Sriram Sokalingam

A Study on the Effect of Surface Lysine to Arginine Mutagenesis on Protein Stability and Structure Using Green Fluorescent Protein

Two positively charged basic amino acids, arginine and lysine, are mostly exposed to protein surface, and play important roles in protein stability by forming electrostatic interactions. In particular, the guanidinium group of arginine allows interactions in three possible directions, which enables arginine to form a larger number of electrostatic interactions compared to lysine. The higher pKa of the basic residue in arginine may also generate more stable ionic interactions than lysine. This paper reports an investigation whether the advantageous properties of arginine over lysine can be utilized to enhance protein stability. A variant of green fluorescent protein (GFP) was created by mutating the maximum possible number of lysine residues on the surface to arginines while retaining the activity. When the stability of the variant was examined under a range of denaturing conditions, the variant was relatively more stable compared to control GFP in the presence of chemical denaturants such as urea, alkaline pH and ionic detergents, but the thermal stability of the protein was not changed. The modeled structure of the variant indicated putative new salt bridges and hydrogen bond interactions that help improve the rigidity of the protein against different chemical denaturants. Structural analyses of the electrostatic interactions also confirmed that the geometric properties of the guanidinium group in arginine had such effects. On the other hand, the altered electrostatic interactions induced by the mutagenesis of surface lysines to arginines adversely affected protein folding, which decreased the productivity of the functional form of the variant. These results suggest that the surface lysine mutagenesis to arginines can be considered one of the parameters in protein stability engineering.

Conjugation of proteins by installing BIO-orthogonally reactive groups at their N-termini.

N-terminal site-specific modification of a protein has many advantages over methods targeting internal positions, but it is not easy to install reactive groups onto a protein in an N-terminal specific manner. We here report a strategy to incorporate amino acid analogues specifically in the N-terminus of a protein in vivo and demonstrate it by preparing green fluorescent protein (GFP) having bio-orthogonally reactive groups at its N-terminus. In the first step, GFP was engineered to be a foldable, internal methionine-free sequence via the semi-rational mutagenesis of five internal methionine residues and the introduction of mutations for GFP folding enhancement. In the second step, the N-terminus of the engineered protein was modified in vivo with bio-orthogonally functional groups by reassigning functional methionine surrogates such as L-homopropargylglycine and L-azidohomoalanine into the first methionine codon of the engineered internal methionine-free GFP. The N-terminal specific incorporation of unnatural amino acids was confirmed by ESI-MS analysis and the incorporation did not affect significantly the specific activity, refolding rate and folding robustness of the protein. The two proteins which have alkyne or azide groups at their N-termini were conjugated each other by bio-orthogonal Cu(I)-catalyzed click chemistry. The strategy used in this study is expected to facilitate bio-conjugation applications of proteins such as N-terminal specific glycosylation, labeling of fluorescent dyes, and immobilization on solid surfaces.

In silico Study on the Effect of Surface Lysines and Arginines on the Electrostatic Interactions and Protein Stability

Charged amino acids are mostly exposed on a protein surface, thereby forming a network of interactions with the surrounding amino acids as well as with water. In particular, positively charged arginine and lysine have different side chain geometries and provide a different number of potential electrostatic interactions. This study reports a comparative analysis of the difference in the number of two representative electrostatic interactions, such as salt-bridges and hydrogen bonds, contributed by surface arginine and lysine, as well as their effect on protein stability using molecular modeling and dynamics simulation techniques. Two in silico variants, the R variant with all arginines and the K variant with all lysines on the protein surface, were modeled by mutating all the surface lysines to arginines and the surface arginines to lysines, respectively, for each of the 10 model proteins. A structural comparison of the respective two variants showed that the majority of R variants possessed more salt-bridges and hydrogen bond interactions than the K variants, indicating that arginine provides a higher probability of electrostatic interactions than lysine owing to its side chain geometry. Molecular dynamics simulations of these variants revealed the R variants to be more stable than the K variants at room temperature but this effect was not prominent under protein denaturating conditions, such as 353 and 333 K with 8 M urea. These results suggest that the arginine residues on a protein surface contribute to the protein stability slightly more than lysine by enhancing the electrostatic interactions.

A comparative study on the stability and structure of two different green fluorescent proteins in organic co-solvent systems

Green fluorescent protein (GFP) has been used as a reporter marker in a wide range of biological and bioengineering studies. The expanded use of GFP in the field of biosensors, biochips and bio-conjugations requires the stability of GFP in organic co-solvent systems. This prompted us to examine the kinetic stability of two different GFP sequences, n-GFP and s-GFP, showing different folding robustness and thermodynamic stability, under a range of organic co-solvent systems. n-GFP and s-GFP are variants whose biophysical properties are comparable to wild type and super folder GFPs, respectively. The stability of n-GFP and s-GFP in 50% water-miscible organic solvents showed that s-GFP with higher thermodynamic stability exhibited much higher stability against organic solvents than n-GFP, which has lower thermodynamic stability. s-GFP was quite stable even in 90% organic solvents. Circular dichroism analysis confirmed that s-GFP maintained its native structure in organic co-solvent systems, whereas n-GFP showed structural variations under these conditions. Four highly fluctuating loop regions were identified from molecular dynamic simulations under the organic cosolvent conditions. A structural comparison of n-GFP and s-GFP suggested that the improved kinetic stability of s-GFP was due to its larger number of hydrogen bonds and salt-bridges that were present in four loop regions. This study suggests that thermodynamically stable s-GFP can be a good choice for use under harsh organic co-solvent conditions.

Deletional Protein Engineering Based on Stable Fold

Diversification of protein sequence-structure space is a major concern in protein engineering. Deletion mutagenesis can generate a protein sequence-structure space different from substitution mutagenesis mediated space, but it has not been widely used in protein engineering compared to substitution mutagenesis, because it causes a relatively huge range of structural perturbations of target proteins which often inactivates the proteins. In this study, we demonstrate that, using green fluorescent protein (GFP) as a model system, the drawback of the deletional protein engineering can be overcome by employing the protein structure with high stability. The systematic dissection of N-terminal, C-terminal and internal sequences of GFPs with two different stabilities showed that GFP with high stability (s-GFP), was more tolerant to the elimination of amino acids compared to a GFP with normal stability (n-GFP). The deletion studies of s-GFP enabled us to achieve three interesting variants viz. s-DL4, s-N14, and s-C225, which could not been obtained from n-GFP. The deletion of 191–196 loop sequences led to the variant s-DL4 that was expressed predominantly as insoluble form but mostly active. The s-N14 and s-C225 are the variants without the amino acid residues involving secondary structures around N- and C-terminals of GFP fold respectively, exhibiting comparable biophysical properties of the n-GFP. Structural analysis of the variants through computational modeling study gave a few structural insights that can explain the spectral properties of the variants. Our study suggests that the protein sequence-structure space of deletion mutants can be more efficiently explored by employing the protein structure with higher stability.

Deciphering the factors responsible for the stability of a GFP variant resistant to alkaline pH using molecular dynamics simulations

Charged amino acids having ionizable side chains play crucial roles in maintaining the solubility and stability of a protein. These charged amino acids are mostly exposed on protein surface and participate in electrostatic interactions with neighboring charged amino acids as well as with solvent. Therefore, the change in the solvent pH affects the protein stability in most cases. Previously, we reported a GFP variant, GFP14R having 14 surface lysines replaced with arginines, that showed enhanced stability under alkaline pH. Here, we analyzed the factors that contribute to the stability of the GFP14R under alkaline pH quantitatively using molecular dynamics simulations. Protonation state of the charged amino acids of GFP14R and control GFP under neutral pH and alkaline pH were modeled, and molecular dynamics simulations were performed. This comparative analysis revealed that the GFP14R with more arginine frequency on the surface maintained the stability under both pH conditions without much change in their salt-bridge interactions as well as the hydrogen bond interactions with solvent. On the other hand, these interactions were significantly reduced for the control GFP under alkaline pH due to the deprotonated lysine side chains. These results suggest that the advantageous property of arginine over lysine can be considered one of the parameter for the protein stability engineering under alkaline pH conditions.

Computational screening of potential non-immunoglobulin scaffolds using overlapped conserved residues (OCR)-based fingerprints

Cystatins and lipocalins have attracted considerable interest for their potential applications in non-immunoglobulin protein scaffold engineering. In the present study, their potential homologs were screened computationally from non-redundant protein sequence database based on the overlapped conserved residues (OCR)-fingerprints, which can detect the protein family with low sequence identity, such as cystatins and lipocalins. Two types of OCR-fingerprints for each family were designed and showed very high detection efficiency (>90%). The protein sequence database was scanned by the fingerprints, which yielded the hypothetical sequences for cystatins and lipocalins. The hypothetical sequences were validated further based on their sequence motifs and structural models, which allowed an identification of the potential homologs of cystatins and lipocalins.

In Silico Study on Retinoid-binding Modes in Human RBP and ApoD Lipocalins

Lipocalins are proteins with highly homologous structures but diverse sequences that are potential candidates for scaffold protein engineering with novel ligand-binding functions. Numerous crystal structures of lipocalin-ligand complexes have been identified and used in the study of their binding modes. On the other hand, crystallization studies cannot meet the increasing demand for novel lipocalin-ligand complexes in scaffold engineering, which requires rapid computational analyses of their binding modes in parallel. Human retinol-binding protein (RBP) and apolipoprotein D (apoD) are sequentially very distant proteins, but they show tight binding against retinoids, such as retinol and retinoic acid. In the present study, complexes of the two lipocalins with retinol and retinoic acid were modeled computationally by a molecular docking simulation, and their ligand-binding modes were analyzed at a molecular level. The models identified the crucial residues of lipocalins that interact with the ligands and revealed the similarities and differences in their retinoid-binding modes as well as in the specific interactions of the retinoid species within the same lipocalin. An analysis of the amino acid propensity of the retinoid-binding residues suggested that the evolutionary preference of the residues is restricted to the binding pocket rather than the entire protein. The distribution of charged residues around the terminus of retinoic acid showed a huge difference between RBP and ApoD, which might be a factor for the different binding affinities of lipocalins against retinoic acid. This in silico study is expected to be applied to scaffold protein engineering for novel retinoid-binding lipocalins.

Generation of anti-c-met single domain antibody fragment based on human stable frameworks

The size reduction is an important issue in the biomedical application of antibody and single domain antibody fragment is recognized as very attractive tool. However, it is very time-consuming and laborious to generate the fragment antibody with targeted binding function. Here, we investigated the possibility to prepare single domain antibody (sdAb) by a simple grafting method based on stable human consensus framework sequences. The complementarity determining region sequences in VH domain of anti-c-Met scFv from rabbit were grafted with the human VH3 consensus framework sequences, which generated the anti-c-Met single domain antibody showing almost same binding activity to its scFv form. The generated single domain antibody could be produced as functional form in oxidizing cytoplasm of E. coli, but produced as inactive form in reducing cytoplasm. The structural analysis of the homology models gave us the insight on the stability of the single domain antibody. In this report, we have demonstrated that the very stable human consensus framework sequence can be used for the generation of active anti-c-Met sdAb via complementarity determining regions grafting. We expect that this kind of grafting method for the generation of sdAb may provide us with the opportunities to prepare sdAbs based on the known antibody sequences.