Potential conflict of interest: Nothing to report.
Bile acid-CoA:amino acid N-acyltransferase (BAAT) conjugates bile salts to glycine or taurine, which is the final step in bile salt biosynthesis. In addition, BAAT is required for reconjugation of bile salts in the enterohepatic circulation. Recently, we showed that BAAT is a peroxisomal protein, implying shuttling of bile salts through peroxisomes for reconjugation. However, the subcellular location of BAAT remains a topic of debate. The aim of this study was to obtain direct proof for reconjugation of bile salts in peroxisomes. Primary rat hepatocytes were incubated with deuterium-labeled cholic acid (D4CA). Over time, media and cells were collected and the levels of D4CA, D4-tauro-CA (D4TCA), and D4-glyco-CA (D4GCA) were quantified by liquid chromatography-tandem mass spectrometry (LC/MS/MS). Subcellular accumulation of D4-labeled bile salts was analyzed by digitonin permeabilization assays and subcellular fractionation experiments. Within 24 hours, cultured rat hepatocytes efficiently (>90%) converted and secreted 100 μM D4CA to D4TCA and D4GCA. The relative amounts of D4TCA and D4GCA produced were dependent on the presence of glycine or taurine in the medium. Treatment of D4CA-exposed hepatocytes with 30-150 μg/mL digitonin led to the complete release of D4CA, D4GCA, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (cytosolic marker). Full release of D4TCA, catalase, and BAAT was only observed at 500 μg/mL digitonin, indicating the presence of D4TCA in membrane-enclosed organelles. D4TCA was detected in fractions of purified peroxisomes, which did not contain D4CA and D4GCA. Conclusion: We established a novel assay to study conjugation and intra- and transcellular transport of bile salts. Using this assay, we show that cholic acid shuttles through peroxisomes for taurine-conjugation. (HEPATOLOGY 2010)
Bile salts are synthesized in the liver and are the driving force of bile flow. Bile is crucial for intestinal absorption of fats and fat-soluble vitamins, as well as the elimination of excess cholesterol and waste products from the body. In the terminal ileum, 90%-95% of the bile salts is reabsorbed and transported back to the liver. Import and export of bile salts in hepatocytes and enterocytes is mediated by well-characterized transmembrane substrate transporters.1
Fecal loss of bile salts is compensated by de novo bile salt synthesis in the liver. Hepatic bile salt synthesis involves at least 13 different enzymes that are located in different subcellular compartments, including the endoplasmic reticulum (ER), mitochondria, peroxisomes, and the cytosol. Bile salts are made from cholesterol and this requires three key modifications of the cholesterol backbone: (1) hydroxylation of the steroid nucleus, (2) shortening of the side chain, and (3) conjugation/amidation of glycine or taurine to the side chain. The last two modifications take place in peroxisomes.2 Bile acid-coenzyme A:amino acid N-acyltransferase (BAAT) catalyzes the third and final modification of bile salts before they enter the enterohepatic cycle.3
The activity of BAAT remains crucial also during enterohepatic cycling of bile salts. A significant portion of bile salts is deconjugated by intestinal bacteria. Approximately one-third of the bile salt pool may undergo deconjugation on a daily basis in healthy humans.4 Still, unconjugated bile salts are also reabsorbed in the enterohepatic circulation and transported back to the liver. However, the fraction unconjugated bile salts in the total bile salt pool is very low, indicating an efficient reconjugation process. The efficiency of bile salt reconjugation is further stressed by the fact that over 97% of the therapeutic bile salt ursodeoxycholate (UDCA) is conjugated to taurine or glycine after a single pass through isolated perfused rat livers.5
Unconjugated bile salts are first activated with coenzyme A (CoA) by the fatty acid transport protein 5 (FATP5; SLC27A5), which is located at the basolateral membrane of hepatocytes.6, 7 Next, the CoA-activated C24-bile salts are the substrate for BAAT. It has been postulated that a cytosolic pool of BAAT is responsible for reconjugating the recycling pool of unconjugated bile salts.8-10 However, we recently applied digitonin permeabilization assays and immunofluorescence microscopy on endogenous and green fluorescent protein (GFP)-tagged human BAAT/rat BAAT and found that it is predominantly, if not solely, present in peroxisomes of hepatocytes.11 An exclusive peroxisomal location of BAAT implies that CoA-activated unconjugated bile acids need to be transported into peroxisomes, followed by glycine/taurine conjugation and export out of these organelles, a yet unexplored bile salt transport process.
In this study we sought further proof for the transit of unconjugated bile salts through peroxisomes. For that purpose we established a novel assay that allows the detection of (un)conjugated bile salts in peroxisomes. Rat hepatocytes were exposed to deuterated cholic acid (D4CA). Over time, the concentrations of taurine- and glycine-conjugated D4CA in cells and medium were determined. At peak intracellular accumulation of D4TCA and D4GCA, digitonin permeabilization assays and cell fractionation experiments were performed. Our data show for the first time that unconjugated bile salts shuttle through peroxisomes to become conjugated to taurine.
Specified pathogen-free male Wistar rats (220-250 g; Charles River Laboratories, Wilmington, MA) were housed under standard laboratory conditions with free access to standard laboratory chow and water. Experiments were performed following the guidelines of the local Committee for Care and Use of Laboratory Animals.
Primary Cells and Culture Conditions.
Rat hepatocytes were isolated and cultured in William's medium E in a humidified incubator at 37°C and 5% CO2, as described.12 In selected experiments (see below), hepatocytes were cultured in minimal essential medium (MEM, Invitrogen, Breda, The Netherlands). Hepatocyte viability and purity were over 90%.
Uptake and Conversion of Deuterated Cholic Acid by Primary Rat Hepatocytes.
Primary rat hepatocytes were plated at a density of 1.0 × 105 cells/cm2. After a 24-hour attachment period, cells were incubated with 25, 100, or 300 μM [2,2,4,4-D]Cholic acid (D4CA; isotopic purity 98%, ISOTEC, Miamisburg, OH) for 0 to 24 hours. For taurine or glycine conjugation preference assays, hepatocytes were cultured in MEM, supplemented with 666 μM glycine (Sigma-Aldrich, St. Louis, MO) and/or 666 μM taurine (Sigma-Aldrich) in the presence of 100 μM D4CA. At indicated timepoints, media and cells were collected followed by subcellular fractionation or immediate storage at −20° C.
Subcellular Fractionation and Isolation of Peroxisomes.
The subcellular fractionation and isolation of peroxisomes from rat liver was performed essentially as described13 using PEG1500-containing homogenization buffer (isolation medium-3). Peroxisomes were purified from the 500g supernatant (postnuclear supernatant [PNS]) using Nycodenz density gradient centrifugation according to the method described by Verheyden et al.14 Twelve mL PNS was loaded on top of a discontinuous Nycodenz gradient (2 mL 56%, 3 mL 45%, 15 mL 30%, and 5 mL 18%) and spun in a vertical rotor (Sorvall, SV288, Thermo Fisher Scientific, Waltham, MA) at 20,000 rpm for 2 hours at 4°C in a slow acceleration/deceleration mode. Equal volumes of all supernatants, pellets, and gradient fractions were analyzed by western blotting or were further purified for mass spectrometry.
Digitonin assays were performed essentially as described,11 with the basic difference that digitonin treatments were performed on rat hepatocytes attached to collagen-coated culture discs instead of treated in suspension. Equal volumes of supernatant and pellet fractions were analyzed by western blotting or further processed for mass spectrometry.
Quantification of Cell Death.
Apoptotic cell death was visualized by acridine orange nuclear staining15 and quantified by determining caspase-3 activity.16 The arbitrary fluorescence unit (AFU) was corrected for the amount of total protein in the cell lysate. Necrotic cell death was quantified by determining lactate dehydrogenase (LDH) leakage17 and Sytox green (Invitrogen) according to the supplier's protocol.
Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) and Western Blotting.
Protein samples were separated by SDS-PAGE and analyzed by western blotting according to established procedures.11 Protein concentrations were determined using the Bio-Rad Protein Assay system (Bio-Rad, Hercules, CA) using bovine serum albumin as standard. Primary antibody dilutions used in this study are shown in Supporting Table S1. Proteins signals were detected and quantified in a ChemiDoc XRS system (Bio-Rad). Protein band intensities were quantified using Quantity One software (Bio-Rad).
Purification of Bile Salts for Mass Spectrometry.
Media samples (0.1-1.5 mL) from D4CA-exposed hepatocytes were collected at indicated timepoints and diluted 1,000-fold in an internal standard (IS) solution (containing 100 ng/mL each of CA, TCA, and GCA in H2O). Hepatocytes were washed twice in 1× ice-cold Hank's buffered salt solution (HBSS; Invitrogen), scraped in 250 μL 75% (v/v) methanol and 1,000-fold diluted in IS solution. Digitonin assay samples (1 mL, both representing supernatant and pellet fraction) were diluted 200-fold in IS solution. Nycodenz gradient fractions were 1,000-fold diluted in IS solution. Total bile salts were purified using reversed phase C18 columns (Sep-Pak C18 cartridge; Waters, Milford, MA) as described.18
LC/MS/MS Analysis of Bile Salts.
A detailed description of the LC/MS/MS analysis of bile salts is given in the Supporting Material and Methods. In short, LC/MS/MS analysis was performed using a triple quadrupole mass spectrometer API 3000 (Applied Biosystems, Foster City, CA) using ESI ionization in the negative mode. CA and D4CA were detected using single ion monitoring at m/z 407 and m/z 411, respectively. Detection of GCA, D4GCA, TCA, and D4TCA was performed using the selected reaction-monitoring mode. Two LC-200 HPLC pumps (Perkin-Elmer, Waltham, MA) coupled to a series 200 autosampler (Perkin-Elmer) were used. Chromatography was performed with a Luna C18(2) (Phenomenex, Torrance, CA) analytical column (50 × 2.0 mm; particle size 3 μm).
Calculation and Interpretation of LC/MS/MS Data.
The peak area for the D4-labeled bile salt was determined and related to the corresponding unlabeled bile salt added as IS. This ratio was corrected for the natural isotope abundance of the IS. For the calculation of intracellular bile salt concentrations, the cellular volume of one million hepatocytes was set at 20 μL, being the higher limit of estimations reported by others.19-24
All numerical results are reported as the mean of at least three independent experiments ± standard error of the mean.
In Vitro Cultured Rat Hepatocytes Efficiently Convert Exogenously Added Cholate to Taurocholate and Glycocholate.
We first determined the rate and specificity by which primary rat hepatocytes convert exogenously added CA to TCA and/or GCA. The 24-hour attached hepatocytes were exposed to various concentrations of deuterated CA (25, 100, and 300 μM D4CA; Fig. 1, left, middle and right panels, respectively). Media (Fig. 1A) and cells (Fig. 1B) were collected after 3 and 24 hours of incubation. D4TCA and D4GCA and the input-bile salt (D4CA) were readily detectable in the medium after 3 hours of incubation (Fig. 1A). D4CA concentrations were below input levels (7, 60, and 225 μM for the input of 25, 100, and 300 μM, respectively). The presence of D4TCA (6, 10, and 10 μM, respectively) and D4GCA (6, 15, and 15 μM, respectively) in the medium after 3 hours exposure time indicates that D4CA is taken up, CoA-activated, taurine/glycine conjugated by and exported from the hepatocytes. After 24 hours, D4CA was absent in medium of cells exposed to 25 μM (Fig. 1A,B, left panels). Instead, D4TCA (12 μM) and D4GCA (10 μM) were detected in these samples. Only small amounts (5 μM) of D4CA were detected in the medium of hepatocytes exposed to 100 μM D4CA for 24 hours, which was effectively converted to D4TCA (23 μM) and D4GCA (68 μM) (Fig. 1A, middle panel). After 24-hour exposure of hepatocytes to 300 μM D4CA, 70 μM D4CA, 40 μM D4TCA, and 160 μM D4GCA were detected in the medium (Fig. 1A, right panel).
Intracellular Accumulation of Taurocholate and Glycocholate in Cholate-Exposed Rat Hepatocytes.
Simultaneously with the media samples, hepatocytes were harvested in order to determine intracellular bile salt accumulation (Fig. 1B). After 3 hours exposure to D4CA, a large intracellular accumulation of conjugated D4-labeled bile salts was detected. D4TCA concentrations were ≈200 μM for all three conditions, whereas D4GCA levels (120, 400, and 600 μM, respectively) were dependent on the D4CA input concentration (25, 100, 300 μM, respectively. Fig 1B, left, middle, and right panels, respectively). D4CA was undetectable in cells exposed to 25 μM D4CA, whereas the cellular concentrations of this bile salt (80 and 310 μM, respectively) were close to the input levels of the other conditions (100 and 300 μM, respectively). After 24 hours the cellular concentrations of all these bile salts were strongly reduced again (Fig. 1B).
To study the dynamic changes in intracellular and extracellular D4-bile salts, hepatocytes were exposed to 100 μM D4CA and medium and hepatocytes were harvested at additional timepoints from 5 minutes to 24 hours (Fig. 2). Medium concentrations of conjugated D4-bile salts steadily increased in the first 4 hours (10 μM D4TCA and 21 μM D4GCA) (Fig. 2A). Almost complete conversion of D4CA to D4TCA and D4GCA was detected after 24 hours. Maximum intracellular accumulation of D4TCA (200 μM) and D4GCA (400 μM) was detected after 3 hours exposure to D4CA. Notably, in the first hour only D4TCA was detected in the medium and hepatocytes, whereas D4GCA started to appear after 1 hour and increased to higher levels compared to D4TCA (Fig. 2).
D4CA Does Not Induce Cell Death in Cultured Rat Hepatocytes.
Specific bile salts may be toxic for hepatocytes inducing either apoptotic or necrotic cell death.25 We analyzed the caspase-3 activity in cultured rat hepatocytes exposed to 100 μM D4CA (Fig. 3A). After 3 hours of incubation with 100 μM D4CA, we observed no significant increase in caspase-3 activity, whereas 50 μM glycochenodeoxycholic acid (GCDCA) induced a very strong apoptotic response (13-fold induction). In line with these findings, many apoptotic cells were detected after 24 hours of GCDCA exposure by acridine orange staining, which were absent in the D4CA-exposed hepatocyte cultures (Fig. 3C). In addition, no cellular leakage of LDH was observed in hepatocytes treated for 4 hours with 100 μM D4CA, indicating that no significant induction of necrotic cell death had occurred (Fig. 3B). These findings were confirmed by Sytox green staining (see Supporting Fig. S1).
Taurine Is the Preferred Substrate for Conjugation to Cholate in Rat Hepatocytes.
Taurine-conjugated bile salts predominate in the bile salt pool of rats. The standard culture medium for rat hepatocytes (Williams' E medium) contains high concentrations of glycine (666 μM) with no additional taurine present, which may result in the high D4GCA formation, especially at later timepoints. To determine whether the presence of glycine and taurine in the medium affects D4TCA and/or D4TCA production, we exposed rat hepatocytes for 24 hours to 100 μM D4CA in minimal medium with or without 666 μM glycine and/or 666 μM taurine (Fig. 4). Excess of either glycine or taurine in the culture medium leads to a concomitant D4GCA and D4TCA production, respectively, both extracellularly (Fig. 4A) and intracellularly (Fig. 4B). When both glycine and taurine were present in excess in the medium, D4CA was predominantly converted to D4TCA (70 μM, compared to only 2 μM D4GCA).
Selective Membrane Permeabilization of Rat Hepatocytes Reveals Peroxisomes as Bile Salt Accumulation Sites.
Peak accumulation of D4TCA (200 μM) and D4GCA (400 μM) in hepatocytes was observed after 3 hours exposure to D4CA (Fig. 2). Hepatocytes exposed to these conditions were analyzed by digitonin permeabilization assays to determine whether D4-labelled bile salts accumulate in membrane-enclosed intracellular compartments. Low concentrations of digitonin (30 μg/mL) disrupt the plasma membrane and cytosolic components are effectively released from the cellular fraction (Fig. 5A; glyceraldehyde 3-phosphate dehydrogenase [GAPDH] is shown as a cytosolic marker protein, quantification in Fig. 5B). D4CA and D4GCA are fully released from hepatocytes at this concentration (Fig. 5B, shown only for D4CA). The peroxisomal membrane is more resistant to digitonin permeabilization and is only fully permeabilized at 500 μg/mL. Partial release of the peroxisomal marker proteins catalase and BAAT is observed at digitonin concentrations of 30 and 150 μg/mL (Fig. 5A, quantification in Fig. 5B). The digitonin-extractability of D4TCA lies between the profile for GAPDH/D4CA and catalase (Fig. 5B), suggesting that D4TCA accumulates, at least partly, in membrane-enclosed organelles with peroxisomal characteristics.
D4TCA Is Detected in Peroxisomes of D4CA-Exposed Hepatocytes.
To obtain further evidence for the accumulation of D4TCA in peroxisomes, we purified these organelles from a PNS fraction of D4CA-exposed rat hepatocytes (Fig. 6). After Nycodenz density gradient centrifugation of the PNS, all 20 gradient fractions were analyzed for the presence of D4TCA, D4CA and markers for various cellular compartments. A PMP70/BAAT-enriched peak was detected at high density fractions 3-5, separated from mitochondria (Cyt C; fractions 10-11) and cytosol (GAPDH; fractions 15-20) (Fig. 6A). The highest concentrations of D4TCA were detected at the top of the gradient (Fig. 6B). In addition, minor but significant amounts of D4TCA were detected in fractions 3-5, revealing a similar concentration profile as the peroxisomal marker proteins (Fig. 6C). In contrast, D4CA and D4GCA were not detected in the peroxisome-enriched fractions.
In this study we established a novel assay that allows the study of transcellular and intracellular transport and conjugation of bile salts by rat hepatocytes in vitro. Primary rat hepatocytes effectively convert exogenously added D4CA to its D4TCA and D4GCA. Using digitonin permeabilization assays and peroxisome isolations, we demonstrate that D4TCA transiently resides in peroxisomes. This provides direct evidence that unconjugated bile salts shuttle through peroxisomes to become (re-)conjugated.
Reconjugation of deconjugated bile salts is an important process. Over 30% of the total bile salt pool may become deconjugated by intestinal bacteria on a daily basis.4 Our previous study showed that the enzyme catalyzing bile salt conjugation, BAAT, is localized predominantly, if not solely, in peroxisomes.11 This suggests that bile salts need to shuttle through peroxisomes for reconjugation. However, the possible existence of a cytosolic pool of BAAT remains a matter of debate.8-11 Therefore, we developed this novel assay to study trans- and intracellular transport and conjugation of D4CA. The use of D4CA allowed us to specifically follow its (re-)conjugation route, independent from endogenous-produced CA.
In the intact liver, bile salt reconjugation is highly efficient, because 97% of infused UDCA is conjugated to taurine or glycine after a single pass through rat livers.5 The in vitro cultured hepatocytes also perform this function with significant efficiency with less than 10% of the D4CA unaccounted for after 24 hours. This part of deuterium-labeled bile salts may still reside in the hepatocytes, be present as (CoA-)intermediate, or is metabolized to other products. The ratio between the amounts of GCA and TCA formed can be strongly manipulated by the levels of glycine and taurine in the growth medium. A high preference for TCA formation is observed when both amino acids are present in excess. This is in line with the predominant presence of taurine-conjugated bile salts in rats.26 However, in the presence of only glycine GCA is efficiently formed. These data show that the availability of glycine and/or taurine strongly determines the final conjugation profile. Glycine-conjugated bile salts are generally more cytotoxic compared to their taurine equivalents.25 Taurine supplementation in cholestatic diseases may therefore limit liver damage. This is especially relevant in humans who have a predominance of glycine-conjugated bile salts.
Cultured rat hepatocytes accumulated high concentrations of D4TCA and D4GCA after exposure to 100 μM D4CA. The cellular volume of a hepatocyte is estimated to be between 4 and 20 pL.19-24 Assuming 20 pL as the volume of a rat hepatocyte, we estimated that the peak intracellular concentrations of D4TCA and D4GCA were ≈200 μM and 400 μM, respectively, whereas D4CA levels were in the range of the concentrations in the medium. The intracellular accumulation of conjugated bile salts is transient, peaks at 3 hours exposure, and then declines. This suggests that export of D4TCA and D4GCA is limiting and cannot keep up with the production of conjugated bile salts. In our model with peroxisomal BAAT, transport of TCA and GCA occurs across the peroxisomal membrane and the plasma membrane (Fig. 7). Digitonin permeabilization and cell fractionation experiments indicate the presence of a peroxisomal and a cytosolic pool of D4TCA. This suggests that transport of D4TCA is a limiting factor at both the peroxisomal membrane and the plasma membrane. Only small amounts of D4TCA were detected in peroxisomes. This may be expected, as D4TCA only transiently resides in peroxisomes. Moreover, D4TCA may disappear from the peroxisomal fraction during their isolation due to (1) mechanical rupture of the peroxisomal membrane, and (2) maintained export of D4TCA without new production in peroxisomes. However, at present we are not able to discriminate between tauro/glyco-CA formed in the peroxisomes followed by transport to the cytosol and tauro/glycol-CA that is formed in the cytosol directly. This requires manipulation of the peroxisomal bile salt shuttle, either by inhibiting the to-be-identified-peroxisomal bile salt transporters or manipulating peroxisome biogenesis. It is relevant to note that this is the first report that demonstrates the presence of a specific product of peroxisomal metabolism in the peroxisome-enriched fractions after a full cell fractionation procedure. Mechanical breakage of peroxisomes was kept to a minimum by using optimized protocols that stabilize these organelles,13 which also further reconfirmed the predominant peroxisomal location of BAAT because it remained (almost) undetectable in the cytosol-enriched fractions after Nycodenz gradient centrifugation. Remarkably, significant amounts of BAAT, catalase, and PMP70 were also detected in low density gradient fractions cofractionating, in part, with mitochondria. In these fractions also D4TCA was detected. It remains to be determined whether these fractions contain a subpopulation of peroxisomes that may be involved in the bile salt conjugation as well.
To obtain independent evidence for the peroxisomal shuttle we also analyzed the subcellular distribution of several variants of 4-nitrobenzo-2-oxa-1,3-diazole (NBD)-labeled cholic acid (with the NBD group at the 3alpha, 3beta, 7alpha, and 7beta position, respectively)27 by fluorescence microscopy. Only 3alpha-NBD-cholate was taurine-conjugated and exported to the medium by cultured rat hepatocytes. However, the efficiency of conjugation is much lower (>90%) compared to D4CA. Interestingly, a clear accumulation of 3alpha-NBD-CA in subcellular structures was detected at early timepoints (see Supporting Fig. S2). Unfortunately, due to technical limitations we were unable so far to identify these subcellular structures (see Supporting data for details). Still, the detection of a clear punctuate staining pattern for 3alpha NBD-cholate in hepatocytes supports our data that bile salts (transiently) accumulate in membrane enclosed organelles.
Remarkably, we did not detect D4GCA in the peroxisomal fractions. Still, BAAT is believed to be responsible for both taurine and glycine-conjugation of bile salts.3 This may indicate that the peroxisomal bile salt exporter in rat hepatocytes has a higher affinity for GCA compared to TCA. Clearly, D4TCA and D4GCA also accumulate in the cytosol of hepatocytes, implying that the export of conjugated bile salts from hepatocytes by the bile salt export pump (BSEP) is also a rate-limiting step in our assay. This is most likely the result of a suboptimal location and/or activity of BSEP in cultured hepatocytes.
The identity of the peroxisomal bile salt transporters (importer and exporter) is unknown to date. A possible importer of CoA-activated C24 bile salts is the 70-kDa peroxisomal membrane protein (PMP70/ABCD3). PMP70 is an ATP-binding cassette transporter that is highly expressed in the liver.28 It has been proposed to transport long chain fatty acids into peroxisomes.29, 30 Recent research suggests that it may also transport bile acid intermediates, although thorough experimental evidence has not been presented yet.31 Importantly, the protein-mediated transport of conjugated bile salts across the peroxisomal membrane has recently been demonstrated in vitro.32 The characteristics of the transport activity, e.g., ATP-independent, make it unlikely that PMP70, or another peroxisomal ABC-transporter, is involved in this step.
Zellweger syndrome patients have no (functional) peroxisomes and accumulate intermediates of bile salt biosynthesis in their serum, variable amounts of which are conjugated.33, 34 This suggests that BAAT is (partially) active in the cytosol of these patients and is able to conjugate the accumulated bile salt intermediates. Recent studies using peroxisome-deficient Pex2−/− mice indeed show that the efficiency of conjugation of both C24 bile acids and C27 intermediates is reduced, but not absent, under normal conditions in these mutants. Moreover, bile acid conjugation is further impaired when these animals are fed a cholate-containing diet.35 Thus, reconjugation of bile salts may not strictly depend on the shuttle of bile salts through peroxisomes. Rather, it strongly increases the efficiency of the process.
In summary, we provide evidence that unconjugated bile salts shuttle through peroxisomes for taurine or glycine conjugation. Defects in the shuttle of bile salts through these organelles may lead to yet unrecognized cholestatic disorders.