Increased expression of the Abcg2 transporter during erythroid maturation plays a role in decreasing cellular protoporphyrin IX levels

Abstract

ABCG2/BCRP is a member of the adenosine triphosphate-binding cassette (ABC) transporter family and is expressed in intestine, kidney, and liver, where it modulates the absorption and excretion of xenobiotic compounds. ABCG2 is also expressed in hematopoietic stem cells and erythroid cells; however, little is known regarding its role in hematopoiesis. Abcg2 null mice have increased levels of protoporphyrin IX (PPIX) in erythroid cells, yet the mechanism for this remains uncertain. We have found that Abcg2 mRNA expression was up-regulated in differentiating erythroid cells, coinciding with increased expression of other erythroid-specific genes. This expression pattern was associated with significant amounts of ABCG2 protein on the membrane of mature peripheral blood erythrocytes. Erythroid cells engineered to express ABCG2 had significantly lower intracellular levels of PPIX, suggesting the modulation of PPIX level by ABCG2. This modulating activity was abrogated by treatment with a specific ABCG2 inhibitor, Ko143, implying that PPIX may be a direct substrate for the transporter. Taken together, our results demonstrate that ABCG2 plays a role in regulating PPIX levels during erythroid differentiation and suggest a potential role for ABCG2 as a genetic determinant in erythropoietic protoporphyria.

Fig 1. Expression of Abcg2 in MEL cells before and after induction with DMSO, and in primary erythroblasts during differentiation

(A) Non-induced MEL cells and MEL cells induced with 2% DMSO for 4 days were analyzed for Abcg2 mRNA by Northern blot, GAPDH probe served as loading control. (B) Flow cytometry analysis of MEL cells after staining with anti-Abcg2 antibody Bxp-53. (C) Hoechst 33342 fluorescence in cells analyzed by flow cytometry. Shaded area in B and C: non-induced MEL cells. Solid line in B and C: MEL cells induced with DMSO. (D) Murine splenic proerythroblast were cultured for differentiation, samples were taken at different time points and measured for Abcg2 mRNA using microarray.

Fig 2. Expression of ABCG2 protein on peripheral mature red blood cells

Peripheral blood from mice, rhesus macaques and humans are collected, stained with primary antibodies recognizing respective ABCG2, and analyzed by flow cytometry. (A) red blood cells from Abcg2−/− mice (shaded area) or wild type mice (solid line) stained with anti-Abcg2 antibody Bxp-53. (B) rhesus macaque red blood cells stained with either isotype control antibody (shaded area) or anti-ABCG2 antibody Bxp-21 (solid line). (C) Human red blood cells stained with either isotype control antibody (shaded area) or anti-ABCG2 antibody 5D3 (solid line).

Fig 3. Efflux of exogenous PPIX by ABCG2

K562 cells or K562 cells engineered to overexpress ABCG2 (K562/ABCG2) are incubated with Hoechst 33342 (A, B) or PPIX (C, D). Cell were also coincubated with the ABCG2 inhibitor Ko143 (B, D). Shaded area: K562 cells. Solid line: K562/ABCG2 cells.

Fig 4. Efflux of endogenous PPIX by ABCG2

(A) K562 cells or K562/ABCG2 cells were incubated with 1mM ALA for 21 hours to induce endogenous PPIX, with or without 1uM Ko143, and analyzed for PPIX fluorescence. K562+ALA (Blue line); K562/ABCG2+ALA (Red line); K562/ABCG2+ALA+Ko143 (Brown line); Non-treated K562 cells (Green line); Non-treated K562/ABCG2 cells (Black line). (B) MEL cells were incubated with 1mM ALA for 21 hours with or without 1uM Ko143 and analyzed in flow cytometry for PPIX fluorescence. MEL+ALA (Red line); MEL+ALA+Ko143 (Blue line); Non-treated MEL cells (Black line). (C) K562 or K562/ABCG2 cells were incubated with 1mM ALA for 21 hours and pelleted. The amount of PPIX in pelleted cells was measured in a HPLC assay. (D) An aliquot of cells was also analyzed in a flow cytometry for fluorescence. PPIX was undetectable in non-treated K562 cells as shown in the bottom panels.

Fig 4. Efflux of endogenous PPIX by ABCG2

(A) K562 cells or K562/ABCG2 cells were incubated with 1mM ALA for 21 hours to induce endogenous PPIX, with or without 1uM Ko143, and analyzed for PPIX fluorescence. K562+ALA (Blue line); K562/ABCG2+ALA (Red line); K562/ABCG2+ALA+Ko143 (Brown line); Non-treated K562 cells (Green line); Non-treated K562/ABCG2 cells (Black line). (B) MEL cells were incubated with 1mM ALA for 21 hours with or without 1uM Ko143 and analyzed in flow cytometry for PPIX fluorescence. MEL+ALA (Red line); MEL+ALA+Ko143 (Blue line); Non-treated MEL cells (Black line). (C) K562 or K562/ABCG2 cells were incubated with 1mM ALA for 21 hours and pelleted. The amount of PPIX in pelleted cells was measured in a HPLC assay. (D) An aliquot of cells was also analyzed in a flow cytometry for fluorescence. PPIX was undetectable in non-treated K562 cells as shown in the bottom panels.

Fig 5. ABCG2 effluxes endogenous PPIX into medium

K562 or K562/ABCG2 cells were incubated with 1mM ALA for 7 hours and the PPIX levels in both medium and within cells were measured by HPLC. The peak of PPIX is indicated in each panel. The peak on the left represents a component in the culture medium (not produced by cells) and serves as an internal control.

Fig 6. Red blood cells efflux PPIX due to expression of ABCG2

(A) Murine peripheral red blood cells were incubated with PPIX, with Ko143 (solid line) or without Ko143 (shaded area) and analyzed for PPIX fluorescence. (B) Murine peripheral red blood cells were incubated with (solid line) or without (shaded area) 2-deoxyglucose and sodium azide and analyzed for PPIX fluorescence.