High mortality in pancreatic cancer patients is partly due to resistance to chemotherapy. We describe that human pancreatic cancer cells acquire drug resistance by a novel mechanism in which they expel and remove chemotherapeutic drugs from the microenvironment via microvesicles (MVs). Using human pancreatic cancer cells that exhibit varied sensitivity to gemcitabine (GEM), we show that GEM exposure triggers the cancer cells to release MVs in an amount that correlates with that cell line's sensitivity to GEM. The importance of MV-release in gaining drug resistance in GEM-resistant pancreatic cancer cells was confirmed when the inhibition of MV-release sensitized the cells to GEM treatment, both in vitro and in vivo. Mechanistically, MVs remove drugs that are internalized into the cells and that are in the microenvironment. The differences between the drug-resistant and drug-sensitive pancreatic cancer cell lines tested here are explained based on the variable content of influx/efflux proteins present on MVs, which directly dictates the ability of MVs either to trap GEM or to allow GEM to flow back to the microenvironment.

Doxorubicin is an effective anticancer agent that is limited by numerous adverse effects, cardiotoxicity causing the most concern. Its alcohol metabolite, doxorubicinol, and free radicals have been implicated in the aetiology of this toxicity. This study was based on the premise that inhibition of aldo-keto reductases would improve the efficacy of doxorubicin by reducing its toxic metabolites and modifying its pharmacokinetics. We assessed the effect of in-vitro inhibition of aldo-keto reductases on the metabolism of doxorubicin in cytosolic fractions of heart and liver of rats in the presence of Na-phenobarbital. The inhibition was confirmed by a significant reduction in the formation of doxorubicinol. The results of the in-vitro study were further evaluated in-vivo. The concentrations of doxorubicin in plasma, bile and urine and its major metabolites in bile and urine were measured in Na-phenobarbital-pretreated rats. Each rat received 100 mg kg(-1)/day intraperitoneal injection of sodium phenobarbital for three days followed by a single intravenous dose of 10 mg kg(-1) [14C-14]doxorubicin (sp. act. 0.2 microCi mg(-1)) on the fourth day. The levels of drug in all biological samples were measured by HPLC. The pretreatment resulted in an increase in biological half-life (5.8 +/- 1.5 vs 3.7 +/- 0.93 h control group, P < 0.05) and area under plasma concentration-time curve (19.6 +/- 1.7 vs 14.65 +/- 1.68 mg h L(-1) control group, P < 0.05). The cumulative amount of doxorubicinol in the bile and urine of pretreated animals was reduced significantly. In terms of % dose, the amount in the bile declined from 4.2 +/- 0.8% in control to 2.4 +/- 0.3% and in urine from 0.18 +/- 0.08% to 0.12 +/- 0.07%. There were no significant changes in doxorubicin aglycone and doxorubicinol aglycone. Serum creatine kinase levels were measured as a biomarker of damage to cardiac muscle. The area under creatine kinase level-time curve was reduced by approximately 50% in phenobarbital-pretreated animals. The results indicate that the inhibition of aldo-keto reductase could provide a useful approach to improve the safety of doxorubicin by reducing its alcohol metabolite. Furthermore, if the reduction in the area under the serum creatine kinase-time curve represents a reduced damage to heart muscle, it can be concluded that doxorubicinol plays an important role in this injury.

1. Influx and efflux proteins play a major role in the overall uptake and efficacy of chemotherapeutic agents and cellular chemo-resistance. 2. The present study investigated the time course and dose dependency of the induction of three efflux proteins, P-gp, MRP1 and MRP5, in response to gemcitabine exposure in Capan-2 pancreatic cancer cell line at transcriptional and translational levels. The influence of exposure on the influx protein (ENT1), the net cellular uptake of the gemcitabine, the overall ATPase activity and the cell death rate were also measured. 3. The time course of the expression exhibited an initial rise, toward a plateau level. The estimated Km and Vmax confirmed that MRP5 and to a lesser extent MRP1 are the prominent proteins for efflux of gemcitabine. Both mRNA and protein expression demonstrated the time and concentration dependency of the induction; and the elevated ATPase activity validated that the induced efflux proteins are functionally active. 4. The results of the study revealed that the efficacy window of gemcitabine as it relates to the function of the efflux proteins is concentration and temporal dependent and is well correlated to the first 60 min of exposure.

Adriamycin has a wide spectrum of antitumour activity with dose-related cardiotoxicity as a major side effect. This cardiotoxicity has been suggested to result from the generation of oxygen free radicals. The objective of the present study was to investigate the influence of the antioxidant, butylated hydroxyanisole co-therapy on the cardiotoxicity of adriamycin, in-vivo. The aqueous solubility of butylated hydroxyanisole was enhanced by inclusion complex formation with hydroxypropyl-beta-cyclodextrin. The extent of drug-induced myocardial damage in rats was assessed using intravenous 111In-labelled antimyosin Fab and chronological changes in serum creatine kinase levels. There was a dose-related increase in myocardial antimyosin uptake in rats, which reached a plateau at an adriamycin dose of 10 mg kg-1. The antimyosin uptake at this dose (% dose g-1 = 0.1942 +/- 0.0150, n = 8) was significantly reduced by co-administration of butylated hydroxyanisole with adriamycin (10 mg kg-1 of each) to 0.1462 +/- 0.0116 (n = 5, P < 0.05). Assessment of cardiotoxicity in the rats was also performed by measuring serial changes in serum creatine kinase levels. Increasing doses of adriamycin caused an increase in serum creatine kinase levels with peak values obtained between 2 and 8 h after dosing. These values decreased upon co-administration of butylated hydroxyanisole with adriamycin at 10 mg kg-1, each and 30 mg kg-1 each by 29 and 41%, respectively. On the other hand, butylated hydroxyanisole did not inhibit the tumouricidal activity of adriamycin as investigated in-vitro using the NMU rat mammary adenocarcinoma cell-line. The significant reduction in anthracycline cardiotoxicity by butylated hydroxyanisole coadministration may result from its scavenging action on adriamycin-mediated free-radical formation or its enhancement of activity of enzymes involved in the metabolism of adriamycin.

Modulators of P-glycoprotein (P-gp) can enhance or limit the permeability of a number of therapeutic agents that are considered substrates of this efflux pump protein. The modulatory effect of amlodipine (4-dihydropyridine calcium antagonist) on P-gp efflux activity has not been fully elucidated. We have studied the concentration dependency of its modulatory effect and compared it qualitatively with tamoxifen (a non-esteroid anti-estrogen). The investigation was conducted on transmembrane efflux of doxorubicin at a fixed concentration of 5 microM across a Caco-2 monolayer in the presence of various concentrations of amlodipine or tamoxifen. The maximum flux of doxorubicin from basolateral to apical (ba) occurred at 4.5 microM amlodipine and at 0.02 microM tamoxifen. At higher concentrations, the apical to basolateral (ab) flux and the net flux of doxorubicin (ba - ab) declined steadily in a concentration-dependent manner. We analysed the observed net flux data by fitting different mathematical models to the data. A composite sigmoidal Emax/Imax (stimulatory/inhibitory) model was found to be the most appropriate to define the system. The observed and calculated parameters supported the modulatory role of both compounds and clearly indicated that the stimulation and inhibition of transmembrane efflux occurred simultaneously in the presence of amlodipine or tamoxifen. It was concluded that amlodipine, similar to tamoxifen, modulated the transporter-dependent transmembrane flux of the P-gp substrate in a concentration-dependent manner.