In-vitro-Modelle zur Erforschung des Absorptionsverhaltens von (Cyclo-)Peptiden im Magen-Darm-Trakt : quantitative Studien über zelluläre Aufnahme, Transport und Metabolisierung von Cyclosporin A und Analogen mit Hilfe von LC-LC-ESI-MS/MS

The course of a drug, from its discovery to its commercialisation is a long and difficult process which a lot of candidates end in a cul-de-sac. Among its difficulties are the form of administration (formulation) and the route of application. Absorption is mostly the time and extent determining parameter during oral administration until a drug reaches the systemic circulation. The intestinal epithelium is a passive as well as an active barrier to drug absorption of peptides, proteins and xenobiotics in general. The metabolic enzyme CYP3A4 and the efflux pump P-glycoprotein play a crucial role in this process in limiting actively intestinal absorption.
The goal of this thesis was to study in vitro the factors limiting the intestinal absorption to help develop a strategy that could optimise drug permeability (absorption).
A specific, sensitive, accurate and precise LC-LC-ESI-MS/MS-method with a good degree of reproducibility was developed for the quantitative measurement of the model substance (Cyclosporin A, CsA) and its three metabolites, as well as a new anti-HIV drug (D-MeAla3EtVal4Cs) in cell supernatant and cell extract. The bioanalytical method which was used to analyse the biological measurements described below, was capable of estimating quantitatively 0.3 ng/ml CsA in HBSS (linear range: 0.3 – 80 ng/ml, LOD: 40 pg/ml), as well as 0.6 ng/ml CsA and its three major metabolites in LS-180 cell extract (linear range: 0.6 – 10 ng/ml, LOD: 0.2 resp. 0.3 ng/ml, r ≥ 0.99, inprecision and accuracy-deviation ≤ 15 resp. 20% at LOD).
This bioanalytical method is the first unpublished LC-MS(/MS) method that can measure quantitatively the major CsA metabolites (AM1, AM9 und AM4N) in HBSS and in cell extract (LS-180 cells) with the lowest yet reported LOD and LLOQ.
In the first biological part of this work, we were able to show in three independent experiments that LS-180 cells cultured in the presence of calcitriol (induction) demonstrated a significant increase in CYP3A4 activity compared to untreated control cells. Significantly elevated metabolism of CsA in the calcitriol-treated cells (compared to untreated control cells) correlated with an elevated CYP3A4 activity and formed metabolites consistent with CYP3A4-mediated metabolism. The two
major CsA hydroxy-metabolites AM1 and AM9 (AM1 > AM9) were predominantly detected in the cells and in their supernatants, whereas the CsA N-desmethyl metabolite AM4N (<< AM1/AM9) was found almost exclusively in cell supernatants and not inside the cells. Amounts of extracellular AM4N were significantly higher in calcitriol treatet cells too. Ketoconazole, a well known CYP3A4 inhibitor, significantly decreased the formation of AM1, AM9 and AM4N back to the level of untreated control cells. Talinolol a P-gp substrate didn’t significantly affect the CsA metabolisation of AM1, AM9 and AM4N or their cellular uptake.
Our data demonstrate for the first time the direct correlation between CsA metabolite formation and CYP3A4 mRNA formation in calcitriol treated LS-180 cells compared to non-induced control cells.
These results indicate that LS-180 cells treated with calcitriol represent an interesting model to study intestinal interactions with CYP3A4 and conditional P-glycoprotein. Furthermore we found evidence for the formation of one or several not yet known CsA N-desmethyl metabolites in LS-180 cells, maybe by an alternative pathway to CYP3A4.
In the second biological part of this work, it will be shown that it was feasible to measure quantitatively CsA and D-MeAla3EtVal4Cs (an anti-HIV clinical trial phase II candidate and CsA derivate) directly in formulations by LC-LC-ESI-MS/MS (taking ion suppression into account). Furthermore a joined project with an industrial partner (Debiopharm Galenic Unit) was performed under the banner of the CTI. In this project in vitro and in vivo permeability studies of different formulations of CsA and D-MeAla3EtVal4Cs were investigated. A transwell system with Caco-2 cell monolayers was used as an in vitro model and SD rats were used as an in vivo model.
Both studies, in vitro and in vivo, were capable of differentiating bad from good formulation candidates: the « absorptions » lay in the same relative range as compared to the reference formulation. Two formu-lations were considered equal when comparing in vitro with in vivo data. Two other formulations were not. In the in vitro study these two formu-lations seemed to be better « absorbed » than the reference formulation whereas in the in vivo study they were less absorbed (compared to the reference formulation).
For reasons of economy the in vivo study was performed only once and it remains unclear which study from the in vitro or the in vivo study has
got the true answer. It has also to be said that when cmax is taken as « absorption » marker all formulations are equivalent. Only the « bad » formulation remains bad and the reference formulation is almost two times better. However it would be a big advantage to have an alternative to the time and cost intensive animal studies to evaluate formulation candidates for good bioavailability. The in vitro method described in this thesis is a human in vitro cell model (Caco-2) that can distinguish bad formulations from good ones and constitutes an interesting alternative to animal studies. The method allows a quick and easy screening of formulation candidates.