Biochimica et Biophysica Acta, 1091 (1991) 173-178 © 1991 Elsevier Science Publishers B.V. (Biomedical Division) 0167-4889/91/$03.50 ADONIS 0167488991000842
Cytotoxic effect and uptake mechanism by isolated rat hepatocytes of lithocholate and its glucuronide and sulfate Hajime Takikawa, Jun Tomita, Takahiro Takemura and Masami Yamanaka Department of Medicine, Teikyo University School of Medicine, Tokyo (Japan) (Received 13 July 1990)
Key words: Cytotoxicity; Hepatotoxicity; Bile Salt: Giucuronidation
The hepatotoxicity and uptake mechanism of lithocholate and its giucuronide and sulfate were studied nsing isolated rat hepatocytes. Cytotoxicity was in the order of lithocholate > lithocholate-glueuronide > lithocholate-suffate; their 50% cytotoxic concentrations on hepatocytes were 50, 150 and 700 #M, respectively. Thus, glucuronidation as well as sulfation acted to detoxify lithocholate, not relating to the previously reported higher cholestatic effect of iithocholateglucuronide than lithocholate. Lithocholate uptake was linear up to 50 #M, whereas the uptakes of lithocholateglucuronide and sulfate were saturable with an apparent K m and Vm,,, of 32 # M and 6.4 nmol/min per 10 6 cells for lithocholate-glucuronide and 26 #M and 11.8 nmol/min per 106 cells for iithocholate-sulfate. Na + replacement by choline + had no effect on the uptake of iithocholate and lithocholate-glucuronide, whereas it slightly inhibited lithocholate-sulfate uptake. Lithocholate-glucuronide uptake was inhibited by iithocholate-sulfate and sulfobromophthalein, whereas lithocholate-glucuronide and sulfobromophthalein had no effect on lithocholate-sulfate uptake. These data indicate that hepatic lithocholate uptake is inediated by simple diffusion, and that hepatic uptake of lithocholateglueuronide and sulfate is mainly mediated by a Na+-independent carrier.
Lithocholate is known to be toxic bile acid which causes cholestasis . Although glucuronidation and sulfation have been reported to be methods of detoxification of bile acids [2-7], Oelberg et al. reported that lithocholate-glucuronide has a higher cholestatic potency than lithocholate . In our previous report, lithocholate-glucuronide was more cholestatic in rats than lithocholate when infused at a rate of 0.29 ttmol/min per 100 g body wt. for 40 min, whereas lithocholatesulfate did not cause cholestasis at the same infusion rate and time period . Many studies about bile acid uptake by the liver have been reported using liver perfusion  and isolated or cultured hepatocytes [11-13]. The uptake of taurocholate is also recognized to be mediated by a Na+-de pendent mechanism using basolateral membrane vesicles [14,15]. Furthermore, recent studies have suggested that a protein of 48 or 49 kDa is the Na+-dependent bile acid carrier [16,17]. Correspondence: H. Takikawa, Department of Medicine, Teikyo University School of Medicine, Kaga 2-11-1, Itabashi, Tokyo 173, Japan.
A Na+-independent carrier for bile acid uptake has also been reported and a recent report suggeste6 that a carrier of 54 kDa is responsible for this mechanism . BStr.er et al. speculated that unconjugated, cholate is taken up by hepatocytes by exchange with OH- at the sinusoidal membrane , although this is controversial . In contrast, sulfobromophthalein (BSP) has also been reported to be taken up by a Na+-independent mechanism  and its uptake is considered to share the Na+-independent bile acid carrier for uptake . The third mechanism of bile acid uptake by the liver is simple diffusion as suggested for unconjugated hydrophobic bile acids . The mechanism of uptake of bile acid glucuronide and sulfate has been poorly understood except for a report by Bartholomew et al. using chenodeoxycholatesulfate . In the present study, (i) the direct cytotoxic effect of lithocholate and its glucuronide and sulfate on isolated rat hepatocytes was compared to study whether glucuronidation and sulfation are working as detoxication of lithocholate, and (ii) uptake mechanism of lithocholate and its glucuronide and sulfate by isolated rat hepatocytes was studk ~1.
174 Materials and Methods
Materials Lithocholate, lithocholate-3-sulfate, collagenase (type IV), NADH and mineral oil were purchased from Sigma; glycolithocholate and taurolithocholate were obtained from Calbiochem; silicon oil was obtained from Aldrich; and sulfobromophthalein (BSP) was obtained from Daiichi Kagaku, Tokyo. Lithocholate-3-O-glucuronide was kindly supplied by Tokyo Tanabe, Tokyo. The other reagents were all of analytical grade. [14C]Lithocholate (56 mCi/mmol) and [14C]cholate (50 mCi/mmol) were purchased from Amersham, U.K. and [14C]lithocholate-3-O-glucuronide and [14C]lithocholate-3-sulfate were synthesized from [~4C]lithocholate as previously reported . [3/3-3H]Lithocholate (> 95~ pure) was obtained by reduction by NaB[3H4] (19.6 Ci/mmol, Amersham) of 3-oxo-5/]-cholanic acid (Ste,'aloid) followed by purification by TLC as previously reported . Isolation of rat hepatocytes i-lepatocytes were isolated from male Wistar rats (280 g) by collagenase perfusion according to the method of Berry and Friend  and suspended in Tris-buffered balanced salt solution (131 mM NaCI, 5.2 mM KCI, 0.9 mM MgSO4, 1 mM CaCI 2, 10 mM Tris, 3 mM NaH2POt, pH 7.4). The viability of cells was > 90% by Trypan blue exclusion. All the following experiments were performed between 30-120 rain after cell isolations. Toxicity of bile acids on hepatocytes Bile acids dissolved in dimethylsulfoxide were added to hepatocyte suspensions (1.106 cells/ml) and incubated at 37 °C for 60 rain. The final concentration of dimethylsuifoxide was < 5%, which caused no cell damage. Cytotoxicity was measured by LDH release from hepatocytes and was shown by the percentages of complete cell lysis caused by sonication of cells in distilled water. Metabolism of lithocholate conjugates by hepatocytes (1.10 6 ceUs/ml) was analyzed by TLC using the appropriate t4C-labelled bile acids. The final concentrations of lithocholate conjugates in the medium were either those only with isotopes or 20 #M for lithechelate, 50 #M for lithocholate-giucuronide and 200 #M for lithocholate-sulfate, which are maximum concentrations not causing cell damage. After incubation for 60 rain, cells were separated from the medium by centrifugation and kept at room temperature for 24 h after the addition of 3 M NaOH. Neutralized digested cell solution and medium were applied to a Sep-Pak Cts cartridge and the bile acids were eluted with methanol . Aliquots of methanol extracts were counted for radioactivities and furiher analyzed by TLC. The plates (Kieselgel 60, 20 × 20 cm, Merck) were developed with
a solvent system of chloroform/methanol/acetic acid/ water (65 : 24:15 : 9) (solvent A). Samples to which lithocholate was added were also analyzed with a solvent system of ethanol/ethyl acetate/ammonia (45:45:15) (solvent B) to separate lithocholate-glucuronide from other conjugates . The plates were analyzed using a radiochromatoscanner (Imaging Scanner System 200, Bioscan, Washington D.C.) followed by spraying with 50% sulfuric acid and heating at 120 ° C.
Uptake of lithocholate conjugates The uptake of bile acids by hepatocytes (1.10 6 cells/ml) was studied by rapid filtration . After incubation at 37 °C for 10, 20 and 30 s with lithocholate and for 15, 30 and 45 s with lithocholate-glucuronide and sulfate, 100 t~l of the incubation mixture was put into a 400-#1 plastic tube which contained 50/~1 3 M KOH overlaid with 100 ~tl of a mixture of mineral and silicon oil ( d = 1.025). Cells were then separated from the medium by centrifugation. Neutralized cell solutions, after storing them at room temperature for 24 h, and medium were counte0 for radioactivities. The kinetics of uptake was calculated from the following equations, v=r.s
v = Vm~x"S / ( Km + S)
where V is uptake velocity, K is diffusion constant, Vm.,x is apparent maximum uptake velocity, S is bile acid concentration and K m is apparent half saturation constant. Eqn. 1 was used for lithocholate uptake and Eqn. 2 was for the uptake of lithocholate-glucuronide and sulfate using the non-linear least-squares method . Uptake of lithocholate conjugates (2 and 20 #M) was studied by replacing Na + with choline + . Furthermore, the effects of other lithocholate conjugates, taurochelate and BSP on the uptake of lithocholate conjugates were studied. In the experiments with lithechelate versus lithocholate-glucuronide or sulfate, the uptake of [3H]lithocholate and 14C-labeled lithocholateglucuronide or sulfate was simultaneously examined. Results and Discussion
The cytotoxic effect on hepatocytes was highest for lithocholate, followed by lithocholate-glucuronide and then sulfate (Fig. 1). The 50% cytotoxic concentrations were about 50 #M for lithocholate, 150 t~M for lithocholate-glucuronide and 700 #M for lithocholate-sulfate. These data agree with those previously reported suggesting that glucuronidation is acting to detoxify toxic bile acids [2-7], and indicate that the higher cholestatic effect of lithocholate-glucuronide than lithocholate [8,9] does not relate to the cytotoxic potency.
106 cells/me / hr incubation !
,~-F'ront "=-LC q"GLC
q'-TLC ,m-.TL.C ,~--LC '~-GLC
2bo s6o iobo 2 oo
Fig. 1. Cytotoxic effects of lithocholate conjugates on isolated rat hepatocytes. Data are means+ S.D. of three to six experiments. Cells (1.106/ml) were incubated at 37°C for 1 h. LC; lithocholate, LC-gl; lithocholate-glucuronide, LC-S; lithocholate-sulfate.
We examined the biotransformation of lithocholate conjugates using isolated hepatocytes to understand which conjugates were actually causing cytotoxic effects in these experiments. Fig. 2 shows examples of TLC analysis of cell extracts with trace amounts of lithocholate conjugates. With lithocholate, although taurine conjugates were predominant, most bands corresponding to taurine conjugates had a lower R F than taurolithocholate suggesting further hydroxylation of taurolithocholate. To detect the glucuronide, the sample was also analyzed using solvent system B showing production of a small amount of glucuronide. With lithocholate-glucuronide, more than half was conjugated with
Fig. 2. Thin-layer chromatography of the radiolabeled bile acids in the cells. A trace amount of [14C]lithocholate was incubated with cell suspension (1-106 cells/ml). (a) and (b), lithocholate (LC); (c) lithocholate-ghicuronide (LC-gl); (d) lithocholate-sulfate (LC-S). Solvent: a, c and d; chloroform/methanol/acetic acid/water (65 : 24:15 : 9), b; ethanol/ethyl acetate/ammonia (45:45:15). GLC; glycolithocholate, TLC; tautolithocholate, TLC-gl; taurolithocholate-glucuronide, TLCS; taurolithocholate-sulfate.
taurine and an unknown product with an R F of 0.43 appeared suggesting a hydroxylated form of lithocholate-glucuronide. With lithocholate-sulfate, the major
Metabolism of iithocholate conjugates by isolated rat hepatoo,tes Values are mean + S.D. of three different experiments. Distribution (%) Conjugate (%) Lithocholate Trace cell buffer 20/tM cell buffer
82 + 1 18 + 1 78+2 22 + 2
87+1 13+1 65+3 35 + 3
Lithocholate-sulfate Trace cell buffer 200/tM cell buffer
glycine lithocholate-glucuronide others conjugate
2+ 3 18 + 16 2+ 1 23_+ 3
79 + 62 + 78+ 51 +
6 10 12 19
7 12 3 3
-+2 _+2 _+2 _+1
lithocholate-glucuronide taurolithocholate-glucuronide others
Lithocholate-glucuronide Trace cell buffer 50pM cell buffer
93+7 7+ 7 27+4 73 + 4
38+ 9 67-+13 86_+ 1 93-+ 1
49_+ 9 22-+10 11+ 1 5_+ 1
13 10 4 2
_+1 -+3 5=1 + 0.2
55:1 46 + 39 88+ 1 97 + 0.3
90+ 1 49 + 35 9+ 1 2 + 0.3
5 +1 6 +4 3 +0.3 0.4 + 0.2
5-+6 5_+4 6_+1 5+0.4
8-+1 5_+1 2+1 1+0.1
176 conjugate was the major metabolite which was more abundant in cells than in medium. With 200/~M lithocholate-sulfate, only a little taurine conjugation occurred which was also more abundant in cells than in the medium. The slower taurine conjugation of these conjugates than lithocholate, which was observed in our in vivo experiments , may explain the lower level of taurine conjugation with a large amount of lithocholate-glucuronide and sulfate. The uptake of lithocholate was linear up to 30 s and that of lithocholate-glucuronide and sulfate was linear up to 45 s (Fig. 3a). The y-axis of the figures, which main~.y corresponded to non-specific binding of bile acids to the cell surface, was the highest with lithocholate which showed linear uptake up to 50/zM (Fig. 3b). These results confirmed a previous report ,
product was taurolithocholate-sulfate and a minor product with an R r of 0.46 was observed suggesting a hydroxylated form of lithocholate-sulfate. Table I summarizes the metabolism of lithocholate conjugates. With a tarce amount of lithocholate, the major metabolites were taurine conjugates both in the cells and medium. With 20 #M lithocholate, taurine conjugates were still the major metabolites in the cells and medium. The glucuronide was less than 10% in both the cells and medium. With lithocholate-glucuronide, some taurine conjugate was observed and the percentage of the taurine conjugate of lithocholateglucuronide was higher with 50 pM than with a trace amount. At both concentrations, the percentage of the taurine conjugate was higher in cells than in the medium. With a trace amount of lithocholate-sulfate, the taurine
~z,...---"- WOpM S,M ~--, IpM
I0, ~ ~ ~ . . . . ~
5pM 2pM ".' I ~M
o . . . . .
5"o (~ W60,,M
20 LC cemmtration
0 i')5 I'0
2"0 5() ~ |O0/~M LC-S concentration
Fig. 3. Uptake of lithocholate (LC) and its glucuronide (LC-gl) and sulfate (LC-S) by isolfated rat hepatocytes. (a) Change with time; (b) uptake velocity at different concentrations.
177 suggesting that lithocholate is taken up by simple diffusion. In contrast, saturated uptake was observed with lithocholate-glucuronide and sulfate (Fig. 3b). The diffusion constant of lithocholate uptake was 0.35 nmol/min per 106 cells. The apparent K m of the uptake of lithocholate-glucuronide and sulfate were similar at 32 +_ 10 and 26 +_ 8 lgM (mean _+ S.D.), respectively, whereas the apparent Vmax of the uptake of lithocholate-glucuronide (6.4 4-2.2 nmol/min per 106 cells) was about half that of lithocholate-sulfate (11.8 42.2 nmol/min per 106 cells). These data suggest that lithocholate-sulfate was more effectively taken up by hepatocytes than lithocholate-glucuronide. However, Gartner et al. reported that the rate of uptake of lithocholate-glucuronide is about twice that of lithocholatesulfate by the multiple indicator dilution method using perfused rat liver . We think that the discrepancy between our data and theirs is because they used 2.5%
albumin in the perfusate. The higher binding affinity of lithocholate-sulfate compared to lithocholate-glucuronide (Takikawa, et al. unpublished data) may have caused a lower apparent uptake of lithocholate-sulfate. When N a + was replaced by choline +, the uptake of lithocholate and lithocholate-glucuronide remained unchanged, whereas that of lithocholate-sulfate was slightly, but significantly inhibited (79 and 72% of control, with 2 and 20/~M lithocholate-sulfate, respectively (Fig. 4). The uptake of [3H]lithocholate and either 4C]lithocholate-glucuronide or sulfate was not affected by each other both at 2 and 20 #M (Fig. 4). The uptake of [3H]lithocholate was not affected by [~4C]taurocholate (Fig. 4) and that of [~4C]taurocholate was not affected by [3Hllithocholate (data not shown). The uptake of lithocholate-glucuronide (2 #M) was inhibited by lithocholate-sulfate (66 and 557o of control
J :f: ~.
u~ m +
m m +