Protein Engineering Design and Selection - Advance Access

A sulfoglycolipid beta-sulfoquinovosyldiacylglycerol ({beta}SQDG) binds to Met1-Arg95 region of murine DNA polymerase lambda (Mmpol {lambda}) and inhibits its nuclear transit

Takakusagi, K., Takakusagi, Y., Ohta, K., Aoki, S., Sugawara, F., Sakaguchi, K. — Fri, 20 Nov 2009 06:24:16 PST

Beta-sulfoquinovosyldiacylglycerol (βSQDG) is a synthetic sulfoglycolipid that shows inhibitory activity of DNA polymerase (pol ). Here we identified a βSQDG binding region within murine pol (Mmpol ) using T7 phage display technology. We compared the binding intensity of βSQDG with recombinant phages (phages 1–6) that displayed different segments of Mmpol . The binding assay clearly showed that phage 1, which displayed the non-structural Met1-Arg95 region including the nuclear localization signal (NLS) and part of the BRCT domain, bound more strongly to βSQDG than the other recombinant phages. Binding assays using recombinant proteins gave similar results, showing specific βSQDG binding to Met1-Arg95 with a K_D value of 9.9 nM. Furthermore, in a cell-based assay, nuclear localization of EGFP-pol was inhibited in the presence of βSQDG possibly due to binding of βSQDG to NLS. These experiments clearly show that the binding region of βSQDG within Mmpol could be successfully identified using T7 phage display technology. We suggest that the strategy we describe here will be of value for identifying the binding site within a protein for small ligands, and will provide information that cannot be obtained using other experimental techniques due to their inherent technical limitations.

Rational design of a GH1 {beta}-glycosidase to prevent self-condensation during the transglycosylation reaction

Tran, V., Hoffmann, L., Rabiller, C., Tellier, C., Dion, M. — Thu, 19 Nov 2009 05:39:33 PST

Mutant N282T of a thermostable β-glycosidase from GH1 family (TtβGly) presenting a high transglycosidase activity was previously obtained by directed evolution. However, it displays a self-condensation activity with the donor 2-nitrophenyl-β-d-galactopyranoside (oNPGal) which competes with the condensation reaction and entails undesirable effects. In order to prevent this reaction, we rationally modified this enzyme at the [+1]/[+2] subsites so that oNPGal would bind less tightly. Molecular modeling (MM) suggested the mutation A221W, which decreased the affinity of the donor at these sites and moved it away from the bound galactose at the –1 subsite. A single (A221W) and a double mutant (A221W/N282T) were constructed, and they gave rise to a drastic decrease in self-condensation. The A221W mutant had no transglycosylation activity whereas the A221W/N282T mutant still displayed a condensation activity, comparable to that of the N282T mutant for the transfer on pNPGlcNAc. MM revealed that the double mutant A221W/N282T could induce the synthesis of a glycosidic bond between a donor and an acceptor displaying an equatorial 4-position. Moreover, it is suggested that mutation N282T could change the orientation of residue N219, leading to a stabilization of the acceptor with a new hydrogen bond. This finding opens the way to further improvements of evolved transglycosidases.

A fast and accurate method for predicting pKa of residues in proteins

Huang, R.-B., Du, Q.-S., Wang, C.-H., Liao, S.-M., Chou, K.-C. — Wed, 18 Nov 2009 20:24:58 PST

Predicting the pH-activities of residues in proteins is an important problem in enzyme engineering and protein design. A novel predictor called ‘Pred-pK_a’ was developed based on the physicochemical properties of amino acids and protein 3D structure. The Pred-pK_a approach considers the influence of all other residues of the protein to predict the pK_a value of an ionizable residue. An empirical equation was formulated, in which the pK_a value was a distance-dependent function of physicochemical parameters of 20 amino acid types, describing their electrostatic and van der Waals interaction, as well as the effects of hydrogen bonds and solvation. Two sets of coefficients, {a} and {b_l}, were used in the predictor: {a} is the weight factors of 20 amino acid types and {b_l} is the weight factors of physicochemical properties of amino acids. An iterative double least square procedure was proposed to solve the two sets of weight factors alternately and iteratively in a training set. The two coefficient sets {a} and {b_l} thus obtained were used to predict the pK_a values of residues in a protein. The average predictive error is ± 0.6 pH in less than a minute in common personal computer.

The relationship between water bridges and the polyproline II conformation: a large-scale analysis of molecular dynamics simulations and crystal structures

Law, P. B., Daggett, V. — Mon, 16 Nov 2009 08:08:03 PST

It has been suggested that denatured proteins are predisposed toward the left-handed polyproline II (P_II) conformation. One possible source of P_II stability in the denatured state is water bridges. Water bridges are networks of water molecules that link nearby hydrogen bond acceptors and/or donors on proteins. On the basis of the proposed behavior of P_II and water bridges, the propensity of a residue to participate in water bridges should be correlated with its P_II propensity. To test this hypothesis, we analyzed the following data sets: 2351 high-resolution crystal structures, and the native and denatured states of 188 different proteins from all-atom, explicit-solvent molecular dynamics (MD) simulations, which are part of our Dynameomics effort. We found that water bridges do not explain the high frequency of P_II in denatured states; such bridges are less frequent around P_II than around other conformations. Thus, this analysis casts doubt on water bridges as a dominant factor determining the residue-based P_II propensities.

Functional chimera of porcine pepsin prosegment and Plasmodium falciparum plasmepsin II

Parr-Vasquez, C. L., Yada, R. Y. — Thu, 12 Nov 2009 05:21:34 PST

Proplasmepsin II (zPMII) represents a unique member of the aspartic proteinase family, with a prosegment–enzyme interaction that is thus far unique among the pepsin-like proteases. The role of the prosegment in aspartic proteinase structure and function was investigated by generating two chimeric proteins, one with the pepsinogen prosegment fused to the mature region of PMII (pepproPMII) and a second with the prosegment of PMII fused to pepsin (PMIIpropep). Both chimeras were expressed using Escherichia coli; however, PMIIpropep was extremely unstable suggesting protein misfolding. Alternatively, pepproPMII was capable of both autoactivation and hydrolysis of a synthetic substrate. Similarly, when the PMII enzyme was expressed without a prosegment, it too exhibited activity against the synthetic enzyme. CD measurements indicated that pepproPMII had reduced thermal stability when compared with zPMII. This reduction of temperature stability may have resulted from the inability of the pepsinogen prosegment to stabilize the C-terminal domain of the PMII enzyme. The ability of PMII to fold in the presence of a completely non-homologous prosegment and in its absence suggests that prosegment is not critical to obtaining a functional enzyme in all pepsin-like enzymes but likely plays a role in protein stabilization.

Bacterial display enables efficient and quantitative peptide affinity maturation

Kenrick, S. A., Daugherty, P. S. — Tue, 10 Nov 2009 05:46:36 PST

A quantitative screening method was developed to enable isolation and affinity maturation of peptide ligands specific for a given target from peptide libraries displayed on the outer surface of Escherichia coli using multi-parameter flow cytometry. From a large, random 15-mer peptide library, screening identified a core motif of W-E/D-W-E/D that conferred binding to vascular endothelial growth factor (VEGF). One cycle of affinity maturation resulted in the identification of several families of VEGF-binding peptides having distinct consensus sequences, from which a preferred disulfide constraint emerged. In the second affinity maturation cycle, high affinity peptides were favored by the addition of a decoy protein that bound an adjacent epitope on the display scaffold. The decoy apparently reduced rebinding or avidity effects, and the resulting peptides exhibited consensus at 12 of 19 amino acid positions. Peptides identified and affinity matured using bacterial display were remarkably similar to the best affinity matured using phage display and exhibited comparable dissociation constants (within 2-fold; K_D = 4.7 x 10^–7 M). Screening of bacterial-displayed peptide libraries using cytometry enabled optimization of screening conditions to favor affinity and specificity and rapid clonal characterization. Bacterial display thus provides a new quantitative tool for the discovery and evolutionary optimization of protein-specific peptide ligands.

USER friendly DNA recombination (USERec): a simple and flexible near homology-independent method for gene library construction

Villiers, B.R.M., Stein, V., Hollfelder, F. — Fri, 06 Nov 2009 05:11:13 PST

USER friendly DNA recombination (USERec) is introduced as a near homology-independent method that allows the simultaneous recombination of an unprecedented number of 10 DNA fragments (~40–400 bp) within a day. The large number of fragments and their ease of preparation enables the creation of libraries of much larger genetic diversity (potentially ~10¹⁰–10¹¹ sequences) than current alternative methods based on DNA truncation (ITCHY, SCRATCHY and SHIPREC) or type IIb restriction enzymes (SISDC). At the same time, the frequency of frameshifts in the recombined library is low (90% of the recombined sequences are in frame). Compared to overlap extension PCR, USERec also requires much reduced crossover sequence constraints (only a 5'-AN_4–8T-3' motif) and fewer experimental steps. Based on its simplicity and flexibility, and the accessibility of large and high quality recombined DNA libraries, USERec is established as a convenient alternative for the combinatorial assembly of gene fragments (e.g. exon or domain shuffling) and for a number of applications in gene library construction, such as loop grafting and multi-site-directed or random mutagenesis.