Logo: to the web site of Uppsala University

Aim: Proteolysis targeting chimeras (PROTACs) exhibit a unique and promising pharmacology. However, this comes with molecular properties exceeding the 'drug-like' rule of five chemical space, which often limits gastrointestinal absorption. This in vivo study aimed to investigate the contribution of luminal and plasma stability, intestinal effective permeability, P-glycoprotein (P-gp) efflux, and bile excretion, on the rat intestinal absorption and systemic exposure of two PROTACs, ARV-110 (812 Da, LogD_7.4 4.8) and ARV-471 (724 Da, LogD_7.4 4.6).

Methods: Luminal stability and effective intestinal permeability were determined directly from luminal disappearance using single-pass intestinal perfusion, with and without a protease inhibitor, or a P-gp/Cytochrome P450 CYP3A inhibitor (ketoconazole) in rats. Plasma stability was tested by in vitro incubations. Intestinal absorption, systemic exposure, and biliary excretion were examined after intraduodenal and intravenous dosing with ketoconazole or the P-gp selective inhibitor (encequidar).

Results and discussion: Both PROTACs were degraded in the intestinal lumen and in plasma by peptidases. The intestinal effective permeability in rats was moderate for ARV-110 (0.62 x 10^-4 cm/s) and low for ARV-471 (0.23 x 10^-4 cm/s). P-gp inhibition increased the permeability 1.6-and 2.3-fold for ARV-110 and ARV-471, respectively. After intraduodenal dosing with the P-gp inhibitors a corresponding increase in systemic exposure was observed for both PROTACs. There was only a minor difference in the increased systemic exposure induced by the two inhibitors, suggesting that the mechanisms were primarily P-gp inhibition, rather than gut-wall and hepatic extraction. Biliary excretion was a minor pathway and did not affect the absorption and systemic exposure of the PROTACs to a large extent.

Conclusion: In the rat, ARV-110 and ARV-471 were enzymatically degraded in the intestinal lumen and in plasma, and their intestinal permeability and systemic exposure seem to be reduced due to P-gp efflux.

The use of herbal alternatives in modern agriculture and healthcare offers a sustainable approach to animal and human health, but raises concerns about safety, particularly regarding phytochemical interactions at membrane transport proteins. Phytochemicals – bioactive compounds found in herbs - can influence the function of the breast cancer resistance protein (BCRP/ABCG2). BCRP is a key efflux transporter in mammals and in particularly abundant in the blood-milk-barrier of lactating animals, with the potential to affect milk quality and safety. This study investigates species-specific interactions of eight selected phytochemicals with BCRP orthologs in humans, cows, sheep and goats, using a transport assay with membrane vesicles derived from species-specific BCRP-overexpressing HEK293 cells. Five phytochemicals, namely apigenin, berberine, kaempferol, N-isobutyldodeca-2E4E8Z10E/Z-tetraenamide and quercetin inhibited BCRP-mediated transport of its prototypic substrate [³H]estrone sulfate by more than 30 % in all species. IC₅₀ values and, assuming competitive inhibition, Ki values were calculated for these compounds. Apigenin and kaempferol were the most potent inhibitors with Ki values <0.1 μM in all species, while N-isobutyldodeca-2E4E8Z10E/Z-tetraenamide was the weakest inhibitor (Ki > 30 μM). No significant interspecies differences in inhibitory potency were observed. Understanding the cross-species effects of bioactive compounds on BCRP activity is essential for predicting their broader biological impacts. Similar inhibition trends across species facilitate interspecies extrapolation and minimize the amount of animal studies needed to further investigate the effect of these phytochemicals on milk composition.

The pharmacokinetics and plasma protein binding of amoxicillin in cats has not been thoroughly investigated. In a single-group sequential designed experimental study, amoxicillin was administered to six healthy cats intravenously, orally, and subcutaneously. Repeated blood samples were drawn after each administration, and amoxicillin concentrations were determined using High Performance Liquid Chromatography coupled to Triple Quadrupole Mass Spectrometry. Plasma amoxicillin data were subjected to population pharmacokinetic analysis, and pharmacokinetic parameters were estimated. The population clearance was 0.18 L/h·kg, the volume of the central compartment was 0.12 L/kg, the highly perfused compartment was 0.009 L/kg, and the poorly perfused compartment was 0.002 L/kg. The bioavailability was 33% and 69% after oral and subcutaneous administration, respectively. After subcutaneous administration of a slow-release formulation, there was absorption rate-limited pharmacokinetics. The plasma protein binding was 0%-24%. The results increase the understanding of the amoxicillin pharmacokinetics in cats. Further studies combining the results with pharmacodynamic data and in silico simulations are warranted.

A variety of enabling formulations has been developed to address poor oral drug absorption caused by insufficient dissolution in the gastrointestinal tract. As the in vivo performance of these formulations is a result of a complex interplay between dissolution, digestion and permeation, development of suitable in vitro assays that captures these phenomena are called for. The enabling-absorption (ENA) device, consisting of a donor and receiver chamber separated by a semipermeable membrane, has successfully been used to study the performance of lipid-based formulations. In this work, the ENA device was prepared with two different setups (a Caco-2 cell monolayer and an artificial lipid membrane) to study the performance of a lipid-based formulation (LBF), an amorphous solid dispersion (ASD) and the potential benefit of combining the two formulation strategies. An in vivo pharmacokinetic study in rats was performed to evaluate the in vitro-in vivo correlation. In the ENA, high drug concentrations in the donor chamber did not translate to a high mass transfer, which was particularly evident for the ASD as compared to the LBF. The solubility of the polymer used in the ASD was strongly affected by pH-shifts in vitro, and the ph_dependence resulted in poor in vivo performance of the formulation. The dissolution was however increased in vitro when the ASD was combined with a blank lipid-based formulation. This beneficial effect was also observed in vivo, where the drug exposure of the ASD increased significantly when the ASD was co-administered with the blank LBF. To conclude, the in vitro model managed to capture solubility limitations and strategies to overcome these for one of the formulations studied. The correlation between the in vivo exposure of the drug exposure and AUC in the ENA was good for the non pH-sensitive formulations. The deconvoluted pharmacokinetic data indicated that the receiver chamber was a better predictor for the in vivo performance of the drug, however both chambers provided valuable insights to the observed outcome in vivo. This shows that the advanced in vitro setting used herein successfully could explain absorption differences of highly complex formulations.

Colonic mucus is a key factor in the colonic environment because it may affect drug absorption. Due to the similarity of human and canine gastrointestinal physiology, dogs are an established preclinical species for the assessment of controlled release formulations. Here we report the development of an artificial colonic mucus model to mimic the native canine one. In vitro models of the canine colonic environment can provide insights for early stages of drug development and contribute to the implementation of the 3Rs (refinement, reduction, and replacement) of animal usage in the drug development process. Our artificial colonic mucus could predict diffusion trends observed in native mucus and was successfully implemented in microscopic and macroscopic assays to study macromolecular permeation through the mucus. The traditional Transwell set up was optimized with the addition of a nylon filter to ensure homogenous representation of the mucus barrier in vitro. In conclusion, the canine artificial colonic mucus can be used to study drug permeation across the mucus and its flexibility allows its use in various set ups depending on the nature of the compound under investigation and equipment availability.

Orthologs of breast cancer resistance protein (BCRP/ABCG2), an ATP-binding cassette (ABC) efflux transmembrane transporter, are present in several species. The list of compounds known to interact with BCRP is growing, and many questions remain concerning species-specific variations in substrate specificity and affinity and the potency of inhibitors. As the most abundant efflux transporter known to be present in the blood–milk barrier, BCRP can increase the elimination of certain xenobiotics to milk, posing a risk for suckling offspring and dairy product consumers. Here we developed a model that can be employed to investigate species-specific differences between BCRP substrates and inhibitors. Membrane vesicles were isolated from transiently transduced human embryonic kidney (HEK) 293 cells, overexpressing BCRP, with human, bovine, caprine, and ovine cDNA sequences. To confirm BCRP transport activity in the transduced cells, D-luciferin efflux was measured and to confirm transport activity in the membrane vesicles, [³H] estrone-3-sulfate ([³H]E₁S) influx was measured. We also determined the Michaelis–Menten constant (Km) and Vmax of [³H]E₁S for each species. We have developed an in vitro transport model to study differences in compound interactions with BCRP orthologs from milk-producing animal species and humans. BCRP transport activity was demonstrated in the species-specific transduced cells by a reduced accumulation of D-luciferin compared with the control cells, indicating BCRP-mediated efflux of D-luciferin. Functionality of the membrane vesicle model was demonstrated by confirming ATP-dependent transport and by quantifying the kinetic parameters, Km and Vmax for the model substrate [³H]E₁S. The values were not significantly different between species for the model substrates tested. This model can be insightful for appropriate inter-species extrapolations and risk assessments of xenobiotics in lactating woman and dairy animals.

Lactational elimination has been described mathematically for nearly 50 years. Over 40 published articles, containing >50 physiologically based kinetic (PBK) lactation models were included in the systematic review. These PBK models described the lactational elimination of xenobiotic compounds in humans, rats, mice, and dairy cows and goats. A total of 78 compounds have been modelled, ranging from industrial chemicals, pesticides, to pain medication, antibiotics, and caffeine. Few models included several species or compounds, and models were thus generally not translational or generic. Three dairy cow models mechanistically described the intramammary disposition of pharmaceuticals after intramammary administration, including volume changes caused by milking, while empirically describing the remaining pharmacokinetics. The remaining models were semi- or whole body PBK models, describing long-term exposure of environmental pollutants, or short-term exposure of pharmaceuticals. The absolute majority described the disposition to the mammary gland or milk with perfusion limited compartments, but permeability limited models were available as well. With long-term exposure, models often included changes in milk volume and/or consumption by the offspring, and changes in body weight of offspring. Periodic emptying of the mammary gland, as with feeding or milking, was sparsely applied. Rodent models used similar physiological parameters, while values of physiological parameters applied in human models could range widely. When milk composition was included in the models, it most often included the fat content. The review gives an extensive overview of the applied functions and modelling strategies of PBK lactation models.

Gastrointestinal (GI) mucus is continuously secreted and lines the entire length of the GI tract. Essential for health, it keeps the noxious luminal content away from the epithelium. Our aim was to characterize the composition and structure of mucus throughout the various GI segments in dog.

Mucus was collected from the stomach, small intestine (duodenum, jejunum, ileum), and large intestine (cecum, proximal and distal colon) from dogs. Composition was determined by multi-omics. Structural properties were investigated using cryoSEM and rheology.

GI mucus contained 74-95% water and maintained a pH around 6.5. The proteome was similar across the different GI segments. The highest abundant secreted gel-forming mucin in the gastric mucus was mucin 5AC, whether mucin 2 had highest abundance in the intestinal mucus. Lipid and metabolite abundance was generally higher in the jejunal mucus than the colonic mucus. CryoSEM microscopy revealed smaller pore size in small intestinal mucus, which increased in the large intestine. All mucus samples showed shear-thinning behavior and characteristics of gel-like structure.

In conclusion, the mucus is a highly viscous and hydrated material. These data provide an important baseline for future studies on human and canine intestinal diseases and the dog model in drug absorption.

There is a growing body of research about subcutaneously administered biologics, emphasizing the need for optimized bioavailability predictions. It is important to inform both translational and in silico models with properties of the drug products and compounds. However, the pharmaceutical, therapeutic and physicochemical properties of market authorized drug products for subcutaneous administration are currently not collated in the public domain. We provide an overview of subcutaneous administered drug products for humans and animals market authorized in EU, Canada, and the US. Data on the drug products were collected from the respective authorities, i.e. European Medicines Agency, Health Canada, and U.S. Food and Drug Administration. Physicochemical properties of active substances were gathered from DrugBank. Human drug products were often indicated for treatment of diabetes and anemia. EU veterinary drug products were often immunologicals. Canadian and US veterinary drug products often acted as antiinfectives for systemic use, on the genito-urinary system or as sex hormones. The final dataset with >1700 subcutaneous drug products is provided. In EU drug products, the majority of active substances were biologics. In the US, drug products most often contained small molecules. Solutions, emulsions and suspensions were the most common dosage forms. A minority of subcutaneous drug products were also registered for intramuscular or intravenous administration. The analysis presented here could aid further research, exploring formulation properties, prescription or sales of market authorized SC drug products and development of inclusive in silico models.

Subcutaneous injection is a commonly used route of drug administration for both small molecules and biologics. To facilitate the development of new subcutaneously administered drugs, methods for prediction of drug absorption from the injection site are essential. For this purpose, in silico models have increasingly been used. This report summarize the current state of in silico models for description and prediction of subcutaneous drug absorption. Original articles on physiologically based models describing subcutaneous administration published from 2010 and onward were reviewed. Eighteen physiologically based models were identified: eleven for small molecules and seven for biologics. Most models described the PK of one drug and for one species. In models for small molecules, the subcutaneous administration site was most often described as a depot compartment with first-order absorption into the plasma or blood. Most models for biologics divided administration and organ compartments into vascular and interstitial subcompartments. Mass transfer to these compartments was frequently described with convection and diffusion, according to the one- or two-pore theory. Tremendous improvement in the quantitative aspects of subcutaneous administration and subsequent absorption of physiologically based models has occurred the last decade. However, improvements related to data translation and generalization of these models were identified.