Investigation of the coordinated functional activities of cytochrome P450 3A4 and P-glycoprotein in limiting the absorption of xenobiotics in Caco-2 cells

The coordination of the functional activities of intestinal CYP3A4 and P-gp in limiting the absorption of xenobiotics in Caco-2 cells was investigated. Growing Caco-2 cells were exposed to increasing concentrations of doxorubicin (1-2 μM) in plastic flasks to encourage a subpopulation of cells, that displayed an intrinsically higher multidrug resistance (mdr) phenotype than the parent cells, to survive and grow. Doxorubicin-exposed (hereinafter referred to as type I cells) and nonexposed Caco-2 cells (parent cells) on collagen-coated inserts were also treated with either 0 (control) or 0.25 μM 1α,25-dihydroxyvitamin D3 to promote cellular CYP3A4 expression. Increased P-gp protein expression, as detected by Western blotting, was noted in type I cells (213±54.35%) compared to that of parent cells (100±6.05%). Furthermore, they retained significantly less [3H]vincristine sulphate (p<0.05), a P-gp substrate, after efflux (272.89±11.86 fmol/mg protein) than the parent cells (381.39±61.82 fmol/mg protein). The expression of CYP3A4 in parental cells after 1α,25-dihydroxyvitamin D3 treatment was quantified to be 76.2±7.6 pmol/mg protein and comparable with that found in human jejunal enterocytes (70.0±20.0 pmol/mg protein). Type I cells, however, expressed a very low quantity of CYP3A4 both before and after the treatment that was beyond the minimum detection limit of Western blotting. Functionally, the rates of 1-hydroxylation of midazolam by CYP3A for both cell types ranged from 257.0±20.0 to 1057.0±46.0 pmol/min/mg protein. Type I cells, although having a higher P-gp expression and activity comparatively, metabolized midazolam less extensively than the parent cells. The results suggested that there were noncoordinated functional activities of intestinal CYP3A4 and P-gp in Caco-2 cells, although they both functioned independently to minimize intestinal epithelial absorption of xenobiotics. © 2002 Wiley-Liss, Inc. and the American Pharmaceutical Association.

Nitric oxide in the rat gastrointestinal tract

This study concerns the nature of nitric oxide synthase (NOS) and the role of nitric oxide (NO) in the rat gastrointestinal tract. The major objectives were (i) to characterise NOS isoforms in the gastric glandular mucosa, (ii) to localise NOS isoforms in the rat gastric glandular mucosa, (iii) to investigate the role of NO in carbachol-stimulated gastric mucus secretion, (iv) to investigate the nature of NOS and small intestine. Immunoblotting was performed using polyclonal antisera raised against two peptides found in the rat brain NOS sequence and commercial monoclonal antibodies directed against neuronal and endothelial isoforms of NOS. A160kDa band was detected in brain and gastric mucosal samples with antibodies and antisera directed against neuronal NOS sequences, and a 140kDa band was detected in gastric mucosal samples using an anti-endothelial NOS antibody. An intense 160kDa neuronal NOS band was detected in a high-density fraction of gastric mucosal cells separated on a Percoll gradient. Detection of neuronal NOS by a carboxyl-terminal antiserum in samples of brain, but not of gastric mucosa, could be blocked by the peptide (20g/ml) against which the antibody was raised. After affinity purification, recognition of gastric mucosal NOS was blocked by peptide. Particulate neuronal NOS was found in the brain by immunoblotting while 94% of gastric mucosal enzyme was soluble. Gastric mucosal endothelial NOS was 95% particulate. 95% of NOS activity in the gastric mucosa was due to neuronal NOS. Paraformaldehyde- and acetone-fixed gastric mucosal sections were subject to immunocytochemistry using the above antibodies. Neuronal NOS was localised to the surface mucosal epithelial cells while endothelial NOS was associated with microvessels at the base of the mucosa and to larger vessels in the submucosa. Intragastric administration of carbachol or 16, 16-dimethyl prostaglandin E2 increased the thickness of the rat gastric mucus layer. The NOS inhibitor NG-nitro-L-arginine methyl ester dose-dependently, and selectively, prevented the stimulatory effect of carbachol. Ca2+-independent NOS activity in rat ileal, jejunal and colonic muscle was increased after LPS induction. Ca2+-dependent activity was not affected. Distribution of inducible NOS protein paralleled Ca2+ -independent activity. LPS treatment did not affect the content of neuronal NOS in colonic muscle.

Intestinal absorption barriers as modelled by p-glycoprotein and cytochrome p450 a34 in caco2 cells

The passage number and origin of two populations of Caco-2 cells influence their enterocyte-like characteristics. Caco-2 cells of passage number >90 from Novartis pharmaceutical company possess higher levels of expression of alkaline phosphatase and P-glycoprotein and a greater cellular uptake of Gly-1.-Pro than those of passage number <40 from the American Type Tissue Culture collection. High P-gp expressing Caco-2 cells have been developed through stepwise selection of the cells with doxonibicin. This newly-developed cell line (hereafter referred to as Type I) possesses approximately twice as much P-gp protein than non-exposed cells, restricts the transepithelial transport of vincristine in the apical-to-basolateral direction whilst facilitating its transport in the reverse direction and accumulates less vincristine than non-exposed cells. There is no apparent evidence of the co-existence of the multidrug resistance protein (MIT) in Type I cells to account for the above-listed observations. Stopping the exposure for more than 28 days decreases the P-gp protein expression in previously doxorubicin-exposed Type I Caco-2 cells and reduces the magnitude of vincristine transepithelial fluxes in both directions to the levels that are almost similar to those of non-exposed cells. Exposing Caco-2 cells to 0.25 JAM la, 25-dihydroxyvitamin D3 induces their expression of cytochrome P450 3A4 protein to the level that is equivalent to that from isolated human jejunal cells. Under the same treatment, doxorubiein-exposed (Type I) cells metabolise naidazolam poorly and less extensively compared to non-exposed cells, suggesting that there is no such co-regulation of P-gp and CYP3A4 in Caco-2 cells. However, there is evidence which suggests CYP3A metabolises mida_zolam into 1- and 4-hydroxymidazolam, the latter may possibly be a P-gp substrate and is transported extracellularly by P-gp, supporting the hypothesis of P-gp-CYP3A4 synergistic roles in keeping xenobiotics out of the body. Doxoru.bicin-exposed (Type I) cells are less effective in translocating L-proline and glycyl-L-proline across the cell mono layers.

The targeting of corticosteroids to inflamed tissues

In recent years, much interest has focused on the beneficial effects of administering potentially harmful therapeutic agents in drug carriers so as to reduce their toxic side effects. Rheumatoid arthritis is a chronic systemic disease with progressive destruction of the Joints and long term patient disability, Corticosteroids have been shown to retard the progression of Joint destruction but are limited in their use due to adverse side effects,This project, following the line of investigation started by other workers, was designed to study the use of microspheres to deliver corticosteroids to inflamed tissues by both the oral and intravenous routes. Hydrocortisone (HC)-loaded albumin microspheres were prepared by three different methods, by direct incorporation of HC within the particles, by indirect incorporation of HC by the enzymatic conversion of hydrocortisone-21-phosphate (H-21-P) to HC within the particles, and by the adsorption of HC onto the surface. HC was also loaded with PLA microspheres. The level of corticosteriod loading and in vitro release from microspheres was determined by HPLC analysis. A reversed-phase, ion-pairing HPLC method was developed to simultaneously measure both HC and H-21-P. The highest level of corticosteroid loading was achieved using the incorporation of H-21-P with enzymatic conversion to HC method. However, HPLC analysis showed only 5% of the incorporated steroid was HC. In vitro release rates of steroid from albumin microspheres showed >95% of incorporated steroid was released within 2 hours of dissolution. Increasing the protein:steroid ratio, and the temperature and duration of microsphere stabilization, had little effect on prolonging drug release. In vivo studies, using the carrageenan-induced rat hind-paw model of inflammation, indicated steroid-incorporated microspheres administered both orally and intraperitoneally were not therapeutically advantageous when compared to equivalent free steroid doses. The ability of orally and intravenously dosed [125I]~albumin microspheres (2.67 μm mean diameter) to accumulate in acutely and chronically inflamed tissues was investigated, The subcutaneous air-pouch was the model of inflammation used, with carrageenan as the inflammatory stimulus. Acute and chronic inflammation was shown to be consistently formed  in pouch tissues in terms of cell infiltration and fluid exudate formation in the pouch cavity. Albumin microspheres were shown to accumulate in the inflamed tissues and pouch fluids after both oral and intravenous administration. Preliminary, confirmatory studies using latex microspheres and quantitation by GPC analysis, also indicated microsphere accumulation in both acutely and chronically inflamed air-pouch tissues. tntl lUr"'poucbtis,sues; The results indicate the uptake and transfer of microspheres across the gastrointestinal tract into the circulation and their migration through disrupted endothelium and basement membranes at the inflamed sites. , .

The mode of action of bradykinin

The action of bradykinin on transepithelial transfer of sodium and water in isolated rat jejunum and on smooth muscle contraction of rat terminal ileum has been investigated. (1) Bradykinin was shown to stimulate transfer at low control transfer, inhibit transfer at high control transfer and have no effect at intermediate transfer in rat jejunal sacs. Stimulation of transfer occurred only when bradykinin was in the serosal solutiun while inhibition of transfer occurred whether bradykinin was in the aerosal or mucosal solution. Bradykinin-induced stimulation of transfer was not affected by adrenalectomy, nephrectomy, combined adrenalectomy-nephrectomy,  nor maintenance on 1% saline drinking solution or low sodium diet pretreatment. Meclofenamic acid abolished the bradykinin-induced inhibition of water transfer while prostaglandins A1, E1 aud F2α all potentiated this action. Theophylline inhibited water transfer and potentiated the bradykinin-induced inhibition of water transfer. Cyclic AMP and dibutyryl cyclic AMP both inhibited water transfer and the bradykinin-induced inhibition of water transfer was potentiated by the latter. ( 2 ) Bradykinin-induced contractions of rat terminal ileum were little affected by hyoscine while those of acetylcholine were abolished. Anoxia reduced markedly responses tv bradykinin while those of acetylcholine were little affected . Theophylline reduced the responses of rat terminal ileum to bradykinin significantly more than those to acetylcholine. Aspirin and indomethacin reduced markedly the responses to bradykinin while not affecting those to acetylcholine and PGT2. Meslofenamic acid at a concentration of 3.4 µM blocked bradykinin-induced contractions but had no effect on those to acctylcholine, PGE2 or PGF2 and at a concentration of 17. 0 µM drastically reduced bradykinin responses but also reduced those to acetylcholine, PGE2 and PGF2α• Flufenamic acid drastically reduced responses to bradykinin while not affecting those to acetylcholine and PGE2 and slightly affecting those to PGF2α. Polyphloretin phosphate reduced responses to bradykinin, PGF2α and PGE2 but not acetylcholine . Diphloretin phosphate reduced responses to bradykinin, PGF2 and PGE2 in a dose dependent manner but not those to acetylcholine. SC 19220 , in a dose dependent manner, inhibited responses to bradykinin and PGE2 but not to acetylcholine and PGF2. 7 oxa - 13 -prostynoic acid non specifically reduced responses to acetylcholine, bradykinin and PGE2. Bradykinin, in the presence of SQ 20881 , increased the release of prostaglandin-like activity from rat terminal ileum and this was reduced or abolished in the presence of indomethacin, aspirin, meclofenamic acid or flufenamio acid. The extract of PG-like activity did not appear as PGE, PGA or PGFon TLC, but included a substance with similar mobility as 15-Keto-prosta-glandin E2.

Adrenoceptors and intestinal fluid and electrolyte transport in the rat

Noradrenaline was found to significantly stimulate fluid and Na absorption across everted sacs of rat jejunum. Of a number of a1, and 2-adrenoceptor antagonists tested only prazosin significantly inhibited the stimulant effect of noradrenaline and further experiments revealed an antiabsorptive effect of prazosin alone. Theophylline reduced jejunal fluid and Na absorption and this effect was not reversed by 2-adrenoceptor stimulation in contrast to previous findings in vivo. Evidence suggests the everted sac preparation is not appropriate to the study of intestinal fluid and electrolyte transport. The investigation of Jejunal ion transport in vitro was continued using an Ussing chamber preparation. Selective 2-adrenoceptor stimulation was found to depress electrogenic anion secretion, as neurotoxin tetrodotoxin indicated that this was a direct epithelial effect. 2-adrenoceptor agonists have considerable therapeutic value as antisecretory agents and the model of rat jejunum in vitro represents a convenient experimental model for research in this area. The selective 2-adrenoceptor antagonist ICI 118551 decreased basal SCC and inhibited increases in SCC in response to isoprenaline or salbutamol indicating the presence of a 2-adrenoceptor mechanism mediating both secretory tone and increases in secretory processes. Many intestinal secretagogues elicit electrolyte secretion via the stimulation of intramural secretory nervous pathways. If these pathways involve the activation of 2-adrenoceptorsthe 2-adrenoceptor antagonists may be useful in the treatment of diarrhoeal diseases. A single pass lumen perfusion technique was used to investigate possible sympathetic tone over colonic fluid and electrolyte absorption in the rat colon in vivo. The technique employed appeared to lack the necessary resolution for this study and alternative approaches are discussed

Aspects of signal transduction in the gastrointestinal tract

This study was undertaken to increase knowledge of the mechanisms of inter- and intracellular signalling in the gastrointestinal tract. Specific aims were: to use cell lines to elucidate factors affecting growth of gastric cells, to investigate the distribution and aspects of function of isoforms of protein kinase C in a gastric cell line and in the rat gastrointestinal tract and to determine the presence and regulation of nitric oxide synthase in gastrointestinal tissues from the rat and in cell lines. The gastric cancer cell line HGT-1 was used to investigate control of growth. Increases in cell number were found to be dependent on the seeding density of the cells. In cells plated at low density insulin, epidermal growth factor and gastrin all increased cell number. Gastrin produced a bell-shaped dose response curve with a maximum activity at 5nM. No effect of gastrin was apparent in cells plated at high density. α and β isoforms of protein kinase C were found, by immunoblotting procedures, to be widespread in the gastrointestinal tract of the rat, but protein kinase Cε was confined to the gastric mucosa and gastrointestinal smooth muscle. HGT-1 cells contained protein kinase C α and ε but β or γ were not detected. Preincubation of HGT-1 cells for 24h with 1μM phorbol-12,13-dibutyrate down-regulated protein kinase C α but not ε. The inhibition by the activator of protein kinase C, 12-O-tetradecanoylphorbol 13-acetate (TPA) of the histamine-stimulated increase in cAMP in HGT-1 cells was down regulated by phorbol-12,13-dibutyrate. Inhibition of histamine-stimulation of adenylate cyclase by TPA was Ca2+-dependent and inhibited by the addition of an antibody to protein kinase C α. A role for protein kinase C α in modulating the effect of histamine on adenylate cyclase in HGT-1 cells is suggested. No nitric oxide synthase activity was detected in the gastrointestinal cell lines HGT-l, MKN-45 or CaCo-2. Ca2+-dependent nitric oxide synthase activity was observed in the gastric mucosa and the gastrointestinal smooth muscle from stomach to colon. The gastric: mucosal enzyme was soluble and showed half-maximal activity at 400nM Ca2+. Pretreatment of rats with endotoxin (3mg/kg body weight) induced nitric oxide synthase activity in both jejunal, ileal and colonic mucosa and muscle. A major portion of the induced activity in ileal and colonic mucosa was Ca2+-independent. Nitric oxide synthase activity in a high-density fraction of gastric mucosal cells was inhibited in a dose-dependent fashion by L-nitroarginine, NG-monomethyl-L-arginine, trifluoperazine and L-canavanine (in descending order of potency). Preincubation with okadaic acid and addition of ATPlMg2+ to the homogenisation buffer inhibited enzyme activity, which implies that phosphorylation inhibits gastric mucosal nitric oxide synthase.

Adrenoceptor influences on fluid and electrolyte transport in the rat intestine

An investigation of rat jejunal and distal colonic electrolyte transport in-vitro was undertaken using an Ussing chamber prepartion. Selective α2-adrenoceptor stimualtion in the jejunum was found to depress theo-phylline elevated anion secretion, as evidenced by decreases in short- circuit current (SCC). or α1 -Adrenoceptor stimulation, after α2 -adrenoceptor antagonism in the jejunum, evoked transient increases in basal anion secretion, as reflected by transient increases in basal SCC. The use of the neurotoxin tetrodotoxin indicated that this was a direct epithelial secretory effect. 5-hydroxytryptamine (5-HT) on the jejunum elicited transient increases in basal anion secretion, as demonstrated by transient increases in basal SCC. The use of tetrodotoxin, reserpine and α1 -adrenoceptor antagonists, indicated that a major component of this epithelial secretory effect by 5-HT, was associated with activation of intramural nervous pathways of the sympathetic nervous system, ultimately stimulating α1-adrenoceptors. This might represent an important secretory mechanism by 5-HT in the jejunum. β2-Adrenoceptor stimulation in the distal colon was found to decrease basal SCC, as evidenced by the metoprolol resistant effect of the selective β2- adrenoceptor agonist salbutamol, and lack of effect of the selective β1-adrenoceptor agonist prenalterol. An investigation of rat distal colonic fluid and electrolyte transport in-vivo was undertaken using an colonic loop technique. Although a basal colonic absorption of Na+ and Cl-, and a secretion of K+ were observed, these processes were not under tonic α-adrenergic regulation, as evidenced by the lack of effect of selective α-adrenoceptor antagonism. The secretory effects of prostaglandin-E2 were inhibited by α-adrenoceptor activation, whereas such stimulation did not evoke pro-absorptive responses upon basal transport, unlike noradrenaline.

Drug targeting with phagocytic polymorphonuclear leucocytes

1. Phagocytic polymorphonuclear leucocytes (PMNLs) or neutrophils have a marked avidity for the uptake of particulate material and are the first cell type to respond to inflammatory stimuli in vivo. 2. By harnessing these pathophysiological characteristics the inherent targeting capacity of the PMNL could be exploited to carry drug loaded particles to these sites. 3. In vitro chemotaxis of PMNLs was studied in response to N-formyl-L-methionyl-L-leucyl-L-phenylalanine (FMLP) in the Blindwell chamber assay. 4. After phagocytosis of 1.1m polystyrene latex (PSL) beads at a range of incubation concentrations (5,10,20, and 30 beads/cell) the migration of the PMNL population was not significantly different from control, without beads. 5. The distribution of the beads within the filter showed that a disproportionately large number of PSL (50%) were associated with the cells on the surface of the filter that had not penetrated the filter. Eighty per cent of the PMNL population migrated and despite containing less PSL beads/cell, 50% of the dose was carried into the filter. Between 5 and 10% of these PSL were carried beyond 60m in the assay. 6. These results suggested heterogeneity of the PMNL population and to achieve efficient targeting with these cells preferential selection of the migratory sub-population would be needed. 7. The air-pouch model was then developed to study the focal accumulation of PMNLs in vivo. The PMNL isolated did not survive long enough in the circulation due to the trauma of the isolation procedure used; an alternative method will have to be employed.

Clonal composition of human adamantinomatous craniopharyngiomas and somatic mutation analyses of the patched (PTCH), Gsα and Gi2α genes

Craniopharyngioma is the most common childhood tumor and thought to arise from embryonic remnants of Rathke's pouch. The paucity of published data on the molecular basis of these tumors prompted us to examine 22 adamantinomatous craniopharyngiomas looking for genetic abnormalities. Using the X-linked polymorphic androgen receptor gene as a tool for X-chromosome inactivating analysis, we found that a subset of craniopharyngiomas are monoclonal and therefore are probably due to acquired somatic genetic defects. Thus, we investigated these tumours for mutations within three candidate genes, Gsα, Gi2α and patched (PTCH). Using single stranded conformational polymorphism (SSCP), denaturing gradient gel electrophoresis and direct sequencing, the presence of somatic mutations in these genes could not be demonstrated in any tumor. Our data indicate that a subset of craniopharyngiomas are monoclonal and the mutations in the PTCH, Gsα, and Gi2α contribute little if any to cranipharyngioma development.

Comparative genomic hybridization analysis of craniopharyngiomas

Object. Craniopharyngioma is the most common childhood brain tumor and is thought to arise from embryonic remnants of the Rathke pouch. Some craniopharyngiomas are monoclonal in origin and hence presumably harbor somatic genetic alterations, although the precise molecular mechanisms involved in craniopharyngioma development are unknown. The goal of this study was to identify genetic alterations in craniopharyngiomas. Methods. To gain insight into the molecular mechanisms involved in development of these tumors, the authors analyzed nine adamantinomatous craniopharyngiomas by using comparative genomic hybridization. Six tumors (67%) displayed at least one genomic alteration, and three had six or more alterations. Only two tumors displayed a decrease in DNA copy number, and in all others an increase in DNA copy number was noted. Conclusions. The authors conclude that a subset of craniopharyngiomas consists of monoclonal tumors arising from activation of oncogenes located at specific chromosomal loci.