Chemical synthesis and biological properties of novel fluorescent antifolates in Pgp- and MRP-overexpressing tumour cell lines.

We have synthesised a series of fluorescent analogues of methylbenzoprim, a diaminopyrimidine antifolate which we have previously shown to exhibit in vivo antitumour activity in a methotrexate (MTX) "transport-resistant" tumour cell line. The analogues bear the dansyl, nitrobenzoxodiazole or methoxycoumarin fluorophores. The cytotoxicity of the compounds was evaluated using the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) colorimetric assay against two human lung cancer cell lines, together with their multidrug resistant (MDR) sublines. H69/P is a small cell line and its multidrug resistant subline H69/LX4 overexpresses P-glycoprotein (Pgp). COR-L23/P is a large cell line and its multidrug resistant subline COR-L23/R overexpresses the multidrug resistance associated protein (MRP). IC50 values for the compounds (i.e. concentration to reduce cell growth by 50%) in the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assay ranged from 0.20 to 0.81 microM in the H69 parental line and from 0.83 to 5.10 microM in the COR-L23 parent line. The MDR sublines both showed clear cross-resistance to each of the compounds, with resistance factors (ratio of IC50 value in resistant vs parental cell line) ranging from 16 to 137 in H69/LX4 and from 5 to 16 in COR-L23/R. For compounds (10) and (11) where drug accumulation was studied using flow cytometry, resistance was associated with an approximately 10-fold reduction in cellular drug accumulation over a period of 30 min. The drug resistance modifiers verapamil (used at 6.6 microM) and cyclosporin A (used at 4.2 microM) were tested for their ability to sensitise the resistant lines. Whereas verapamil showed little activity, cyclosporin A partially restored the activity of compound (10), and fully restored the activity of compound (11) in H69/LX4 cells. This sensitisation of H69/LX4 by cyclosporin A was associated with a partial restoration of the drug accumulation deficit in this line. Hence, these novel lipophilic antifolates appear to be substrates for both the P-glycoprotein and MRP resistance mechanisms. Therefore, although they have been designed to overcome one mechanism of methotrexate resistance, namely impaired drug transport, this has been achieved only at the cost of rendering them susceptible to alternative mechanisms.

Reversal of P-glycoprotein-mediated multidrug resistance by XR9051, a novel diketopiperazine derivative.

XR9051 (N-(4-(2-(6,7-Dimethoxy-1,2,3,4-tetrahydro-2-isoquinolyl)ethyl)phe nyl)-3-((3Z,6Z)-6-benzylidene-1-methyl-2,5-dioxo-3-pipera zinylidene) methylbenzamide) was identified as a potent modulator of P-glycoprotein-mediated multidrug resistance (MDR) following a synthetic chemistry programme based on a natural product lead compound. The activity of XR9051 was determined using a panel of human and murine drug-resistant cell lines (H69/LX4, 2780AD, EMT6/AR 1.0, MC26 and P388/DX Johnson). XR9051 was able to reverse resistance to a variety of cytotoxic drugs, including doxorubicin, etoposide and vincristine, which are associated with classical MDR. At a concentration of 0.3-0.5 microM, XR9051 was able to fully sensitize resistant cells to cytotoxics, whereas little or no effect was observed on the corresponding parental cell lines. No effect of XR9051 was observed on the response of cells to non-MDR cytotoxics such as methotrexate and 5-fluorouracil. XR9051 was consistently more potent than cyclosporin A (CsA) and verapamil (Vpm) in all assays used. XR9051 inhibited the efflux of [3H]daunorubicin from preloaded cells and, unlike CsA and Vpm, remained active for several hours after removal of resistance-modifying agent. In photoaffinity labelling experiments employing [3H]azidopine, XR9051 was able to displace binding to P-glycoprotein. In binding studies using [3H]vinblastine, XR9051 was shown to be a potent inhibitor of the binding of the cytotoxic to P-glycoprotein (EC50 = 1.4 +/- 0.5 nM). Taken together, the results indicate that XR9051 reverses the MDR phenotype through direct interaction with P-glycoprotein.

Intermittent exposure to doxorubicin in vitro selects for multifactorial non-P-glycoprotein-associated multidrug resistance in RPMI 8226 human myeloma cells.

The purpose of the present study was to evaluate whether intermittent exposure to a constant dose of doxorubicin selects for multidrug resistance (MDR) in RPMI 8226 human myeloma cells and, if so, to determine the molecular mechanism. In an attempt to approximate clinical doxorubicin treatment in vitro, cells were exposed to a fixed dose of doxorubicin for 4 d alternating with growth in drug-free medium for 17 d. An MDR subline emerged, termed 8226/DOXint5, which was 3-4-fold resistant to doxorubicin, etoposide and m-AMSA, and 1.6-fold resistant to vincristine. Sensitivity to docetaxel, melphalan and cisplatin was normal. Verapamil normalized vincristine sensitivity but had little effect on resistance to the other agents. Cellular uptake and retention of daunorubicin and vincristine were reduced by approximately 10%. The 8226/DOXint5 cells showed diminished DNA topoisomerase IIalpha expression and increased expression of the multidrug resistance protein MRP. Expression of MDR1/P-glycoprotein was not detected. Immunostaining showed 70% of the cells to over-express the lung-resistance protein LRP. This new MDR myeloma cell line may prove to be a useful model for the development of strategies to overcome low-level, multifactorial MDR, which might be a common phenomenon in clinical myeloma treated with doxorubicin.

Transport proteins in drug resistance: biology and approaches to circumvention.

At least two transport proteins, P-glycoprotein (Pgp) and the multidrug resistance associated protein (MRP), are believed to play a significant role in clinical resistance to cytotoxic therapy. These proteins are expressed at relatively high levels in a number of malignant diseases including various types of leukaemias. They are variably expressed on both the plasma membrane and intracellular vesicular membranes resulting in cellular drug efflux or vesicular drug sequestration, respectively. The action of MRP as a drug transporter depends on intracellular levels of glutathione. A number of strategies for circumvention of these drug resistance mechanisms have been developed and some of these are now in clinical trial.

[Alkaloids in Sophora alopecuroides seed and relevant tests for activity].

Seven alkaloids were isolated from the seed of Sophora alopecuroides and identified to be oxymatrin, oxysophocarpine, cytisine, matrine, sophocarpine, sophoridine and nicotine respectively by comparing chromactographic and spectral characteristics with authentic known compounds. Nicotine was isolated from Sophora for the first time. The activity of extracts and alkaloids against cancer, virus, dermatophytes and bacteria was carried out in vitro.

Acceleration of MRP-associated efflux of rhodamine 123 by genistein and related compounds.

Multidrug resistance (MDR), caused by overexpression of either P-glycoprotein or the multidrug resistance protein (MRP), is characterised by a decreased cellular drug accumulation due to an enhanced drug efflux. In this study, we examined the effects of genistein and structurally related (iso)flavonoids on the transport of rhodamine 123 (Rh123) and daunorubicin in the MRP-overexpressing MDR lung cancer cell lines COR-L23/R and MOR/R. Genistein, genistin, daidzein and quercetin showed major differences in effects on Rh123 vs daunorubicin transport in the MRP-mediated MDR cell lines: the accumulation of daunorubicin was increased, whereas the accumulation of Rh123 was decreased by the flavonoids. The depolarisation of the membrane potential caused by genistein might be involved in the acceleration of the efflux of Rh123 measured in the MRP-overexpressing cell lines. These observations should be taken into account when using fluorescent dyes as probes for determination of transporter activity as a measure of MDR.

Rapid recovery of a functional MDR phenotype caused by MRP after a transient exposure to MDR drugs in a revertant human lung cancer cell line.

Prior studies have shown that, in some human tumour cells, increased expression of the multidrug resistance gene MDR1 can be induced in response to certain stress conditions such as a transient exposure to cytotoxic agents. Little is known about the possibility of increasing the expression of the recently cloned multidrug resistance-associated protein (MRP) in response to a transient exposure to cytotoxic drugs. In order to examine this possibility, we have used sensitive assays (RT-PCR, flow cytometry) and the sensitive large cell lung cancer cell line, COR-L23/P, and the revertant line (COR-L23/Rev), generated by growing the doxorubicin-selected, MRP-overexpressing resistant variant COR-L23/R without drug exposure for 24-28 weeks. COR-L23/Rev overexpresses MRP, but to a lesser extent than COR-L23/R. COR-L23/Rev rapidly recovered similar levels of MRP mRNA, protein expression, resistance and drug accumulation deficit as COR-L23/R after a 48-72 h exposure to cytotoxic concentrations of doxorubicin or vincristine but not cisplatin. The increase in MRP mRNA could only be detected 3 to 4 days after the transient exposure to drugs. However, when the parental line, COR-L23/P, was exposed to equitoxic doses of doxorubicin, vincristine or cisplatin, no increase in the levels of MRP mRNA could be observed at higher doses (5- to 10-fold the IC50) of doxorubicin or vincristine (but not of cisplatin), we detected a transient increase in the levels of MDR1 mRNA immediately after short-term exposure. In conclusion, we have shown that a human revertant lung cancer cell line (COR-L23/Rev) has the ability to recover quickly, similar levels of MRP expression and resistance as COR-L23/R after a transient exposure to the MDR-drugs doxorubicin and vincristine.

Modulation by acrolein and chloroacetaldehyde of multidrug resistance mediated by the multidrug resistance-associated protein (MRP).

Acrolein (AC) and chloroacetaldehyde (CHA) are metabolites of the non-multidrug resistance cytotoxic drugs cyclophosphamide and ifosfamide. It has previously been reported that both metabolites can induce extensive depletion of glutathione (GSH) in vitro and in vivo and that this depletion occurs at drug concentrations in the micromolar range. A link between the function of the multidrug resistance-associated protein (MRP) and the intracellular concentration of GSH has also been demonstrated. To determine whether AC and CHA can modulate the function of MRP by inducing GSH depletion, we used two human lung cancer cell lines overexpressing MRP: the large cell carcinoma cell line COR-L23/R and the adenocarcinoma cell line MOR/R0.4, along with their respective sensitive parental lines, COR-L23/P and MOR/P. We showed that micromolar concentrations of AC and millimolar concentrations of CHA are able to deplete GSH concentrations in the cell lines studied. In addition, concentrations of 50 micrometer AC and 5 mm CHA could completely reverse the daunorubicin (DNR) and vinblastine accumulation deficit present in COR-L23/R and partially reverse the DNR accumulation deficit in MOR/R0.4. In contrast, AC and CHA did not reverse the drug accumulation deficit in the P-glycoprotein-overexpressing lung cancer cell line H69/LX4. The effect of CHA and AC on drug accumulation was related to the GSH depletion, as we found a concentration-dependent relationship between the GSH levels and the reversal of the accumulation deficit for both AC and CHA. To substantiate further this correlation, we increased cellular GSH content in AC- and CHA-treated cells with the GSH ethyl ester. An increase in cellular GSH levels in CHA- and AC-treated COR-L23/R cells was accompanied by a restoration of the DNR accumulation deficit. No significant effect of the GSH ethyl ester was detected on DNR accumulation in COR-L23/P parental cells. In conclusion, treatment with AC or CHA can reverse the drug accumulation deficit of MRP-overexpressing cells, and this effect appears to be mediated by GSH depletion.

Hypotonicity-induced anion fluxes in cells expressing the multidrug-resistance-associated protein, MRP.

Anion transport in human multidrug-resistant large cell lung tumour cells (COR-L23/R) which overexpress the multidrug-resistance-associated protein (MRP) has been compared with that in cells of the parent line (COR-L23/P). Whole-cell patch-clamp recordings reveal variability between individual cells in basal anion conductance and in anion conductance increases following exposure to hypotonic media. The increase of stimulated over basal conductance is significantly larger for resistant cells than for parent cells. The chloride channel blocker, diisothiocyanatostilbene-2-2'-disulphonic acid (DIDS), rapidly and reversibly inhibits the increase in outward but not inward conductance when applied externally at 10(-4) M during recording, but it is without effect when introduced into the cells via the patch pipette. Preincubation with DIDS greatly reduces both inward and outward conductance. 125I- efflux has been used to measure anion movement in cell populations. Basal efflux is similar in the two cell lines, but following a hypotonic challenge, the increase in rate constant for efflux from COR-L23/R cells is at least double that from COR-L23/P cells. This increase in efflux is greatly reduced by incubation with DIDS at 10(-4) M. Replacement of external chloride by gluconate does not affect efflux, thus excluding the possible involvement of DIDS-sensitive chloride exchange. Results from both techniques suggest that DIDS-sensitive, hypotonicity-induced anion channel activity is augmented in COR-L23/R multidrug-resistant variant cells which overexpress MRP. This augmentation may be caused by MRP itself or by other genes coexpressed with MRP.

Phase I study of etoposide with SDZ PSC 833 as a modulator of multidrug resistance in patients with cancer.

PURPOSE: To determine the maximum-tolerated dose (MTD) and toxicity of PSC 833 infusion administered with etoposide for 5 days in patients with cancer, and to determine the effect of PSC 833 on etoposide pharmacokinetics. PATIENTS AND METHODS: Thirty-five patients were entered onto the study, one of whom was ineligible. Etoposide was delivered from day 1 as a 2-hour infusion over 5 consecutive days at a dose of 75 to 100 mg/m2/d. PSC 833 was administered from day 2 as a 2-hour loading dose and as a 5-day continuous infusion. Doses were escalated from 1 to 2 mg/kg (loading dose) and 1 to 15 mg/kg/d (continuous infusion). RESULTS: Thirty-four patients were treated with 53 cycles of PSC 833 and etoposide. Steady-state blood PSC 833 levels more than 1,000 ng/mL were achieved in all patients treated at PSC 833 doses > or = 6.6 mg/kg/d by continuous infusion. Myelosuppression was the most common toxicity. The major dose-related toxicity of PSC 833 was reversible hyperbilirubinemia, which occurred in 83% of cycles. The dose-limiting toxicity of PSC 833 was severe ataxia, which occurred in two of nine patients treated at 12 mg/kg/d and in both of the single patients treated at 13.5 and 15 mg/kg/d. PSC 833 concentrations more than 2,000 ng/mL resulted in an increase in etoposide area under the curve (AUC) of 89%, a decrease in etoposide clearance (Cl) of 45%, a decrease in volume of steady-state distribution (Vss) of 41%, and an insignificant increase in alpha half-life (t 1/2 alpha) and significant increase of beta half-life (t 1/2 beta) of 19% and 77%, respectively. CONCLUSION: PSC 833 can be administered in combination with etoposide with acceptable toxicity. The recommended continuous infusion dose of PSC 833 for this schedule is 10 mg/kg/d over 5 days. PSC 833 results in an increase in etoposide exposure and etoposide doses should be reduced in patients receiving PSC 833.

Sensitive and rapid bioassay for analysis of P-glycoprotein-inhibiting activity of chemosensitizers in patient serum.

Clinical studies of agents capable of reversing P-glycoprotein (Pgp)-mediated multidrug resistance have attracted much attention in recent years. One question of interest in such studies is whether the concentrations achieved by chemosensitizers are sufficient to inhibit Pgp function. The goal of the present study was to develop a reliable ex vivo bioassay for analysis of the Pgp-inhibiting activity of chemosensitizer-containing patient serum. The fluorescent Pgp substrates daunorubicin (DNR) and rhodamine 123 (R123) were used as probes for Pgp function. The 8226/DOX6 human myeloma cell line, which expresses Pgp at levels that can be detected in clinical cancers, was used as a model system. The index chemosensitizers tested were dexverapamil (DVPM) and cyclosporin A, with particular focus on DVPM. Using flow cytometry, chemosensitizer effects on 1-h drug accumulation and on drug retention at 30 min were evaluated. In the studies using pooled human serum spiked in vitro with graded chemosensitizer concentrations, the order of assay sensitivity was R123 retention >>> R123 accumulation > DNR retention equal to DNR accumulation. Keeping serum spiked with DVPM for several hours at room temperature or 4 degreesC or for several months at -80 degreesC had no effect on Pgp-blocking activity. Sixteen blood samples from patients with metastatic breast cancer receiving DVPM to overcome epirubicin resistance were analyzed for Pgp-inhibiting activity and for levels of DVPM and nor-DVPM, the major metabolite of verapamil. Each patient sample was found capable of increasing R123 retention in the 8226/DOX6 cells, with activity factors of 3- to 8-fold and good agreement between DVPM blood levels and bioassay activity (r = 0.7168; two-sided P = 0.0018). The R123 retention assay developed and validated in this study seems to be a sensitive, reproducible, and easy-to-use method for analysis of Pgp-inhibiting activity of chemosensitizer-containing human serum. The assay seems capable of estimating DVPM blood levels and could prove to be a valuable tool for monitoring chemosensitizer treatment in cancer patients.

On the relationship between the probenecid-sensitive transport of daunorubicin or calcein and the glutathione status of cells overexpressing the multidrug resistance-associated protein (MRP).

Cells exposed to calcein acetoxymethyl ester (calcein AM) in the growth medium become fluorescent following cleavage of calcein AM by cellular esterases to produce the fluorescent derivative calcein. It has previously been shown by others that multidrug resistant cells which overexpress P-glycoprotein accumulate much less fluorescent calcein than the corresponding parental cells. We have now examined the transport of calcein in multidrug resistant cells which overexpress an alternative transporter, the multidrug resistance-associated protein (MRP). Accumulation of calcein fluorescence was greatly reduced in the MRP-overexpressing human lung cancer cell lines COR-L23/R and MOR/R compared with their parental lines. Energy depletion resulted in a considerably increased accumulation in the resistant lines. Treatment of resistant cells with buthionine sulfoximine (BSO), which depletes cellular glutathione (GSH), did not affect calcein accumulation, in marked contrast to our previous results for daunorubicin or the fluorescent probe rhodamine 123. Genistein, verapamil, cyclosporin A and ouabain were also each able to modify, to some extent, accumulation of daunorubicin, whilst having essentially no effect on calcein accumulation. However, the organic anion transport inhibitor probenecid was able to increase accumulation of both calcein and daunorubicin in the resistant cells. Genistein and verapamil treatment preferentially reduced the GSH content of resistant cells, whilst probenecid did not. However, probenecid caused a clear decrease in release of GSH from resistant cells into the medium.

Localisation of the multidrug resistance-associated protein, MRP, in resistant large-cell lung tumour cells.

The drug transport protein, P-glycoprotein, confers multidrug resistance (MDR) by expelling drugs across the cell surface. The structurally similar multidrug resistance-associated protein, or MRP, is also involved with drug efflux. In MDR variants of the human lung tumour cell line COR-L23 that overexpress MRP, there are also changes in intracellular drug distribution. To ascertain whether MRP could be involved in either process, experiments were performed to identify where MRP was located in these cells. Following separation of membranes by sucrose gradient centrifugation, MRP was found predominantly in the lighter membrane fractions containing plasma membrane enzyme activity. Immunofluorescent staining with a monoclonal antibody raised against MRP confirmed that MRP is present at the cell surface of these MDR lung tumour cells.

Regulation by glutathione of drug transport in multidrug-resistant human lung tumour cell lines overexpressing multidrug resistance-associated protein.

Previous studies have shown that multidrug resistance (MDR) in the doxorubicin-selected lung tumour cell lines COR-L23/R, GLC4/ADR and MOR/R is associated with overexpression of the MRP gene. In this study we report that resistance to daunorubicin, vincristine and rhodamine 123 can be partially reversed in these cell lines by exposing the cells to buthionine sulphoximine (BSO), an inhibitor of glutathione (GSH) synthesis. This effect of BSO on drug resistance was associated with an increased intracellular accumulation of daunorubicin and rhodamine 123, owing to inhibition of the enhanced drug efflux. In contrast, the accumulation of daunorubicin was not increased by BSO treatment in a P-glycoprotein (P-gp)-mediated MDR cell line. BSO treatment (25 microM, 20 h) of the cell lines resulted in 60-80% depletion of cellular GSH levels. The effects of BSO on daunorubicin accumulation in the COR-L23/R and GLC4/ADR cells were associated with cellular GSH depletion. In addition, increase of cellular GSH levels in BSO-treated COR-L23/R and GLC4/ADR cells as a result of incubation with 5 mM GSH ethyl ester restored the accumulation deficit of daunorubicin. However, the transport of daunorubicin did not increase the GSH release in any of the cell lines. These results demonstrate that drug transport in MRP- but not in P-gp-overexpressing MDR tumour cell lines can be regulated by intracellular GSH levels.

Expression of vascular endothelial growth factor and its receptors flt and KDR in ovarian carcinoma.

BACKGROUND: Two thirds of patients with ovarian carcinoma have advanced disease at diagnosis and have poor prognoses because of the presence of highly invasive carcinoma cells and rapidly accumulating ascitic fluid. Vascular endothelial growth factor (VEGF), a potent mitogen of endothelial cells, is produced in elevated amounts by many tumors, including ovarian carcinomas. The known human receptors for VEGF, flt and KDR, are both cell surface tyrosine kinases and are expressed predominantly on endothelial cells. Acting through these receptors, VEGF may stimulate angiogenesis and promote tumor progression. PURPOSE: We aimed to clarify the function of VEGF in tumor development by identifying the cells in ovarian carcinoma tissue that express VEGF and its receptors. METHODS: VEGF, flt, and KDR expression was localized by in situ hybridization and immunohistochemistry in frozen sections of primary tumors from five patients with ovarian carcinoma and from metastases of ovarian carcinoma from three different patients. Reverse transcription followed by polymerase chain reaction (RT-PCR) and an enzyme-linked immunosorbent assay were used to analyze VEGF, flt, and KDR expression in six epithelial cell lines derived from ovarian carcinoma ascites from five additional patients. RESULTS: Messenger RNAs (mRNAs) encoding VEGF, flt, and KDR were detected in primary ascitic cells and in three of four ovarian carcinoma cell lines examined by RT-PCR. Two novel complementary DNAs that may encode truncated, soluble forms of flt were cloned from one primary source. VEGF levels of 20-120 pM were found in culture media conditioned by the cell lines. Elevated expression of VEGF mRNA was found in all primary tumors and metastases, especially at the margins of tumor acini. VEGF immunoreactivity was concentrated in clusters of tumor cells and patches of stromal matrix. flt immunoreactivity was confined to tumor blood vessels, but flt mRNA was not detected by in situ hybridization. In contrast, KDR mRNA was detected not only in vascular endothelial cells but also in tumor cells at primary malignant sites. CONCLUSIONS: VEGF is expressed by tumor cells in primary and metastatic ovarian carcinoma and accumulates in the stromal matrix. Its receptors, flt and KDR, are expressed by some tumor cells that coexpress VEGF. This is the first localization of KDR expression in nonendothelial cells. IMPLICATIONS: Coexpression of VEGF and KDR by tumor cells in ovarian carcinoma raises the possibility of autocrine stimulation and of therapeutic strategies targeting this receptor-ligand interaction.

Mechanisms of MRP over-expression in four human lung-cancer cell lines and analysis of the MRP amplicon.

Some multidrug resistant cell lines over-express the gene encoding the multidrug-resistance-associated protein (MRP). In all cell lines reported thus far, over-expression is associated with gene amplification. We have studied the predominant mechanisms of MRP over-expression in 4 human lung-cancer cell lines that cover a range of drug-resistance levels, and we have analyzed the MRP amplicon. In the SW-1573-derived, weakly resistant cell line 30.3M, MRP mRNA is elevated 3-fold in the absence of gene amplification. Run-on analysis shows that the increased MRP gene expression in this cell line is due to transcriptional activation. In the highly resistant GLC4/ADR and COR-L23/R cells, MRP gene amplification predominates, whereas in the moderately resistant MOR/R cells, gene amplification is combined with a mechanism resulting in an additional increase in the level of MRP mRNA. Fluorescence in situ hybridization shows that, in the GLC4/ADR cells, amplified MRP sequences are present both in double minute chromosomes (DM) and in homogeneously staining regions (HSR). By pulsed-field gel electrophoresis we show that the MRP-containing DM are 1 Mb in length. Chromosome-16-specific repetitive sequences adjacent to the MRP gene are also present in the DM and HSR, compatible with the involvement of these sequences in recombination events underlying MRP gene amplification. Our results show that low levels of drug resistance may arise by transcriptional activation of the MRP gene, whereas at high levels of drug resistance amplification of the MRP gene predominates, possibly facilitated by the presence of recombination-prone sequences.

The activity of deoxyspergualin in multidrug-resistant cells.

Deoxyspergualin, a synthetic analogue of the immunosuppressive anti-tumour antibiotic spergualin, has been shown to possess potent in vitro and in vivo anti-tumour activity and is currently in the National Cancer Institute (NCI) decision network. Deoxyspergualin shows similarities in properties and mechanisms of action to the natural-product immunosuppressive agents cyclosporin A and FK506, each of which can act as a modifier of multidrug resistance. We therefore decided to examine the comparative activity of deoxyspergualin in parent and multidrug-resistant cells. Deoxyspergualin contains the polyamine spermidine within its structure. Bovine serum copper amine oxidase catalyses the oxidative deamination of spermidine to produce an aminoaldehyde, ammonia and hydrogen peroxide. These aminoaldehydes are believed to be responsible for the toxicity of polyamines in vitro in the presence of bovine serum. For this reason, all experiments were carried out in medium containing bovine serum and in medium containing horse serum (which is low in copper amine oxidase content). We used the tetrazolium (MTT) colorimetric assay to determine drug sensitivity and tritiated daunorubicin accumulation together with inhibition of azidopine binding to study specific mechanisms of resistance modulation. The murine cell lines EMT6/P and EMT6/AR1.0 and the human cell lines H69/P and H69/LX4 were, respectively, 32-, 32-, 372- and 483-fold more sensitive to spermidine and 175-, 133-, 321- and 444-fold more sensitive to spermine in the presence of calf serum than in the presence of horse serum. However, these large differential effects were not seen for deoxyspergualin. It appears that in the presence of horse serum, deoxyspergualin may exert its effect by a mechanism other than polyamine oxidation. Deoxyspergualin did not enhance the accumulation of [3H]-daunorubicin in EMT6/AR1.0 cells. Furthermore, deoxyspergualin (1-20 microM) did not restore the sensitivity of EMT6/AR1.0 or H69/LX4 cells to that of the parent lines. P-glycoprotein (Pgp) in membranes prepared from H69/LX4 cells was photo-affinity-labeled with [3H]-azidopine. Deoxyspergualin did not inhibit this labeling. Although deoxyspergualin appears to exert its immunosuppressive effect via a mechanism similar to that of cyclosporin A and FK506, it does not share their ability to modify Pgp-mediated multidrug resistance. However, its lack of cross-resistance and potent in vivo anti-tumour activity make deoxyspergualin a promising candidate for development as an anti-cancer agent.

Modification by brefeldin A, bafilomycin A1 and 7-chloro-4-nitrobenz-2-oxa-1,3-diazole (NBD) of cellular accumulation and intracellular distribution of anthracyclines in the non-P-glycoprotein-mediated multidrug-resistant cell line COR-L23/R.

We have investigated the effects of H(+)-ATPase inhibitors, bafilomycin A1 and 7-chloro-4-nitro-benz-2-oxa-1,3 diazole (NBD), and the Golgi inhibitor, brefeldin A, on daunorubicin accumulation and doxorubicin intracellular distribution in the non-P-glycoprotein-mediated multidrug-resistant cell line COR-L23/R. This cell line overexpress a 190 kDa protein which is probably the product of the MRP gene and shows an anthracycline accumulation defect and a drastically altered intracellular anthracycline distribution from the parental cell line COR-L23/P. We found that all three agents could selectively increase the cellular accumulation of daunorubicin in resistant cells. However, these effects were only seen at doses of the modifiers which were equal to or greater than the IC50 of the modifier alone. Effects of the modifiers on the intracellular distribution of doxorubicin fluorescence could, however, be seen at doses lower than those required to produce significant effects on daunorubicin accumulation. However, when used in a continuous MTT chemosensitivity assay none of the agents, used at maximum non-toxic doses, was able to sensitise COR-L23/R cells to doxorubicin or to colchicine. Although these lead compounds are unlikely to be useful as clinical modifiers, development of more selective analogues may prove useful in the modification of non-P-glycoprotein-mediated multidrug resistance.