In vitro folding of KvAP, a voltage-gated K+ channel

Abstract

In this contribution, we report in vitro folding of the archaebacterial voltage-gated K(+) channel, K(v)AP. We show that in vitro folding of the K(v)AP channel from the extensively unfolded state requires lipid vesicles and that the refolded channel is biochemically and functionally similar to the native channel. The in vitro folding process is slow at room temperature, and the folding yield depends on the composition of the lipid bilayer. The major factor influencing refolding is temperature, and almost quantitative refolding of the K(v)AP channel is observed at 80 °C. To differentiate between insertion into the bilayer and folding within the bilayer, we developed a cysteine protection assay. Using this assay, we demonstrate that insertion of the unfolded protein into the bilayer is relatively fast at room temperature and independent of lipid composition, suggesting that temperature and bilayer composition influence folding within the bilayer. Further, we demonstrate that in vitro folding provides an effective method for obtaining high yields of the native channel. Our studies suggest that the K(v)AP channel provides a good model system for investigating the folding of a multidomain integral membrane protein.

Fig. 1. Structure of the K_vAP channel

(A) Top view of the tetrameric K_vAP channel. The structural model of the K_vAP channel as described in (24) is shown. (B) Structure of a single subunit of the K_vAP channel. Each subunit consists of two domains, the sensor domain and the pore domain. SDS-PAGE gel showing glutaraldehdye crosslinking of the native K_vAP channel (C) and the unfolded protein (D). Lanes 1) without crosslinker; 2) 0.025%; 3) 0.05% and 4) 0.1% glutaraldehyde. The oligomeric nature (1×, 2×, 3× & 4×) of the crosslinked bands is indicated.

Fig. 2. In vitro folding of the K_vAP channel

Refolding was tested after dilution of the unfolded protein into detergents (DM, Fos-12, DDM and OG; 2% w/v), asolectin lipid vesicles (Aso) or asolectin lipid vesicles solubilized with DM or Fos-12 (2% w/v). Glutaraldehyde crosslinking followed by SDS-PAGE was used to determine formation of tetrameric species. (−) without crosslinker and (+) 0.1% glutaraldehyde.

Fig. 3. Characterization of the refolded K_vAP channel

(A) Size exclusion chromatography of the refolded K_vAP channel. Elution volume observed = 12.5 ml. Inset: Glutaraldehyde crosslinking of the peak fraction. Lanes 1) without crosslinker; 2) 0.025%; 3) 0.05% and 4) 0.1% glutaraldehyde. (B) CD spectra of the refolded K_vAP channel. CD spectra reported as mean residue ellipiticity (MRE) for the native (red) and the refolded K_vAP channel (green) in 10mM sodium phosphate buffer pH 7.5, 150mM KCl, 0.25% DM. A protein concentration of 450 µg/ml was used. (C) Fluorescence spectra of the refolded K_vAP channel. Intrinsic fluorescence spectra (excitation = 280 nm) for the native (red), refolded (green) recorded in 50 mM HEPES pH 7.5, 150 mM KCl and 0.25% DM and unfolded K_vAP in 100 mM sodium phosphate pH 7.5, 1% SDS (black). A protein concentration of 25 µg/ml was used.

Fig. 4. Functional characterization of the refolded K_vAP channel

(A) Single channel trace for the refolded K_vAP channel recorded at +100 mV. (B) Single channel current as a function of voltage for the native (red) and refolded (green) K_vAP channels. (C) Voltage activated macroscopic currents from the refolded K_vAP channel recorded using the voltage protocol shown (inset) (D) Voltage gating of the refolded K_vAP channel. The fraction of the maximal current observed was plotted as a function of the test potential (see methods). The smooth line corresponds to a Boltzmann function with a V_0.5 = −36.1 ± 4.3 and z =1.86 ± 0.36. (E) Slow inactivation of the refolded K_vAP channel. K_vAP currents were elicited from a holding potential of −150 mV by a depolarization to +100 mV. The inactivation time constant was determined by fitting the decay in current after peak activation to a single exponential function. The recordings were carried out in 150 mM KCl, 10 mM HEPES-KOH pH 7.5. For Panels B and D, the error bars indicate the standard deviation for 3 or more experiments.

Fig. 5. Refolding of the K_vAP channel in various lipids

(A) SDS-PAGE gel showing glutaraldehyde crosslinking of the K_vAP channel refolded in POPC, POPC: POPG (75: 25) and DPhPC lipid vesicles for 2 hours at room temperature (−, without crosslinker and +, 0.1% glutaraldehdye) and (B) bar graph comparing the refolding yields in the various lipids. Refolding yields of K_vAP in asolectin vesicles (Aso) under similar refolding conditions is also shown. (C) SDS-PAGE gel and (D) bar graph showing the effect of increasing the ratio of POPE in POPC: POPE lipid vesicles on refolding of the K_vAP channel. The extent of refolding after 2 hours at room temperature is shown. (E) Time course of refolding of the K_vAP channel in POPC (circles), DPhPC (triangles) and POPC: POPE (60: 40, squares) lipid vesicles. The smooth lines for DPhPC and POPC: POPE corresponds to a single exponential fit while a line joining points is shown for POPC. For Panels B, C and E, the error bars indicate the standard deviation for 3 or more experiments.

Fig. 6. Effect of temperature on refolding of the K_vAP channel

(A) SDS-PAGE gel showing refolding of K_vAP in DPhPC lipid vesicles after 10 min incubation at the indicated temperatures. (B) The refolding yields after 10 min are plotted as a function of temperature. Error bars indicate standard deviation (n=3). Time course of refolding of the K_vAP channel in DPhPC (C) and POPC (D) at 25°C (squares) and 80°C (circles). The solid lines represent single exponential fits to the data.

Fig. 7. Insertion of the K_vAP polypeptide into the lipid bilayer during refolding

(A) Cartoon depicting the location of Cys residues introduced into the K_vAP channel for the PEGylation experiment. (B) Accessibility of Cys residues to PEGylation immediately after dilution of unfolded protein into lipid vesicles or into buffer containing 2% (w/v) Fos12 at room temperature. SDS-PAGE gels showing modification of the single Cys mutants of K_vAP by PEG-2K-mal. The slow migrating PEGylated protein (solid arrows) and fast migrating protein without modification (open arrows) are indicated. (C) Summary of the PEGylation experiment. Upon dilution of the unfolded K_vAP polypeptide into lipid vesicles, the Cys residues in the transmembrane segments (red) became inaccessible to PEGylation while the Cys residues in the loops (blue) remain accessible. In contrast, all the Cys residues are accessible to PEGylation when the unfolded K_vAP polypeptide is diluted into detergent micelles or into lipid vesicles solubilized by Fos12 (2%, w/v).