Real Time Ligand-Induced Motion Mappings of AChBP and nAChR Using X-ray Single Molecule Tracking

We observed the dynamic three-dimensional (3D) single molecule behaviour of acetylcholine-binding protein (AChBP) and nicotinic acetylcholine receptor (nAChR) using a single molecule tracking technique, diffracted X-ray tracking (DXT) with atomic scale and 100 μs time resolution. We found that the combined tilting and twisting motions of the proteins were enhanced upon acetylcholine (ACh) binding. We present the internal motion maps of AChBP and nAChR in the presence of either ACh or α-bungarotoxin (αBtx), with views from two rotational axes. Our findings indicate that specific motion patterns represented as biaxial angular motion maps are associated with channel function in real time and on an atomic scale.


Preparation of recombinant AChBP
AChBP cDNA, originally isolated from the Aplysia kurodai CNS (S. Kobayashi, B.-K. Kaang and T. Kubo, manuscript in preparation), was amplified by PCR using a forward primer encoding a hexahistidine-(His-) tag sequence and a reverse primer encoding a Met-tag (MGGMGGM, Okada et al. Bioconjugate Chemistry 22:887 (2011)) sequence. The amplified DNA fragment was subcloned into the plasmid pQE-30 (QIAGEN). Met-tagged AChBP was expressed in E. coli XL1Blue MRF (Stratagene, La Jolla, CA) and induced by 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG). The AChBP accumulated in inclusion body pellets was solubilised in solubilisation buffer (50 mM CHAPS pH 11, 0.3% N-lauroylsarcosine and 1 mM DTT). Solubilised AChBP was refolded using a Protein Refolding Kit (Merck, Germany) and purified with Ni-NTA Agarose (QIAGEN), according to the manufacturer's protocol. To check the quality of the refolded and purified AChBP, it was assessed for binding of several nAChR ligands by Biacore (GE Healthcare, NJ). The dissociation constant of AChBP was 29 nM for αBtx, which is comparable to the dissociation constant of nAChR. The immobilisation efficiency of gold nanocrystal for AChBP was enhanced by inserting three Met residues at the C-terminal of AChBP.

Preparation of nAChR
A Torpedo californica electric organ was used to prepare nAChR-rich membrane vesicles, which permeate cations in an agonist-dependent manner (H.-P. P. Moore et al., Proc. Natl. Acad. Sci. U. S. A. 77, 4509-13 (1980), as described by Brisson andUnwin (J. Cell Biol. 99:1202 (1984).), with minor modifications as follows: 40 g of electric organ was homogenised using an Omni Mixer Homogeniser (OMNI International) for 1 min at maximum speed in 50 ml of buffer A (400 mM NaCl, 10 mM N-ethylmaleimide, 20 mM sodium phosphate, pH 7.4) supplemented with 1/2 concentration of Complete Protease Inhibitor Cocktail (Roche). After centrifugation at 6,000 rpm for 10 min at 4°C (Beckman JLA-10.500), the supernatant was filtered through four layers of sterile gauze. The filtrate was centrifuged at 20,000 rpm at 4°C for 30 min in a P45AT rotor (Hitachi). The pellet was resuspended in 20 ml of buffer A using a Potter-Elvehjem homogeniser and centrifuged at 30,000 rpm at 4°C for 50 min using a P45AT rotor. The pellet was resuspended in 50 ml of buffer B (100 mM sodium phosphate, pH 7.0) using a Potter-Elvehjem homogeniser supplemented with 1/2 concentration of Complete Protease Inhibitor Cocktail and then sealed in a glass cylinder tube at 4°C for 7 days to spontaneously form a density gradient. The suspension was divided into 2-ml aliquots, using a disposable pipette tip and taking care not to disturb the remaining portion. The collected fractions were examined by transmission electron microscopy, and fractions containing nAChR-rich membrane vesicles were used for analysis.

Preparation of F(ab)2 fragment antibody against nAChR
The F(ab')2 fragment antibody against nAChR was prepared as follows. Hybridoma cells producing anti-nAChR IgG1 monoclonal antibody, mAb 35, were cultured at 37°C with 7.5% CO2 in Iscove's DMEM (Thermo Scientific) containing 20% foetal bovine serum and 50 μg/ml gentamicin. The culture medium was gradually exchanged to HB basal medium (Irvine Scientific) supplemented with HB101 Lyophilised Supplement (Irvine Scientific). The hybridoma cell culture supernatant was loaded onto a 1 ml Protein G Sepharose 4 Fast Flow column (GE Healthcare). IgG1 was eluted with 0.1 M glycine buffer at pH 2.8, and the eluate was neutralised by adding 1 M Tris-HCl at pH 9.0. The F(ab')2 fragment antibody was produced with a Pierce Mouse IgG1 Fab and F(ab')2 Preparation Kit (Thermo Scientific) according to the manufacturer's protocol. Briefly, IgG1 was incubated overnight at 37°C with ficin that was immobilised on agarose resin. The F(ab')2 fragments were separated from the Fc fragments and undigested IgG1 by size-exclusion chromatography on tandem Superdex 200 HR10/30 columns (GE Healthcare) with running buffer (100 mM NaCl, 20 mM sodium phosphate, pH 7.0) at 4°C. The number of possible positions for the F(ab')2 fragment on nAChR is two (two α-subunits of nAChR), and the F(ab')2 fragment could bind to either of the two sites, considering the reaction efficiency of F(ab')2 fragment for nAChR. A sample of nAChR formed a supported lipid bilayer on the substrate surface. Although the orientation of nAChR on the substrate surface could not be controlled, the gold nanocrystal could only access the extracellular side of the nAChR lipid bilayers.

Angular resolution of DXT used for the measurement
The time-resolved diffraction images were recorded using an X-ray image intensifier (V5445P, Hamamatsu photonics) and CMOS camera (1024 pixel x 1024 pixel, SA 1.1, Photoron). The nominal entrance field of view for the x-ray image intensifier is 150 mm in diameter; and, the effective pixel size is 0.1465 mm. The peak energy of incident X-ray was 15.2 keV and the sample-to-detector distance is 100 mm in our DXT measurement, therefore a one-pixel movement of a diffraction spot in the tilting (θ) direction corresponds to 0.7 mrad/pixel(@15.2 keV). Most of the diffraction spots from gold nanocrystal are obtained at 36.4 mm from the beam centre, considering the d-spacing of Au (111) (d=2.35 Å). The length in 36.4 mm corresponds to 248.5 pixels in our set-up, and the circle that was 248.5 pixels in radius corresponds to approximately 1560 pixels in circumference. Therefore, a one-pixel motion in the twisting (χ) direction corresponds to 4.0 mrad/pixel@15.2 keV.

Low detection probability of diffraction spot from gold nanocrystal
We detected only 3-5 moving diffraction spots from gold nanocrystal in average for each measurement (100 μs/f, 100 frames), the detection probability was low considering the density of gold nanocrystal (about 1 nanocrystal/μm 2 ) on sample surface (Supplementary Figure S2) and the beam size of incident X-ray (40 μm x 150 μm). We think that primary reason for such low detection probability of diffraction spots comes from the low quality of gold nanocrystal. Since large amounts of gold nanocrystals were needed for DXT measurement, we fabricated gold nanocrystals by epitaxial growth on NaCl (100) or KCl (100) surface in 10 -4 Pa vacuum condition (not in ultra-high vacuum condition). At current our fabrication condition, the gold nanocrystals were heterogeneously arranged on the substrate in shape and in size as shown in Figure 5a. We found that normalised intensity of diffraction spots varied in broad range (from 3 to 12 as shown in Figure 5d), and there were simple relationship between tilting or twisting angular velocity and the normalised intensity of diffraction spot, as shown in Figures 5d, 5e and 5f. Therefore we think that diffraction spots we analysed were from certain variation of gold nanocrystals in size. We might detect diffraction spot from larger portion of gold nanocrystals, such as larger 50 nm. The method to provide large amount of high quality of gold nanocrystal with smaller size is to be established in the future. Figure S1 Fast diffracted X-ray tracking at SPring8 BL40XU. The intramolecular dynamics of AChBP and nAChR were monitored by recording the Laue spot trajectories of the gold nanocrystal on the objective protein. X-rays from the beamline (BL40XU, SPring-8, Japan) with energy widths ranging from 14.0-16.5 keV (undulator gap=30.1 mm) were used for DXT measurements. The photon flux density profile for the incident beam from the BL40XU in SPring-8 is shown in the inset graph (upper right, red line). The exposure time for the sample was limited to less than 15 ms to prevent X-ray radiation damage, which was achieved by combining two X-ray shutters, a solenoid shutter and a millisecond shutter (UNIBLITZ, XRS1S2PO). Impairment of protein motion was not observed under these conditions. A fast DXT measurement with 1 μs/f was achieved.  Table S1. Fitting parameters for a 2D-Gaussian function of AChBP and nAChR motions using cluster analysis. The parameters of the 1st, 2nd and 3rd peaks were determined using the following equations: