Cryptosporidium parvum activates nuclear factor kappaB in biliary epithelia preventing epithelial cell apoptosis.

BACKGROUND & AIMS: Our previous studies have shown that Cryptosporidium parvum induces biliary epithelial cell apoptosis in vivo and causes apoptosis in bystander uninfected biliary epithelia in vitro. We analyzed C. parvum-induced nuclear factor kappa B (NF-kappaB) activation in human biliary epithelial cells and assessed its relevance to epithelial cell apoptosis. METHODS: In vitro models of cryptosporidial infection using a human biliary epithelial cell line were used to assay C. parvum- induced NF-kappaB activation and associated apoptosis. RESULTS: Degradation of I(kappa)B and nuclear translocation of the NF-kappaB family of proteins (p65 and p50) were observed in the biliary epithelial cell cultures directly exposed to the parasite. Activation of NF-kappaB was found only in directly infected cells (but not in bystander uninfected cells). A time-dependent secretion of a known NF-kappaB gene product, interleukin 8, from infected cell cultures was detected. C. parvum-induced biliary epithelial cell apoptosis was limited to bystander uninfected cells. In contrast, inhibition of NF-kappaB activation resulted in apoptosis in directly infected cells and significantly enhanced C. parvum-induced apoptosis in bystander uninfected cells. CONCLUSIONS: These observations support the concept that, while C. parvum triggers host cell apoptosis in bystander uninfected biliary epithelial cells, which may limit spread of the infection, it directly activates the NF-kappaB/I(kappa)B system in infected biliary epithelia thus protecting infected cells from death and facilitating parasite survival and propagation.

Upregulation of secretin receptors on cholangiocytes after bile duct ligation.

Secretin not only increases ductular bile secretion in vivo in rats after bile duct ligation (BDL) [1], but also increases cAMP levels and stimulates exocytosis in isolated cholangiocytes [2]. Although we have previously reported that secretin receptor mRNA was upregulated in cholangiocytes after BDL [3], the cholangiocyte secretin receptor has not been functionally characterized or quantified after BDL. In this work, we used a novel, photolabile and biologically active analogue of secretin to quantify and characterize secretin receptors on cholangiocytes isolated from normal and BDL rats. The cholangiocyte secretin receptor bound radioligand with high affinity and in a rapid, reversible, and temperature-dependent manner. While receptors on cholangiocytes from normal and BDL rats were functionally and biochemically identical, receptor density on cholangiocytes was increased 5-fold following BDL. The combination of increased cell number with increased functional secretin receptors per cell is due to the fact that cholangiocyte hyperplasia represents a reactive response to a cholestatic condition and this effort on the part of the organism to maintain bile secretion, explains the increased hormone-responsive choleresis observed after BDL and may reflect an adaptive response of the organism to cholestasis.

Expression of aquaporin-4 water channels in rat cholangiocytes.

We recently reported that secretin induces the exocytic insertion of functional aquaporin-1 water channels (AQP1) into the apical membrane of cholangiocytes and proposed that this was a key process in ductal bile secretion. Because AQP1 is present on the basolateral cholangiocyte membrane in low amounts, we hypothesized that another AQP must be expressed at this domain to facilitate transbasolateral water movement. Thus, we investigated the expression, subcellular localization, possible regulation by secretin, and functional activity of AQP4, a mercury-insensitive water channel expressed in other fluid transporting epithelia. Using reverse transcription-polymerase chain reaction (RT-PCR) on RNA prepared from purified rat cholangiocytes, we amplified a product of 311 bp that was 100% homologous to the reported AQP4 sequence. RNase protection assay confirmed the presence of an appropriate size transcript for AQP4 in cholangiocytes. Immunoblotting detected a band of approximately 31 kd corresponding to AQP4 in basolateral but not apical membranes of cholangiocytes. Secretin did not alter the amount of plasma membrane AQP4 but, as expected, induced AQP1 redistribution from intracellular to apical plasma membranes. Functional studies showed that AQP4 accounts for about 15% of total cholangiocyte membrane water permeability. Our results indicate that: (1) cholangiocytes express AQP4 messenger RNA (mRNA) and protein and (2) in contrast to AQP1, which is targeted to the apical cholangiocyte membrane by secretin, AQP4 is constitutively expressed on the basolateral cholangiocyte membrane and is secretin unresponsive. The data suggest that AQP4 facilitates the basolateral transport of water in cholangiocytes, a process that could be relevant to ductal bile formation.

Secretin induces the apical insertion of aquaporin-1 water channels in rat cholangiocytes.

Aquaporin-1 (AQP1) water channels are present in the apical and basolateral plasma membrane domains of bile duct epithelial cells, or cholangiocytes, and mediate the transport of water in these cells. We previously reported that secretin, a hormone known to stimulate ductal bile secretion, increases cholangiocyte osmotic water permeability and stimulates the redistribution of AQP1 from an intracellular vesicular pool to the cholangiocyte plasma membrane. Nevertheless, the target plasma membrane domain (i.e., basolateral or apical) for secretin-regulated trafficking of AQP1 in cholangiocytes is unknown, as is the functional significance of this process for the secretion of ductal bile. In this study, we used primarily an in vivo model (i.e., rats with cholangiocyte hyperplasia induced by bile duct ligation) to address these issues. AQP1 was quantitated by immunoblotting in apical and basolateral plasma membranes prepared from cholangiocytes isolated from rats 20 min after intravenous infusion of secretin. Secretin increased bile flow (78%, P < 0.01) as well as the amount of AQP1 in the apical cholangiocyte plasma membrane (127%, P < 0.05). In contrast, the amount of AQP1 in the basolateral cholangiocyte membrane and the specific activity of an apical cholangiocyte marker enzyme (i.e., gamma-glutamyltranspeptidase) were unaffected by secretin. Similar observations were made when freshly isolated cholangiocytes were directly exposed to secretin. Immunohistochemistry for AQP1 in liver sections from secretin-treated rats showed intensified staining at the apical region of cholangiocytes. Pretreatment of rats with colchicine (but not with its inactive analog beta-lumicolchicine) inhibited both the increases of AQP1 in the cholangiocyte plasma membrane (94%, P < 0.05) and the bile flow induced by secretin (54%, P < 0.05). Our results in vivo indicate that secretin induces the microtubule-dependent insertion of AQP1 exclusively into the secretory pole (i.e., apical membrane domain) of rat cholangiocytes, a process that likely accounts for the ability of secretin to stimulate ductal bile secretion.

Dynamic measurements of the acute and chronic effects of lysosomotropic agents on hepatocyte lysosomal pH using flow cytometry.

Previous investigators measuring the pH of lysosomes have used digitized video microscopy (DVM) in freshly isolated or cultured cells. Although useful, this technique is time consuming, requires the use of an image analysis system, and is limited by the fact that measurements can be made in only a relatively small number of cells. The aim of this study was to develop and initially apply a technique using flow cytometry to make dynamic measurements of lysosomal pH in a large number of living hepatocytes. Rats were injected intraperitoneally with fluorescein isothiocyanate-dextran (FITC-Dex), a pH-sensitive fluorescent probe that is sequestered into lysosomes. Hepatocytes were isolated 16 hours after injection by collagenase perfusion. Lysosomal pH was measured in 20,000 hepatocytes per animal using flow cytometry with excitation at 488 nm and emission at 530 nm (pH sensitive) and 585 nm (pH insensitive). A standard curve of pH versus the 530/585 nm ratio was generated with FITC-Dex-loaded hepatocytes by equilibrating intralysosomal pH with extracellular pH using ionophores and metabolic inhibitors. The acute effects of chloroquine and methylamine were determined by exposing isolated hepatocytes to these lysosomotropic agents. The effect of chronic administration of chloroquine and Triton WR-1339 (Rutger Chemical, Inc., Irvington, NJ) on lysosomal pH was also measured. Intralysosomal pH was 4.67 + 0.02, nearly identical to the value 4.70 + 0.05 previously measured by us using DVM. Both chloroquine and methylamine caused both rapid (< 1 minute), major (0.5 to 2.0 pH units), and dose-dependent increases in lysosomal pH as well as changes in lysosome morphology.(ABSTRACT TRUNCATED AT 250 WORDS)

Isolation and characterization of rat cholangiocyte vesicles enriched in apical or basolateral plasma membrane domains.

Cholangiocytes, the epithelial cells that line intrahepatic bile ducts, are composed of plasma membranes with discrete apical (lumenal) and basolateral domains. While these domains are thought to contain different transporters, exchangers, channels, and receptors, no methodology currently exists for the isolation of these different membrane compartments. Thus, our aim was to develop a technique to isolated plasma membranes from cholangiocytes enriched in apical or basolateral domains. We isolate a cholangiocyte-enriched population of cells from rats 3 weeks after bile duct ligation (BDL), a maneuver which stimulates selective cholangiocyte proliferation. Using isopycnic centrifugation on linear sucrose gradients, we prepared a mixed cholangiocyte plasma membrane (MCPM) fraction from which we further generated separate apical and basolateral cholangiocyte plasma membrane (ACPM and BCPM, respectively). We characterized these fractions by specific marker enzyme assays, transmission electron microscopy (TEM), lipid analysis, anisotropy measurements, one- and two-dimensional gel electrophoresis, and quantitative immunoblots of the cystic fibrosis transmembrane conductance regulator (CFTR). Marker enzyme assays and TEM revealed that the MCPM fraction was essentially devoid of other organelles but was enriched approximately 70-fold in phosphodiesterase I, a general plasma membrane marker; the ACPM and BCPM were appropriately enriched in the respective apical and basolateral markers. TEM of ACPM and BCPM revealed homogeneous preparations of vesiculated membranes without contamination by other organelles. Lipid analysis, one- and two-dimensional gel electrophoresis, CFTR immunoblots, and anisotropy measurements showed unequivocal differences in lipid and protein composition and in fluidity between the ACPM and BCPM domains.(ABSTRACT TRUNCATED AT 250 WORDS)

Quantitative importance of biliary excretion to the turnover of hepatic lysosomal enzymes.

The turnover rate of an individual protein is a function of the rates of synthesis and loss of that protein. For most intracellular proteins, loss occurs through digestion by lysosomal or cytosolic proteases. Although a significant proportion of hepatic lysosomal enzymes is released from the hepatocyte by excretion into bile, the contribution of biliary excretion to the turnover of hepatic lysosomal enzymes has never been measured. Thus, we used in vivo pulse-labeling to determine the half-lives of two hepatic hydrolases, beta-galactosidase (beta-gal) and beta-glucuronidase (beta-glu). Each enzyme was purified by immunoisolation from hepatic lysosomes that were isolated at various times after injection of rats with 3H-labeled leucine. The decay curves for the specific radioactivities of beta-gal and beta-glu were used to calculate the half-lives of the proteins, which were 3.8 and 5.1 days, respectively. To determine the percent of total hepatic contents of each enzyme that was lost per day by biliary excretion, we collected bile from bile fistula rats for 24 hours and then used radioimmunoassays to quantitate the amounts of beta-gal and beta-glu in bile and liver samples of the same rats. We found that approximately 4% of the total hepatic contents of both beta-gal and beta-glu was excreted into bile per day. Finally, we used these data to calculate that 31% and 41% of hepatic losses of beta-gal and beta-glu, respectively, were due to biliary excretion. These results suggest that extracellular release through biliary excretion is a major mechanism contributing to the turnover of lysosomal hydrolases.

Somatostatin inhibits secretin-induced ductal hypercholeresis and exocytosis by cholangiocytes.

Previous work from our laboratory has implicated hormone-induced plasma membrane movement (i.e., endo- and exocytosis) in water and electrolyte transport by the epithelial cells that line the ducts in the liver (i.e., cholangiocytes). To further explore the cellular mechanisms regulating ductal bile secretion, we infused somatostatin and/or secretin intravenously into rats 2 wk after either bile duct ligation (BDL), a procedure that induces selective proliferation of cholangiocytes, or sham surgery and measured bile flow and biliary constituents. We also determined the effect of somatostatin on basal and secretin-induced exocytosis by purified cholangiocytes isolated from rat liver after BDL. Finally, we studied the expression of the somatostatin receptor gene by both ribonuclease (RNase) protection and nuclear run-on assays using cDNA encoding for two subtypes of the somatostatin receptor gene (i.e., SSTR1 and SSTR2). In vivo, somatostatin infusion caused a dose-dependent bicarbonate-poor decrease (57% maximal decrease below baseline; P < 0.05) in bile flow in BDL but not in sham-operated rats; in contrast, secretin caused a dose-dependent bicarbonate-rich choleresis (228% maximal increase above baseline; P < 0.05) in BDL but not in sham-operated rats. Simultaneous or prior infusion of somatostatin inhibited the secretin-induced hypercholeresis in BDL rats. In vitro, somatostatin had no effect on basal exocytosis by cholangiocytes isolated from BDL rats; however, somatostatin inhitibed (88% maximal inhibition; P < 0.05) secretin-induced exocytosis by cholangiocytes in a dose-dependent fashion. In addition, somatostatin inhibited secretin-induced increases in levels of adenosine 3',5'-cyclic monophosphate (cAMP) in cholangiocytes isolated from BDL rats.(ABSTRACT TRUNCATED AT 250 WORDS)

Time-dependent effects of chloroquine on pH of hepatocyte lysosomes.

In vivo administration of chloroquine to rats caused an increase in the pH of hepatocyte lysosomes within 1 hr after administration with a return to baseline pH values by 3 hr; continued administration of chloroquine for up to 12 days was unaccompanied by any further changes in hepatocyte lysosomal pH. We interpret these data as evidence against a major role for an increase in the pH of hepatocyte lysosomes in CAC-induced phospholipidosis.

Circadian rhythms of biliary protein and lipid excretion in rats.

To gain insight into the mechanisms by which hepatocytes release lipids and proteins into bile, we studied extended, steady-state secretion of bile, lipids, and lysosomal and canalicular membrane proteins in freely moving, unanesthetized rats with chronic bile fistulas. We found circadian rhythms of biliary secretion for all measured constituents. In the basal state (nocturnal feeding), two distinct secretory patterns emerged: type 1, characterized by a peak at midnight and a nadir at noon; and type 2, characterized by a peak at 8 A.M. and a nadir at 8 P.M. We observed parallel, type 1 circadian rhythms of excretion for bile, biliary lipids (bile acid, cholesterol, phospholipid), and a canalicular membrane enzyme (alkaline phosphodiesterase I). In contrast, a type 2 circadian rhythm was observed for the outputs of two lysosomal enzymes. Hepatic lysosomal enzyme concentrations and the number of pericanalicular lysosomes decreased (P less than 0.05) by 15 and 35%, respectively, at the nadir of their biliary output relative to the time of their peak outputs. In response to daytime feeding, major shifts in the circadian rhythms of excretion of biliary constituents occurred such that secretion of bile, lipids, and the canalicular membrane protein adopted a type 2-like rhythm, whereas the biliary secretion of the lysosomal proteins exhibited a type 1-like pattern. These results indicate that bile flow and biliary excretion of individual lipids and proteins exhibit distinct circadian rhythms that are altered by feeding. Secretory events at the canaliculus that depend on the transmembrane flux of bile acids, such as water and lipid movement or the solubilization of membrane proteins, display a common rhythm.(ABSTRACT TRUNCATED AT 250 WORDS)

Purification and immunological quantification of rat liver lysosomal glycosidases.

Although lysosomal enzyme activities are known to vary in response to numerous physiological and pharmacological stimuli, the relationship between lysosomal enzyme activity and enzyme concentration has not been systematically studied. Therefore we developed radioimmunoassays for two lysosomal glycosidases in order to determine lysosomal enzyme concentration. beta-Galactosidase and beta-glucuronidase were purified from rat liver 2780-fold and 1280-fold respectively, by using differential centrifugation, affinity chromatography, ion-exchange chromatography and molecular-sieve chromatography. Polyclonal antibodies to these enzymes were raised in rabbits, and two radioimmunoassays were established. Antibody specificity was shown by: (i) selective immunoprecipitation of enzyme activity; (ii) identical bands of purified enzyme on SDS/polyacrylamide-gel electrophoresis and immunoelectrophoresis; (iii) single immunoreactive peaks in molecular-sieve chromatography experiments. Sensitivities of the assays were such that 15 ng of beta-galactosidase and 45 ng of beta-glucuronidase decreased the ratio of bound to free radiolabel by 50%; minimal detectable amounts of immunoreactive enzymes were 2 ng and 10 ng respectively. The assays were initially used to assess the effects of physiological perturbations (i.e. fasting and age) on enzyme concentrations in rat liver; these experiments showed that changes in enzyme concentrations do not always correlate with changes in enzyme activities. This represents the first report of radioimmunoassays for lysosomal glycosidases. The results suggest that these radioimmunoassays provide useful technology for the study of regulatory control mechanisms of the concentrations of lysosomal glycosidases in mammalian tissues.

Development and validation of a method for measuring the glycine and taurine conjugates of bile acids in bile by high-performance liquid chromatography.

We developed and validated a simple method for measuring the individual glycine and taurine conjugates of bile acids in bile by high-performance liquid chromatography with a C18 reversed-phase column using an isocratic solvent system of acidified methanol--potassium phosphate. Without preliminary derivatization or purification, complete separation of the ten major conjugated bile acids present in bile could be achieved in 65 min. Total bile acid concentrations were identical when measured enzymatically and by summing the individual bile acids determined by high-performance liquid chromatography. Bile acid composition determined by gas-liquid chromatography correlated with results by high-performance liquid chromatography. Finally, measurements of individual glycine and taurine conjugates in human bile and in mixtures of bile acid standards by high-performance liquid chromatography and thin-layer chromatography gave similar results. This high-performance liquid chromatographic system permits simultaneous quantification of total and individual bile acids and their glycine and taurine conjugates in bile.

Role of vitamin D glucosiduronate in calcium homeostasis.

Evidence has been presented suggesting the presence of vitamin D(3) 3beta-glucosiduronate and 1,25-dihydroxyvitamin D(3) glucosiduronate in rat bile. To evaluate the role of vitamin D glucosiduronates in calcium and phosphorus homeostasis, we synthesized vitamin D(3) 3beta-glucosiduronate and tested its biological activity in calcium- and vitamin D-deficient rats. After the intravenous administration of vitamin D(3) 3beta-glucosiduronate to rats maintained on a low calcium diet, there was an increase in duodenal calcium transport and an increase in serum calcium. Vitamin D(3) 3beta-glucosiduronate, however, was less active than equimolar amounts of vitamin D(3). At doses of less than 0.65-1 nmol per rat, the conjugate exhibited no activity. When vitamin D(3) 3beta-glucosiduronate was administered to vitamin D-deficient rats, 25-hydroxyvitamin D was detected in the serum; the increase in serum 25-hydroxyvitamin D levels was less than that observed after the administration of an equimolar amount of vitamin D(3). Vitamin D(3) 3beta-glucosiduronate showed no detectable activity in the induction of calcium binding protein in chick embryonic duodena, a system in which no endogenous steroid beta-glucuronidase activity is detectable. These data demonstrate that vitamin D(3) 3beta-glucosiduronate is biologically active in vivo and that the observed activity is due to hydrolysis of the conjugate to vitamin D(3). As vitamin D(3) 3beta-glucosiduronate is excreted in the bile of rats, it is possible that this conjugate is reutilized in vivo after hydrolysis to free vitamin D(3). These results suggest the existence of a mechanism for reutilization of the biliary products of vitamin D(3).