REPOSITORY > RESULTS

Abstract

The main problem of the current routine in vitro genotoxicity testing for in vivo genotoxicity and carcinogenicity assessment, which is in the EU required for nearly all chemicals prior to marketing, is relatively low predictivity of rodent cell-based models used for this purpose. The consequence is unnecessary in vivo follow-up testing, or the discontinued development of promising chemicals. Unacceptably low predictivity is ascribed to a number of deficiencies that the commonly used cell lines possess, wherein their abnormal metabolic capacity is certainly the substantially contributing factor. The most advantageous approach to overcome the problem of poor predictivity of the in vitro genotoxicity test systems would be a cell line of human origin, which reflects in vivo xenobiotic metabolism, has a stable karyotype, is p53 and DNA-repair proficient, and is able to respond to direct and indirect acting chemicals. The use of several human hepatic tumor-derived cell lines with preserved activity of certain metabolic enzymes in vitro has been shown to be promising, but a much larger number of other human cell lines derived from hepatic tumors is available on the market that have never been used in genotoxicity studies, and some of them may be even more suitable for this purpose. However, some cancerous features of such cells might significantly affect their sensitivity and specificity compared to normal cells. In that respect, new cell-based models with metabolic capacities similar to primary hepatocytes are needed.
In the first part of this work, the characteristics relevant for the potential use in genotoxicity testing have been investigated in a panel of human liver-derived cell lines. The findings suggested HepG2 and HuH6 hepatic cancer lines as potentially the best cell models. The two cell lines have comparable doubling times, express functional wild type p53, their chromosome number is in narrow range and they both showed similar sensitivity to detect DNA damage induced by all tested model pro-carcinogens. The basal expression of selected liver-specific transcription factors and genes involved in the metabolism of xenobiotics has been analysed in both lines and compared to that of primary human hepatocytes (pHep) from two donors, to get insight into their metabolic profiles. Furthermore, the transcriptional regulation of the selected genes was determined after the 24 h exposure to non-cytotoxic concentrations of three model pro-carcinogens. The comparison of metabolic profiles of hepatic tumor-derived cells to that of primary hepatocytes elucidated several limitations of metabolic capacity of HepG2 and HuH6 cells and none of them has been shown to be more comparable to pHep. However, the findings represent valuable information that can be used for improvements of both cancer lines to be more convenient for in vitro genotoxicity studies and for the better understanding and interpretation of toxicological results obtained with HepG2 or HuH6 line.
In the second part, the development of metabolically active differentiated hepatic progenies (hDHP) from human adipose tissue-derived mesenchymal stem cells (hASC) using a three-step hepatic differentiation procedure has been performed. Except morphology, the hASC-derived hepatic progenies expressed typical hepatic functional characteristics, such as glycogen storage and albumin secretion. The results of the comet assay revealed comparable sensitivity of hASC and hDHP towards direct acting genotoxic agent tert-butyl hydroperoxide. The exposure to model indirect acting genotoxins did not induce DNA damage in hASC, while hDHP cells have responded to B(a)P and AFB1 exposure indicating capability of metabolic activation of the two genotoxic agents. The gene and protein expression analysis confirmed the presence of several phase I and phase II xenobiotic metabolic enzymes in an inducible form in hDHP cells. The generated hDHP cells have been further immortalized with transfection with hTERT. The transfected cells lost metabolic capacity, but can be easily matured to metabolically active cells with sensitivity comparable to that of non-immortalized hDHP, providing limitless supply of hDHP cells.
The findings suggest many advantages of hASC-derived immortalized hDHP cells and several limitations of HepG2 and HuH6 cell lines. All three cell models are worth further investigations in terms of development of a test model, which could contribute to the reduction of the use of laboratory animals and to more efficient safety evaluation of new chemical products needed for better human health protection.