Protein engineering最新文献

Stimulation of T-cells by IL-2 has been exploited for treatment of metastatic renal carcinoma and melanoma. However, a narrow therapeutic window delimited by negligible stimulation of T-cells at low picomolar concentrations and undesirable stimulation of NK cells at nanomolar concentrations hampers IL-2-based therapies. We hypothesized that increasing the affinity of IL-2 for IL-2Ralpha may create a class of IL-2 mutants with increased biological potency as compared with wild-type IL-2. Towards this end, we have screened libraries of mutated IL-2 displayed on the surface of yeast and isolated mutants with a 15-30-fold improved affinity for the IL-2Ralpha subunit. These mutants do not exhibit appreciably altered bioactivity at 0.5-5 pM in steady-state bioassays, concentrations well below the IL-2Ralpha equilibrium binding constant for both the mutant and wild-type IL-2. A mutant was serendipitously identified that exhibited somewhat improved potency, perhaps via altered endocytic trafficking mechanisms described previously.

The influence of different factors acting on Escherichia coli periplasmic expression of recombinant human growth hormone (hGH) in shake flask cultures has been investigated. Bacterial vectors containing the phage lambdaP(L) promoter, which is temperature activated, were utilized. Four different signal peptides were compared: DsbA, npr, STII and one derived from the natural hGH signal peptide, this last used as a reference. Other factors such as medium composition, optimized induction and expression conditions, and different bacterial strains were also studied. The determination of hGH, carried out directly in osmotic shock fluids, was based on an isocratic reversed-phase high-performance liquid chromatography method, which allows direct, rapid evaluation of the quality and quantity of hGH being secreted in the bacterial periplasmic space immediately after or even during fermentation. The level of hGH production increased 2.5-fold compared with the reference vector, reaching a level of 3.9 +/- 0.63 micro g/ml/A(600) (n = 6; coefficient of variation = 16.2%). The expression level was affected by the signal peptide and by the induction conditions, being more effective when activation started in the early logarithmic phase which, however, exhibited remarkably different optical density (OD) according to medium composition. Our results thus indicate that 6 h activation at 40-42 degrees C, starting with an OD (A(600)) of approximately 3 in a very rich medium, were conditions capable of providing the maximum secretion level for a vector utilizing the DsbA signal sequence and E.coli W3110 or RB791 as host cells.

One of the most important and challenging tasks in protein modelling is the prediction of loops, as can be seen in the large variety of existing approaches. Loops In Proteins (LIP) is a database that includes all protein segments of a length up to 15 residues contained in the Protein Data Bank (PDB). In this study, the applicability of LIP to loop prediction in the framework of homology modelling is investigated. Searching the database for loop candidates takes less than 1 s on a desktop PC, and ranking them takes a few minutes. This is an order of magnitude faster than most existing procedures. The measure of accuracy is the root mean square deviation (RMSD) with respect to the main-chain atoms after local superposition of target loop and predicted loop. Loops of up to nine residues length were modelled with a local RMSD <1 A and those of length up to 14 residues with an accuracy better than 2 A. The results were compared in detail with a thoroughly evaluated and tested ab initio method published recently and additionally with two further methods for a small loop test set. The LIP method produced very good predictions. In particular for longer loops it outperformed other methods.

An interesting observation was found during our continued studies on the hydrolysis of ibuprofen esters by Candida rugosa lipase (CRL). An important role is played by pH in the stereospecific hydrolysis of these esters. The flap region of CRL plays a significant role in the access of the substrate to the active site of the enzyme. At pH 5.6, 48% of the methyl ester and 5% of the butyl ester of ibuprofen were hydrolysed in 5.5 h, whereas at pH 7.2, 9% of methyl ester and 45% of the butyl ester of ibuprofen was hydrolysed in a identical reaction time using CRL. This lead us to assume that CRL prefers the methyl ester of ibuprofen as a substrate at an acidic pH and the butyl ester of ibuprofen at a neutral pH. Therefore, in order to understand the role of pH in the substrate selection by CRL for the esters of ibuprofen we used the crystallographic coordinates of the open form of the CRL (1CRL) for molecular dynamics (MD) simulations under acidic and neutral conditions for 2 ns using GROMACS. The final structures obtained after simulation in acidic and neutral conditions were compared with the energy-minimized structure, and the root-mean-square deviations (r.m.s.ds) were calculated. The r.m.s.d. of the CRL flap at neutral pH was found to be greater than that of the CRL flap at acidic pH. The extent to which the flap opens at neutral pH allowed the bulkier substrate, the butyl ester of ibuprofen, to diffuse into the active site and provides the best enzyme-substrate fit for this specific substrate. At acidic pH there is a decreased opening of the flap thereby accommodating a more compact substrate, namely the methyl ester of ibuprofen. Thus, simulation experiments using MD provide reasonable insight for the pH-dependent substrate selectivity of CRL in aqueous environments.

Cyclodextrins are cyclic oligosaccharides with the shape of a hollow truncated cone. Their exterior is hydrophilic and their cavity is hydrophobic, which gives cyclodextrins the ability to accommodate hydrophobic molecules/moieties in the cavity. This special molecular arrangement accounts for the variety of beneficial effects cyclodextrins have on proteins, which is widely used in pharmacological applications. We have studied the interaction between beta-cyclodextrin and four non-carbohydrate-binding model proteins: ubiquitin, chymotrypsin inhibitor 2 (CI2), S6 and insulin SerB9Asp by NMR spectroscopy at varying structural detail. We demonstrate that the interaction of beta-cyclodextrin and our model proteins takes place at specific sites on the protein surface, and that solvent accessibility of those sites is a necessary but not compelling condition for the occurrence of an interaction. If this behaviour can be generalized, it might explain the wide range of different effects of cyclodextrins on different proteins: aggregation suppression (if residues responsible for aggregation are highly solvent accessible), protection against degradation (if point of attack of a protease is sterically 'masked' by cyclodextrin), alteration of function (if residues involved in function are 'masked' by cyclodextrin). The exact effect of cyclodextrins on a given protein will always be related to the particular structure of this protein.

Ligand-targeted anticancer therapeutics represent an opportunity for the selective and efficient delivery of drugs to tumours. The chemical coupling of ligands to drugs or drug carrier systems is, however, often hampered by the presence of multiple reactive groups within the ligand, for example, epsilon-NH(2) groups in lysine side chains. In this paper, we describe the isolation by phage display of human epidermal growth factor (EGF) variants without lysine and a reduced number of arginine residues. The selection on A431 carcinoma cells also revealed that R41 is indispensable for EGF binding activity as all EGF variants contained an arginine residue at this position. One EGF variant (EGFm1) with K28Q, R45S, K48S and R53S mutations was expressed in bacteria and showed an identical binding activity as wild-type EGF. EGFm1 could be labelled with fluorescein isothiocyanate demonstrating the accessibility of the N-terminal amino group for coupling reagents. Furthermore, coupling of EGFm1 to PEGylated liposomes resulted in target cell-specific binding and internalization of the liposomes. These human EGF variants should be advantageous for the generation of anticancer therapeutics targeting the EGF receptor, which is overexpressed by a wide variety of different tumours.

Recently, we designed a short alpha-helical fibril-forming peptide (alphaFFP) that can form alpha-helical nanofibrils at acid pH. The non-physiological conditions of the fibril formation hamper biomedical application of alphaFFP. It was hypothesized that electrostatic repulsion between glutamic acid residues present at positions (g) of the alphaFFP coiled-coil sequence prevent the fibrillogenesis at neutral pH, while their protonation below pH 5.5 triggers axial growth of the fibril. To test this hypothesis, we synthesized alphaFFPs where all glutamic acid residues were substituted by glutamines or serines. The electron microscopy study confirmed that the modified alphaFFPs form nanofibrils in a wider range of pH (2.5-11). Circular dichroism spectroscopy, sedimentation, diffusion and differential scanning calorimetry showed that the fibrils are alpha-helical and have elongated and highly stable cooperative tertiary structures. This work leads to a better understanding of interactions that control the fibrillogenesis of the alphaFFPs and opens opportunities for their biomedical application.

We compare the modelling accuracy of two common rotamer libraries, the Dunbrack-Cohen and the 'Penultimate' rotamer libraries, with that of a novel library of discrete side chain conformations extracted from the Protein Data Bank. These side chain conformer libraries are extracted automatically from high-quality protein structures using stringent filters and maintain crystallographic bond lengths and angles. This contrasts with traditional rotamer libraries defined in terms of chi angles under the assumption of idealized covalent geometry. We demonstrate that side chain modelling onto native and near-native main chain conformations is significantly more successful with the conformer libraries than with the rotamer libraries when solely considering excluded-volume interactions. The rotamer libraries are inadequate to model side chains without atomic clashes on over 20% of targets if the backbone is held fixed in the native conformation. An algorithm is described for simultaneously modelling both main chain and side chain atoms during discrete ab initio sampling. The resulting models have equivalent root mean square deviations from the experimentally determined protein loops as models from backbone-only ensembles, indicating that all-atom modelling does not detract from the accuracy of conformational sampling.

The capability of predicting folding and conformation of a protein from its primary structure is probably one of the main goals of modern biology. An accurate prediction of solvent accessibility is an intermediate step along this way. A new method for predicting solvent accessibility from single sequence and multiple alignment data is described. The method is based on probability profiles calculated on an amino acid sequence centred on the residue whose accessibility has to be predicted. A profile is constructed for each exposure category considered so as to calculate the probability of a sequence being generated by the different profiles. Prediction accuracy was tested on a variety of protein sets with two- and three-state models. Different thresholds were used according to those adopted by the authors proposing the data sets. The prediction accuracy is significantly improved over existing methods.

PHOSPHO1 is a recently identified phosphatase whose expression is upregulated in mineralizing cells and is implicated in the generation of inorganic phosphate for matrix mineralization, a process central to skeletal development. The enzyme is a member of the haloacid dehalogenase (HAD) superfamily of magnesium-dependent hydrolases. However, the natural substrate(s) is as yet unidentified and to date no structural information is known. We have identified homologous proteins in a number of species and have modelled human PHOSPHO1 based upon the crystal structure of phosphoserine phosphatase (PSP) from Methanococcus jannaschii. The model includes the catalytic Mg(2+) atom bound via three conserved Asp residues (Asp32, Asp34 and Asp203); O-ligands are also provided by a phosphate anion and two water molecules. Additional residues involved in PSP-catalysed hydrolysis are conserved and are located nearby, suggesting both enzymes share a similar reaction mechanism. In PHOSPHO1, none of the PSP residues that confer the enzyme's substrate specificity (Arg56, Glu20, Met43 and Phe49) are conserved. Instead, we propose that two fully conserved Asp residues (Asp43 and Asp123), not present in PSPs contribute to substrate specificity in PHOSPHO1. Our findings show that PHOSPHO1 is not a member of the subfamily of PSPs but belongs to a novel, closely related enzyme group within the HAD superfamily.