The glutamine/amino acid transporter (ASCT2) reconstituted in liposomes: Transport mechanism, regulation by ATP and characterization of the glutamine/glutamate antiport

The glutamine/amino acid transporter solubilized from rat renal apical plasma membrane (brush-border membrane) with C₁₂E₈ and reconstituted into liposomes has been previously identified as the ASCT2 transporter. The reconstituted transporter catalyses an antiport reaction in which external glutamine and Na⁺ are cotransported in exchange with internal glutamine (or other amino acids). The glutamine-Na⁺ cotransport occurred with a 1:1 stoichiometry. The concentration of Na⁺ did not influence the Km for glutamine and vice versa. Experimental data obtained by a bi-substrate analysis of the glutamine-Na⁺ cotransport, together with previous report on the glutamine_ex/glutamine_in pseudo bi-reactant analysis, indicated that the transporter catalyses a three-substrate transport reaction with a random simultaneous mechanism. The presence of ATP in the internal compartment of the proteoliposomes led to an increase of the Vmax of the transport and to a decrease of the Km of the transporter for external Na⁺. The reconstituted glutamine/amino acid transporter was inhibited by glutamate; the inhibition was more pronounced at acidic pH. A kinetic analysis revealed that the inhibition was competitive with respect to glutamine. Glutamate was also transported in exchange with glutamine. The external Km of the transporter for glutamate (13.3 mM) was slightly higher than the internal one (8.3 mM). At acidic pH the external but not the internal Km decreased. According with the Km values, glutamate should be transported preferentially from inside to outside in exchange for external glutamine and Na⁺.     

Characterization of functional intermediates of endoglucanase from Aspergillus aculeatus during urea and guanidine hydrochloride unfolding

Low concentrations of urea and GuHCl (2 M) enhanced the activity of endoglucanase (EC 3.1.2.4) from Aspergillus aculeatus by 2.3- and 1.9-fold, respectively. The K_m values for controls, in the presence of 2 M urea and GuHCl, were found to be 2.4 ± 0.2 × 10⁻⁸ mol L⁻¹, 1.4 ± 0.2 × 10⁻⁸ mol L⁻¹, and 1.6 ± 0.2 × 10⁻⁸ mol L⁻¹, respectively. The dissociation constant (K_d) showed changes in the affinity of the enzyme for the substrate with increases in the K_cat suggesting an increased turnover number in the presence of urea and GuHCl. Fluorescence studies showed changes in the microenvironment of the protein. The increase in the activity of this intermediate state was due to conformational changes accompanied by increased flexibility at the active site.     

Detection of an intermediate during the unfolding process of the dimeric ketosteroid isomerase

Failure to detect the intermediate in spite of its existence often leads to the conclusion that two-state transition in the unfolding process of the protein can be justified. In contrast to the previous equilibrium unfolding experiment fitted to a two-state model by circular dichroism and fluorescence spectroscopies, an equilibrium unfolding intermediate of a dimeric ketosteroid isomerase (KSI) could be detected by small angle X-ray scattering (SAXS) and analytical ultracentrifugation. The sizes of KSI were determined to be 18.7A in 0M urea, 17.3A in 5.2M urea, and 25.1A in 7M urea by SAXS. The size of KSI in 5.2M urea was significantly decreased compared with those in 0M and 7M urea, suggesting the existence of a compact intermediate. Sedimentation velocity as obtained by ultracentrifugation confirmed that KSI in 5.2M urea is distinctly different from native and fully-unfolded forms. The sizes measured by pulse field gradient nuclear magnetic resonance (NMR) spectroscopy were consistent with those obtained by SAXS. Discrepancy of equilibrium unfolding studies between size measurement methods and optical spectroscopies might be due to the failure in detecting the intermediate by optical spectroscopic methods. Further characterization of the intermediate using (1)H NMR spectroscopy and Kratky plot supported the existence of a partially-folded form of KSI which is distinct from those of native and fully-unfolded KSIs. Taken together, our results suggest that the formation of a compact intermediate should precede the association of monomers prior to the dimerization process during the folding of KSI.     

Probing the steric barrier of nonionic surfactant vesicles with melittin

The role of the surface polymer brush of nonionic surfactant vesicles (NSV) in inhibiting interactions with small membrane-perturbing molecules was investigated using the bee venom peptide melittin as a probe. The interaction between melittin and NSV was compared with that of distearoylphosphatidylcholine (DSPC) vesicles and sterically stabilised liposomes (SSL) containing 5 mol% pegylated distearoylphosphatidylethanolamine (DSPE.E₄₄). The degree of melittin interaction with the various vesicles was determined by measuring peptide binding and folding, using intrinsic tryptophan fluorescence and circular dichroism respectively, in addition to monitoring the release of encapsulated carboxyfluorescein dye. NSV composed of 1,2-di-O-octadecyl-rac-glyceryl-3-(ω-dodecaethylene glycol) (2C₁₈E₁₂) showed a strong affinity for melittin, whilst exhibiting ~ 50% less bound peptide than SSL. 2C₁₈E₁₂:Chol vesicles showed reduced melittin interaction, in a manner consistent with Chol incorporation into DSPC vesicles. These results are discussed with respect to the effect of Chol on the in-plane order of 2C₁₈E₁₂ bilayers and consequent attenuation of hydrophobic interactions with the peptide. NSV formed from equimolar mixtures of polyoxyethylene-n-stearoyl ethers C₁₈E₂ and C₁₈E₂₀ showed a greater interaction with melittin than 2C₁₈E₁₂. However, replacing C₁₈E₂₀ with C₁₈E₁₀ was sufficient to achieve an attenuation of melittin interaction similar to that observed in 2C₁₈E₁₂:Chol vesicles. This indicates that the presence of surface polymer brush alone may confer resistance to melittin, provided hydrophobic interactions between the peptide and the vesicles can be minimised, through improved in-plane bilayer order.     

Off-pathway aggregation can inhibit fibrillation at high protein concentrations

Ribosomal protein S6 fibrillates readily at slightly elevated temperatures and acidic pH. We find that S6 fibrillation is retarded rather than favored when the protein concentration is increased above a threshold concentration of around 3.5 mg/mL. We name this threshold concentration C_FR, the concentration at which fibrillation is retarded. Our data are consistent with a model in which this inhibition is due to the formation of an off-pathway oligomeric species with native-like secondary structure. The oligomeric species dominates at high protein concentrations but exists in dynamic equilibrium with the monomer so that seeding with fibrils can overrule oligomer formation and favors fibrillation under C_FR conditions. Thus, fibrillation competes with formation of off-pathway oligomers, probably due to a monomeric conversion step that is required to commit the protein to the fibrillation pathway. The S6 oligomer is resistant to pepsin digestion. We also report that S6 forms different types of fibrils dependent on protein concentration. Our observations highlight the multitude of conformational states available to proteins under destabilizing conditions.     

A method for solution NMR structural studies of large integral membrane proteins: Reverse micelle encapsulation

The structural study of membrane proteins perhaps represents one of the greatest challenges of the post-genomic era. While membrane proteins comprise over 50% of current and potential drug targets, their structural characterization lags far behind that of soluble proteins. Nuclear magnetic resonance (NMR) offers great potential not only with respect to structural characterization of integral membrane proteins but may also provide the ability to study the details of small ligand interactions. However, the size limitations of solution NMR have restricted comprehensive structural characterization of membrane protein NMR structures to the relatively small β-barrel proteins or helical proteins of relatively simple topology. In an effort to escape the barriers presented by slow molecular reorientation of large integral membrane proteins solubilized by detergent micelles in water, we have adapted the reverse micelle encapsulation strategy originally developed for the study of large soluble proteins by solution NMR methods. Here we review a novel approach to the solubilization of large integral membrane proteins in reverse micelle surfactants dissolved in low viscosity alkane solvents. The procedure is illustrated with a 54 kDa construct of the homotetrameric KcsA potassium channel.     

Full-sequence computational design and solution structure of a thermostable protein variant

Computational protein design procedures were applied to the redesign of the entire sequence of a 51 amino acid residue protein, Drosophila melanogaster engrailed homeodomain. Various sequence optimization algorithms were compared and two resulting designed sequences were experimentally evaluated. The two sequences differ by 11 mutations and share 22% and 24% sequence identity with the wild-type protein. Both computationally designed proteins were considerably more stable than the naturally occurring protein, with midpoints of thermal denaturation greater than 99 degrees C. The solution structure was determined for one of the two sequences using multidimensional heteronuclear NMR spectroscopy, and the structure was found to closely match the original design template scaffold.     

Effect of cholesterol content on the structural and dynamic membrane properties of DMPC/DSPC large unilamellar bilayers

In this study, we report the effect of cholesterol content on the dynamic and structural properties of a dimyristoyl-phosphatidylcholine and distearoyl-phosphatidylcholine mixture in large unilamellar vesicles. The range of cholesterol concentrations studied varied around approximately 33.3 mol%, where it has been postulated that an abrupt change in bilayer organization occurs. Steady-state fluorescence measurements demonstrated a typical behavior; at low temperatures in the main phase transition, the cholesterol concentration did not affect the gel phase, but at 37.5 °C (phase coexistence) and in the liquid crystalline phase, the presence of cholesterol produced an increase in the fluorescence anisotropy of DPH and the generalized polarization of Laurdan. The greater effect was observed in the liquid crystalline phase, in which the bilayer became a mixture of fluid-like and liquid-ordered phases. The results obtained at approximately 33.3 mol% of Cholesterol demonstrated that the Generalized Polarization of Laurdan, the DPH lifetime, the limiting anisotropy and the rotational correlation time, as well as the fluorescence quenching of DPH by TEMPO, are at maxima, while the fluorescence intensity of dehydroergosterol and the lipid solubility in TritonX-100 are at minima. These results correlate well with the hypothesis of domain segregation in the DMPC/DSPC/Cholesterol LUV system. In this context, we postulate that at 33.3 mol% of Cho, the proportion of ordered domains reaches a maximum.     

Implication of a critical residue (Glu175) in structure and function of bacterial luciferase

Structural properties of a bacterial luciferase mutant, evolved by random mutagenesis, have been investigated. Bacterial luciferases (LuxAB) can be readily classed as slow or fast decay luciferases based on their rates of luminescence decay in a single turnover assay. By random mutagenesis, one of the mutants generated by a single mutation on LuxA at position 175 (E175G) resulted in the "slow decay" Xenorhabdus luminescens luciferase was converted into a luciferase with a significantly more rapid decay rate [Hosseinkhani, S., Szittner, R. and Meighen, E.A. (2005) Biochemical Journal 385, 575-580]. A single mutation (E175G), in a loop that connects alpha helix 5 and beta sheet 5 brought about changes in the kinetic and structural properties of the enzyme. Enhancement of tryptophan fluorescence was observed with a lower degree of fluorescence quenching by acrylamide upon mutation. Near- and far-UV circular dichroism spectra of the native and mutant forms suggested formation of an intermediate structure, further supported by 8-anilino-1-naphthalene-sulphonic acid (ANS) fluorescence which indicated lower exposure of hydrophobic residues as a result of mutation. Fluorescence quenching studies utilizing acrylamide indicated a more accessible fluor for the native form. Thus, the E175G point mutation appears to change the enzymatic decay rate by inducing a substantial tertiary structural change, without a large effect on secondary structural elements, as revealed by Fourier transform IR spectroscopy. Overall, the mutation caused structural changes that go beyond the simple change in orientation of Glu175.     

The predicted transmembrane fragment 17 of the human multidrug resistance protein 1 (MRP1) behaves as an interfacial helix in membrane mimics

The human multidrug resistance protein MRP1 (or ABCC1) is one of the most important members of the large ABC transporter family, in terms of both its biological (tissue defense) and pharmacological functions. Many studies have investigated the function of MRP1, but structural data remain scarce for this protein. We investigated the structure and dynamics of predicted transmembrane fragment 17 (TM17, from Ala₁₂₂₇ to Ser₁₂₅₁), which contains a single Trp residue (W₁₂₄₆) involved in MRP1 substrate specificity and transport function. We synthesized TM17 and a modified peptide in which Ala₁₂₂₇ was replaced by a charged Lys residue. Both peptides were readily solubilized in dodecylmaltoside (DM) or dodecylphosphocholine (DPC) micelles, as membrane mimics. The interaction of these peptides with DM or DPC micelles was studied by steady-state and time-resolved Trp fluorescence spectroscopy, including experiments in which Trp was quenched by acrylamide or by two brominated analogs of DM. The secondary structure of these peptides was determined by circular dichroism. Overall, the results obtained indicated significant structuring (∼50% α-helix) of TM17 in the presence of either DM or DPC micelles as compared to buffer. A main interfacial location of TM17 is proposed, based on significant accessibility of Trp₁₂₄₆ to brominated alkyl chains of DM and/or acrylamide. The comparison of various fluorescence parameters including λ_max, lifetime distributions and Trp rotational mobility with those determined for model fluorescent transmembrane helices in the same detergents is also consistent with the interfacial location of TM17. We therefore suggest that TM17 intrinsic properties may be insufficient for its transmembrane insertion as proposed by the MRP1 consensus topological model. This insertion may also be controlled by additional constraints such as interactions with other TM domains and its position in the protein sequence. The particular pattern of behavior of this predicted transmembrane peptide may be the hallmark of a fragment involved in substrate transport.     

Cooperative unfolding of a metastable serpin to a molten globule suggests a link between functional and folding energy landscapes

Alpha-1 antitrypsin (alpha(1)-AT) is a member of the serpin class of protease inhibitors, and folds to a metastable state rather than its thermodynamically most stable native state. Upon cleavage by a target protease, alpha(1)-AT undergoes a dramatic conformational change to a stable form, translocating the bound protease more than 70 A to form an inhibitory protease-serpin complex. Numerous mutagenesis studies on serpins have demonstrated the trade-off between the stability of the metastable state on the one hand and the inhibitory efficiency on the other. Studies of the equilibrium unfolding of serpins provide insight into this connection between structural plasticity and metastability. We studied equilibrium unfolding of wild-type alpha(1)-AT using hydrogen-deuterium/exchange mass spectrometry to characterize the structure and the stability of an equilibrium intermediate that was observed in low concentrations of denaturant in earlier studies. Our results show that the intermediate observed at low concentrations of denaturant has no protection from hydrogen-deuterium exchange, indicating a lack of stable structure. Further, differential scanning calorimetry of alpha(1)-AT at low concentrations of denaturant shows no heat capacity peak during thermal denaturation, indicating that the transition from the intermediate to the unfolded state is not a cooperative first-order-like phase transition.. Our results show that the unfolding of alpha(1)-AT involves a cooperative transition to a molten globule form, followed by a non-cooperative transition to a random-coil form as more guanidine is added. Thus, the entire alpha(1)-AT molecule consists of one cooperative structural unit rather than multiple structural domains with different stabilities. Furthermore, our results together with previous mutagenesis studies suggest a possible link between an equilibrium molten globule and a functional intermediate that may be populated during the protease inhibition.     

Pore characteristics of nontypeable Haemophilus influenzae outer membrane protein P5 in planar lipid bilayers

The structure of outer membrane protein P5 of NTHi, a homolog of Escherichia coli OmpA, was investigated by observing its pore characteristics in planar lipid bilayers. Recombinant NTHi P5 was overexpressed in E. coli and purified using ionic detergent, LDS-P5, or nonionic detergent, OG-P5. LDS-P5 and OG-P5 could not be distinguished by their migration on SDS-PAGE gels; however, when incorporated into planar bilayers of DPhPC between symmetric aqueous solutions of 1 M KCl at 22 degrees C, LDS-P5 formed narrow pores (58 +/- 6 pS) with low open probability, whereas OG-P5 formed large pores (1.1 +/- 0.1 nS) with high open probability (0.99). LDS-P5 narrow pores were gradually and irreversibly transformed into large pores, indistinguishable from those formed by OG-P5, at temperatures >or=40 degrees C; the process took 4-6 h at 40 degrees C or 35-45 min at 42 degrees C. Large pores were stable to changes in temperatures; however, large pores were rapidly converted to narrow pores when exposed to LDS at room temperatures, indicating acute sensitivity of this conformer to ionic detergent. These studies suggest that narrow pores are partially denatured forms and support the premise that the native conformation of NTHi P5 is that of a large monomeric pore.     

The impact of membrane lipid composition on antimicrobial function of an alpha-helical peptide

VP1, a putative alpha-helical antimicrobial peptide (alpha-AMP) inhibited growth of Bacillus subtilis and Escherichia coli at 500microM. The peptide induced stable surface pressure changes in monolayers formed from B. subtilis native lipid extract (circa 4.5mNm(-1)) but transient pressure changes in corresponding E. coli monolayers (circa 1.0mNm(-1)), which led to monolayer disintegration. Synthetic lipid monolayers mimetic of the extracts were used to generate compression isotherms. Thermodynamic analysis of B. subtilis isotherms indicated membrane stabilisation by VP1 (DeltaG(Mix)<0), via a mechanism dependent upon the phosphatidylglycerol to cardiolipin ratio. Corresponding analysis of E. coli isotherms indicated membrane destabilisation by the peptide (DeltaG(Mix)>0). Destabilisation correlated with PE levels present and appeared to involve a mechanism resembling those used by tilted peptides. These data emphasise that structure/function analysis of alpha-AMPs must consider not only their structural characteristics but also the lipid make-up of the target microbial membrane.     

Solubilization,purification and reconstitution of Ca-ATPase from bovine pulmonary artery smooth muscle microsomes by different detergents: Preservation of native structure and function of the enzyme by DHPC

The properties of Ca²⁺-ATPase purified and reconstituted from bovine pulmonary artery smooth muscle microsomes {enriched with endoplasmic reticulum (ER)} were studied using the detergents 1,2-diheptanoyl-sn-phosphatidylcholine (DHPC), poly(oxy-ethylene)8-lauryl ether (C₁₂E₈) and Triton X-100 as the solubilizing agents. Solubilization with DHPC consistently gave higher yields of purified Ca²⁺-ATPase with a greater specific activity than solubilization with C₁₂E₈ or Triton X-100. DHPC was determined to be superior to C₁₂E₈; while that the C₁₂E₈ was determined to be better than Triton X-100 in active enzyme yields and specific activity. DHPC solubilized and purified Ca²⁺-ATPase retained the E1Ca−E1*Ca conformational transition as that observed for native microsomes; whereas the C₁₂E₈ and Triton X-100 solubilized preparations did not fully retain this transition. The coupling of Ca²⁺ transported to ATP hydrolyzed in the DHPC purified enzyme reconstituted in liposomes was similar to that of the native micosomes, whereas that the coupling was much lower for the C₁₂E₈ and Triton X-100 purified enzyme reconstituted in liposomes. The specific activity of Ca²⁺-ATPase reconstituted into dioleoyl-phosphatidylcholine (DOPC) vesicles with DHPC was 2.5-fold and 3-fold greater than that achieved with C₁₂E₈ and Triton X-100, respectively. Addition of the protonophore, FCCP caused a marked increase in Ca²⁺ uptake in the reconstituted proteoliposomes compared with the untreated liposomes. Circular dichroism analysis of the three detergents solubilized and purified enzyme preparations showed that the increased negative ellipticity at 223 nm is well correlated with decreased specific activity. It, therefore, appears that the DHPC purified Ca²⁺-ATPase retained more organized and native secondary conformation compared to C₁₂E₈ and Triton X-100 solubilized and purified preparations. The size distribution of the reconstituted liposomes measured by quasi-elastic light scattering indicated that DHPC preparation has nearly similar size to that of the native microsomal vesicles whereas C₁₂E₈ and Triton X-100 preparations have to some extent smaller size. These studies suggest that the Ca²⁺-ATPase solubilized, purified and reconstituted with DHPC is superior to that obtained with C₁₂E₈ and Triton X-100 in many ways, which is suitable for detailed studies on the mechanism of ion transport and the role of protein–lipid interactions in the function of the membrane-bound enzyme.     

Temporary membrane distortion of vascular smooth muscle cells is responsible for their apoptosis induced by platelet-activating factor-like oxidized phospholipids and their degradation product, lysophosphatidylcholine

To obtain information about the mechanism of apoptosis induced by oxidized low density lipoproteins (oxLDL) in atherosclerotic plaques, we examined the effects of lysophosphatidylcholine (LPC) and platelet-activating factor (PAF)-like lipids (PAF-LL), which can be derived from oxLDL, on rat vascular smooth muscle cells (VSMC). All the lipids with different structures examined induced apoptosis of VSMC, so we studied the mechanism of induction of apoptosis by LPC. LPC-induced apoptosis was inhibited by alpha-tocopherol (alpha-T) and cholesterol (Chol), but not by other antioxidants such as palmitoyl ascorbic acid and PAF receptor antagonist. The cells temporarily became spherical and highly permeable before induction of apoptosis, and their change in shape was prevented by alpha-T and Chol. From these results, we suggest that the apoptosis induced by oxLDL-derived phospholipids in VSMC is caused by temporary membrane distortion, not through specific receptors.     

Crystal structure of a thermally stable rhodopsin mutant

We determined the structure of the rhodopsin mutant N2C/D282C expressed in mammalian cells; the first structure of a recombinantly produced G protein-coupled receptor (GPCR). The mutant was designed to form a disulfide bond between the N terminus and loop E3, which allows handling of opsin in detergent solution and increases thermal stability of rhodopsin by 10 deg.C. It allowed us to crystallize a fully deglycosylated rhodopsin (N2C/N15D/D282C). N15 mutations are normally misfolding and cause retinitis pigmentosa in humans. Microcrystallographic techniques and a 5 microm X-ray beam were used to collect data along a single needle measuring 5 microm x 5 microm x 90 microm. The disulfide introduces only minor changes but fixes the N-terminal cap over the beta-sheet lid covering the ligand-binding site, a likely explanation for the increased stability. This work allows structural investigation of rhodopsin mutants and shows the problems encountered during structure determination of GPCRs and other mammalian membrane proteins.     

The funnel approach to the precrystallization production of membrane proteins

Challenges in the production of integral membrane proteins for structural studies include low expression levels, incorrect membrane insertion, aggregation and instability. In this report, we describe a “funnel approach” to overcoming these difficulties and demonstrate its efficacy in a case study of 36 prokaryotic P-type transporters. A diverse ensemble of modified constructs is generated and tested for expression in Escherichia coli, membrane localization, detergent extraction, and homogeneity. High-throughput methodologies are implemented throughout the process to facilitate identification of promising targets. We find that the choice of promoter, the choice of source organism providing the cloned gene, and, most importantly, the position of the affinity tag have a great effect on successful production. The latter had pronounced effects at all tested levels, from expression levels observed in whole cells to the extent of membrane insertion, and even on protein function. Following the initial streamlined screening, we were able to fine-tune and produce 9 of the 36 targets as materials suitable for crystallization or other structural studies.     

Functional analysis of propeptide as an intramolecular chaperone for in vivo folding of subtilisin nattokinase

Here, we show that during in vivo folding of the precursor, the propeptide of subtilisin nattokinase functions as an intramolecular chaperone (IMC) that organises the in vivo folding of the subtilisin domain. Two residues belonging to β-strands formed by conserved regions of the IMC are crucial for the folding of the subtilisin domain through direct interactions. An identical protease can fold into different conformations in vivo due to the action of a mutated IMC, resulting in different kinetic parameters. Some interfacial changes involving conserved regions, even those induced by the subtilisin domain, blocked subtilisin folding and altered its conformation. Insight into the interaction between the subtilisin and IMC domains is provided by a three-dimensional structural model.     

Stability enhancement of cytochrome c through heme deprotonation and mutations

The chemical denaturation of Pseudomonas aeruginosa cytochrome c(551) variants was examined at pH 5.0 and 3.6. All variants were stabilized at both pHs compared with the wild-type. Remarkably, the variants carrying the F34Y and/or E43Y mutations were more stabilized than those having the F7A/V13M or V78I ones at pH 5.0 compared with at pH 3.6 by ~3.0-4.6 kJ/mol. Structural analyses predicted that the side chains of introduced Tyr-34 and Tyr-43 become hydrogen donors for the hydrogen bond formation with heme 17-propionate at pH 5.0, but less efficiently at pH 3.6, because the propionate is deprotonated at the higher pH. Our results provide an insight into a stabilization strategy for heme proteins involving variation of the heme electronic state and introduction of appropriate mutations.     

Characteristics affecting expression and solubilization of yeast membrane proteins

Biochemical and structural analysis of membrane proteins often critically depends on the ability to overexpress and solubilize them. To identify properties of eukaryotic membrane proteins that may be predictive of successful overexpression, we analyzed expression levels of the genomic complement of over 1000 predicted membrane proteins in a recently completed Saccharomyces cerevisiae protein expression library. We detected statistically significant positive and negative correlations between high membrane protein expression and protein properties such as size, overall hydrophobicity, number of transmembrane helices, and amino acid composition of transmembrane segments. Although expression levels of membrane and soluble proteins exhibited similar negative correlations with overall hydrophobicity, high-level membrane protein expression was positively correlated with the hydrophobicity of predicted transmembrane segments. To further characterize yeast membrane proteins as potential targets for structure determination, we tested the solubility of 122 of the highest expressed yeast membrane proteins in six commonly used detergents. Almost all the proteins tested could be solubilized using a small number of detergents. Solubility in some detergents depended on protein size, number of transmembrane segments, and hydrophobicity of predicted transmembrane segments. These results suggest that bioinformatic approaches may be capable of identifying membrane proteins that are most amenable to overexpression and detergent solubilization for structural and biochemical analyses. Bioinformatic approaches could also be used in the redesign of proteins that are not intrinsically well-adapted to such studies.