Review of Arginine-482 Is Not Essential for the Transport of Antibiotics, Primary Bile Acids and Unconjugated Sterols by the Human Breast Cancer Resistance Protein (abcg2)

The breast cancer resistance protein has been found to be a transporter that extrudes various anticancer drugs from cells, resulting in multi-drug resistance. Recent studies on tissue distribution of BCRP (breast cancer resistance protein) showed that the protein is expressed in the apical membrane of cells in tissues with excretory functions, such as luminal membrane of villous epithelial cells in small and large intestines, the apical pole of the trophoblast cells in the placenta, and apical membranes of capillary vessels in the blood-brain barrier, ducts and lobules of the breast. These findings are consistent with the belief that BCRP plays a role in the disposition and pharmacological activity of a broad range of compounds. Other early studies on BCRP revealed that some drug-selected mammalian cell lines expressing BCRP demonstrate resistance to cationic rhodamine 123 and anthracyclines due to the presence of a neutral G (glycine), T (threonine), S (serine) or M (methionine) residue at position 482 of BCRP while those expressing wild-type, R (with an arginine at 482), remain sensitive. In this study, investigations on sterol transport were expanded to BCRP-R in L. lactis. In addition, the specificities of BCRP-G and -R for antibiotics and primary bile acids were compared in detail to analyze the role of Arg-482 in the BCRP-mediated transport of these substrates.

To examine whether BCRP-G and -R are active as antibiotic efflux systems the interaction of BCRP with antibiotics was tested. When lactococcal cells expressing BCRP-G or -R were exposed to erythromycin, both exhibited significantly increased survival ratio (the number of colony-forming units in the presence of erythromycin / number of colony-forming units without the antibiotic) at increasing concentrations of erythromycin when compared to the non-expressing control cells. High relative resistance factors of 82 and 78 were determined for BCRP-G- and -R-expressing cells in growth experiments. BCRP-G-associated ATPase activity was stimulated by the interaction of erythromycin with BCRP-G. However, erythromycin did not stimulate ATPase activity in the non-expressing control. This data suggests that endogenous ABC multidrug transporters, which may be expressed at low levels in L. lactis, do not contribute to the observed erythromycin resistance and transport in BCRP-R- or BCRP-G-expressing cells. Also, BCRP-R- or BCRP-G-expressing cells exhibited a significant decrease in the amount of cell-associated erythromycin, whereas an accumulation of erythromycin was observed in the non-expressing control. Therefore, BCRP-G/R-expressing cells exhibited greater active efflux of [14C]erythromycin than in intact control cells.

The specificity of BCRP for other antibiotics was also tested. The expression of BCRP-R significantly enhanced the growth of L. lactis in the presence of tetracycline or rifampicin, giving a relative resistance factor (IC50 of BCRP-expressing cells/IC50 of control cells) of 2.0 and 4.5 respectively. Furthermore, BCRP-G revealed a high level of BCRP-G-associated resistance in L. lactis to the macrolide dirithromycin with a relative resistance factor of 28 and a significant relative resistance of 2 for tetracycline, fluoroquinolones such as ofloxacin, the antiprotozoal drug quinacrine and other drugs such as cephalothin and rifampicin. These drugs also stimulated the BCRP-G-associated ATPase activity by 2-5-fold at SC50 values in the lower micromolar concentration range. BCRP-G expression did not confer a significant resistance on lactococcal cells to aminoglycosides, b-lactams, lincosamides, sulphacetamine, gramicidin, metronidazole, vancomycin. These observations suggest that BRCP-G and -R interact with a similar range of antibiotics in L. lactis.

Another factor that was examined was whether the transport of methotrexate and rhodamine 123 is affected by the R to G substitution at position 482 in BCRP. Mammalian cell lines expressing the BCRP-G exhibit an enhanced efflux of cationic anthracyclines and rhodamine 123, but a decreased efflux of anionic methotrexate when compared with cell lines expressing BCRP-R. Experiments were performed to test whether these differences in drug selectivity of BCRP-G and -R were retained in L. lactis. The experiments showed that BCRP-G mediated the transport of rhodamine 123 and the rhodamine analogue TMR, but BCRP-R was unable to handle these substrates. The interaction of BCRP-G/R with methotrexate was studied in competition assays on the basis of the BCRP-mediated transport of the fluorescent substrate Hoechst 33342. BCRP-R showed an enhanced interaction with methotrexate compared with BCRP-G, showing the different specificities of BCRP-R and -G for charged substrates. Together, these observations suggest that the differences in the specificity of BCRP-G and -R for methotrexate and rhodamine 123, as observed in mammalian cells, are at least partly conserved in L. lactis.

The final factor that was examined was whether BCRP-G and -R mediate the transport of oestradiol and bile acids. In a previous study on BCRP-G in L. lactis, it was observed that the BCRP-G-associated ATPase was stimulated by sterols including oestradiol and cholesterol, the natural steroids