Functional Rescue of Trafficking-Impaired ABCB4 Mutants by Chemical Chaperones

Abstract

Multidrug resistance protein 3 (MDR3, ABCB4) is a hepatocellular membrane protein that mediates biliary secretion of phosphatidylcholine. Null mutations in ABCB4 gene give rise to severe early-onset cholestatic liver disease. We have previously shown that the disease-associated mutations p.G68R, p.G228R, p.D459H, and p.A934T resulted in retention of ABCB4 in the endoplasmic reticulum, thus failing to target the plasma membrane. In the present study, we tested the ability of two compounds with chaperone-like activity, 4-phenylbutyrate and curcumin, to rescue these ABCB4 mutants by assessing their effects on subcellular localization, protein maturation, and phospholipid efflux capability. Incubation of transfected cells at a reduced temperature (30°C) or exposure to pharmacological doses of either 4-PBA or curcumin restored cell surface expression of mutants G228R and A934T. The delivery of these mutants to the plasma membrane was accompanied by a switch in the ratio of mature to inmature protein forms, leading to a predominant expression of the mature protein. This effect was due to an improvement in the maturation rate and not to the stabilization of the mature forms. Both mutants were also functionally rescued, displaying bile salt-dependent phospholipid efflux activity after addition of 4-PBA or curcumin. Drug-induced rescue was mutant specific, given neither 4-PBA nor curcumin had an effect on the ABCB4 mutants G68R and A934T. Collectively, these data indicate that the functionality of selected trafficking-defective ABCB4 mutants can be recovered by chemical chaperones through restoration of membrane localization, suggesting a potential treatment for patients carrying such mutations.

Fig 1. Low temperature restores apical membrane localization of particular ABCB4 mutants.

(A) Representative double immunofluorescence staining for ABCB4 (red) and the ER marker calnexin (green) in MDCK-II cells expressing wild-type ABCB4 after incubation at 37°C or 30°C. Areas of yellow in the merged images indicate colocalization of ABCB4 and calnexin. Optical horizontal x,y (upper) and vertical x,z sections (lower) are shown. (B) Confocal microscopy images of cells expressing ABCB4 mutants. For simplicity, only the merged images are shown. Reducing temperature improved cell surface localization for the mutants G228R and A934T, as revealed by the appearance of red signal around the apical membrane. All images represent three independent transfections. Bars, 20 μm.

Fig 2. Chemical chaperones rescue cell surface expression of the ABCB4 mutants G228R and A934T.

Transfected MDCK-II cells were treated with 0.1% DMSO (vehicle), 1 mM 4-PBA or 1 μM curcumin during 24 h. The cells were double stained for ABCB4 and calnexin and analyzed by confocal microscopy. The merged images of ABCB4 and calnexin are presented. Co-localization (yellow) of calnexin with G68R or D459H mutants was seen after addition of any of the chaperone drugs. Positive apical staining for ABCB4 (red) was found in cells expressing the G228R and A934T mutants following treatment with either 4-PBA or curcumin. The images are representative of four independent experiments, in which at least two different batches of each plasmid were used. Bars, 20 μm.

Fig 3. Effect of chemical chaperones on the processing of mutant ABCB4 proteins.

(A) Western blots with anti-ABCB4 (top) and anti-Na/K-ATPase (bottom) antibodies of cell lysates (30 μg) from AD-293 transfected cells untreated or treated with 1 mM 4-PBA or 1 μM curcumin. The images are representative of four independent experiments. Numbers below the blots represent the mean±SD of the ratio between mature and immature forms of ABCB4, as determined by densitometric quantification of the signals. (B) Real-time quantitative PCR analysis. ABCB4 mRNA measurements were normalized to β-actin. Values represent the mean±SD of four independent experiments and are expressed relative to untreated cells (Control).

Fig 4. Effects of 4-PBA on the stability and maturation of the G228R and A934T mutants.

(A) AD-293 cells expressing wild-type ABCB4 or the G228R or A934T mutants were incubated with cycloheximide (50 μg/ml) for 8 or 16 h in the presence or absence of 1mM 4-PBA. Equal amounts of proteins (30 μg) were fractionated by 6% SDS-PAGE and subjected to western blotting. (B) Transfected cells were untreated (control) or treated for 2 h with BFA (1 μg/ml). BFA-treated cells were subsequently rinsed and further incubated for 8 h with or without 1 mM 4-PBA. The total amount of protein loaded onto each lane was 30 μg. The images are representative of three independent experiments.

Fig 5. Chemical chaperones rescue phospholipid efflux activity of the ABCB4 mutants G228R and A934T.

Transfected AD-293 cells were labeled with 2 μCi/mL of [3H]-choline and incubated for 3 h in EMEM with or without 1mM NaTC, as indicated. Aliquots of the culture medium were harvested each 30 min and the radioactivity released was measured by liquid scintillation counting. Values were normalized to the levels of ABCB4 mRNA and are expressed relative to total cell protein. (A and B) Time course of [³H] release in the presence or absence of NaTC in cells expressing wild-type ABCB4 (A) or ABCB4 mutants (B). (C) NaTC-dependent efflux of radioactivity in cells expressing wild-type ABCB4 previously treated for 24 h with 0.1% DMSO (control), 1 mM 4-PBA, or 1 μM curcumin. (D) NaTC-dependent efflux of radioactivity in cells expressing ABCB4 mutants after treatment with 1 mM 4-PBA or 1 μM curcumin; H₂O and 0.1% DMSO were used as vehicle, respectively. A representative experiment out of four is shown. For cells expressing the mutants G228R and A934T, differences in NaTC-dependent radioactivity efflux rate between vehicle- and 4-PBA- or curcumin-treated cells were statistically significant (p < .05).