Defining new criteria for selection of cell-based intestinal models using publicly available databases.

BACKGROUND: The criteria for choosing relevant cell lines among a vast panel of available intestinal-derived lines exhibiting a wide range of functional properties are still ill-defined. The objective of this study was, therefore, to establish objective criteria for choosing relevant cell lines to assess their appropriateness as tumor models as well as for drug absorption studies. RESULTS: We made use of publicly available expression signatures and cell based functional assays to delineate differences between various intestinal colon carcinoma cell lines and normal intestinal epithelium. We have compared a panel of intestinal cell lines with patient-derived normal and tumor epithelium and classified them according to traits relating to oncogenic pathway activity, epithelial-mesenchymal transition (EMT) and stemness, migratory properties, proliferative activity, transporter expression profiles and chemosensitivity. For example, SW480 represent an EMT-high, migratory phenotype and scored highest in terms of signatures associated to worse overall survival and higher risk of recurrence based on patient derived databases. On the other hand, differentiated HT29 and T84 cells showed gene expression patterns closest to tumor bulk derived cells. Regarding drug absorption, we confirmed that differentiated Caco-2 cells are the model of choice for active uptake studies in the small intestine. Regarding chemosensitivity we were unable to confirm a recently proposed association of chemo-resistance with EMT traits. However, a novel signature was identified through mining of NCI60 GI50 values that allowed to rank the panel of intestinal cell lines according to their drug responsiveness to commonly used chemotherapeutics. CONCLUSIONS: This study presents a straightforward strategy to exploit publicly available gene expression data to guide the choice of cell-based models. While this approach does not overcome the major limitations of such models, introducing a rank order of selected features may allow selecting model cell lines that are more adapted and pertinent to the addressed biological question.

Covalent homodimers of murine secretory component induced by epitope substitution unravel the capacity of the polymeric Ig receptor to dimerize noncovalently in the absence of IgA ligand.

Recombinant secretory immunoglobulin A containing a bacterial epitope in domain I of the secretory component (SC) moiety can serve as a mucosal delivery vehicle triggering both mucosal and systemic responses (Corthésy, B., Kaufmann, M., Phalipon, A., Peitsch, M., Neutra, M. R., and Kraehenbuhl, J.-P. (1996) J. Biol. Chem. 271, 33670-33677). To load recombinant secretory IgA with multiple B and T epitopes and extend its biological functions, we selected, based on molecular modeling, five surface-exposed sites in domains II and III of murine SC. Loops predicted to be exposed at the surface of SC domains were replaced with the DYKDDDDK octapeptide (FLAG). Another two mutants were obtained with the FLAG inserted in between domains II and III or at the carboxyl terminus of SC. As shown by mass spectrometry, internal substitution of the FLAG into four of the mutants induced the formation of disulfide-linked homodimers. Three of the dimers and two of the monomers from SC mutants could be affinity-purified using an antibody to the FLAG, mapping them as candidates for insertion. FLAG-induced dimerization also occurred with the polymeric immunoglobulin receptor (pIgR) and might reflect the so-far nondemonstrated capacity of the receptor to oligomerize. By co-expressing in COS-7 cells and epithelial Caco-2 cells two pIgR constructs tagged at the carboxyl terminus with hexahistidine or FLAG, we provide the strongest evidence reported to date that the pIgR dimerizes noncovalently in the plasma membrane in the absence of polymeric IgA ligand. The implication of this finding is discussed in terms of IgA transport and specific antibody response at mucosal surfaces.

Evaluation of uptake and transport of cationic and anionic ultrasmall iron oxide nanoparticles by human colon cells.

Nanoparticles (NPs) are in clinical use or under development for therapeutic imaging and drug delivery. However, relatively little information exists concerning the uptake and transport of NPs across human colon cell layers, or their potential to invade three-dimensional models of human colon cells that better mimic the tissue structures of normal and tumoral colon. In order to gain such information, the interactions of biocompatible ultrasmall superparamagnetic iron oxide nanoparticles (USPIO NPs) (iron oxide core 9-10 nm) coated with either cationic polyvinylamine (aminoPVA) or anionic oleic acid with human HT-29 and Caco-2 colon cells was determined. The uptake of the cationic USPIO NPs was much higher than the uptake of the anionic USPIO NPs. The intracellular localization of aminoPVA USPIO NPs was confirmed in HT-29 cells by transmission electron microscopy that detected the iron oxide core. AminoPVA USPIO NPs invaded three-dimensional spheroids of both HT-29 and Caco-2 cells, whereas oleic acid-coated USPIO NPs could only invade Caco-2 spheroids. Neither cationic aminoPVA USPIO NPs nor anionic oleic acid-coated USPIO NPs were transported at detectable levels across the tight CacoReady? intestinal barrier model or the more permeable mucus-secreting CacoGoblet? model.

Toll-interacting protein modulates colitis susceptibility in mice.

BACKGROUND: The intestinal epithelium accommodates with a myriad of commensals to maintain immunological homeostasis, but the underlying mechanisms regulating epithelial responsiveness to flora-derived signals remain poorly understood. Herein, we sought to determine the role of the Toll/interleukin (IL)-1 receptor regulator Toll-interacting protein (Tollip) in intestinal homeostasis. METHODS: Colitis susceptibility was determined after oral dextran sulfate sodium (DSS) administration or by breeding Tollip on an IL-10 background. The intestinal flora was depleted with 4 antibiotics before DSS exposure to assess its contribution in colitis onset. Bone marrow chimeras were generated to identify the cellular compartment, whereby Tollip may negatively regulate intestinal inflammation in response to DSS. Tollip-dependent epithelial barrier functions were studied in vitro by using Tollip-knockdown in Caco-2 cells and in vivo by immunohistochemistry and fluorescein isothiocyanate-labeled dextran gavage. RESULTS: Genetic ablation of Tollip did not lead to spontaneous intestinal inflammatory disorders. However, Tollip deficiency aggravated spontaneous disease onset in IL-10 mice and increased susceptibility to DSS colitis. Increased colitis severity in Tollip-deficient mice was not improved by bacterial flora depletion using broad-spectrum antibiotics. In addition, DSS exposure of bone marrow chimeric mice revealed a protective role for Tollip in nonhematopoietic cells. Knockdown of Tollip in epithelial cells led to exaggerated NFκ-B activity and proinflammatory cytokine secretion. Finally, DSS-treated Tollip mice showed enhanced intestinal permeability and increased epithelial apoptosis when compared with wild-type controls, a finding that coincided with tight junction alterations on injury. CONCLUSION: Overall, our data show an essential role for Tollip on colitis susceptibility in mice.

L-tyrosine and nitric oxide synergize to prevent cytotoxic effects of superoxide.

We found previously that the nitric oxide donor DEA/NO enhanced lipid peroxidation, DNA fragmentation, and cytotoxicity in human bronchial epithelial cells (BEAS-2B) when they were cultured in LHC-8 medium containing the superoxide-generating system hypoxanthine/xanthine oxidase (HX/XO). We have now discovered that DEA/NO's prooxidant action can be reversed by raising the L-tyrosine concentration from 30 to 400 microM. DEA/NO also protected the cells when they were cultured in Dulbecco's Modified Eagle's Medium (DMEM), whose standard concentration of L-tyrosine is 400 microM. Similar trends were seen with the colon adenoma cell line CaCo-2. Since HPLC analysis of cell-free DMEM or LHC-8 containing 400 microM L-tyrosine, DEA/NO, and HX/XO revealed no evidence of L-tyrosine nitration, our data suggest the existence of an as-yet uncharacterized mechanism by which L-tyrosine can influence the biochemical and toxicological effects of reactive nitrogen species.

Structural and functional properties of two human FXYD3 (Mat-8) isoforms.

Six of 7 FXYD proteins have been shown to be tissue-specific modulators of Na,K-ATPase. In this study, we have identified two splice variants of human FXYD3, or Mat-8, in CaCo-2 cells. Short human FXYD3 has 72% sequence identity with mouse FXYD3, whereas long human FXYD3 is identical to short human FXYD3 but has a 26-amino acid insertion after the transmembrane domain. Short and long human FXYD3 RNAs and proteins are differentially expressed during differentiation of CaCo-2 cells. Long human FXYD3 is mainly expressed in nondifferentiated cells and short human FXYD3 in differentiated cells and both FXYD3 variants can be co-immunoprecipitated with a Na,K-ATPase antibody. In contrast to mouse FXYD3, which has two transmembrane domains for lack of cleavage of the signal peptide, human FXYD3 has a cleavable signal peptide and adopts a type I topology. After co-expression in Xenopus oocytes, both human FXYD3 variants associate stably only with Na,K-ATPase isozymes but not with H,K-ATPase or Ca-ATPase. Similar to mouse FXYD3, short human FXYD3 decreases the apparent K(+) and Na(+) affinity of Na,K-ATPase over a large range of membrane potentials. On the other hand, long human FXYD3 decreases the apparent K(+) affinity only at slightly negative and positive membrane potentials and increases the apparent Na(+) affinity of Na,K-ATPase. Finally, both short and long human FXYD3 induce a hyperpolarization activated current, similar to that induced by mouse FXYD3. Thus, we have characterized two human FXYD3 isoforms that are differentially expressed in differentiated and non-differentiated cells and show different functional properties.

Cyclosporine A kinetics in brain cell cultures and its potential of crossing the blood-brain barrier.

There is an increasing need to develop improved systems for predicting the safety of xenobiotics. However, to move beyond hazard identification the available concentration of the test compounds needs to be incorporated. In this study cyclosporine A (CsA) was used as a model compound to assess the kinetic profiles in two rodent brain cell cultures after single and repeated exposures. CsA induced-cyclophilin B (Cyp-B) secretion was also determined as CsA-specific pharmacodynamic endpoint. Since CsA is a potent p-glycoprotein substrate, the ability of this compound to cross the blood-brain barrier (BBB) was also investigated using an in vitro bovine model with repeated exposures up to 14days. Finally, CsA uptake mechanisms were studied using a parallel artificial membrane assay (PAMPA) in combination with a Caco-2 model. Kinetic results indicate a low intracellular CsA uptake, with no marked bioaccumulation or biotransformation. In addition, only low CsA amounts crossed the BBB. PAMPA and Caco-2 experiments revealed that CsA is mostly trapped to lipophilic compartments and exits the cell apically via active transport. Thus, although CsA is unlikely to enter the brain at cytotoxic concentrations, it may cause alterations in electrical activity and is likely to increase the CNS concentration of other compounds by occupying the BBBs extrusion capacity. Such an integrated testing system, incorporating BBB, brain culture models and kinetics could be applied for assessing neurotoxicity potential of compounds.

Junctional adhesion molecule-2 (JAM-2) promotes lymphocyte transendothelial migration.

The molecular mechanisms underlying lymphocyte extravasation remain poorly characterized. We have recently identified junctional adhesion molecule-2 (JAM-2), and have shown that antibodies to JAM-2 stain high endothelial venules (HEVs) within lymph nodes and Peyer patches of adult mice. Here we show that mouse lymphocytes migrate in greater numbers across monolayers of endothelioma cells transfected with JAM-2. The significance of these findings to an understanding of both normal and pathologic lymphocyte extravasation prompted us to clone the human homologue of JAM-2. We herein demonstrate that an anti-JAM-2 antibody, or a soluble JAM-2 molecule, blocks the transmigration of primary human peripheral blood leukocytes across human umbilical vein endothelial cells expressing endogenous JAM-2. Furthermore, we show that JAM-2 is expressed on HEVs in human tonsil and on a subset of human leukocytes, suggesting that JAM-2 plays a central role in the regulation of transendothelial migration.