Functional reconstitution into liposomes of purified human RhCG ammonia channel

Abstract

Background: Rh glycoproteins (RhAG, RhBG, RhCG) are members of the Amt/Mep/Rh family which facilitate movement of ammonium across plasma membranes. Changes in ammonium transport activity following expression of Rh glycoproteins have been described in different heterologous systems such as yeasts, oocytes and eukaryotic cell lines. However, in these complex systems, a potential contribution of endogenous proteins to this function cannot be excluded. To demonstrate that Rh glycoproteins by themselves transport NH(3), human RhCG was purified to homogeneity and reconstituted into liposomes, giving new insights into its channel functional properties.

Methodology/principal findings: An HA-tag introduced in the second extracellular loop of RhCG was used to purify to homogeneity the HA-tagged RhCG glycoprotein from detergent-solubilized recombinant HEK293E cells. Electron microscopy analysis of negatively stained purified RhCG-HA revealed, after image processing, homogeneous particles of 9 nm diameter with a trimeric protein structure. Reconstitution was performed with sphingomyelin, phosphatidylcholine and phosphatidic acid lipids in the presence of the C(12)E(8) detergent which was subsequently removed by Biobeads. Control of protein incorporation was carried out by freeze-fracture electron microscopy. Particle density in liposomes was a function of the Lipid/Protein ratio. When compared to empty liposomes, ammonium permeability was increased two and three fold in RhCG-proteoliposomes, depending on the Lipid/Protein ratio (1/300 and 1/150, respectively). This strong NH(3) transport was reversibly inhibited by mercuric and copper salts and exhibited a low Arrhenius activation energy.

Conclusions/significance: This study allowed the determination of ammonia permeability per RhCG monomer, showing that the apparent Punit(NH3) (around 1x10(-3) microm(3)xs(-1)) is close to the permeability measured in HEK293E cells expressing a recombinant human RhCG (1.60x10(-3) microm(3)xs(-1)), and in human red blood cells endogenously expressing RhAG (2.18x10(-3) microm(3)xs(-1)). The major finding of this study is that RhCG protein is active as an NH(3) channel and that this function does not require any protein partner.

Figure 1. Quality control of RhCG purification.

A 10% SDS PAGE was divided in two parts (A and B) after migration. A, Silver staining. lanes 1 to 3: aliquots of the three Washes (W1, W2, W3); lane 4: HA peptide Eluted fraction (Elu); lane 5: molecular weight Standards (Std) from Invitrogen. B, Western blot probed with an anti HA and revealed with Amersham ECL Western blotting detection reagent (GE Healthcare, Buckinghamshire, UK), RhCG can be detected in starting material (SM) on lane 6 and in the HA peptide Eluted fraction (Elu) on line 8 but not in the flow through (FT) on lane 7. Arrows indicate the two bands of 50 kDa and 110 kDa detected by both silver staining and western blot analysis, most likely corresponding to monomeric and oligomeric forms of RhCG.

Figure 2. TEM of negatively stained RhCG particles.

The electron micrograph reflects the homogeneity of the purified protein. Scale bar is 50 nm. Arrows indicate some top-views of RhCG. In the insert, class averages corresponding to top-views of RhCG are displayed. Scale bar is 10 nm.

Figure 3. TEM of RhCG incorporated into liposomes.

A, Cryo-electron microscopy of vitrified proteoliposomes. Arrows show unilamelar structures. Scale bar is 250 nm. B and C, Freeze fracture electron microscopy of empty liposomes and proteoliposomes (LPR 150) respectively. The arrows indicate RhCG particles. Scale bar is 250 nm.

Figure 4. Stopped-flow analysis of ammonia (A) or methylamine (B) fluxes in liposomes and RhCG-proteoliposomes.

Individual time courses of fluorescence change were obtained after submitting the empty liposomes and proteoliposomes containing purified RhCG protein to a 10 meq inwardly directed gradient of ammonium (A) or methylammonium (B). In (A) two different LPRs (Lipid/Protein ratio) were used: 150 and 300. Insert in A represents the Arrhenius plots of temperature-dependent ammonia permeation allowing the determination of activation energy (Ea) values from the slopes.

Figure 5. Stopped flow analysis of the influence of divalent cations on NH₃ fluxes.

The effect of Hg⁺⁺ and Cu⁺⁺ on individual time courses of fluorescence change were analysed in RhCG-proteoliposomes after incubation with 0.1 mM HgCl₂ (A) or 0.21 mM CuSO₄ (B). For reversion, the β-mercaptoethanol (βme at 5 mM) and the Gly-Gly-His peptide (GGH at 2 mM) were used. Results (means of several independent experiments +/− SEM) regarding the effect of divalent cations (A:+Hg⁺⁺ and B:+Cu⁺⁺) and reversions (A: +βme and B:+GGH) on transport activities (corresponding to alkalinisation rate constants after removal of the constant value of empty liposomes) are reported in inserts. The number of experiments is mentioned above each bar.

Figure 6. Stopped-flow analysis of water osmotic variations in liposomes and RhCG-proteoliposomes.

Individual time courses of fluorescence quenching in RhCG-proteoliposomes were obtained at 20°C with empty liposomes and RhCG-proteoliposomes submitted to a 50 mosmol/kg H₂O osmotic gradient. Inserts represents the Arrhenius plots of temperature-dependent water permeation allowing the determination of activation energy (Ea) values from the slopes.