Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Carrier proteins located in the basolateral and apical plasma membranes of hepatocytes are responsible for bile acid (BA) uptake and secretion into bile (Meier and Stieger, 2002). During intrauterine life, when the liver is not yet mature, efficient transfer of fetal BAs across the placenta, together with normal maternal hepatobiliary function, maintains fetal BA levels within the physiological range. Previous studies have shown that carriers located in the trophoblast plasma membranes play a role in the vectorial fetal-to-maternal translocation of cholephilic compounds, such as BAs (Marin et al., 1990, 1995) and bilirubin (Serrano et al., 2002). Maternal hypercholanemia may challenge transplacental elimination of fetal BAs through the creation of inversely directed gradients as compared with the physiological situation (Monte et al., 1995), and by impairing the placental ability to carry out vectorial BA transfer (Macias et al., 2000). Moreover, obstructive cholestasis during pregnancy (OCP) in rats results in an accumulation of BAs in fetuses, a situation that is also transiently observed later in juvenile animals born from cholestatic mothers (Monte et al., 1996). Exposure of the fetal liver to high BA levels is probably involved in the delayed maturation of hepatobiliary function observed in these young rats (Monte et al., 1996; El-Mir et al., 1997).

In humans, intrahepatic cholestasis of pregnancy (ICP) is accompanied, among other maternal and fetal alterations (Reid et al., 1976; Mullally and Hansen, 2002), by a reduced ability to transport BAs across the trophoblast plasma membrane (Serrano et al., 1998). Treatment with ursodeoxycholic acid (UDCA), which has beneficial effects in several cholestatic liver diseases (Poupon and Poupon, 1995; Kowdley, 2000; Lazaridis et al., 2001), is able to mitigate pruritus and enzyme elevations in the serum of women with ICP (Palma et al., 1992; Floreani et al., 1994; Brites et al., 1998). Because UDCA was also found to have a positive effect on ICP placentas, at least as far as BA transport by trophoblast plasma membrane vesicles was concerned (Serrano et al., 1998), in the present study we intended to further characterize structural and functional aspects of the beneficial effects of UDCA on maternal hypercholanemia-induced impairment of the rat placenta-maternal liver tandem excretory pathway for fetal BAs.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Experimental Groups. Nonfasting pregnant Wistar CF rats (Animal House, University of Salamanca, Spain) were used. The experiments were approved by the Local Ethical Committee. On day 14 of pregnancy, all animals were anesthetized with sodium pentobarbital (50 mg/kg body weight, i.p.; Abbot, Madrid, Spain) to either sham-operate them (control) or produce a complete obstructive cholestasis (OCP) by implantation of a blunted catheter in the common bile duct (Macias et al., 2000). During the following week some of these animals received daily intragastric administration of 60 µg of UDCA in 60 µl of 75 mM NaCl, 50 mM Na₂CO₃, pH 8.3, per 100 g b.wt. (OCP + UDCA group, n = 60). The rest of the cholestatic rats (OCP group, n = 70) received only the vehicle.

"In Vivo" Experiments. On day 21 the animals were anesthetized again. Cannulation of the common bile duct was performed in controls. In the OCP and OCP + UDCA groups, the blunted tips of the catheters implanted were cut, thereby allowing free bile flow. This was measured gravimetrically. BA concentrations were determined enzymatically (Talalay, 1960). The doses of [¹⁴C]glycocholate ([¹⁴C]GC; specific radioactivity 46.7 mCi/mmol) used in different types of experiments were selected based on previous studies and were administered once BA output had reached an approximate steady state after bile drainage (Macias et al., 2000). To evaluate BA uptake and secretion by the maternal liver, 5 nmol [¹⁴C]GC in 150 mM NaCl, 5 mM glucose was administered through the maternal jugular vein as a 5-min bolus (50 µl/min). Transfer of [¹⁴C]GC across the placenta-maternal liver tandem was measured using "in situ" single-pass perfused placenta. [¹⁴C]GC (250 nmol) was administered with the perfusate [137 mM NaCl, 2.7 mM KCl, 1.05 mM MgCl₂, 1.80 mM CaCl₂, 12 mM NaHCO₃, 0.4 mM NaH₂PO₄, 5 mM glucose, 0.05% (w/v) heparin, 10 mM Hepes, pH 7.40 at 37°C] through the umbilical artery over 5 min (500 µl/min), and radioactivity in maternal serum and bile was determined (Briz et al., 1998; Macias et al., 2000). Placental [¹⁴C]GC content was measured in similar experiments, except that 2 or 10 min after finishing [¹⁴C]GC administration, the perfused placenta and the remaining nonperfused placentas and their fetuses were collected to measure radioactivity. The amount of tissue able to carry out placental exchange was determined by measuring the magnitude of the diffusional pathway for antipyrine. Fetal/maternal serum antipyrine concentration ratios were assayed as previously described following administration through the maternal jugular vein (100-mg bolus plus 0.5 mg/min infusion) (Macias et al., 2000).

Experiments on Plasma Membrane Vesicles. Maternal-facing trophoblast plasma membrane (mTPM) vesicles were isolated from rat placenta as previously described (Bravo et al., 1995; Macias et al., 2000). Enrichment of mTPM associated marker alkaline phosphatase activity over the homogenate did not differ significantly among control (12.0 ± 3.1-fold, n = 5), OCP (16.2 ± 2.2-fold, n = 5), and OCP + UDCA (15.8 ± 1.3-fold, n = 5) groups. Low contamination with fetal-facing trophoblastic basal plasma membrane fragments was indicated by similar de-enrichments of dihydroalprenolol binding in control (0.5 ± 0.2-fold, n = 5), OCP (0.4 ± 0.1-fold, n = 5), and OCP + UDCA (0.5 ± 0.1-fold, n = 5) groups. Membrane vesicles were resuspended in buffer A (250 mM sucrose, 0.2 mM CaCl₂, 10 mM MgCl₂, 100 mM KNO₃, 10 mM Hepes/Tris, pH 7.40) and stored in liquid nitrogen. Before carrying out the experiments, membranes were first thawed, then diluted with buffer A to approximately 5 mg of protein/ml and vesiculated by six passages through a 25-gauge needle. Protein was determined (Markwell et al., 1978) using bovine serum albumin as standard. Using a rapid filtration technique (Marin et al., 1990, Macias et al., 2000), ATP-dependent [¹⁴C]GC uptake by mTPM vesicles was measured during the initial linear uptake phase (30 s) at 37°C.

Determination of Gene Expression. RNA from snap-frozen placentas and livers was isolated using RNeasy spin columns from QIAGEN (Izasa, Barcelona, Spain) and measured with the RiboGreen RNA-Quantitation kit (Molecular Probes, Leiden, The Netherlands). Reverse transcription was carried out with 1 µg of total RNA, using random nanomers and an Enhanced Avian RT-PCR kit (Sigma-Genosys, Cambridge, UK).

Histological Examination and Data Analysis. Light and transmission electron microscopic examinations of placental tissue were carried out as previously reported (Macias et al., 2000). Morphometric analysis was performed using NIH Image V1.62 software (National Institutes of Health, Bethesda, MD).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Morphological, Biochemical, Histological, and Molecular Changes. Morphological and biochemical changes are in agreement with previous studies using OCP rats (Monte et al., 1996; El-Mir et al., 1997; Macias et al., 2000). Some of them were prevented by UDCA (Table 2). Atrophy of trophoblastic tissue in OCP placentas with dilated maternal vascular lacunae, to approximately 189% of control, was seen. These alterations were in part corrected in OCP + UDCA placentas in which morphometric analysis revealed that fetal tissue was restored to 92% of control value (Fig. 1). At subcellular level (Fig. 2) syncytiotrophoblastic cells from the OCP group displayed hydroponic degeneration. Cells appeared with a very dilated pericanalicular space, a highly vacuolar cytoplasm, and a lack of microvilli on mTPM. These alterations were partly prevented by UDCA. The proportion of trophoblast in the placentas, as determined by placental lactogen type II (rPLII), a specific trophoblastic marker during the last stages of rat gestation (Shah et al., 1998; Faria et al., 1990), was reduced by OCP but restored by UDCA (Fig. 3B).

View larger version (71K): [in this window] [in a new window]   Fig. 1.   Light microscopy images from hematoxylin-eosin staining of tissue slices obtained from placentas collected from 21-day pregnant rats. Samples were from animals representative of each of the experimental groups, which were controls, and rats undergoing surgical common bile duct obstruction on day 14 that, for the remaining 7 days, received the vehicle (OCP) or 60 µg of UDCA/100 g b.wt./day (OCP + UDCA). C, connective tissue; CT, cytotrophoblastic cells; F, fetal vascular spaces containing fetal red blood cells; M, maternal vascular spaces; ST, syncytiotrophoblastic cells.

View larger version (86K): [in this window] [in a new window]   Fig. 2.   Transmission electron microscopy images from tissue slices obtained from placentas collected from 21-day pregnant rats. Samples were from animals representative of each of the experimental groups, which were controls, and rats undergoing surgical common bile duct obstruction on day 14 that, for the remaining 7 days, received the vehicle (OCP) or 60 µg of UDCA/100 g b.wt./day (OCP + UDCA). Original magnification, 4000×.

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Expression level of the trophoblast-specific gene, rat placental lactogen II (rPLII), the organic anion transporters Oatp1, Oatp2, and Oatp4; and the multidrug resistance-associated proteins Mrp1, Mrp2, and Mrp3, in placentas at day 21 of pregnancy. A, representative Western blot analysis for Oatp2 and Mrp2 in placental cell membranes from pregnant rats of each experimental group, i.e., controls, OCP (surgical common bile duct obstruction on day 14 of pregnancy), and OCP + UDCA (biliary obstruction on day 14, followed by treatment with 60 µg of UDCA/day/100 g b.wt. for the remaining 7 days). As a positive control, male rat liver bLPM and cLPM were included. B, relative abundance of mRNA determined using the real-time quantitative reverse transcription-PCR and expressed as percentages of the control group. C, relative abundance of mRNA corrected by the expression level of rPLII, which was considered as representative of the amount of trophoblast tissue. Values (means ± S.D.) were from total RNA isolated from placentas belonging to control (n = 5), OCP (n = 5), and OCP + UDCA (n = 5) groups. , p < 0.05, on comparing OCP and OCP + UDCA with control by the Bonferroni method of multiple range testing.

In Vitro Functional Studies. Kinetic parameters for ATP-dependent [¹⁴C]GC uptake by mTPM versus substrate concentrations (Fig. 4A) indicated that, in agreement with previous studies (Macias et al., 2000), the transport efficiency, as defined by the V_max-to-K_M ratio, was reduced in mTPM from OCP placentas. This was mainly caused by an increase of the K_M. UDCA treatment did not significantly modify V_max but restored the K_M value to normal and, hence, reversed the efficiency of this transport to values close to those of control placentas (Fig. 4B).

View larger version (30K): [in this window] [in a new window]   Fig. 4.   A, ATP-dependent [14C]GC uptake by apical plasma membrane vesicles obtained from placentas collected from 21-day pregnant rats. Experimental groups were controls (n = 8) and rats undergoing surgical common bile duct obstruction on day 14 that, for the remaining 7 days, received the vehicle (OCP group, n = 8) or 60 µg of UDCA/100 g b.wt./day (OCP + UDCA group, n = 7). Values were obtained from experiments carried out in triplicate for each substrate concentration. ATP-dependent uptake was calculated by subtracting uptake values in the absence of ATP from uptake values in the presence of ATP. B, kinetic parameters calculated from fitting the results to a Michaelis-Menten equation for a single-carrier system: Vo = (Vmax · S)/(KM + S), where Vo is the initial velocity of [14C]GC uptake, S is substrate concentration in the incubation medium, KM is the apparent affinity constant or Michaelis-Menten constant, and Vmax is the maximal transport velocity. Values are means ± S.D. , p < 0.05 on comparing OCP and OCP + UDCA with controls by the Bonferroni method of multiple range testing.

In Vivo Functional Studies. Once bile flow was allowed on day 21, this and BA output were initially higher in OCP and OCP + UDCA than in controls (Fig. 5). This was followed by a progressive decrease in BA output below control levels, probably due to the fact that, in the absence of a normal enterohepatic circulation, most of the BA pool is located in the liver. By contrast, after an initial decrease, bile flow was maintained in both OCP and OCP + UDCA at a level similar to that found in controls. Cholangiocyte proliferation and changes in its secretory function associated with rat biliary obstruction for 1 week (Alpini et al., 2002) might well account for an increased contribution of ductular components to bile formation.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Time course of bile flow (A) and bile acid output (B) in anesthetized pregnant rats at term (day 21) for a 180-min bile collection through a catheter implanted into the common bile duct in control rats (closed squares; n = 21) and in rats that underwent surgical common bile duct obstruction on day 14 and for the remaining 7 days received the vehicle (OCP group, open circles; n = 17) or 60 µg of UDCA/100 g b.wt./day during the last week of gestation (OCP + UDCA group, closed circles; n = 11). Values are means ± S.E.M. , p < 0.05 on comparing controls with the OCP group; , p < 0.05 on comparing controls with the OCP + UDCA group, by the Bonferroni method of multiple range testing.

View larger version (28K): [in this window] [in a new window]   Fig. 6.   Fetal-to-maternal ratios of serum antipyrine concentrations during its administration through the jugular vein (100-mg bolus followed by 0.5 mg/min continuous infusion) to 21-day pregnant rats. Experimental groups were controls (closed squares, n = 8) and rats undergoing surgical common bile duct obstruction on day 14 that, for the remaining 7 days, received the vehicle (OCP group, open circles; n = 5) or 60 µg of UDCA/100 g b.wt./day (OCP + UDCA group, closed circles; n = 4). After bile drainage for 180 min, antipyrine administration started immediately after ligation of the common bile duct in all groups. The inset depicts a schematic representation of the experiment. Values are means ± S.E.M. , p < 0.05 on comparing controls with the OCP group; , p < 0.05 on comparing controls with the OCP + UDCA group, by the Bonferroni method of multiple range testing.

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Bile output of [14C]GC following administration (250 nmol over 5 min) through the umbilical artery of one of the in situ perfused rat placentas in 21-day pregnant rats. Experimental groups were controls (closed squares; n = 8) and rats undergoing surgical common bile duct obstruction on day 14 that, for the remaining 7 days, received the vehicle (OCP group, open circles; n = 8) or 60 µg of UDCA/100 g b.wt./day (OCP + UDCA group, closed circles; n = 6). The inset depicts a schematic representation of the experiment. Values are means ± S.E.M. , p < 0.05 on comparing controls with the OCP group; , p < 0.05 on comparing controls with the OCP + UDCA group, by the Bonferroni method of multiple range testing.

View this table: [in this window] [in a new window]   TABLE 3 Glycocholate placental uptake and bile secretion by maternal liver The amount of [14C]GC retained by the placenta after administration (250 nmol over 5 min) through the umbilical artery of in situ perfused rat placentas in 21-day pregnant rats was measured after a 2- or 10-min washing period with [14C]GC-free perfusion medium. In two separate sets of animals, values of [14C]GC bile output into maternal bile over 3 h after administration of [14C]GC through the umbilical artery (250 nmol over 5 min) or through the maternal jugular artery (5 nmol over 5 min) were used to calculate AUC (area under the curve or accumulated output with no extrapolation to infinity) and MRT (mean residence time) by numerical integration using the trapezoidal rule. The half-life time (t1/2) of [14C]GC in the maternal compartment was calculated as Ln2/Ke, where Ke (elimination constant) was 1/MRT (Yamaoka et al., 1978). Groups were control, and rats undergoing surgical common bile duct obstruction on day 14 that, for the remaining 7 days, received the vehicle (OCP group) or 60 µg of UDCA/100 g b.wt./day (OCP + UDCA group). Values are means ± S.E.M. (n  5).

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Bile output of [14C]GC following administration (5 nmol over 5 min) through the jugular vein of 21-day pregnant rats. Experimental groups were control (closed squares; n = 8) and rats undergoing surgical common bile duct obstruction on day 14 that, for the remaining 7 days, received the vehicle (OCP group, open circles; n = 8) or 60 µg of UDCA/100 g b.wt./day (OCP + UDCA group, closed circles; n = 6). The inset depicts a schematic representation of the experiment. Values are means ± S.E.M. , p < 0.05 on comparing controls with the OCP group; , p < 0.05 on comparing controls with the OCP + UDCA group, by the Bonferroni method of multiple range testing.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In agreement with previous findings by our group and others (Klaassen, 1974; Macias et al., 2000), diminished and slower secretion of exogenously administered bile acids to rats after releasing obstructive cholestasis was observed. This is consistent with the known reduction occurring after BDL in the expression of rat sinusoidal BA carriers, whereas canalicular exporter pumps, Mrp2, and bile salt export pump are decreased or partially maintained, respectively (Trauner et al., 1999).

Although improvement of the drug-induced cholestasis in the rat by UDCA has been reported (Jacquemin et al., 1993), its beneficial effect after BDL is limited (Poo et al., 1992; Hinz et al., 1997; Purucker et al., 2001). In the present study, UDCA administration to OCP rats also resulted in only a partial beneficial effect on liver ability to secrete [¹⁴C]GC. Nevertheless, this probably contributed to enhancing the elimination of [¹⁴C]GC when this radiolabeled BA was administered through the umbilical artery.

One of the most striking findings of the present study was the effect of UDCA on trophoblasts. Morphological, molecular, and functional measurements indicated a dramatic reduction in the amount of functional trophoblasts in OCP placentas. As has been reported by others in several tissues (Abdel-Aziz et al., 1991), inflammatory processes associated with BA accumulation probably account for the fibrogenesis, accumulation of extracellular matrix, and necrosis also present in OCP placentas (data not shown), all of which may contribute to reducing transplacental exchange. UDCA prevented the OCP-induced loss of trophoblast; structural alterations were limited and the functional test based on antipyrine diffusion was partly restored.

Another important contribution is the confirmation of preliminary observations (St-Pierre et al., 1999) regarding the presence of Oatp1, Oatp2, and Oatp4 in rat placenta. Assuming that these carriers are located only in the trophoblast, it could be calculated that the relative abundance of their mRNA as compared with rPLII was enhanced in OCP and OCP + UDCA. However, because the amount of trophoblast was reduced by OCP, the total levels of whole placentas in this group were not markedly higher than in control. By contrast, because UDCA restored the amount of trophoblast, total placental mRNA levels for these carriers were enhanced in a moderate-to-marked range in the OCP + UDCA group. This could in part account for the partial recovery of the efficiency of placental [¹⁴C]GC uptake in this group. The fact that [¹⁴C]GC uptake was markedly reduced in OCP suggests that "adaptive" changes in the level of mRNA for these carriers in the remaining trophoblast were insufficient to compensate for the negative effects due to OCP-induced structural and functional alterations.

Whereas placental [¹⁴C]GC uptake occurred over 5 min, its output toward the mother and then into bile took much longer, especially in the OCP animals, supporting the concept that an export mechanism located in the apical membrane of the trophoblast is probably the limiting step in the overall process of BA elimination by the placenta-maternal liver tandem (Briz et al., 1998).

The marked OCP-induced impairment in [¹⁴C]GC transfer across the apical membrane of the trophoblast was consistent with the lower efficiency of ATP-dependent transport found in studies with plasma membrane vesicles. This transport is probably mediated by one or several members of the superfamily of ATP-binding cassette proteins. Some of the human MRP orthologs have been identified in human placenta (St-Pierre et al., 2000). Similar to what happens in the liver (except for Mrp2) and kidney (Lee et al., 2001; Soroka et al., 2001; Pei et al., 2002), the adaptive up-regulation of several Mrp transporters in the rat trophoblast was stimulated by OCP.

Increases in the levels of mRNA for uptake and export carriers were probably accompanied by enhancements in the amount of these proteins in trophoblastic cells, as indicated by Western blot analysis of two representative carriers, Oatp2 and Mrp2. However, the lack of normal microvilli in trophoblast plasma membranes in OCP placentas, together with the fact that, even when purified apical membranes were used, the efficiency of ATP-dependent [¹⁴C]GC transport was reduced in this group, suggests that carrier overexpression was not enough to counterbalance other negative changes in the plasma membrane composition/structure or in the maturation/sorting process affecting the function of the carriers, and that such negative effects would be responsible for the reduction in carrier ability to export BAs.

Moreover, the possibility that part of the discrepancy between enhanced expression levels and reduced transport ability could be attributed to the fact that some of these transporters may be present in intracellular compartments, which are not actively involved in BA transport across the trophoblast, should be considered. Treatment with UDCA did not reduce the relative expression of Mrps in the trophoblast, which together with the UDCA-induced recovery of the amount and structure of this tissue contributed to restoring [¹⁴C]GC transfer into the maternal compartment.

The overall UDCA-induced beneficial effects had important consequences on the conceptus development. Thus, the intrauterine growth retardation observed in OCP fetuses was prevented by UDCA. It should be kept in mind that OCP resulted in a diminished presence of BAs in the maternal gut that was corrected by gavage supplementation with UDCA. This may determine better intestinal lipid digestion and liposoluble vitamin absorption (Mullally and Hansen, 2002), which together with displacement of more toxic BA species in the maternal-fetal BA pool and the beneficial effect on the placenta and maternal liver functions, all probably play an important role in normal fetal growth and the number of fetuses per pregnancy found in the OCP + UDCA group. This is consistent with the clinical results of UDCA therapy in ICP, where the compound reduces the number of stillbirths and perinatal mortality (Davies et al., 1995; Palma et al., 1997).