fmol X 10(6) cells-1 X min-1 X pmol/L-1; p less than 0.001). Fractional rates of efflux of the study compound from hepatocytes preincubated with 99mTc-PIPIDA were significantly decreased in experiments using ethinyl estradiol (p less than 0.005) but did not differ significantly from controls in studies of bile duct ligation. 99mTc-PIPIDA uptake was significantly decreased in the presence of high bile salt concentrations (100 to 200 mumol/L), but unconjugated bilirubin concentrations as high as 200 mumol/L (approximately 12 mg/dl) had no effect on hepatocyte uptake of the ligand. The finding that cholestasis significantly impairs hepatocyte uptake of 99mTc-PIPIDA provides a possible explanation for the clinical observation that a patent biliary tree and normal serum bilirubin level are not always sufficient to ensure normal hepatobiliary imaging. These data also suggest that elevation of bile acid levels during cholestasis may either contribute to impaired uptake of hepatobiliary imaging agents or serve as a marker of cholestasis-induced abnormalities in the liver functions responsible for hepatic transport of these compounds.     

The effect of thyroid hormone on bile salt-independent bile flow and Na+, K+ -ATPase activity in liver plasma membranes enriched in bile canaliculi.

The relationship between bile salt-independent canalicular flow and ATPase activity in liver plasma membranes (LPM) enriched in bile canaliculi, was studied in control, hyperthyroid, and hypothyroid rats. Canalicular bile production was significantly increased in hyperthyroid rats (3.19 +/- 0.23 mul/min per g liver) compared to controls (2.27 +/- 0.24 mul/min per g liver), while it diminished in hypothyroid animals (1.58 +/- 0.17 mul/min per g liver). Although bile salt excretion was also increased in hyperthyroid animals (62.4 +/- 13.3 vs. 41.2 +/- 8.4 nmol/min per g liver), the stimulation in canalicular secretion was primarily related to enhancement of the bile salt-independent fraction of flow (2.47 mul/min per g liver in hyperthyroid rats vs. 1.67 mul/min per g liver in controls). LPM Na+, K+-ATPase activity doubled in hyperthyroid animals (21.5 +/- 5.8 vs. 10.7 +/- 3.1 mumol Pi/mg protein per h) while Mg++-ATPase activity remained unchanged and 5'-nucleotidase activity increased to a small but significant extent. In hypothyroid rats, bile salt excretion remained unchanged from control values so that the reduced secretion was entirely secondary to an inhibition of bile salt-independent secretion (1.19 mul/min per g liver). Na+, K+-ATPase activity in the LPMs from hypothyroid animals decreased by nearly 50% (5.4 +/- 1.6 mumol Pi/mg protein per h), although comparable reductions in the specific activity of Mg++-ATPase and 5'-nucleotidase were also observed. Administration of L-thyroxine to hypothyroid animals restored both bile salt-independent canalicular secretion and membrane enzymes to control values within 2 and 4 days, respectively. Sodium dodecyl sulfate gel electrophoresis demonstrated no significant changes in LPM protein fractions from any of the treatment groups. These studies indicate that thyroid hormone has a parallel effect on bile salt-independent canalicular secretion and LPM Na+, K+-ATPase activity, supporting the hypothesis that Na+ transport and Na+, K+-ATPase may be determinants of bile salt-independent canalicular flow.     

Stimulation of hepatic sodium and potassium-activated adenosine triphosphatase activity by phenobarbital. Its possible role in regulation of bile flow.

Since phenobarbital administration produces a profound increase in bile flow without changing bile acid secretion, we examined whether this drug increases the activity of hepatic sodium-potassium-activated ATPase [Na+-K+)-ATPase], the postulated regulating enzyme in the secretion of bile salt independent bile flow. After freeze-thawing to increase substrate accessibility, (Na+-K+) ATPase activity was determined by ouabain inhibition of total ATPase activity. Its activity was highest in isolated liver surface membrane fractions enriched in bile canalicult. Phenobarbital administration significatly increased (Na+-K+)-ATPase activity in both liver surface membrane fractions as well as liver homogenates. This enhanced activity is apparently selective for other membrane phosphatases and the enzyme activity in other tissues is either unaltered or decreased. Kinetic analysis of (Ka+-K+)-ATPase indicates that phenobarbital treatment increased maximum velocity and half-maximum activation constant was unchanged, consistent with activation of latent molecules or an increased number of enzyme molecules. The latter process seems more likely because cycloheximide prevented phenobarbital induction and activators were not demonstrated in vitro. Examination of the full time course of phenobarbital induction to determine whether phenobarbital increased synthesis or decreased degradation was consistent with increased synthesis since the apparent degradation rates were similar with or without phenobarbital treatment. The apparent half-life for (Na+-K+)-ATPase was estimated to be approximately 2.5 days, consistent with liver surface membrane protein turnover. The correlation of changes in bile flow with (Na+-K+)-ATPase was examined under several experimental situations. Phenobarbital caused a parallel increase in each during the 1st 2 days of greatment: thereafter other factors become rate limiting for flow, since enzyme activity doesn't reach a new steady state until 4-days. Consistent with increased sodium-potassium exchange, bile sodium was unchanged while potasium concentrations were significantly reduced. Changes in both bile flow and (Na+-K+)-ATPase induced by phenobarbital are independent of thyroid hormone. These studies support the postulate that (Na+-K+)-ATPase is an important factor in regulation of bile flow. In addition, phenobarbital enhancement of both bile flow and (Na+-K+)-ATPase is dependent upon de novo protein synthesis.     

Effect of chlorpromazine on hepatic perfusion and bile secretory function in the isolated perfused rat liver.

The hepatotoxicity of CPZ was studied in the isolated perfused rat liver in order to more closely define possible mechanisms of phenothiazine-induced cholestasis. Perfusate concentrations of CPZ were increased from 5 x 10(-6) M to 5 x 10(-4) M until bile secretion was significantly inhibited. Measurements were then made of determinants of bile secretory function, including the magnitude of lobar distribution of perfusate flow, BAIF, and liver plasma membrane enzyme activity, Na+,K+-ATPase, Mg++-ATPase and 5'-nucleotidase. BAIF diminished significantly from control values of 1.76 +/- 0.07 microliter min-1gm-1 of liver to 1.34 +/- 0.15 and 0.80 +/- 0.09 following 2.5 and 5 x 10(-4) M CPZ, respectively. Perfusate flow also diminished from 5.64 +/- 0.44 to 1.24 +/- 0.12 ml min-1 gm-1 of liver 20 min following 5 x 10(-4) M CPZ and was associated with reduced flow to peripheral areas of the hepatic lobes as demonstrated by Tc-HAM. By 30 min, perfusate flow had returned to baseline values. CPZ also transiently diminished the excretion of bile acids in livers receiving a constant infusion of 40 mumol hr-1 sodium taurocholate. Defects in hepatic perfusion could not account entirely for the impairment in BAIF, since comparable mechanical restriction of perfusate flow in controls only diminished BAIF to 1.49 +/- 0.08 microliter min-1gm-1 of liver. CPZ signofocamt;u rediced tje secofoc actovotu pf Mg++-ATPase and 5'-nucleotidase but did not affect Na+,K+-ATPase in liver plasma membrane isolated 20 min after 5 x 10(-4) M CPZ. CPZ also resulted in a profound shift in the recovery of protein in isolated liver plasma membrane fractions from the light (density = 1.16) to heavier (density = 1.18) fractions. These findings, together with previous observations demonstrating alterations in hepatic ultrastructure, indicate that CPZ interacts in a complex manner with hepatocyte plasma and cytoplasmic membrane components and suggest that these drug-membrane interactions independently result in diminished hepatic perfusion, impairment of bile acid excretion, and inhibition of bile acid-independent bile secretion.     

Alteration of lipid composition of hepatic membranes associated with manganese-bilirubin induced cholestasis

Summary— One hypothesis concerning the pathogenesis of manganese-bilirubin (Mn-BR)-induced cholestasis is that the molecular organization of the bile canalicular membrane is altered. The purpose of the present study was to evaluate lipid composition and fluidity of hepatic membranes during cholestasis in male Sprague-Dawley rats. To induce cholestasis, manganese (Mn, 4.5 mg/kg, intravenously [iv]) was given IS min before bilirubin (BR, 25 mg/kg, iv). The rats were killed 30 min after BR injection, at which time bile flow was decreased by approximately 40% compared to control values. Liver cell plasma membranes enriched in canalicular fractions (BCM) and plasma membranes enriched in sinusoidal and lateral fractions (PM), microsomes, mitochondria and cytosol were isolated by differential centrifugation. Total lipids were extracted and measured colorimetrically. To assess fluidity, membranes were incubated in vitro with fluorescent probes [1,6-diphenyl-1,3,5-hexatriene and 1-(4'-trimethyl-ammonium-phenyl)-6-phenyl-1,3,5-hexatriene]. After Mn-BR treatment, BCM cholesterol incorporation increased markedly (about 3-fold) accompanied by a decrease in fluidity. BCM phospholipid content was unaltered by the cholestatic challenge. In PM-enriched fractions, the changes in cholesterol and phospholipid content after Mn-BR treatment were not statistically significant ( P > 0.05) compared to controls. Furthermore, the biochemical alterations in PM were not accompanied by changes in membrane fluidity. These results support the hypothesis that altered lipid composition and fluidity of BCM are involved in the pathogenesis of Mn-BR cholestasis.     

Ursodeoxycholic acid improves ethinyl estradiol–induced cholestasis in the rat

Abstract. The effect of oral chronic administration of ursodeoxycholic acid has been examined in rats with cholestasis induced by ethinyl estradiol. Ursodeoxycholic acid at the dose of 25 mg kg^-1 per day during 4 days, did not improve the decrease in basal bile flow and bile acid secretion induced by ethinyl estradiol alone. In contrast, when ursodeoxycholic acid was given at the same dose during 10 days, basal bile flow was significantly improved and basal bile acid secretion was restored to control values. When ursodeoxycholic acid was given at the dose of 500 mg kg^-1 per day, basal bile flow and bile acid output were not further improved. However, bile flow and bile acid output under taurocholate infusion were restored to control values. Bile of rats treated with ursodeoxycholic acid was enriched with this bile acid. These results show a significant improvement of ethinyl estradiol-induced cholestasis in rats after chronic administration of ursodeoxycholic acid and support the use of this bile acid in intrahepatic cholestasis in man.     

S-Adenosylmethionine protects against cyclosporin A-induced alterations in rat liver plasma membrane fluidity and functions.

We studied the effect of cyclosporin A (CyA) on liver plasma membrane (LPM) composition, fluidity, and functions and on hepatic glutathione (GS) and oxidative status. We also evaluated the ability of S-adenosylmethionine (SAMe) to antagonize the CyA-induced disturbances in rats. The animals were randomly divided into four groups and treated daily with saline, CyA vehicle, CyA, and SAMe plus CyA, respectively, for 1 week. Bile, blood, and liver samples and LPM vesicles were obtained at the end of the treatments. CyA-induced cholestasis was associated with alterations in LPM composition and fluidity. The contents of total phospholipids, phosphatidylcholine, and proteins were decreased and cholesterol and the cholesterol/phospholipid molar ratio increased. Na(+), K(+)-ATPase activity was decreased, whereas those of 5'-nucleotidase, Mg(2+)-ATPase, and gamma-glutamyltransferase increased. The hepatic contents of proteins and GS and the reduced/oxidized glutathione molar ratio were decreased and hepatic malondialdehyde increased. SAMe cotreatment 1) significantly improved or abolished the CyA-induced changes in LPM fluidity and composition and the changes in the activity of the carrier and enzymes tested, 2) counteracted the hepatic depletion of GS and proteins caused by CyA and normalized the reduced/oxidized glutathione ratio, and, as expected, 3) prevented cholestasis and the inhibitory effect of CyA on hepatobiliary transport of the major bile components. We conclude that CyA-induced cholestasis and hepatotoxicity in the rat is associated with changes in LPM composition and fluidity, liver GS depletion, and oxidative stress. SAMe cotreatment significantly improves or totally protects against these hepatotoxic effects.     

Effects of chronic nitric oxide activation or inhibition on early hepatic fibrosis in rats with bile duct ligation

Hepatic fibrosis or increased liver collagen contents drive functional abnormalities that, when extensive, may be life threatening. The purpose of this study was to assess the effects of the chronic stimulation or inhibition of nitric oxide synthesis in rats with hepatic fibrosis induced by permanent common bile duct ligation (3 weeks) and the role of expression of the different nitric oxide synthase isoforms. Bile duct ligation led to an important accumulation of collagen in the hepatic parenchyma, as shown both histologically and by the hydroxyproline contents of livers. Bilirubin and serum enzyme activities (measured as markers of cholestasis) increased several-fold after bile duct ligation. The area of fibrotic tissue, liver hydroxyproline content and serum markers of cholestasis were clearly related in obstructed rats. The absence of modifications in haemodynamic parameters excludes circulatory changes from being responsible for the development of liver alterations. In animals treated with NG-nitro-L-arginine methyl ester (L-NAME) the area of fibrosis was similar to that of untreated animals, the signs of cholestasis and cellular injury being more evident. In rats treated with L-arginine the area of fibrosis was almost three times larger than that found in bile duct ligated rats and in L-NAME-treated bile duct ligated rats, although the observed biochemical changes were similar to those seen in rats treated with L-NAME. Our results with inducible nitric oxide synthase, obtained by Western blots and immunohistochemistry, indicate a greater expression of the inducible enzyme in bile duct ligated and L-arginine-treated animals and a lower expression in the L-NAME and control groups. Constitutive nitric oxide synthase expression, obtained by Western blots, was very similar in all groups, except for the L-arginine-treated rats in which it was lower. These results suggest that nitric oxide production may be a key factor in the development of fibrosis in bile duct ligated rats. They also support the hypothesis of a dual role for nitric oxide; one beneficial, mediated by its circulatory effects, and the second negative, through its local toxic effects.     

Extrahepatic obstructive cholestasis reverses the bile salt secretory polarity of rat hepatocytes.

To elucidate the consequences of extrahepatic cholestasis on the structure and function of hepatocytes, we studied the effects of bile duct ligation on the turnover, surface distribution, and functional activity of the canalicular 100-kD bile salt transport protein (cBSTP). Basolateral (blLPM) and canalicular (cLPM) liver plasma membrane vesicles were purified to the same degree from normal and cholestatic rat livers and the membrane bound cBSTP identified and quantitated using polyclonal anti-cBSTP antibodies. Cholestasis of 50 h resulted in an increased release of cBSTP into bile, thereby decreasing its in vivo half-life from 65 to 25 h. Furthermore, a significant portion of cBSTP accumulated at the basolateral surface and in intracellular vesicles of cholestatic hepatocytes. This redistribution of cBSTP was functionally paralleled by decreased and increased electrogenic taurocholate anion transport in cLPM and blLPM vesicles, respectively. These results demonstrate that biliary obstruction causes a reversal of the bile salt secretory polarity of rat hepatocytes. The resulting increase in basolateral (sinusoidal) bile salt efflux might protect hepatocytes from too high an accumulation of toxic bile salts within the cell interior.     

Regulation of hepatic transport of bile salt. Effect of protein synthesis inhibition on excretion of bile salts and their binding to liver surface membrane fractions.

The overall transport of bile salts across the hepatocyte is characterized as a carrier-mediated process whose rate-limiting step is biliary secretion. Specific bile salt binding proteins have been identified in liver surface membrane fractions and were postulated to represent the initial interaction in bile salt translocation across both the sinusoidal and canalicular membranes. To test this hypothesis, cycloheximide was administered to rats to inhibit hepatic protein synthesis. 16 h after cycloheximide administration [14C]leucine incorporation into hepatic protein was inhibited by 93% at 1 h and 47% at 12 h. However, values of liver function tests were not increased, although serum albumin, serum alanine amino-transferase, and alkaline phosphatase were significantly decreased. Light and electron microscopy did not demonstrate necrosis or fat accumulation. The latter demonstrated minimal disorganization of rough endoplasmic reticulum and occasional lamellar whorls. 16 h after cycloheximide administration bile salt independent bile flow, basal bile salt excretion, and basal bile flow were unaltered, but the maximum bile salt transport capacity was reduced to 62% of control and 24 h later to 38%. Decreased bile salt transport was reversible, for it returned to control values after 48 h, when hepatic protein synthesis was also normal. Maximum bromosulfophthalein (BSP) transport, on the other hand, was reduced after 16 h to only 85% of control. Both bile salt and BPS maximum transport capacities decreased with time during inhibition of protein synthesis, apparently following first order kinetics. It was estimated that their half-lives are 20 h for bile salt transport and 55 h for BSP transport. These different turnover rates suggest that cycloheximide does not decrease active transport through generalized hepatic dysfunction or alteration of high energy sources possibly required for transport. The maximum number of [14C]cholic acid binding sites in liver surface membrane fractions was determined by an ultrafiltration assay. They were reduced to 68% of control after 16 h of cycloheximide and to 25% after 24 h. This reduction in the number of binding sites is apparently selective, for the activities of the liver surface membrane enzymes (Na+-K+)ATPase, Mg++-ATPase, and 5'-nucleotidase were not significantly changed. The associated alterations in bile salt transport and the maximum number of binding sites after cycloheximide administration suggests that these receptors may be the bile salt carriers.     

The effect of complete biliary obstruction on bile flow and bile acid excretion: postcholestatic choleresis in the rat.

Bile secretory function was studied in rats subjected to a 7-day obstructive cholestasis induced by complete common duct obstruction. Bile flow and bile acid excretion were examined during bile depletion, following the release of the biliary obstruction, and during the infusion of sodium taurocholate at submaximal and saturating rates. A highly significant increase, greater than 100%, in bile flow was evident in cholestatic rats at any bile acid excretory rate, when compared to control sham-operated rats. 14C-erythritol clearance measurements performed during bile depletion and during the infusion of taurocholate suggest that bile flow was mainly of canalicular origin in cholestatic rats. Estimated taurocholate transport maximum (mumol/min per rat) was not statistically different between cholestatic and control rats. However, significantly greater taurocholate plasma levels at Tm in cholestatic rats suggest a decreased efficiency of the bile acid transport process. In addition, the relationship between bile flow and bile acid excretion was found to be nonlinear at low bile acid excretory rates in cholestatic rats. Thus important changes in bile formation occurred in rats subjected to temporary obstructive cholestatis, which differ from those observed in other models of cholestasis that are associated to a reduction in bile flow and bile acid transport capacity.     

Viral gene delivery of superoxide dismutase attenuates experimental cholestasis-induced liver fibrosis in the rat

Hydrophobic bile acids lead to generation of oxygen free radicals in mitochondria. Accordingly, this study investigated if gene delivery of superoxide dismutase (SOD) would reduce hepatic injury caused by experimental cholestasis. Rats were given adenovirus (Ad; 3 x 10(9) p.f.u., i.v.) carrying the bacterial control gene lacZ, mitochondrial Mn-SOD or cytosolic Cu/Zn-SOD genes 3 days before bile duct ligation. Both Mn- and Cu/Zn-SOD activity was increased in the liver about four-fold 3 days after viral infection. Serum alanine transaminase increased to about 710 U/l after bile duct ligation, which was blunted by about 70% in rats receiving Ad-Mn-SOD, but by only 30% in rats receiving Ad-Cu/Zn-SOD. Bile duct ligation caused focal necrosis, apoptosis and fibrosis in the liver and increased collagen alpha1 mRNA about 20-fold. These effects were reduced significantly by Ad-Mn-SOD, but not by Ad-Cu/Zn-SOD. In addition, bile duct ligation increased 4-hydroxynonenal, a product of lipid peroxidation, activated NF-kappaB and increased synthesis of TNF(alpha) and TGF-beta. These effects were also blunted significantly by Ad-Mn-SOD, but not by Ad-Cu/Zn-SOD. Taken together, it is concluded that cholestasis causes liver injury by mechanisms involving mitochondrial oxidative stress. Gene delivery of mitochondrial Mn-SOD blocks formation of oxygen radicals and production of toxic cytokines thereby minimizing liver injury caused by cholestasis.     

Plasma alkaline phosphodiesterase I in intrahepatic cholestasis induced by alpha-naphthylisothiocyanate in rats

1. Administration of alpha-naphthylisothiocyanate (ANIT) to rats produced dose-dependent increases in plasma bile acid and bilirubin concentrations. Similar increases in plasma bile acid and bilirubin concentrations were evident in bile duct ligated rats, indicating that the severity of cholestasis is almost identical in both models. 2. Plasma alkaline phosphodiesterase I was increased by only 50-80% while alkaline phosphatase was increased more than threefold after ANIT administration. This is in contrast to an earlier study [S. R. Simpson, K. Rahman & D. Billington (1984) Clinical Science 67, 647-652] where, after bile duct ligation, serum alkaline phosphodiesterase I was elevated sixfold before any increase in alkaline phosphatase activity became apparent. Thus, plasma alkaline phosphodiesterase I does not offer as sensitive a marker of intrahepatic cholestasis (induced by ANIT) as it does of extrahepatic cholestasis (induced by bile duct ligation). 3. Hepatic alkaline phosphodiesterase I was unaffected by ANIT pretreatment while hepatic alkaline phosphatase was increased up to seven times. It is suggested that raised plasma alkaline phosphodiesterase I is due to regurgitation of the biliary enzyme rather than overspill of the enzyme from liver into blood. 4. Gel filtration showed that 24 h and 96 h after ANIT administration, rat serum contained a high molecular weight form of alkaline phosphodiesterase I, suggesting a different isoenzyme profile.     

Studies of Relationships among Bile Flow, Liver Plasma Membrane NaK-ATPase, and Membrane Microviscosity in the Rat

Liver plasma membrane (LPM) NaK-ATPase activity, LPM fluidity, and bile acid-independent flow (BAIF) were studied in rats pretreated with one of five experimental agents. Compared with controls, BAIF was increased 24.6% by thyroid hormone and 34.4% by phenobarbital, decreased by ethinyl estradiol, but unchanged by propylene glycol and cortisone acetate. Parallel to the observed changes in BAIF, NaK-ATPase activity also was increased by thyroid hormone (40.8%) and decreased by ethinyl estradiol (26.2%). In contrast, NaK-ATPase activity failed to increase after phenobarbital but did increase 36% after propylene glycol and 34.8% after cortisone acetate. Thus BAIF and NaK-ATPase activity did not always change in parallel. The NaK-ATPase K(m) for ATP was not affected by any of these agents.LPM fluidity, measured by fluorescence polarization using the probe 1,6-diphenyl-1,3,5-hexatriene, was found to be increased by propylene glycol, thyroid hormone, and cortisone acetate, decreased by ethinyl estradiol, and unaffected by phenobarbital. Thus in these cases, induced changes in LPM fluidity paralleled those in NaK-ATPase activity. In no case did Mg-ATPase or 5'-nucleotidase activities change in the same direction as NaK-ATPase, and the activity of neither of these enzymes correlated with LPM fluidity, thus indicating the selective nature of the changes in LPM enzyme activity caused by the agents.These findings indicate that LPM fluidity correlates with NaK-ATPase activity and may influence the activity of this enzyme. However, the nature of the role of LPM NaK-ATPase in bile secretion is uncertain and needs further study.