Improving extraction and post-purification concentration of membrane proteins.

Membrane proteins (MPs), despite being critically important drug targets for the pharmaceutical industry, are difficult to study due to challenges in obtaining high yields of functional protein. Most current extraction efforts use specialized non-ionic detergents to solubilize and stabilize MPs, with MPs being concentrated by ultrafiltration (UF). However, many detergents are retained during the UF step, which can destabilize MPs and/or interfere with their characterization. Here, we studied the influence of detergent selection on the extraction and UF-based concentration of biomedically-relevant MPs, the light-driven sodium and chloride transporters, KR2 and halorhodopsin (pHR) which are also model proteins for more complex mammalian rhodopsins. We also designed a flat-bottomed centrifugal filter that can concentrate MPs with enhanced removal of free detergents by promoting concentration polarization (CP). We tested the performance of this new filter using four commonly employed MP detergents, octyl-ß-D maltoside (OM), decyl-ß-D maltoside (DM), dodecyl-ß-D maltoside (DDM) and octyl-ß-D glucoside (OG), over a range of detergent and salt concentrations. Detergent passage is significantly higher for the flat-bottomed filter achieving up to 2-fold greater sieving of detergent in DM-solubilized pHR system due to the high degree of CP. We observe more efficient, up to 5-fold higher extraction of KR2 in the presence of a longer 12-carbon alkyl chain detergent, DDM compared to a shorter 8-carbon detergent, OM. Assuming complete binding and elution of the extracted protein, DDM-based extraction of KR2 could lead to a potential 7-fold improvement in purification yields compared to conventional methods which yield ∼1 mg MP per liter of cell culture. However, the longer chain detergents like DDM form larger micelles that are difficult to remove by UF. Thus, there exists a trade-off between choosing a detergent that will enable efficient extraction of MP while showing easier removal during subsequent UF. The extraction efficiency and UF-based separation of detergent micelles provide insights for other applications involving detergent-mediated separation/extraction.

Engineered-membranes and engineered-micelles as efficient tools for purification of halorhodopsin and bacteriorhodopsin.

We describe two alternative and complementary purification methods for halorhodopsin and bacteriorhodopsin. The first relies on a unique form of detergent micelles which we have called engineered-micelles. These are specifically conjugated in the presence of [hydrophobic chelator:Fe(2+)] complexes and form detergent aggregates into which membrane proteins partition, but hydrophilic water-soluble proteins do not. The approach was tested on the membrane protein, bacteriorhodopsin (bR), with five non-ionic detergents (OG, OTG, NG, DM, and DDM), commonly used in purification and crystallization of membrane proteins, in combination with the commercially available bathophenanthroline or with one of the three synthesized phenanthroline derivatives (Phen-C10, Phen-C8 and Phen-C6). Our results show that bR is extracted efficiently (60-86%) and directly from its native membrane into diverse detergent aggregates with preservation of its native conformation, while 90-95% of an artificial contaminating background is excluded. For implementation of the second method, based on engineered-membranes, the use of detergents, which in some cases may produce protein denaturation, is not required at all. Protein-containing membranes are conjugated via the same hydrophobic [chelator:metal ion] complexes but maintain the membrane protein in its native bilayer environment throughout the process. This method is demonstrated on the membrane protein halorhodopsin from Natronomonas pharaonis (phR) and leads to good recovery yields (74-89%) and removal of >85% of artificial background impurities while preserving the native state of phR. The detailed structure of the hydrophobic chelator used has been found to have a marked effect on the purity and yield of both methods.

Solubilization of membrane proteins with novel N-acylamino acid detergents.

N-Acylamino acids are a new family of versatile biological surfactants capable of extracting integral membrane proteins of various topologies from the biological membrane, in many instances surpassing the efficiency of commercial detergents.

Spectroscopic studies of a sensory rhodopsin I homologue from the archaeon Haloarcula vallismortis.

Sensory rhodopsin I (SRI) functions as a dual receptor regulating both negative and positive phototaxis. It transmits light signals through changes in protein-protein interactions with its transducer protein, HtrI. The phototaxis function of Halobacterium salinarum SRI (HsSRI) has been well characterized using genetic and molecular techniques, whereas that of Salinibacter ruber SRI (SrSRI) has not. SrSRI has the advantage of high protein stability compared with HsSRI and, therefore, provided new information about structural changes and Cl(-) binding of SRI. However, nothing is known about the functional role of SrSRI in phototaxis behavior. In this study, we expressed a SRI homologue from the archaeon Haloarcula vallismortis (HvSRI) as a recombinant protein which uses all-trans-retinal as a chromophore. Functionally important residues of HsSRI are completely conserved in HvSRI (unlike in SrSRI), and HvSRI is extremely stable in buffers without Cl(-). Taking advantage of the high stability, we characterized the photochemical properties of HvSRI under acidic and basic conditions and observed the effects of Cl(-) on the protein under both conditions. Fourier transform infrared results revealed that the structural changes in HvSRI were quite similar to those in HsSRI and SrSRI. Thus, HvSRI can become a useful protein model for improving our understanding of the molecular mechanism of the dual photosensing by SRI.

Halorhodopsin from natronomonas pharaonis forms a trimer even in the presence of a detergent, dodecyl-beta-D-maltoside.

Halorhodopsin (HR) is a transmembrane seven-helix retinal protein, and acts as an inward light-driven Cl(-) pump. HR from Natronomonas pharaonis (NpHR) can be expressed in Escherichia coli inner membrane in large quantities. Here, we showed that NpHR forms the trimer structure even in the presence of 0.1% (2 mm) to 1% (20 mm) dodecyl-beta-d-maltoside (DDM), whose concentrations are much higher than the critical micelle concentration (0.17 mm). This conclusion was drawn from the following observations. (1) NpHR in the DDM solution showed an exciton-coupling circular dichroism (CD) spectrum. (2) From the elution volume of gel filtration, the molecular mass of the NpHR-DDM complex was estimated. After evaluation of the mass of the bound DDM molecules, the mass of NpHR calculated was approximately equal to that of the trimer. (3) The cross-linked NpHR by glutaraldehyde gave the SDS-PAGE corresponding to the trimer. Mass spectra of these samples also support the notion of the trimer. Using the membrane fractions expressing NpHR (Escherichia coli and Halobacterium salinarum), CD spectra showed exciton-coupling, which suggests strongly the trimer structure in the cell membrane.

Stopped-flow analysis on anion binding to blue-form halorhodopsin from Natronobacterium pharaonis: comparison with the anion-uptake process during the photocycle.

Pharaonis halorhodopsin (phR), the light-driven chloride ion pump from Natronobacterium pharaonis with C-terminal histidine tag, was expressed in Escherichia coli cells. The protein was solubilized with 0.1% n-dodecyl beta-D-maltopyranoside and purified with a nickel column. Removal of Cl- from the medium yields blue phR (phR(blue)) that has lost Cl- near the chromophore. Addition of Cl- converts phR(blue) to a red-shifted Cl--bound form (phR(Cl)). Circular dichroic spectra of phR(blue) and phR(Cl) exhibited a bilobe in the visual region, indicating specific oligomerization of the phR monomers. The order of anion concentration which induced a shift from phR(blue) to phR(X) was Br- < Cl- < NO3- < N3-, which was the same as in the case of phR purified from N. pharaonis membranes. Chloride binding kinetics was measured by time-resolved absorption changes with stopped-flow rapid mixing. Rates of Cl- binding consisted of fast and slow components, and the amplitude of the fast component was about 90% of the total changes. The rate constant of the fast component at 100 mM NaCl at 25 degrees C was 260 s(-1) with an apparent activation energy of 35 kJ/mol. These values are in good agreement with the process of Cl- uptake in the photocycle (O --> hR' reaction) reported previously [Váró et al. (1995) Biochemistry 34, 14500-14507]. In addition, the Cl- concentration dependence on both rates was similar to each other. These observations suggest that the O-intermediate is similar to phR(blue) and that Cl- uptake during the photocycle may be ruled by a passive process.

The halo-opsin gene. I. Identification and isolation.

Halorhodopsin (HR), the light-driven chloride pump in halobacteria, was purified in the denatured as well as in the native state and chemically cleaved into peptide fragments. Isolation of peptide and liquid phase sequencing yielded approximately 20% of the halo-opsin (HO) structure in non-overlapping peptides. Chemically synthesized oligodeoxynucleotides corresponding to a peptide sequence obtained from both HR preparations were used to screen a cosmid gene bank of Halobacterium halobium strain L-33. A positive clone contained cosmid pAB H47 which by subcloning and nucleotide sequencing was shown to encode at least part of the HO gene.