Structural basis for the modulation of MRP2 activity by phosphorylation and drugs

Abstract

Multidrug resistance-associated protein 2 (MRP2/ABCC2) is a polyspecific efflux transporter of organic anions expressed in hepatocyte canalicular membranes. MRP2 dysfunction, in Dubin-Johnson syndrome or by off-target inhibition, for example by the uricosuric drug probenecid, elevates circulating bilirubin glucuronide and is a cause of jaundice. Here, we determine the cryo-EM structure of rat Mrp2 (rMrp2) in an autoinhibited state and in complex with probenecid. The autoinhibited state exhibits an unusual conformation for this class of transporter in which the regulatory domain is folded within the transmembrane domain cavity. In vitro phosphorylation, mass spectrometry and transport assays show that phosphorylation of the regulatory domain relieves this autoinhibition and enhances rMrp2 transport activity. The in vitro data is confirmed in human hepatocyte-like cells, in which inhibition of endogenous kinases also reduces human MRP2 transport activity. The drug-bound state reveals two probenecid binding sites that suggest a dynamic interplay with autoinhibition. Mapping of the Dubin-Johnson mutations onto the rodent structure indicates that many may interfere with the transition between conformational states.

Fig. 1. Cryo-EM structure of nucleotide-free rMrp2.

a Chromatogram and SDS-PAGE analysis of Size Exclusion Chromatography of purified rMrp2. Representatives from three purifications. b Effect of drugs on the basal ATPase activity of rMrp2 reconstituted in destabilised liposomes. The basal ATPase activity of rMrp2 can be stimulated by 1 mM probenecid (PRB) and 0.1 mM methotrexate (MTX). The E1458Q mutant displays no ATPase activity. Results are represented as means with individual data points indicated by asterisc, from three independent experiments. Significantly different as estimated by one way ANOVA (p = 2.32 ×10⁻⁷) followed by Tukey test p < 0.001 (red triangle: p < 0.01)). c MRP2 topology with the key domains highlighted. d Ab initio cryo-EM map of rMrp2 at 3.21 Å resolution. Continuous density can be observed for the different rMrp2 domains. The CHS molecules and GDN micelle are shown in pink and transparent grey, respectively. e The nucleotide-free rMrp2 atomic model displays an inward-facing conformation with the TMD cavity occupied by the R-domain; the helical R-domain is shown in orange. Source data are provided as a Source Data file.

Fig. 2. rMrp2 exhibits an autoinhibited conformation.

a Continuous helical density is observed for the R-domain (shown in orange cartoon for reference). b The R-domain is stabilised by several hydrogen bonds and van der Waals interactions within the TMD. The side chains involved in hydrogen bonds (shown in black dashed lines) are shown in stick. c Comparison of the conformation of the R-domain between rMrp2 (partially phosphorylated; (R-domain in orange)), CFTR (dephosphorylated (left panel; PDBID: 5UAK) and phosphorylated (right panel; PDBID: 6MSM); R-domain in red) and Ycf1 (phosphorylated (PDBID: 7M69); R-domain and resolved phosphosites (spheres) in red).

Fig. 3. Effect of phosphorylation on the activity of rMrp2.

a The phosphorylation state of rMrp2 was analysed by Pro-Q Diamond Phosphoprotein Stain (top panel) and Coomassie blue (bottom panel) of the SDS-PAGE. Densitometry analysis shows a 48% enhancement of phosphorylation upon in vitro treatment with the HEK293 kinase extract; white and black bars correspond to the top and bottom panels of the SDS-PAGE, respectively. Phosphorylation state abbreviations: rMrp2_R-* (partially phosphorylated), rMrp2_R-de (fully dephosphorylated) and rMrp2_R-pho (fully phosphorylated). Representative SDS-PAGE from three repeats. b (left panel) Mass spectrometry-based phosphorylation mapping of rMrp2 protein showing presence or absence of detection across different conditions. The phosphorylated residues have been mapped onto the rMrp2 structure (right panel). The phosphorylated residues that are present in both the rMrp2_R-* and rMRP2_R-pho samples are shown as blue spheres. The sites that are unique to the rMrp2_R-pho sample are shown as red spheres; the additionally phosphorylated R-domain residues are labelled for clarity (a star indicates a phosphorylated R-domain residue that has not been built in the structure). c Effect of probenecid on the ATPase activity of rMrp2 in different phosphorylation states. Dephosphorylation by λ-protein phosphatase does not affect the basal ATPase (white bars), whereas full phosphorylation increases the basal ATPase by 2-fold. Probenecid can stimulate the basal ATPase activity of rMrp2_R-* and rMRP2_R-de but it has no additional effect on rMrp2_R-pho (grey bars). Results are represented as means from three independent experiments. Significantly different significantly different as estimated by one way ANOVA (p = 2 ×10⁻⁷) followed by Tukey test p < 0.025 (red, yellow and green triangles:not significant). d rMrp2-dependent uptake of CDF in proteoliposomes. Transport was initiated by adding (empty bars and light grey bars) or not (dark grey bars) ATP-MgCl₂, to the samples containing (light grey) or not (empty and dark grey) 1 mM probenecid. rMrp2_R-pho displays significantly increased transport activity relative to the rMrp2_R-* and rMrp2_R-de proteins. Probenecid inhibits CDF transport. Results are represented as means from three independent experiments. Significantly different significantly different as estimated by one way ANOVA (p = 1.6 ×10⁻¹⁹) followed by Tukey test p < 0.025 (red, yellow and green triangles: not significant). Source data are provided as a Source Data file.

Fig. 4. Human MRP2 activity is reduced by kinase inhibition in hepatocyte-like cells.

a Live cell imaging of CDF (green) accumulation in the bile canaliculi of iHEPs. Nuclei are stained with Hoechst 33342 (blue). Single grayscale images and merged multicolour images are shown. b Integrated density image analysis shows a 50% reduction in CDF transport into the canaliculi upon treatment with staurosporine. Results are represented as means from 4 iHEP batches (n = 3 images per condition per batch). Significantly different as estimated by one-tail paired t-test (p = 8.01 × 10⁻⁴). Source data are provided as a Source Data file.

Fig. 5. Cryo-EM structure of drug-bound rMrp2.

a Density for two probenecid molecules (shown in orange sticks) and one CHS molecule (shown in grey sticks) within the TMD; their structures are shown for reference. b Binding of probenecid displaces the R-domain from the TMD. c The two probenecid molecules are coordinated by hydrogen bonds and π-stacking interactions within the TMD. Left panel is a view along the membrane and right panel is a view from the cytosol. d The probenecid bound structure defines the two drug binding sites that are in close proximity to each other.

Fig. 6. rMrp2 structural changes upon drug binding.

a Displacement of the R-domain upon drug binding results in movement of both the TMDs and NBDs; both the cytosolic half-end of the TMDs and the NBDs display an 8 Å closure upon drug binding whereas no significant conformational changes are observed towards the drug binding site. b The two drug binding sites partly overlap with the R-domain. The probenecid bound (orange sticks) rMrp2 is shown as pink cartoons whereas the autoinhibited rMrp2 as blue cartoons. c Drug binding sites 1 and 2 overlap to a large extent with the substrate binding site of bMrp1. bMrp1 (grey cartoon) bound with LTC4 (green sticks) has been superimposed onto rMrp2 (pink) bound with probenecid (orange sticks). Drug binding site 2 is below the substrate binding site. The CHS molecule (grey sticks) fills the hydrophobic space of the H-pocket.

Fig. 7. Dubin–Johnson syndrome mutations.

Mutations that cause Dubin–Johnson syndrome have been mapped onto the rMrp2 structure; they are shown as green spheres and labelled according to the MRP2 sequence. X refers to a stop codon and del2 to a two-residue deletion. Most mutations are found on the interface of the coupling helices and the NBDs.