Liver organotropism and biotransformation of a novel platinum-ursodeoxycholate derivative, Bamet-UD2, with enhanced antitumour activity.

BACKGROUND/AIMS: Several members of a novel family of bile acid derivatives with cytostatic and virostatic activity have been synthesized and characterized. The aim of this work was to investigate the liver organotropism and biotransformation of two novel compounds with enhanced DNA-reactivity: Bamet-D3, in which a glycine-polyamine tandem was used as a spacer to separate the glycocholic acid moiety from the platinum(II) atom, and Bamet-UD2, in which cisplatin was directly bound to the carboxylate group of two ursodeoxycholic acid moieties. METHODS: Drug uptake and "in vitro" toxicity were investigated using rat hepatocytes in primary culture. Following i.v. administration of 0.5 mumol cisplatin, Bamet-D3 or Bamet-UD2, bile output, urinary and fecal excretion, organ distribution and pharmacokinetic parameters were determined in short-term (3 h) and long-term (14 days) experiments carried out on anaesthetized and conscious rats, respectively. Liver biotransformation was investigated by HPLC analysis of bile samples. Total platinum was measured by flameless atomic absorption spectroscopy. Using Nude mice, antitumour activity was investigated in subcutaneously implanted Hepa 1-6 mouse hepatoma cells. RESULTS: Uptake by rat hepatocytes was Bamet-UD2 (11.3 nmol/mg protein) > Bamet-D3 (5.6 nmol/mg protein) > cisplatin (2.1 pmol/mg protein). Bamet-UD2 induced "in vitro" cell toxicity, which was not observed for Bamet-D3 or cisplatin. On the contrary, no toxicity "in vivo" for Bamet-UD2 was found which was observed for cisplatin and Bamet-D3. This may be related with the fact that bile output of Bamet-UD2, which occurs with no major biotransformation, was > 10 fold higher than that of cisplatin and 3-fold higher than that of Bamet-D3, which was previously transformed into at least three different metabolites. Fecal excretion was Bamet-UD2 > Bamet-D3 > cisplatin, whereas urinary output was Bamet-D3 > cisplatin > Bamet-UD2. Accordingly, a marked liver- and a reduced kidney-vectoriality for Bamet-UD2, but not for Bamet-D3, was observed. Bamet-UD2 and cisplatin, but not Bamet-D3, were efficient in inhibiting tumour growth whereas, only Bamet-UD2 significantly prolonged survival time. CONCLUSIONS: There results indicate that Bamet-UD2 is a cisplatin-ursodeoxycholate derivative with strong antitumour activity, marked hepatobiliary organotropism, and reduced toxic side-effects as compared to the parent drug cisplatin.

Relationship between DNA-reactivity and cytostatic effect of two novel bile acid-platinum derivatives, Bamet-UD2 and Bamet-D3.

BACKGROUND AND AIMS: Several platinum(II)-bile acid derivatives, named Bamets, have been previously synthesized. Their ability to interact with DNA, their cytostatic activity and their liver organotropic properties have been characterized. Two new compounds of this family, with particular structural properties, have been developed. Bamet-UD2 was formed by two ursodeoxycholic acid moieties bound by the carboxylate groups to cisplatin. In contrast, in Bamet-D3, glycine and a polyamine were used as tandem spacer elements to separate a cholic acid moiety from the platinum(II) atom. The aim of this work was to evaluate how these changes affect the ability of these compounds to interact with DNA and reduce tumour cell growth. MATERIALS AND METHODS: Drug reactivity with DNA was determined by changes in the electrophoretic mobility of the pUC18 plasmid test and by the ethidium bromide (EthBr) displacement assay. Cytostatic activity was measured against two mouse-derived cell lines from lymphocytic leukemia (L1210) and sarcoma (S-180-II). RESULTS: Bamet-UD2, and more markedly Bamet-D3, induced changes in the electrophoretic mobility of pUC18, suggesting the formation of DNA-drug interactions. Bamet-UD2 displaced EthBr from its binding to DNA. This effect was stronger in the case of Bamet-D3. Scatchard plots revealed that pre-incubation with both Bamet-UD2 and Bamet-D3 decreased the number of DNA sites available and their ability to bind EthBr. In spite of the enhanced DNA-reactivity of Bamet-D3, its ability to reduce tumour cell growth was much weaker than that of Bamet-UD2, which was seen to exert a very strong cytostatic effect. CONCLUSION: Although the distance between the platinum atom and the bile acid moiety affects the in vitro Bamet reactivity with DNA, other factors determine the overall cytostatic activity of these compounds.

Comparison of the effects of bile acids on cell viability and DNA synthesis by rat hepatocytes in primary culture.

Bile acid-induced inhibition of DNA synthesis by the regenerating rat liver in the absence of other manifestation of impairment in liver cell viability has been reported. Because in experiments carried out on in vivo models bile acids are rapidly taken up and secreted into bile, it is difficult to establish steady concentrations to which the hepatocytes are exposed. Thus, in this work, a dose-response study was carried out to investigate the in vitro cytotoxic effect of major unconjugated and tauro- (T) or glyco- (G) conjugated bile acids and to compare this as regards their ability to inhibit DNA synthesis. Viability of hepatocytes in primary culture was measured by Neutral red uptake and formazan formation after 6 h exposure of cells to bile acids. The rate of DNA synthesis was determined by radiolabeled thymidine incorporation into DNA. Incubation of hepatocytes with different bile acid species - cholic acid (CA), deoxycholic acid (DCA), chenodeoxycholic acid (CDCA) and ursodeoxycholic acid (UDCA), in the range of 10-1000 microM - revealed that toxicity was stronger for the unconjugated forms of CDCA and DCA than for CA and UDCA. Conjugation markedly reduced the effects of bile acids on cell viability. By contrast, the ability to inhibit radiolabeled thymidine incorporation into DNA was only slightly lower for taurodeoxycholic acid (TDCA) and glycodeoxycholic acid (GDCA) than for DCA. When the effect of these bile acids on DNA synthesis and cell viability was compared, a clear dissociation was observed. Radiolabeled thymidine incorporation into DNA was significantly decreased (-50%) at TDCA concentrations at which cell viability was not affected. Lack of a cause-effect relationship between both processes was further supported by the fact that well-known hepatoprotective compounds, such as tauroursodeoxycholic acid (TUDCA) and S-adenosylmethionine (SAMe) failed to prevent the effect of bile acids on DNA synthesis. In summary, our results indicate that bile acid-induced reduction of DNA synthesis does not require previous decreases in hepatocyte viability. This suggests the existence of a high sensitivity to bile acids of cellular mechanisms that may affect the rate of DNA repair and/or proliferation, which is of particular interest regarding the role of bile acids in the etiology of certain types of cancer.

Effect of maternal cholestasis on the kinetics of bile acid transport across the canalicular membrane of infant rat livers.

A reversible impairment in the ability of the liver to secrete cholephilic compounds has been reported to exist in infant rats born from mothers with surgically induced complete cholestasis during the last third of the pregnancy. Canalicular plasma membranes (CPM) were purified from livers obtained from 4 and 8 week-old offspring of healthy or cholestatic rats. Using radiolabelled glycocholic acid (GC) and a rapid filtration technique, bile acid transport by CPM vesicles in the presence of 3 mM ATP plus an ATP-regenerating system was measured at varying substrate concentrations. Kinetic parameters were calculated by nonlinear regression analysis. Similar values for the apparent affinity constant (Kt) were found in all experimental groups (approximately 350 microM). The value of the maximal velocity of the transport (Vmax) was similar for CPM obtained from control animals at 4 or 8 weeks of age (approximately 1.5 nmol/20 s/mg protein). In the offspring of cholestatic mothers the Vmax value was not different from that found in control animals as far as 4 week-old rats were concerned. However, Vmax in the 8 week-old group from cholestatic mothers was two-fold higher than that found in the rest of the experimental groups. Thus, the efficiency of transport, defined as Vmax/Kt, was very similar in all experimental groups, except in the group of 8 week-old offspring of cholestatic mothers, where this value was 60% higher. Isolated livers obtained from this group were able to secrete a tracer dose of radiolabelled GC (11.25 nmol) into bile significantly faster than isolated livers obtained from control animals of the same age (8 weeks). In sum, these results indicate that, in young infant rats (4 week-old), in which the maximal secretion rate for bile acids was reduced by maternal cholestasis during pregnancy, the kinetics of ATP-dependent bile acid transport across the canalicular membrane were not affected. By contrast, in older infant rats (8 week-old), in which the overall ability of the liver to secrete bile acids seems to be restored to normality, the efficiency of the canalicular transport system was actually enhanced. This suggests the existence of compensation at the level of the canalicular membrane transfer and thus that there is another hitherto unidentified mechanism involved in bile acid secretion.