Humanized liver TK-NOG mice with functional deletion of hepatic murine cytochrome P450s as a model for studying human drug metabolism

Humanized mouse models: A valuable platform for preclinical evaluation of human cancer

Animal models are routinely employed to assess the treatments for human cancer. However, due to significant differences in genetic backgrounds, traditional animal models are unable to meet bioresearch needs. To overcome this restriction, researchers have generated and optimized immunodeficient mice, and then engrafted human genes, cells, tissues, or organs in mice so that the responses in the model mice could provide a more reliable reference for treatments. As a bridge connecting clinical application and basic research, humanized mice are increasingly used in the preclinical evaluation of cancer treatments, particularly after gene interleukin 2 receptor gamma mutant mice were generated. Human cancer models established in humanized mice support exploration of the mechanism of cancer occurrence and provide an efficient platform for drug screening. However, it is undeniable that the further application of humanized mice still faces multiple challenges. This review summarizes the construction approaches for humanized mice and their existing limitations. We also report the latest applications of humanized mice in preclinical evaluation for the treatment of cancer and point out directions for future optimization of these models.

Omics-based investigation of pathological liver injury induced by Echinococcus multilocularis infection in mice

BACKGROUND

Alveolar echinococcosis (AE) can cause severe liver injury and be fatal if left untreated. Currently, there are no effective therapeutic options for AE-induced liver injury. Therefore, by exploring the changes of gene proteins in mice with damaged liver, we attempted to identify the key molecules of liver damage, and provide data that will enable the development of drugs targeting hepatic AE.

METHODS

BALB/c mice were inoculated with protoscoleces via the hepatic portal vein. Three months later, B-ultrasound examination and Hematoxylin-eosin (H&E) staining were used to confirm liver damage in mice. RNA sequencing and Liquid chromatography-mass spectrometry (LC-MS) were used to screen differentially expressed molecules associated with liver damage through bioinformatics, and Quantitative Real-Time PCR (qRT-PCR) was used to verify their expression.

RESULTS

B-ultrasound examination showed liver lesions in the infected group, and H&E staining showed liver inflammation, fibrosis and liver necrosis. RNA sequencing and LC-MS results showed changes in the levels of more than 1000 genes and proteins, with upregulation of immune and inflammation pathways. By contrast, the downregulated genes and proteins were mostly involved in various metabolic reactions. Correlation analysis was conducted between the transcriptome data and proteome data. The results revealed 240 differentially expressed genes, of which 192 were upregulated, and 48 were downregulated. Many of these genes were involved in metabolic reactions, such as Catalase (Cat), fatty acid synthase (Fasn), and IL-16 genes, which may have relevance to liver injury. The results of qRT-PCR were consistent with those of bioinformatics analysis.

CONCLUSIONS

The mechanisms of liver injury in mice infected with Echinococcus multilocularis are complex, involving abnormal metabolism, oxidative stress, inflammatory response, and many other factors. This study provides the data for preliminary exploration for the development of targeted therapies against AE.

Empirical scaling factor for predicting human pharmacokinetic profiles of disproportionate metabolites using the Css-MRTpo method and chimeric mice with humanised livers

Predicting plasma concentration-time profiles of disproportionate metabolites in humans is crucial for evaluating metabolites according to the Safety Testing guidelines. We evaluated Css-MRTpo, an empirical method, using chimeric mice with humanised livers capable of generating human-disproportionate metabolites. Azilsartan and AZ-M2 were administered to humanised chimeric mice, and pharmacokinetic parameters were obtained. Pharmacokinetic data for DS-1971a and DS-M1 in humanised chimeric mice were obtained from the literature. The human plasma concentration-time profiles of these compounds were simulated using the Css-MRTpo method. Azilsartan, DS-1971a, and PF-04937319 produced human disproportionate metabolites, AZ-M2, DS-M1, and PF-M1, respectively. The predicted human pharmacokinetic profiles of PF-04937319 and PF-M1 were obtained from a previous study, and their outcomes were re-evaluated. Our findings revealed that the plasma concentrations of the three metabolites were unexpectedly underpredicted, whereas the three unchanged drugs were reasonably predicted. Further, the introduction of the empirical scaling factor of 3, obtained from six model compounds, improved the predictability of metabolites, suggesting the potential usefulness of the Css-MRTpo method in combination with humanised chimeric mice for predicting the pharmacokinetic profiles of disproportionate metabolites at the early stage of new drug development.

Analysis of the Bile Acid Composition in a Fibroblast Growth Factor 19-Expressing Liver-Humanized Mouse Model and Its Use for CYP3A4-Mediated Drug-Drug Interaction Studies

Numerous biomedical applications have been described for liver-humanized mouse models, such as in drug metabolism or drug-drug interaction (DDI) studies. However, the strong enlargement of the bile acid (BA) pool due to lack of recognition of murine intestine-derived fibroblast growth factor-15 by human hepatocytes and a resulting upregulation in the rate-controlling enzyme for BA synthesis, cytochrome P450 (CYP) 7A1, may pose a challenge in interpreting the results obtained from such mice. To address this challenge, the human fibroblast growth factor-19 (FGF19) gene was inserted into the Fah-/- , Rag2-/- , Il2rg-/- NOD (FRGN) mouse model, allowing repopulation with human hepatocytes capable of responding to FGF19. While a decrease in CYP7A1 expression in human hepatocytes from humanized FRGN19 mice (huFRGN19) and a concomitant reduction in BA production was previously shown, a detailed analysis of the BA pool in these animals has not been elucidated. Furthermore, there are sparse data on the use of this model to assess potential clinical DDI. In the present work, the change in BA composition in huFRGN19 compared with huFRGN control animals was systematically evaluated, and the ability of the model to recapitulate a clinically described CYP3A4-mediated DDI was assessed. In addition to a massive reduction in the total amount of BA, FGF19 expression in huFRGN19 mice resulted in significant changes in the profile of various primary, secondary, and sulfated BAs in serum and feces. Moreover, as observed clinically, administration of the pregnane X receptor agonist rifampicin reduced the oral exposure of the CYP3A4 substrate triazolam. SIGNIFICANCE STATEMENT: Transgenic expression of FGF19 normalizes the unphysiologically high level of bile acids in a chimeric liver-humanized mouse model and leads to massive changes in bile acid composition. These adaptations could overcome one of the potential impediments in the use of these mouse models for drug-drug interaction studies.