Andrew Collett

Comparison of HT29-18-C1 and Caco-2 cell lines as models for studying intestinal paracellular drug absorption

AbstractPurpose. To compare the permeability characteristics of HT29-18-C1 colonic epithelial cell line with Caco-2, an established model of intestinal drug transport. Methods. Cell lines were grown as epithelial monolayers. Permeability was measured over a range of transepithelial electrical resistance (Rt) using a group of drug compounds. Results. HT29-18-C1 develop Rt slowly when grown in culture, allowing permeability to be measured over a wide range (80–600 Ω·cm2). In contrast, Caco-2 monolayers rapidly develop Rt of ≈300 Ω·cm2 and require Ca2+-chelation to generate Rt equivalent to human intestine (60–120 Ω·cm2). Permeability of atenolol, ranitidine, cimetidine, hydrochlorothiazide and mannitol across HT29-18-C1 decreased 4–5 fold as Rt developed from 100–300 Ω·cm2 indicating they permeate via the paracellular route. In contrast, ondansetron showed no difference in permeability with changing Rt consistent with transcellular permeation. Permeability profiles across low Rt HT29-18C1 and pulse EGTA-treated Caco-2 monolayers were the same for all 5 paracellular drugs suggesting that transient Ca2+ removal does not alter selectivity of the tight junctions. Permeabilities of cimetidine, hydrochlorothiazide and atenolol across 100 Ω·cm2 HT29-18-C1 monolayers reflect more closely those reported for the human ileum in vivo than did mature Caco-2 monolayers. Conclusions. HT29-18-C1 monolayers can be used to study drug permeability at Rt values similar to human intestine without the need for Ca2+ chelation. As such, they offer a useful alternative to Caco-2 for modelling intestinal drug absorption.

Predicting P-glycoprotein effects on oral absorption: Correlation of transport in Caco-2 with drug pharmacokinetics in wild-type and mdr1a(-/-) mice in vivo

AbstractPurpose. Cell-based permeability screens are widely used to identify drug-P-glycoprotein (PGP) interaction in vitro. However, their reliability in predicting the impact of PGP on human drug pharmacokinetics is poorly defined. The aim was to determine whether a quantitative relationship exists between PGP-mediated alterations in Caco-2 permeability and oral pharmacokinetics in mice. Methods. Two indicators of drug efflux were measured in Caco-2 for a group of 10 compounds, the ratio of A-B and B-A transport (RB-A/A-B) and the ratio of A-B transport in the presence and absence of a PGP inhibitor, GF120918 (RGF). These data were correlated with ratios of oral plasma levels in either mdr1a(-/-) or mdr1a/1b(-/-) and wild-type mice (RKO/WTinvivo) calculated from literature data on these compounds. Results. A significant, positive correlation (r2= 0.8, p \h 0.01) was observed between RGF and RKO/WTinvivo. In contrast, RB-A/A-B, a more commonly used in vitro measure, showed a much weaker correlation with in vivo data (r2= 0.33, p = 0.11). A strong correlation with RGF was also observed after correction of in vivo data for PGP effects on IV clearance. Conclusion. The increase in A-B drug permeability following inhibition of PGP in Caco-2 allows a reasonable prediction of the likely in vivo impact that PGP will have on plasma drug levels after oral administration.

Early molecular and functional changes in colonic epithelium that precede increased gut permeability during colitis development in mdr1a(-/-) mice.

Background: The early molecular changes preceding the onset of mucosal inflammation in colitis and their temporal relationship with gut permeability remain poorly defined. This study investigated functional and transcriptomic changes in mdr1a(−/−) mice lacking the intestinal transporter P‐glycoprotein, which develop colitis spontaneously when exposed to normal enteric flora. Methods: Mdr1a(−/−) mice were housed in specific pathogen‐free conditions to slow colitis development and compared to congenic controls. Mucosal permeability and cytokine secretion were analyzed in ex vivo colon. Gene expression in colonic mucosal and epithelial preparations was analyzed by microarray and qPCR. Colonocyte responsiveness to bacterial antigens was measured in short‐term culture. Results: Colon from 4–5‐week‐old, disease‐free mdr1a(−/−) mice was histologically normal with no evidence of increased permeability compared to controls. However, these tissues display a distinctive pattern of gene expression involving significant changes in a small number of genes. The majority of upregulated genes were associated with bacterial recognition and the ubiquitin‐proteasome system and were gamma‐interferon (IFN‐&ggr;) responsive. Expression of the antiinflammatory factor pancreatitis‐associated protein (PAP) and the related gene RegIII&ggr; were markedly reduced. Colonocytes from 4–5‐week mdr1a(−/−) exhibit similar transcriptomic changes, accompanied by higher basal chemokine secretion and increased responsiveness to LPS. Significant increases in colonic permeability were associated with older (12–16‐week) mdr1a(−/−) mice displaying molecular and functional evidence of active inflammation. Conclusions: These studies show that early epithelial changes associated with altered responsiveness to bacteria precede increased permeability and mucosal inflammation in this model of colitis, highlighting the importance of P‐glycoprotein in regulating interactions with the commensal microflora.

Influence of morphometric factors on quantitation of paracellular permeability of intestinal epithelia in vitro.

AbstractPurpose. The relative contribution of the small and large intestine to paracellular absorption is a subject of some controversy. Direct comparison of paracellular permeability in different epithelia is complicated by variations in junctional density and/or the absorptive surface area. Methods. This study used a combination of morphometric analyses and in vitro absorption studies to define permeability characteristics in relation to the amount of paracellular pathway present in rat ileum, colon and the model epithelium, Caco-2. Results. Mucosal to serosal amplification was higher in ileum (3.9) than colon (1.9) or Caco-2 (1). Tight junctional density (1p) of ileal crypts was ≈3 fold greater (91 m/cm2) than that measured in ileal villi, colonic surface and crypt cells or Caco-2 monolayers (34−37 m/ cm2). However, when the relative contributions of the crypts and villi was taken into account there was no significant difference in the mean 1p per mucosal area for the three epithelia studied. Using these data to correct for morphometric differences the permeabilities of a range of small hydrophilic molecules (atenolol, D-PheAsp and PEG oligomers MW 282-634) was measured. Permeability of rat ileum and colon were virtually identical for all compounds studied. In contrast, Caco-2 monolayers showed a significantly lower permeability than intestinal tissues with the difference increasing markedly with molecular size. Conclusions. These studies suggest the importance of accounting for morphological variation when comparing the permeability characteristics of different epithelial systems.

Oral absorption of D-oligopeptides in rats via the paracellular route.

AbstractPurpose. This study was undertaken to examine the structural determinants of oral bioavailability in the rat of a set of oligopeptides comprising D-amino acids, which were taken to be absorbed paracellularly based on a pronounced sensitivity of permeability to electrical resistance in Caco-2 cell monolayers. Methods. The study series comprised eleven D-oligopeptides, designed not to be recognised by peptidases or transport proteins, and to have molecular weights between 222 and 406 daltons with different net electrical charges and composition of D-amino acids. All the peptides were [3H]-radiolabelled and analyzed by HPLC with radiometric detection. Bioavailability was estimated based on 24-hr urinary excretion of unchanged peptide after oral and intravenous administrations. Results. As expected, the series proved metabolically stable. Bioavailability was independent of oral dose when varied by a factor of 10,000, suggesting passive absorption. Whereas bioavailability decreased sharply from 30% to 1% with increasing molecular weight, net charge showed little, if any, effect on bioavailabilty. Conclusions. This D-oligopeptide model series served as a useful probe for the structural requirements for paracellular absorption in vivo. A critical determinant of bioavailability is molecular size, expressed as molecular weight in this study; net charged appeared of much lesser importance.

The structure and immunomodulatory activity on intestinal epithelial cells of the EPSs isolated from Lactobacillus helveticus sp. Rosyjski and Lactobacillus acidophilus sp. 5e2.

The Lactic acid bacteria (LAB) Lactobacillus acidophilus sp. 5e2 and Lactobacillus helveticus sp. Rosyjski both secrete exopolysaccharides (EPSs) into their surrounding environments during growth. A number of EPSs have previously been shown to exhibit immunomodulatory activity with professional immune cells, such as macrophages, but only limited studies have been reported of their interaction with intestinal epithelial cells. An investigation of the immunomodulatory potential of pure EPSs, isolated from cultures of Lactobacillus acidophilus sp. 5e2 and Lactobacillus helveticus sp. Rosyjski, with the HT29-19A intestinal epithelial cell line are reported here. For the first time the structure of the EPS from Lactobacillus helveticus sp. Rosyjski which is a hetropolysaccharide with a branched pentasaccharide repeat unit containing d-glucose, d-galactose and N-acetyl-d-mannosamine is described. In response to exposure to lactobacilli EPSs HT29-19A cells produce significantly increased levels of the proinflammatory cytokine IL-8. Additionally, the EPSs differentially modulate the mRNA expression of Toll-like receptors. Finally, the pre-treatment of HT29-19A cells with the EPSs sensitises the cells to subsequent challenge with bacterial antigens. The results reported here suggest that EPSs could potentially play a role in intestinal homeostasis via a specific interaction with intestinal epithelial cells.

Exploring the immunomodulatory potential of microbial-associated molecular patterns derived from the enteric bacterial microbiota.

The human intestinal lumen represents one of the most densely populated microbial niches in the biological world and, as a result, the intestinal innate immune system exists in a constant state of stimulation. A key component in the innate defence system is the intestinal epithelial layer, which acts not only as a physical barrier, but also as an immune sensor. The expression of pattern recognition receptors, such as Toll-like receptors, in epithelial cells allows innate recognition of a wide range of highly conserved bacterial moieties, termed microbial-associated molecular patterns (MAMPs), from both pathogenic and non-pathogenic bacteria. To date, studies of epithelial immunity have largely concentrated on inflammatory pathogenic antigens; however, this review discusses the major types of MAMPs likely to be produced by the enteric bacterial microbiota and, using data from in vitro studies, animal model systems and clinical observations, speculates on their immunomodulatory potential.

Use of in vitro human keratinocyte models to study the effect of cooling on chemotherapy drug-induced cytotoxicity

A highly distressing side-effect of cancer chemotherapy is chemotherapy-induced alopecia (CIA). Scalp cooling remains the only treatment for CIA, yet there is no experimental evidence to support the cytoprotective capacity of cooling. We have established a series of in vitro models for the culture of human keratinocytes under conditions where they adopt a basal, highly-proliferative phenotype thus resembling the rapidly-dividing sub-population of native hair-matrix keratinocytes. Using a panel of chemotherapy drugs routinely used clinically (docetaxel, doxorubicin and the active metabolite of cyclophosphamide 4-OH-CP), we demonstrate that although these drugs are highly-cytotoxic, cooling can markedly reduce or completely inhibit drug cytotoxicity, in agreement with clinical observations. By contrast, we show that cytotoxicity caused by specific combinatorial drug treatments cannot be adequately attenuated by cooling, supporting data showing that such treatments do not always respond well to cooling clinically. Importantly, we provide evidence that the choice of temperature may be critical in determining the efficacy of cooling in rescuing cells from drug-mediated toxicity. Therefore, despite their reductive nature, these in vitro models have provided experimental evidence for the clinically-reported cytoprotective role of cooling and represent useful tools for future studies on the molecular mechanisms of cooling-mediated cytoprotection.

Investigation of Regional Mechanisms Responsible for Poor Oral Absorption in Humans of a Modified Release Preparation of the -Adrenoreceptor Antagonist, 4-Amino-6,7-dimethoxy-2-(5-methanesulfonamido-1,2,3,4 tetrahydroisoquinol-2-yl)-5-(2-pyridyl)quinazoline (UK-338,003): The Rational Use of ex Vivo Intestine to Predict in Vivo Absorption

Modified release (MR) formulations are used to enhance the safety and compliance of existing drugs by improving their pharmacokinetics. Predicting the likely success of MR formulations is often difficult before clinical studies. A systematic in vitro approach using mouse and human tissues was adopted to rationalize the in vivo pharmacokinetics of 9- and 15-h MR formulations of an α-adrenoreceptor antagonist, 4-amino-6,7-dimethoxy-2-(5-methanesulfonamido-1,2,3,4 tetrahydroisoquinol-2-yl)-5-(2-pyridyl)quinazoline (UK-338,003). Immediate release UK-338,003 was well absorbed in humans consistent with moderate Caco-2 cell monolayer permeability. In contrast, 9- and 15-h modified release formulations showed marked reductions in Cmax (47.1 and 68.9%) and AUC0–72 (32.6 and 54.0%). Colonic intubation resulted in 81.3 and 73.8% reductions in Cmax and AUC0–72. Mechanistic studies in isolated mouse tissues showed that colonic UK-338,003 permeability (Papp < 0.5 × 10-6 cm/s) was at least 40 times lower than that for ileum with marked asymmetry. UK-338,003 was found to be a substrate for P-glycoprotein (PGP) with a weaker interaction for multidrug resistance-associated protein-type transporters in mouse intestine. PGP inhibition dramatically increased colonic UK-338,003 permeability to the levels observed in ileum. Low UK-338,003 apical to basolateral permeability was also observed in ex vivo human distal intestine, but both the asymmetry and increase in permeability after PGP inhibition were significantly lower. In conclusion, the poor absorption of MR UK-338,003 in humans can be explained by a combination of PGP-dependent efflux and low intrinsic permeability in the lower bowel. Regional permeability studies in ex vivo tissues used during drug development can highlight absorption problems in the distal bowel and assess the feasibility of developing successful MR formulations.

Potentiation of flagellin responses in gut epithelial cells by interferon-γ is associated with STAT-independent regulation of MyD88 expression

Flagellin acting via TLR5 (Toll-like receptor 5) is a key regulator of the host response to the gut microbial flora in both health and disease. The present study has investigated regulation of flagellin-TLR5 signalling in human colonocytes (HT29-19A) by IFNgamma (interferon-gamma), a cytokine released early in the inflammatory process which has multiple effects on gut epithelial function that may facilitate abnormal responses to enteric bacteria. Flagellin induced a dose-dependent secretion of chemokines CXCL8 and CCL2 in the human colonocyte line, HT29-19A. Exposure to IFNgamma did not induce chemokine secretion, but markedly potentiated responses to flagellin, increasing CXL8 gene expression and protein secretion by approx. 4-fold. Potentiation by IFNgamma was independent of changes in TLR5 and was associated with a rapid, sustained increase in expression of the downstream adaptor molecule MyD88 (myeloid differentiation factor 88). Knockdown of MyD88 expression using siRNA (small interfering RNA) abolished flagellin-dependent CXCL8 secretion and the potentiating effect of IFNgamma. Exposure of non-transformed mouse and human colonocytes to IFNgamma also increased MyD88 expression. STAT (signal transducer and activator of transcription) 1 knockdown and use of the broad-spectrum JAK (Janus kinase)-STAT inhibitor AG490 had no effect on IFNgamma-mediated up-regulation of MyD88. The findings of the present study suggest that IFNgamma sensitizes colonic epithelial cells to bacterial flagellin via a largely STAT-independent up-regulation of MyD88 expression leading to increased secretion of immunomodulatory factors. These results indicate that epithelial responses to flagellin are potentiated by IFNgamma, most likely mediated by increased MyD88 expression. The present study adds to our understanding of the spectrum of effects of this cytokine on gut epithelium that may contribute to bacterial-driven inflammation in the gut.