GLDADec: marker-gene guided LDA modeling for bulk gene expression deconvolution.

Inferring cell type proportions from bulk transcriptome data is crucial in immunology and oncology. Here, we introduce guided LDA deconvolution (GLDADec), a bulk deconvolution method that guides topics using cell type-specific marker gene names to estimate topic distributions for each sample. Through benchmarking using blood-derived datasets, we demonstrate its high estimation performance and robustness. Moreover, we apply GLDADec to heterogeneous tissue bulk data and perform comprehensive cell type analysis in a data-driven manner. We show that GLDADec outperforms existing methods in estimation performance and evaluate its biological interpretability by examining enrichment of biological processes for topics. Finally, we apply GLDADec to The Cancer Genome Atlas tumor samples, enabling subtype stratification and survival analysis based on estimated cell type proportions, thus proving its practical utility in clinical settings. This approach, utilizing marker gene names as partial prior information, can be applied to various scenarios for bulk data deconvolution. GLDADec is available as an open-source Python package at https://github.com/mizuno-group/GLDADec.

Anti-Inflammatory Effects of a Novel Nuclear Factor-κB Inhibitory Derivative Derived from Pyrazolo[3,4-d]Pyrimidine in Three Inflammation Models.

Nuclear factor-κB (NF-κB) plays a central role in inflammatory responses, and its physiologic functions are essential for cell survival and proliferation. Currently, drugs targeting NF-κB inhibition have not yet been applied in clinical practice. We investigated the physiologic effect of a novel NF-κB inhibitory compound, 1H-pyrazolo[3,4-d]pyrimidin-4-amine derivative (INH #1), on three inflammatory animal models. The pharmacokinetics were measured by liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis. Acute hepatitis was induced by administrating lipopolysaccharide (LPS) and D-(+)-galactosamine hydrochloride followed by the analysis of survival time and inflammatory mediators. Collagen-induced arthritis (CIA) was induced by immunization with type II collagen (CII), and serum-transfer arthritis (STA) was caused by injecting K/BxN mice serum. Clinical and histologic scores were evaluated in both arthritis models. Immune cell subset analysis, CII-induced interferon-gamma (IFN-γ) production and proliferation, and measurement of anti-CII IgG antibodies were performed in the CIA model. In the acute hepatitis model, INH #1 suppressed tumor necrosis factor-α (TNF-α) production and prevented early death in a dose-dependent manner. INH #1 significantly attenuated arthritis scores and joint inflammation in both arthritis models. Additionally, in the CIA model, dendritic cells (DCs) in the regional lymph nodes were decreased in the treated mice and antigen-induced IFN-γ production and cell proliferation in splenocytes were inhibited, whereas the titers of anti-CII IgG antibodies were comparable regardless of the treatment. Here we revealed that INH #1 exerted anti-inflammatory effects in vivo via inhibition of inflammatory mediators and suppression of cellular immune responses. This compound could be a novel candidate for inhibition of NF-κB in certain inflammatory diseases. SIGNIFICANCE STATEMENT: A novel nuclear factor-κB (NF-κB) inhibitory compound, 1H-pyrazolo[3,4-d]pyrimidin-4-amine derivative (INH #1), which retains physiologically essential NF-κB bioactivity, suppressed inflammation in three different mouse models: the acute hepatitis model, the collagen-induced arthritis model, and the K/BxN serum-transfer arthritis model. These results suggest that this compound could be a novel and potent anti-inflammatory agent.

Effect of N-o-nitrobenzylation on conformation and membrane permeability of linear peptides.

In this study, we explored the potential of the photoremovable o-nitrobenzyl (oNB) group as a tool to manipulate the membrane permeability and regulate the conformation of linear peptides by means of experimental and computational studies. We found that the introduction of one or more oNB groups markedly increased the permeability and altered the conformation, as compared to the corresponding unmodified peptides. We thoroughly investigated the impact of peptide length, number of oNB group, oNB insertion position, and introduction of N- and C-terminal protecting groups on the passive membrane permeability by means of parallel artificial membrane permeability assay (PAMPA). Photoreaction of peptides containing one or two oNB groups proceeded cleanly in moderate to high yields, releasing the unprotected parent linear peptide. The oNB-modified peptides showed a cis/trans conformational equilibrium, while after photolysis, the unprotected linear peptides showed only the trans-amide conformation. Furthermore, a comprehensive comparison of oNB-modified peptides and N-methylated peptides was conducted, encompassing conformational analysis and physicochemical properties. N-Substituted peptides favored a folded-like structure, which may contribute to the improvement in permeability.

Quantitative Model Analysis and Simulation of Pharmacokinetics and Metastasis-Associated Lung Adenocarcinoma 1 RNA Knockdown Effect After Systemic Administration of Cholesterol-Conjugated DNA/RNA Heteroduplex Oligonucleotide Crossing Blood-Brain Barrier of Mice.

The cholesterol-conjugated heteroduplex oligonucleotide (Chol-HDO) is a double-stranded complex; it comprises an antisense oligonucleotide (ASO) and its complementary strand with a cholesterol ligand. Chol-HDO is a powerful tool for achieving target RNA knockdown in the brains of mice after systemic injection. Here, a quantitative model analysis was conducted to characterize the relationship between the pharmacokinetics (PK) and pharmacodynamics (PD), non-coding RNA metastasis-associated lung adenocarcinoma 1 (Malat1) RNA, of Chol-HDO, in a time-dependent manner. The established PK model could describe regional differences in the observed brain concentration-time profiles. Incorporating the PD model enabled the unique knockdown profiles in the brain to be explained in terms of the time delay after single dosing and enhancement following repeated dosing. Moreover, sensitivity analysis of PK exposure/persistency, target RNA turnover, and knockdown potency identified key factors for the efficient and sustained target RNA knockdown in the brain. The simulation of an adequate dosing regimen quantitatively supported the benefit of Chol-HDO in terms of achieving a suitable dosing interval. This was achieved via sufficient and sustained brain exposure and subsequent strong and sustained target RNA knockdown in the brain, even after systemic injection. The present study provides new insights into drug discoveries and development strategies for HDO in patients with neurogenic disorders. SIGNIFICANCE STATEMENT: The quantitative model analysis presented here characterized the PK/PD relationship of Chol-HDO, enabled its simulation under various conditions or assumptions, and identified key factors for efficient and sustained RNA knockdown, such as PK exposure and persistency. Chol-HDO appears to be an efficient drug delivery system for the systemic administration of desired drugs to brain targets.

Progress in the Quantitative Assessment of Transporter-Mediated Drug-Drug Interactions Using Endogenous Substrates in Clinical Studies.

Variations in drug transporter activities, caused by genetic polymorphism and drug-drug interactions (DDIs), alter the systemic exposure of substrate drugs, leading to differences in drug responses. Recently, some endogenous substrates of drug transporters, particularly the solute carrier family transporters such as OATP1B1, OATP1B3, OAT1, OAT3, OCT1, OCT2, MATE1, and MATE2-K, have been identified to investigate variations in drug transporters in humans. Clinical data obtained support their performance as surrogate probes in terms of specificity and reproducibility. Pharmacokinetic parameters of the endogenous biomarkers depend on the genotypes of drug transporters and the systemic exposure to perpetrator drugs. Furthermore, the development of physiologically based pharmacokinetic models for the endogenous biomarkers has enabled a top-down approach to obtain insights into the effect of perpetrators on drug transporters and to more precisely simulate the DDI with victim drugs, including probe drugs. The endogenous biomarkers can address the uncertainty in the DDI prediction in the preclinical and early phases of clinical development and have the potential to fulfill regulatory requirements. Therefore, the endogenous biomarkers should be able to predict disease effects on the variations in drug transporter activities observed in patients. This mini-review focuses on recent progress in the identification and use of the endogenous drug transporter substrate biomarkers and their application in drug development. SIGNIFICANCE STATEMENT: Advances in analytical methods have enabled the identification of endogenous substrates of drug transporters. Changes in the pharmacokinetic parameters (Cmax, AUC, or CLR) of these endogenous biomarkers relative to baseline values can serve as a quantitative index to assess variations in drug transporter activities during clinical studies and thereby provide more precise DDI predictions.

Impact of Direction of Transport on the Evaluation of Substrate Recognition of Mouse Multidrug and Toxin Extrusion Protein 1.

Multidrug and toxin extrusion protein (MATE/SLC47A) secretes metabolites and xenobiotics into the urine in the proximal tubules of the kidney. Uptake assays have been commonly used for evaluating MATE-mediated transport of new chemical entities in drug development. The purpose of this study was to examine the relationship between in vitro uptake activities by MATEs and the impact of MATE-mediated transport in in vivo renal secretion. In vitro uptake in mouse Mate1 (mMate1)-expressing human embryonic kidney 293 (HEK293) cells and several in vivo parameters from mMate1 knockout and wild-type mice were compared using nine cationic compounds (almotriptan, naratriptan, talinolol, sumatriptan, alogliptin, sitagliptin, rivaroxaban, saxagliptin, and vildagliptin). Compounds that showed statistically significant decrease in secretory clearances with respect to kidney concentrations (CLR,kidney) in mMate1 knockout mice were categorized as in vivo substrates in this study. A good correlation (R2 = 0.637) was observed between the in vitro uptake ratio and the in vivo ratio of CLR,kidney of mMate1 knockout mice and wild-type mice. This study supported the rationale of using an uptake assay to determine whether investigational compounds are the substrate of MATEs and to predict drug-drug interaction risk via renal secretion by MATE from the viewpoint of drug development in pharmaceutical companies. SIGNIFICANCE STATEMENT: We revealed that substrates judged by in vitro experiments using mouse multidrug and toxin extrusion (mMate)1-expressing cells were excreted in urine via mMate1 in vivo, and a good correlation (R2 = 0.637) was observed between in vitro uptake ratio and in vivo ratio of secretory clearance with respect to the kidney concentrations (CLR,kidney) of mMate1 knockout and wild-type mice. This study supported the rationale of using an uptake assay to predict potential human MATE1-mediated drug-drug interaction as a victim.

NRBdMF: A Recommendation Algorithm for Predicting Drug Effects Considering Directionality.

Predicting the novel effects of drugs based on information about approved drugs can be regarded as a recommendation system. Matrix factorization is one of the most used recommendation systems, and various algorithms have been devised for it. A literature survey and summary of existing algorithms for predicting drug effects demonstrated that most such methods, including neighborhood regularized logistic matrix factorization, which was the best performer in benchmark tests, used a binary matrix that considers only the presence or absence of interactions. However, drug effects are known to have two opposite aspects, such as side effects and therapeutic effects. In the present study, we proposed using neighborhood regularized bidirectional matrix factorization (NRBdMF) to predict drug effects by incorporating bidirectionality, which is a characteristic property of drug effects. We used this proposed method for predicting side effects using a matrix that considered the bidirectionality of drug effects, in which known side effects were assigned a positive (+1) label and known treatment effects were assigned a negative (-1) label. The NRBdMF model, which utilizes drug bidirectional information, achieved enrichment of side effects at the top and indications at the bottom of the prediction list. This first attempt to consider the bidirectional nature of drug effects using NRBdMF showed that it reduced false positives and produced a highly interpretable output.

Usefulness of Human Jejunal Spheroid-Derived Differentiated Intestinal Epithelial Cells for the Prediction of Intestinal Drug Absorption in Humans.

This study aimed to demonstrate the usefulness of human jejunal spheroid-derived differentiated intestinal epithelial cells as a novel in vitro model for clarifying the impact of intestinal drug-metabolizing enzymes and transporters on the intestinal absorption of substrate drugs in humans. Three-dimensional human intestinal spheroids were successfully established from surgical human jejunal specimens and expanded for a long period using L-WRN-conditioned medium, which contains Wnt3a, R-spondin 3, and noggin. The mRNA expression levels of intestinal pharmacokinetics-related genes in the human jejunal spheroid-derived differentiated intestinal epithelial cells were drastically increased over a 5-day period after seeding compared with those in human jejunal spheroids and were approximately the same as those in human jejunal tissue over a culture period of at least 13 days. Activities of typical drug-metabolizing enzymes [cytochrome P450 (CYP) 3A, CYP2C9, uridine 5'-diphospho-glucuronosyltransferase 1A, and carboxylesterase 2] and uptake/efflux transporters [peptide transporter 1/solute carrier 15A1], P-glycoprotein, and breast cancer resistance protein) in the differentiated cells were confirmed. Furthermore, intestinal availability (Fg) values estimated from the apical-to-basolateral permeation clearance across cell monolayer showed a good correlation with the in vivo Fg values in humans for five CYP3A substrate drugs (Fg range, 0.35-0.98). In conclusion, the functions of major intestinal drug-metabolizing enzymes and transporters could be maintained in human jejunal spheroid-derived differentiated intestinal epithelial cells. This model would be useful for the quantitative evaluation of the impact of intestinal drug-metabolizing enzymes and transporters on the intestinal absorption of substrate drugs in humans. SIGNIFICANCE STATEMENT: Limited information is available regarding the quantitative prediction of the impact of drug-metabolizing enzymes and transporters on the human intestinal absorption of substrates using in vitro assays with differentiated cells derived from human intestinal spheroids/organoids. This study confirmed the functions of typical drug-metabolizing enzymes and transporters in human jejunal spheroid-derived differentiated intestinal epithelial cells and demonstrated that intestinal availability (Fg) estimated from apical-to-basolateral permeation clearance across cell monolayers showed a good correlation with in vivo human Fg for CYP3A substrates.

A rapid and simple electrochemical detection of the free drug concentration in human serum using boron-doped diamond electrodes.

Monitoring drug concentration in blood and reflecting this in the dosage are crucial for safe and effective drug treatment. Most drug assays are based on total concentrations of bound and unbound proteins in the serum, although only the unbound concentration causes beneficial and adverse events. Monitoring the unbound concentration alone is expected to provide a means for further optimisation of drug treatment. However, unbound concentration monitoring has not been routinely used for drug treatment due to the long analysis time and the high cost of conventional methods. Here, we have developed a rapid electrochemical method to determine the unbound concentration in ultrafiltered human serum using boron-doped diamond (BDD) electrodes. When the anticancer drug doxorubicin was used as the test drug, the catalytic doxorubicin-mediated reduction of dissolved oxygen provided a sensitive electrochemical signal, with a detection limit of 0.14 nM. In contrast, the sensitivity of glassy carbon (GC) was inferior under the same conditions due to interference from the dissolved oxygen reduction current. The signal background ratio (S/B) of BDD and GC was 11.5 (10 nM doxorubicin) and 1.1 (50 nM), respectively. The results show that a fast measurement time within ten seconds is possible in the clinical concentration range. Additionally, in the ultrafiltered human serum, the obtained values of unbound doxorubicin concentration showed good agreement with those quantified by conventional liquid chromatography-mass spectrometry. This approach has the potential for application in clinical settings where rapid and simple analysis methods would be beneficial.

Investigation of a Data Split Strategy Involving the Time Axis in Adverse Event Prediction Using Machine Learning.

Adverse events are a serious issue in drug development, and many prediction methods using machine learning have been developed. The random split cross-validation is the de facto standard for model building and evaluation in machine learning, but care should be taken in adverse event prediction because this approach does not strictly match the real-world situation. The time split, which uses the time axis, is considered suitable for real-world prediction. However, the differences in model performance obtained using the time and random splits are not clear due to the lack of comparable studies. To understand the differences, we compared the model performance between the time and random splits using nine types of compound information as input, eight adverse events as targets, and six machine learning algorithms. The random split showed higher area under the curve values than did the time split for six of eight targets. The chemical spaces of the training and test datasets of the time split were similar, suggesting that the concept of applicability domain is insufficient to explain the differences derived from the splitting. The area under the curve differences were smaller for the protein interaction than for the other datasets. Subsequent detailed analyses suggested the danger of confounding in the use of knowledge-based information in the time split. These findings indicate the importance of understanding the differences between the time and random splits in adverse event prediction and suggest that appropriate use of the splitting strategies and interpretation of results are necessary for the real-world prediction of adverse events. We provide the analysis code and datasets used in the present study at https://github.com/mizuno-group/AE_prediction.

Quantitative prediction of pharmacokinetic properties of drugs in humans: Recent advance in in vitro models to predict the impact of efflux transporters in the small intestine and blood-brain barrier.

Efflux transport systems are essential to suppress the absorption of xenobiotics from the intestinal lumen and protect the critical tissues at the blood-tissue barriers, such as the blood-brain barrier. The function of drug efflux transport is dominated by various transporters. Accumulated clinical evidences have revealed that genetic variations of the transporters, together with coadministered drugs, affect the expression and/or function of transporters and subsequently the pharmacokinetics of substrate drugs. Thus, in the preclinical stage of drug development, quantitative prediction of the impact of efflux transporters as well as that of uptake transporters and metabolic enzymes on the pharmacokinetics of drugs in humans has been performed using various in vitro experimental tools. Various kinds of human-derived cell systems can be applied to the precise prediction of drug transport in humans. Mathematical modeling consisting of each intrinsic metabolic or transport process enables us to understand the disposition of drugs both at the organ level and at the level of the whole body by integrating a variety of experimental results into model parameters. This review focuses on the role of efflux transporters in the intestinal absorption and brain distribution of drugs, in addition to recent advances in predictive tools and methodologies.

Frequency of null genotypes of glutathione S-transferase M1 and T1 in Japanese patients with drug-induced liver injury.

AIM: Previous reports suggest that the null genotype (*0/*0) of glutathione S-transferase (GST) M1 and/or GSTT1 could be risk factors for drug-induced liver injury (DILI). However, multi-institutional pharmacogenetic research with various suspected drugs has rarely been performed in Japan. Therefore, the aim of this study was to investigate the role of GSTM1 and GSTT1 null genotype in the occurrence of DILI in Japanese patients. METHODS: Blood samples of 270 DILI patients from 23 hospitals throughout Japan collected between 2010 and 2018 were subjected to genotyping of null genotypes of GSTM1 and GSTT1 using the SmartAmp-2 method. We also collected information on DILI types, time to onset of DILI, pharmacological classification of suspected drugs and Digestive Disease Week-Japan score, as well as genotypes of GSTM1 and GSTT1 in each patient with DILI. RESULTS: The distribution of a combination of null genotypes of GSTM1 and GSTT1 in Japanese patients with DILI was significantly different from that reported in the general Japanese population. Notably, the incidence of the GSTM1 null genotype in patients with DILI was significantly higher than that of the control population. A significant relationship between the frequency of GSTM1 and GSTT1 null genotypes and pharmacological classification of suspected drugs, clinical laboratory data for liver function, time to onset of DILI, and Digestive Disease Week-Japan scores was not observed. CONCLUSIONS: The GSTM1 null genotype was associated with an increased incidence of DILI in Japanese patients.

A randomized trial to examine the impact of food on pharmacokinetics of 4-phenylbutyrate and change in amino acid availability after a single oral administration of sodium 4-phenylbutyrarte in healthy volunteers.

Urea cycle disorders (UCDs), inborn errors of hepatocyte metabolism, result in the systemic accumulation of ammonia to toxic levels. Sodium 4-phenylbutyrate (NaPB), a standard therapy for UCDs for over 20 years, generates an alternative pathway of nitrogen deposition through glutamine consumption. Administration during or immediately after a meal is the accepted use of NaPB. However, this regimen is not based on clinical evidence. Here, an open-label, single-dose, five-period crossover study was conducted in healthy adults to investigate the effect of food on the pharmacokinetics of NaPB and determine any subsequent change in amino acid availability. Twenty subjects were randomized to one of four treatment groups. Following an overnight fast, NaPB was administered orally at 4.3 g/m2 (high dose, HD) or 1.4 g/m2 (low dose, LD) either 30 min before or just after breakfast. At both doses, compared with post-breakfast administration, pre-breakfast administration significantly increased systemic exposure of PB and decreased plasma glutamine availability. Pre-breakfast LD administration attenuated plasma glutamine availability to the same extent as post-breakfast HD administration. Regardless of the regimen, plasma levels of branched-chain amino acids (BCAA) were decreased below baseline in a dose-dependent manner. In conclusion, preprandial oral administration of NaPB maximized systemic exposure of the drug and thereby its potency to consume plasma glutamine. This finding may improve poor medication compliance because of the issues with odor, taste, and pill burden of NaPB and reduce the risk of BCAA deficiency in NaPB therapy.

Impact of Direction of Transport on the Evaluation of Inhibition Potencies of Multidrug and Toxin Extrusion Protein 1 Inhibitors.

Multidrug and toxin extrusion (MATE) transporters are expressed on the luminal membrane of renal proximal tubule cells and extrude their substrates into the luminal side of the tubules. Inhibition of MATE1 can reduce renal secretory clearance of its substrate drugs and lead to drug-drug interactions (DDIs). To address whether IC50 values of MATE1 inhibitors with regard to their extracellular concentrations are affected by the direction of MATE1-mediated transport, we established an efflux assay of 1-methyl-4-phenylpyridinium (MPP+) and metformin using the human embryonic kidney 293 model transiently expressing human MATE1. The efflux rate was defined by reduction of the cellular amount of MPP+ and metformin for 0.25 minutes shortly after the removal of extracellular MPP+ and metformin. Inhibition potencies of 12 inhibitors toward MATE1-mediated transport were determined in both uptake and efflux assays. When MPP+ was used as a substrate, 8 out of 12 inhibitors showed comparable IC50 values between assays (<4-fold). IC50 values from the efflux assays were higher for cimetidine (9.9-fold), trimethoprim (10-fold), famotidine (6.4-fold), and cephalexin (>3.8-fold). When metformin was used as a substrate, IC50 values of the tested inhibitors when evaluated using uptake and efflux assays were within 4-fold of each other, with the exception of cephalexin (>4.7-fold). IC50 values obtained from the uptake assay using metformin showed smaller IC50 values than those from the efflux assay. Therefore, the uptake assay is recommended to determine IC50 values for the DDI predictions. SIGNIFICANCE STATEMENT: In this study, a new method to evaluate IC50 values of extracellular added inhibitors utilizing an efflux assay was established. IC50 values were not largely different between uptake and efflux directions but were smaller for uptake. This study supports the rationale for a commonly accepted uptake assay with metformin as an in vitro probe substrate for multidrug and toxin extrusion 1-mediated drug-drug interaction risk assessment in drug development.

Characterization of the Human Intestinal Drug Transport with Ussing Chamber System Incorporating Freshly Isolated Human Jejunum.

Intestinal permeability is a critical factor for orally administered drugs. It can be facilitated by uptake transporters or limited by efflux transporters and metabolic enzymes in the intestine. The present study aimed to characterize the Ussing chamber system incorporating human intestinal tissue as an in vitro model for investigating the impact of intestinal uptake/efflux transporters on the intestinal absorption of substrate drugs in humans. We confirmed the functions of major intestinal uptake/efflux drug transporters in freshly isolated human jejunum sections by demonstrating a significant decrease in the mucosal uptake of cefadroxil (peptide transporter 1) and methotrexate (proton-coupled folate transporter), mucosal-to-serosal permeability of ribavirin (concentrative nucleoside transporters/equilibrative nucleoside transporters), and serosal-to-mucosal permeability of P-glycoprotein and breast cancer resistance protein substrates in the presence of their typical inhibitors. The mucosal-to-serosal apparent permeability coefficients (Papp) of 19 drugs, including substrates of drug transporters and cytochrome P450 3A, ranged from 0.60 × 10-6 to 29 × 10-6 cm/s and showed a good correlation with reported fraction of an oral dose that enters the gut wall and passes into the portal circulation with escaping intestinal metabolism (FaFg) values in humans. Furthermore, the Papp values for cefadroxil, methotrexate, and ribavirin in the presence of the corresponding transporter inhibitors underestimated the FaFg of these drugs, which clearly showed that intestinal uptake transporters facilitate their intestinal absorption in humans. In conclusion, the functions of major intestinal uptake/efflux drug transporters could be maintained in freshly isolated human jejunum sections. The Ussing chamber system incorporating human intestinal tissue would be useful for evaluating the impact of intestinal uptake/efflux transporters on the intestinal absorption of various types of drugs in humans. SIGNIFICANCE STATEMENT: Although previous studies have predicted the intestinal absorption of drugs in humans using the Ussing chamber system incorporating human intestinal tissue, there is little systematic information about drug transport mediated by multiple transporters in this system. We confirmed the functions of major intestinal uptake/efflux transporters in freshly isolated human jejunum sections and demonstrated that the mucosal-to-serosal apparent permeability coefficient of various types of drugs showed a good correlation with reported human FaFg values.

Functional Investigation of Solute Carrier Family 35, Member F2, in Three Cellular Models of the Primate Blood-Brain Barrier.

Understanding the mechanisms of drug transport across the blood-brain barrier (BBB) is an important issue for regulating the pharmacokinetics of drugs in the central nervous system. In this study, we focused on solute carrier family 35, member F2 (SLC35F2), whose mRNA is highly expressed in the BBB. SLC35F2 protein was enriched in isolated mouse and monkey brain capillaries relative to brain homogenates and was localized exclusively on the apical membrane of MDCKII cells and brain microvascular endothelial cells (BMECs) differentiated from human induced pluripotent stem cells (hiPS-BMECs). SLC35F2 activity was assessed using its substrate, YM155, and pharmacological experiments revealed SLC35F2 inhibitors, such as famotidine (half-maximal inhibitory concentration, 160 µM). Uptake of YM155 was decreased by famotidine or SLC35F2 knockdown in immortalized human BMECs (human cerebral microvascular endothelial cell/D3 cells). Furthermore, famotidine significantly inhibited the apical (A)-to-basal (B) transport of YM155 in primary cultured monkey BMECs and hiPS-BMECs. Crucially, SLC35F2 knockout diminished the A-to-B transport and intracellular accumulation of YM155 in hiPS-BMECs. By contrast, in studies using an in situ brain perfusion technique, neither deletion of Slc35f2 nor famotidine reduced brain uptake of YM155, even though YM155 is a substrate of mouse SLC35F2. YM155 uptake was decreased significantly by losartan and naringin, inhibitors for the organic anion transporting polypeptide (OATP) 1A4. These findings suggest SLC35F2 is a functional transporter in various cellular models of the primate BBB that delivers its substrates to the brain and that its relative importance in the BBB is modified by differences in the expression of OATPs between primates and rodents. SIGNIFICANCE STATEMENT: This study demonstrated that SLC35F2 is a functional drug influx transporter in three different cellular models of the primate blood-brain barrier (i.e., human cerebral microvascular endothelial cell/D3 cells, primary cultured monkey BMECs, and human induced pluripotent stem-BMECs) but has limited roles in mouse brain. SLC35F2 facilitates apical-to-basal transport across the tight cell monolayer. These findings will contribute to the development of improved strategies for targeting drugs to the central nervous system.

Multi-omics analysis of hiPSCs-derived HLCs matured on-chip revealed patterns typical of liver regeneration.

Maturation of human-induced pluripotent stem cells (hiPSCs)-derived hepatocytes-like cells (HLCs) toward a complete hepatocyte phenotype remains a challenge as primitiveness patterns are still commonly observed. In this study, we propose a modified differentiation protocol for those cells which includes a prematuration in Petri dishes and a maturation in microfluidic biochip. For the first time, a large range of biomolecular families has been extracted from the same sample to combine transcriptomic, proteomic, and metabolomic analysis. After integration, these datasets revealed specific molecular patterns and highlighted the hepatic regeneration profile in biochips. Overall, biochips exhibited processes of cell proliferation and inflammation (via TGFB1) coupled with anti-fibrotic signaling (via angiotensin 1-7, ATR-2, and MASR). Moreover, cultures in this condition displayed physiological lipid-carbohydrate homeostasis (notably via PPAR, cholesterol metabolism, and bile synthesis) coupled with cell respiration through advanced oxidative phosphorylation (through the overexpression of proteins from the third and fourth complex). The results presented provide an original overview of the complex mechanisms involved in liver regeneration using an advanced in vitro organ-on-chip technology.

Decomposition Profile Data Analysis for Deep Understanding of Multiple Effects of Natural Products.

It is difficult to understand the entire effect of a natural product because such products generally have multiple effects. We propose a strategy to understand these effects effectively by decomposing them with a profile data analysis method we developed. A transcriptome profile data set was obtained from a public database and analyzed. Considering their high similarity in structure and transcriptome profile, we focused on rescinnamine and syrosingopine. Decomposed effects predicted clear differences between the compounds. Two of the decomposed effects, SREBF1 activation and HDAC inhibition, were investigated experimentally because the relationship between these effects and the compounds had not yet been reported. Analyses in vitro validated these effects, and their strength was consistent with predicted scores. Moreover, the number of outliers in decomposed effects per compound was higher in natural products than in drugs in the data set, which is consistent with the nature of the effects of natural products.

Analysis of hiPSCs differentiation toward hepatocyte-like cells upon extended exposition to oncostatin.

The capability to produce and maintain functional human adult hepatocytes remains one of the major challenges for the use of in-vitro models toward liver cell therapy and industrial drug-screening applications. Among the suggested strategies to solve this issue, the use of human-induced pluripotent stem cells (hiPSCs), differentiated toward hepatocyte-like cells (HLCs) is promising. In this work, we propose a 31-day long protocol, that includes a final 14-day long phase of oncostatin treatment, as opposed to a 7-day treatment which led to the formation of a hepatic tissue functional for CYP1A2, CYP2B6, CYP2C8, CYP2D6, and CYP3A4. The production of albumin, as well as bile acid metabolism and transport, were also detected. Transcriptome profile comparisons and liver transcription factors (TFs) motif dynamics revealed increased expression of typical hepatic markers such as HNF1A and of important metabolic markers like PPARA. The performed analysis has allowed for the extraction of potential targets and pathways which would allow enhanced hepatic maturation in-vitro. From this investigation, NRF1 and SP3 appeared as transcription factors of importance. Complex epithelial-mesenchymal transition (EMT) and mesenchymal-epithelial transition (MET) patterns were also observed during the differentiation process. Moreover, whole transcriptome analysis highlighted a response typical of the one observed in liver regeneration and hepatocyte proliferation. While a complete maturation of hepatocytes was yet to be obtained, the results presented in this work provide new insights into the process of liver development and highlight potential targets aimed to improve in-vitro liver regeneration.

Impact of P-Glycoprotein-Mediated Active Efflux on Drug Distribution into Lumbar Cerebrospinal Fluid in Nonhuman Primates.

Estimation of unbound drug concentration in the brain (Cu,brain) is an essential part of central nervous system (CNS) drug development. As a surrogate for Cu,brain in humans and nonhuman primates, drug concentration in cerebrospinal fluid (CCSF) collected by lumbar puncture is often used; however, the predictability of Cu,brain by lumbar CCSF is unclear, particularly for substrates of the active efflux transporter P-glycoprotein (P-gp). Here, we measured lumbar CCSF in cynomolgus monkey after single intravenous administration of 10 test compounds with varying P-gp transport activities. The in vivo lumbar cerebrospinal fluid (CSF)-to-plasma unbound drug concentration ratios (Kp,uu,lumbar CSF) of nonsubstrates or weak substrates of P-gp were in the range 0.885-1.34, whereas those of good substrates of P-gp were in the range 0.195-0.458 and were strongly negatively correlated with in vitro P-gp transport activity. Moreover, concomitant treatment with a P-gp inhibitor, zosuquidar, increased the Kp,uu,lumbar CSF values of the good P-gp substrates, indicating that P-gp-mediated active efflux contributed to the low Kp,uu,lumbar CSF values of these compounds. Compared with the drug concentrations in the cisternal CSF and interstitial fluid (ISF) that we previously determined in cynomolgus monkeys, the lumbar CCSF were more than triple for two and all of the good P-gp substrates examined, respectively. Although lumbar CCSF may overestimate cisternal CSF and ISF concentrations of good P-gp substrates, lumbar CCSF allowed discrimination of good P-gp substrates from the weak and nonsubstrates and can be used to estimate the impact of P-gp-mediated active efflux on drug CNS penetration. SIGNIFICANCE STATEMENT: This is the first study to systematically evaluate the penetration of various P-glycoprotein (P-gp) substrates into lumbar cerebrospinal fluid (CSF) in nonhuman primates. Lumbar CSF may contain >3-fold higher concentrations of good P-gp substrates than interstitial fluid (ISF) and cisternal CSF but was able to discriminate the good substrates from the weak or nonsubstrates. Because lumbar CSF is more accessible than ISF and cisternal CSF in nonhuman primates, these findings will help increase our understanding of drug central nervous system penetration at the nonclinical stage.