ABCG2 expression, function, and promoter methylation in human multiple myeloma

Abstract

We investigated the role of the breast cancer resistance protein (BCRP/ABCG2) in drug resistance in multiple myeloma (MM). Human MM cell lines, and MM patient plasma cells isolated from bone marrow, were evaluated for ABCG2 mRNA expression by quantitative polymerase chain reaction (PCR) and ABCG2 protein, by Western blot analysis, immunofluorescence microscopy, and flow cytometry. ABCG2 function was determined by measuring topotecan and doxorubicin efflux using flow cytometry, in the presence and absence of the specific ABCG2 inhibitor, tryprostatin A. The methylation of the ABCG2 promoter was determined using bisulfite sequencing. We found that ABCG2 expression in myeloma cell lines increased after exposure to topotecan and doxorubicin, and was greater in logphase cells when compared with quiescent cells. Myeloma patients treated with topotecan had an increase in ABCG2 mRNA and protein expression after treatment with topotecan, and at relapse. Expression of ABCG2 is regulated, at least in part, by promoter methylation both in cell lines and in patient plasma cells. Demethylation of the promoter increased ABCG2 mRNA and protein expression. These findings suggest that ABCG2 is expressed and functional in human myeloma cells, regulated by promoter methylation, affected by cell density, up-regulated in response to chemotherapy, and may contribute to intrinsic drug resistance.

Figure 1.

ABCG2 expression and function in myeloma cell lines. (A) Flow cytometric analysis of MM cell lines for ABCG2 membrane expression was performed. 8226MR cells express more ABCG2 than wild-type 8226 cells, and H929 cells express very little, as shown by the shift in fluorescence. (B) Immunostaining of myeloma cell lines using an anti-ABCG2 FITC (Chemicon)–labeled antibody (ABCG2 is green and DAPI is blue). (C) Western blot of protein (25 μg/lane) extracted from myeloma cell lines for ABCG2. (D) Functional analysis of ABCG2 using topotecan as a substrate and the ABCG2-specific inhibitor tryprostatin A. Topotecan, a very good ABCG2 substrate and a naturally fluorescent molecule, is effluxed in high (8226MR) and moderate (8226) ABCG2 expressers, but is accumulated by H929 cells (which do not express ABCG2). Tryprostatin A (trypA) blocks the efflux of topotecan, demonstrating that topotecan efflux is ABCG2 dependent.

Figure 2.

Functional assay in patient myeloma cells. Patient samples with high ABCG2 and low ABCG2 mRNA (as measured by QPCR) were assayed for ABCG2 function. The high ABCG2 expresser effluxed topotecan more efficiently than the lower expresser. Efflux was shown to be ABCG2 specific by the addition of tryprostatin A.

Figure 3.

ABCG2 expression increases in response to doxorubicin exposure. ABCG2 functional expression was assayed by flow cytometry in 8226 and 8226MR MM cells after exposure to 1 μM doxorubicin for 20 minutes. Higher ABCG2-expressing 8226MR cells were able to efflux doxorubicin more efficiently than parental 8226 cells. The ABCG2-specific inhibitor tryprostatin A decreased efflux in the 8226MR cell line but not the 8226 parental cell line, indicating that doxorubicin efflux was mediated by ABCG2. Myeloma cells treated with low-dose doxorubicin, 0.1 μM in 8226MR cells and 1.0 μM in 8226 cells, exhibit an increase in protein expression as determined by Western analysis (inset of each graph). Equal amounts of protein (25 μg) were assayed. Both 8226 and 8226MR cells demonstrated a 1.7-fold increase in ABCG2 protein after low-dose doxorubicin treatment.

Figure 4.

ABCG2 expression is elevated in log-phase myeloma cells. (A) Flow cytometric data using an ABCG2 antibody (Chemicon) demonstrate that ABCG2-expressing 8226 cells have increased ABCG2 at log-phase density compared with log-phase H929 cells. (B-C) The FACScan data are confirmed by immunostaining for ABCG2 (B), and by QPCR and Western analyses (C). Densitometry analysis of the immunoblot shows a 4:1 ratio of log-plateau ABCG2 in 8226 cells, and a 2:1 ratio of log-plateau 8226MR ABCG2. Error bars represent the standard deviation for 3 separate experiments. Note, 8226 and 8226MR Western blots were exposed for different time intervals and do not reflect relative protein levels.

Figure 5.

ABCG2 expression increases in response to topotecan chemotherapy. (A-B) Multiple myeloma 8226MR, 8226, and H929 cells treated with low-dose topotecan (B) exhibit an increase in membrane ABCG2 over the no-drug control (A), as shown by immunostaining with ABCG2 antibody (Bcrp1 FITC). (C) Protein expression of ABCG2 was quantified as pixel intensity (from immunofluorescence microscopy), and also assessed by Western analysis (inset of each graph). (D) ABCG2 expression was measured by flow cytometry and showed an increase in log-phase 8226MR and 8226 cells after exposure to low-dose topotecan (1 μM) for 24 hours. The ABCG2-nonexpressing cell line (H929) shows no increase in ABCG2 antibody binding. Red indicates control cells in the absence of topotecan, and green represents myeloma cells exposed to topotecan.

Figure 6.

ABCG2 increases in patient plasma cells after high-dose chemotherapy and at relapse. (A-B) A patient bone marrow aspirate taken before (A) and during (B) HDC with melphalan and topotecan exhibited an increase in immunofluorescence of ABCG2 (green). (C) This same patient demonstrated an increase in ABCG2 expression by immunofluorescence at relapse as well. (D) ABCG2 protein expression before HDC and at relapse assessed by Western blot in 4 patients on the MTV study. Laser densitometry analysis of the immunoblots shows a 1.54- and 1.94-fold increase in ABCG2 during HDC for patients A and B, respectively, and a 3.68- and 1.34-fold increase in ABCG2 at relapse for patients C and D, respectively. In all cases, this is relative to the pre-HDC ABCG2 protein expression.

Figure 7.

ABCG2 promoter methylation. (A) Cells were harvested at plateau phase, and the DNA was extracted and assayed by bisulfite DNA sequencing analysis. The figure shows the methylation status of the putative CpG island of the ABCG2 promoter in 4 cell lines. Filled circles represent methylated groups and the open circles, demethylated CpG. (B) H929/5aza are cells treated with 100 nM 5-aza-2′-deoxycytidine for 72 hours and the ABCG2 promoter was assayed by bisulfite sequencing. 5-Aza-2′-deoxycytidine was able to augment ABCG2 transcription in low ABCG2-expressing H929 cells but had no effect in moderate-expressing 8226 cells. (C) CD138-selected MM patient samples were assayed for ABCG2 promoter methylation after bisulfite conversion using real-time quantitative PCR. The percentage of alleles that was methylated inversely correlated with ABCG2 mRNA expression. Error bars represent the standard deviation for 3 separate experiments.