Laurent P. Rivory

Conversion of irinotecan (CPT-11) to its active metabolite, 7-ethyl-10-hydroxycamptothecin (SN-38), by human liver carboxylesterase

We have investigated the conversion of the novel anti-topoisomerase I agent CPT-11 (irinotecan; 7-ethyl-10[4-(1-piperidino)-1-piperidno]carbonyloxycamptothecin ) to its active metabolite, SN-38 (7-ethyl-10-hydroxycamptothecin), by human liver carboxylesterase (HLC). Production of SN-38 was relatively inefficient and was enzyme deacylation rate-limited with a steady-state phase occurring after 15-20 min of incubation. This later phase followed Michaelis-Menten kinetics with an apparent Km of 52.9 +/- 5.9 microM and a specific activity of 200 +/- 10 mumol/sec/mol. However, the total enzyme concentration estimated from the intercept concentrations of SN-38 was much lower than that estimated directly from the titration of active sites with paraoxon (0.65 vs. 2.0 microM, respectively). Because deacylation rate-limiting kinetics result in the accumulation of inactive acyl-enzyme complex, we postulated that incubation of CPT-11 with HLC would result in an inhibition of the HLC-catalysed hydrolysis of p-nitrophenylacetate (p-NPA), an excellent substrate for this enzyme. Indeed, this was found to be the case although complete inhibition could not be attained. Analysis of possible kinetic schemes revealed that the most likely explanation for the disparity in estimated enzyme concentrations and the incomplete inhibition of p-NPA hydrolysis is that CPT-11 also interacts at a modulator site on the enzyme, which profoundly reduces substrate hydrolysis. Furthermore, loperamide, a drug often used for the treatment of CPT-11-associated diarrhea, was found to inhibit both CPT-11 and p-NPA HLC-catalysed hydrolysis, most likely by a similar interaction. These observations have direct implications for the clinical use of CPT-11.

IRINOTECAN (CPT-11) : A BRIEF OVERVIEW

1. Irinotecan (also known as CPT‐11) is a water soluble, semi‐synthetic analogue of 20(S)camptothecin (CPT) with promising activity against a range of tumour types.

Physiological pharmacokinetics of solutes in the isolated perfused rat hindlimb: Characterization of the physiology with changing perfusate flow, protein content, and temperature using statistical moment analysis

Distribution of Evans Blue (EB), sucrose, and water into the isolated perfused rat hindlimb was studied under various conditions using the multiple indicator dilution (MID) technique. Statistical moment analyses of the outflow profiles for the EB, sucrose, and water were used to define the vascular, extravascular, and total water spaces, respectively. The varied perfusion conditions included albumin content (2, 4.7, and 7%), temperature (25, 37, and 42 C), perfusate flow rate (2, 4, 8, and 12 ml/min) and the presence/absence of red blood cells. The range of studies undertaken were chosen to represent the variety of conditions used in the preparation of both isolated animal and human limbs, the latter being particularly important in cytotoxic therapy for recurrent malignant melanoma. The distribution volumes of EB, sucrose, and water were dependent on the flow rate and the albumin content of perfusate. The normalized variances (CV2) of the markers were of the following order: sucrose (2.18) > water (1.58) > EB (0.68), indicating that some disequilibrium occurs during the capillary exchange of water and sucrose. It is suggested that a Krebs-Henseleit buffer containing 2% BSA is a suitable perfusate for most studies of the isolated rat hindlimb perfusion. The effect of albumin concentration manifests itself only at higher flows.

Kinetics of toxic doses of paraquat and the effects of hemoperfusion in the dog.

Knowledge of the kinetics of an intoxicant is required for designing potential therapies in poisoned patients. In the case of paraquat, elucidating the kinetics has been made difficult by the paraquat-induced renal failure and the consequent dose- and time-dependent elimination of the herbicide. In the current study, we have modelled the plasma and urinary concentrations of paraquat in dogs given a toxic dose, the elimination of which was nonlinear. This enabled us, in turn, to simulate the apparent concentrations of paraquat in the deep tissue compartment, part of which is constituted by the major target organ for paraquat toxicity, the lung. Finally, we defined conditions, if any, under which charcoal hemoperfusion could reduce exposure of the deep compartment to paraquat by > or = 25%. We found that the plasma concentrations of paraquat could be described by a two compartment model with non-linear elimination from the central compartment. Use of a three compartment model did not improve the fit over that for a two compartment. The volume of distribution of paraquat at steady state approximated that of total body water. Simulated hemoperfusion performed for eight or eighty hours did not reduce exposure of the deep compartment to paraquat by > or = 25%, unless begun at times < or = two hours of the infusion commencing. This is consistent with our experimental data in the dog. The lack of efficacy of hemoperfusion is due to the rapid renal elimination of most of the absorbed dose of paraquat over the first 12 hours after its administration, and the later limitation of the rate of removal of paraquat from the body by the slow efflux rate from the deep to central compartment.

Aging, acute oxidative injury and hepatocellular glucose transport in the rat

We studied the effects of aging and acute exposure to hydrogen peroxide on hepatocellular glucose transport to determine whether (1) acute oxidative stress impairs glucose transport, (2) aging is associated with reduced glucose transport and (3) there are similarities between these changes that may provide insight into the aging process. Glucose transport was measured in the perfused livers of young and aged rats using the multiple indicator-dilution method. There were significant reductions in the rate constants for glucose influx (P < 0.001) and a significant increase in the extracellular volume in the livers from aged rats (aged: 0.39 ± 0.05 ml/g, young: 0.27 ± 0.04 ml/g). However, the Km and Vmax for glucose influx in the livers from aged rats (44 ± 22 mM, 7.1 ± 1.4 μmol.s−1g−1, respectively) were not significantly different from the values in young rats (64 ± 20 mM, 8.8 ± 1.3 μmol.s−1g−1). In the livers of young rats, treatment with hydrogen peroxide caused a significant reduction in glucose transport from 1.18 ± 0.22 to 0.49 ± 0.25 μmol.s−1g−1. This was partly restored to 0.69 ± 0.20 μmol.s−1g−1 by occluding the outflow catheter and expanding the extracellular space. Thus, although aging did not influence the rate of glucose transport in the perfused rat liver, this may be due to a compensatory age-related increase in the extracellular volume. In conclusion, the changes observed in hepatocellular glucose transport in the aged liver could be simulated by oxidative injury in the young liver, suggesting a role for oxidative injury in the aging process.

Effects of lipophilicity and protein binding on the hepatocellular uptake and hepatic disposition of two anthracyclines, doxorubicin and iododoxorubicin

Abstractu2002The anthracyclines, in particular doxorubicin (DOX), have been used for the intra-arterial locoregional therapy of liver tumours for over two decades. However, the results obtained with this form of therapy have been disappointing. It is widely recognised that DOX has a slow and limited tissue uptake, and we hypothesised that lipophilic analogues could be more suitable for locoregional administration. Using rat hepatocyte suspensions and the isolated rat liver, we examined the effects of lipophilicity, as determined from the octanol: buffer partition coefficient (Koct:buf), and protein binding of several anthracyclines on hepatocellular uptake. In particular, we compared DOX with 4′-iodo-4′-deoxy-doxorubicin (IDX), which differs only in the substitution of the daunosamine hydroxyl by an iodine molecule. Using a direct spectrofluorimetric method to evaluate cell uptake, we found that the influx rates correlated with the logarithm of Koct:buf and that IDX had the highest rate. However, the addition of bovine serum albumin (BSA) to the medium reduced the hepatocellular uptake of IDX more extensively than that of DOX such that the DOX uptake exceeded that of IDX with 4% BSA. Experiments in the isolated perfused rat liver confirmed these findings. We suggest that a trade-off of cellular uptake for reduced protein binding is desirable in the selection of drugs for intrahepatic administration. This may be accomplished by choosing anthracyclines with intermediate lipophilicity.

Axial tissue diffusion can account for the disparity between current models of hepatic elimination for lipophilic drugs

An assumption of previous models of hepatic elimination is that there is negligible axial diffusion in the liver. We show, by construction of a stochastic model and analysis of published data, that compounds which are readily diffusible and partitioned into hepatocytes may undergo axial tissue diffusion. The compounds most likely to be affected by axial tissue diffusion are the lipophilic drugs for which the cell membranes provide little resistance and which are highly extracted, thereby creating steep concentration gradients along the sinusoid at steady state. This phenomenon greatly modifies the availability of the compound under conditions of altered hepatic blood flow and protein binding. For moderately diffusible compounds, these relationships are similar to those predicted by the simplistic venous-equilibrium model. Hence, the paradoxical ability of the venous-equilibrium model to describe the steady-state kinetics of lipophilic drugs such as lidocaine, meperidine, and propranolol may be finally resolved. The effects of axial tissue diffusion and vascular dispersion on hepatic availability of drugs are compared. Vascular dispersion is of major importance to the availability of poorly diffusible compounds, whereas axial tissue diffusion becomes increasingly dominant for highly diffusive and partitioned substances.

The influence of pH on the interaction of lipophilic anthracyclines with bovine serum albumin : quantitative characterization by measurement of fluorescence quenching

We have investigated the interaction of the lipophilic anthracyclines 4-iodo-4-deoxydoxorubicin (IDX) and 4-demethoxy-daunorubicin (DDN) with bovine serum albumin by the quantitation of fluorescence quenching. The protein binding of IDX was extremely sensitive to the pH of the solution in which the complex was formed and paralleled the effect of pH on dimerization of the drug. The effect of pH on the protein binding and self-association of DDN was less extensive. Both compounds exhibited curvilinear Scatchard plots indicating apparent cooperativity in the binding process. Because of the self-association of the drugs in aqueous solution, we attempted to resolve this cooperativity in terms of the preferential binding of the dimer to the acceptor. However, we found that similar Scatchard plots could be simulated by using slightly erroneous estimates of the fluorescence yield of the complex, rendering any such analysis inconclusive. Consequently, the relationship between acceptor concentration and the fraction of ligand bound was considered to be fitted adequately in terms of a single acceptor site per albumin molecule. The pH dependence of the association constants for bovine serum albumin was described best by the hydrophobic interaction of neutral drug monomer with a binding site with titratable affinity. We postulate that the pH-dependent binding of some anthracyclines with albumin may lead to their enhanced uptake, relative to that of non-target organs, into tumours with an acidotic extracellular milieu.

Semisynthesis of RPR 121056A, a major metabolite of irinotecan (CPT-11)

Abstract The semisynthesis of RPR 121056A ( 4 0, a major metabolite of irinotecan (CPT-11, 2 ), is reported starting from SN-38 ( 3 ) and an appropriate side-chain precusor, and using a 2-step sequence. This semisynthesis is based on the 10-O-acylation of SN-38 with the conveniently protected carbamoylchloride derivative 10 followed by cleavage of the benzylic protecting groups by hydrogenolysis. Preliminary in vitro results show that RPR 121056A displays no cytotoxicity.