Probing binding mechanism of interleukin-6 and olokizumab: in silico design of potential lead antibodies for autoimmune and inflammatory diseases

Computer-aided antibody engineering has been successful in the design of new biologics for disease diagnosis and therapeutic interventions. Interleukin-6 (IL-6), a well-recognized drug target for various autoimmune and inflammatory diseases such as rheumatoid arthritis, multiple sclerosis, and psoriasis, was investigated in silico to design potential lead antibodies. Here, crystal structure of IL-6 along with monoclonal antibody olokizumab was explored to predict antigen–antibody (Ag???Ab)-interacting residues using DiscoTope, Paratome, and PyMOL. Tyr56, Tyr103 in heavy chain and Gly30, Ile31 in light chain of olokizumab were mutated with residues Ser, Thr, Tyr, Trp, and Phe. A set of 899 mutant macromolecules were designed, and binding affinity of these macromolecules to IL-6 was evaluated through Ag???Ab docking (ZDOCK, ClusPro, and Rosetta server), binding free-energy calculations using Molecular Mechanics/Poisson Boltzman Surface Area (MM/PBSA) method, and interaction energy estimation. In comparison to olokizumab, eight newly designed theoretical antibodies demonstrated better result in all assessments. Therefore, these newly designed macromolecules were proposed as potential lead antibodies to serve as a therapeutics option for IL-6-mediated diseases.     

Dynamics and modulation studies of human voltage gated Kv1.5 channel

The voltage gated Kv1.5 channels conduct the ultrarapid delayed rectifier current (IK_ur) and play critical role in repolarization of action potential duration. It is the most rapidly activated channel and has very little or no inactivated states. In human cardiac cells, these channels are expressed more extensively in atrial myocytes than ventricle. From the evidences of its localization and functions, Kv1.5 has been declared a selective drug target for the treatment of atrial fibrillation (AF). In this present study, we have tried to identify the rapidly activating property of Kv1.5 and studied its mode of inhibition using molecular modeling, docking, and simulation techniques. Channel in open conformation is found to be stabilized quickly within the dipalmitoylphosphatidylcholine membrane, whereas most of the secondary structure elements were lost in closed state conformation. The obvious reason behind its ultra-rapid property is possibly due to the amino acid alteration in S4–S5 linker; the replacement of Lysine by Glutamine and vice versa. The popular published drugs as well as newly identified lead molecules were able to inhibit the Kv1.5 in a very similar pattern, mainly through the nonpolar interactions, and formed sable complexes. V512 is found as the main contributor for the interaction along with the other important residues such as V505, I508, A509, V512, P513, and V516. Furthermore, two screened novel compounds show surprisingly better inhibitory potency and can be considered for the future perspective of antiarrhythmic survey.     

Comparison of different methods of determining cell viability after exposure to cytotoxic compounds

Cell-survival (of DON and L1210 cells) after treatment with cytotoxic compounds was assessed by measuring cloning efficiency, exclusion of trypan blue and erythrosin B, [⁵¹Cr] release, and attachment of DON cells to glass. Cell survival as measured by cloning efficiency did not correlate with survival measured by any of the other methods. We found that the stainability of cells after drug exposure depended on the cell line used. For example, after 3 h exposure to tubercidin although 100% of both DON and L1210 cells were killed (on basis of cloning efficiency), only 11% of DON cells and 68% of L1210 cells were dead as indicated by staining with erythrosin B. The stainability of cells also depended on the particular drug used. For example, after 24 h exposure of L1210 cells to adriamycin and tubercidin (both killed >99% of cells on basis of cloning efficiency) 21% of cells exposed to adriamycin and 99% of cells exposed to tubercidin were stained. The results obtained with several other cytotoxic compounds are discussed.     

Modelling family 2 cystatins and their interaction with papain

Cystatins are extensively studied cysteine protease inhibitors, found in wide range of organisms with highly conserved structural folds. S-type of cystatins is well known for their abundance in saliva, high selectivity and poorer activity towards host cysteine proteases in comparison to their immediate ancestor cystatin C. Despite more than 90% sequence similarity, the members of this group show highly dissimilar binding affinity towards papain. Cystatin M/E is a potent inhibitor of legumain and papain like cysteine proteases and recognized for its involvement in skin barrier formation and potential role as a tumor suppressor gene. However, the structures of these proteins and their complexes with papain or legumain are still unknown. In the present study, we have employed computational methods to get insight into the interactions between papain and cystatins. Three-dimensional structures of the cystatins are generated by homology modelling, refined with molecular dynamics simulation, validated through numerous web servers and finally complexed with papain using ZDOCK algorithm in Discovery Studio. A high degree of shape complementarity is observed within the complexes, stabilized by numerous hydrogen bonds (HB) and hydrophobic interactions. Using interaction energy, HB and solvent accessible surface area analyses, we have identified a series of key residues that may be involved in papain–cystatin interaction. Differential approaches of cystatins towards papain are also noticed which are possibly responsible for diverse inhibitory activity within the group. These findings will improve our understanding of fundamental inhibitory mechanisms of cystatin and provide clues for further research.     

Lens-specific regulation of the thioredoxin-1 gene, but not thioredoxin-2, upon in vivo photochemical oxidative stress in the Emory mouse

The angiotensin II (Ang II) type 1 receptor mediates various actions of Ang II, whereas the function of the type 2 (AT2) receptor is not well understood. In the mice lacking the gene encoding the AT2 receptor, the pressor response to Ang II was increased although the underlying mechanism is unknown. We tested the hypothesis that vasoconstrictor response is exaggerated in the AT2 receptor null mice. We measured hemodynamic parameters and evaluated systemic vascular resistance (SVR) in the anesthetized open-chest wild-type and AT2 receptor null mice. Ang II infusion caused dose-dependent increases in SVR in both strains, while the response was significantly higher at 0.5 microgram/kg Ang II in the AT2 receptor null mice (305 +/- 53% of baseline) than in the wild-type mice (179 +/- 27% of baseline). To investigate further the vascular contractility, we examined contraction of aortic rings in vitro. The contraction induced by 1 microM Ang II was increased in the AT2 receptor null mice compared with that in the wild-type mice (0.82 +/- 0.11 vs. 0.54 +/- 0.12 g). Ang II-induced contraction was still greater in the AT2 receptor null mice when calibrated by the maximum tension induced by 90 mM KCl. These data suggest that the AT2 receptor modulates vascular contractility, which may influence blood pressure.     

Backbone dynamics of free barnase and its complex with barstar determined by 15N NMR relaxation study

Backbone dynamics of uniformly ¹⁵N-labeled free barnase and its complex with unlabelled barstar have been studied at 40°C, pH 6.6, using ¹⁵N relaxation data obtained from proton-detected 2D {¹H}-¹⁵N NMR spectroscopy. ¹⁵N spin-lattice relaxation rate constants (R₁), spin-spin relaxation rate constants (R₂), and steady-state heteronuclear {¹H}-¹⁵N NOEs have been measured at a magnetic field strength of 14.1 Tesla for 91 residues of free barnase and for 90 residues out of a total of 106 in the complex (excluding three prolines and the N-terminal residue) backbone amide ¹⁵N sites of barnase. The primary relaxation data for both the cases have been analyzed in the framework of the model-free formalism using both isotropic and axially symmetric models of the rotational diffusion tensor. As per the latter, the overall rotational correlation times (_m) are 5.0 and 9.5 ns for the free and complexed barnase, respectively. The average order parameter is found to be 0.80 for free barnase and 0.86 for the complex. However, the changes are not uniform along the backbone and for about 5 residues near the binding interface there is actually a significant decrease in the order parameters on complex formation. These residues are not involved in the actual binding. For the residues where the order parameter increases, the magnitudes vary significantly. It is observed that the complex has much less internal mobility, compared to free barnase. From the changes in the order parameters, the entropic contribution of NH bond vector motion to the free energy of complex formation has been calculated. It is apparent that these motions cause significant unfavorable contributions and therefore must be compensated by many other favorable contributions to effect tight complex formation. The observed variations in the motion and their different locations with regard to the binding interface may have important implications for remote effects and regulation of the enzyme action.     

The slow folding reaction of barstar: the core tryptophan region attains tight packing before substantial secondary and tertiary structure formation and final compaction of the polypeptide chain

The slow folding of a single tryptophan-containing mutant of barstar has been studied in the presence of 2 M urea at 10 degrees C, using steady state and time-resolved fluorescence methods and far and near-UV CD measurements. The protein folds in two major phases: a fast phase, which is lost in the dead time of measurement during which the polypeptide collapses to a compact form, is followed by a slow observable phase. During the fast phase, the rotational correlation time of Trp53 increases from 2.2 ns to 7.2 ns, and its mean fluorescence lifetime increases from 2.3 ns to 3.4 ns. The fractional changes in steady-state fluorescence, far-UV CD, and near-UV CD signals, which are associated with the fast phase are, respectively, 36 %, 46 %, and 16 %. The product of the fast phase can bind the hydrophobic dye ANS. These observations together suggest that the folding intermediate accumulated at the end of the fast phase has: (a) about 20 % of the native-state secondary structure, (b) marginally formed or disordered tertiary structure, (c) a water-intruded and mobile protein interior; and (d) solvent-accessible patches of hydrophobic groups. Measurements of the anisotropy decay of Trp53 suggest that it undergoes two types of rotational motion in the intermediate: (i) fast (tau(r) approximately 1 ns) local motion of its indole side-chain, and (ii) a slower (tau(r) approximately 7.2 ns) motion corresponding to global tumbling of the entire protein molecule. The ability of the Trp53 side-chain to undergo fast local motion in the intermediate, but not in the fully folded protein where it is completely buried in the hydrophobic core, suggests that the core of the intermediate is still poorly packed. The global tumbling time of the fully folded protein is faster at 5.6 ns, suggesting that the volume of the intermediate is 25 % more than that of the fully folded protein. The rate of folding of this intermediate to the native state, measured by steady-state fluorescence, far-UV CD, and near-UV CD, is 0.07(+/-0.01) min(-1) This rate compares to a rate of folding of 0.03(+/-0.005) min(-1), determined by double-jump experiments which monitor directly formation of native protein; and to a rate of folding of 0.05 min(-1), when determined from time-resolved anisotropy measurements of the long rotational correlation time, which relaxes from an initial value of 7.2 ns to a final value of 5. 6 ns as the protein folds. On the other hand, the amplitude of the short correlation time decreases rapidly with a rate of 0.24(+/-0.06) min(-1). These results suggest that tight packing of residues in the hydrophobic core occurs relatively early during the observable slow folding reaction, before substantial secondary and tertiary structure formation and before final compaction of the protein.     

CD98hc regulates the development of experimental colitis by controlling effector and regulatory CD4 T cells

CD4⁺ T cell activation is controlled by signaling through the T cell receptor in addition to various co-receptors, and is also affected by their interactions with effector and regulatory T cells in the microenvironment. Inflammatory bowel diseases (IBD) are caused by the persistent activation and expansion of auto-aggressive CD4⁺ T cells that attack intestinal epithelial cells. However, the molecular basis for the persistent activation of CD4⁺ T cells in IBD remains unclear. In this study, we investigated how the CD98 heavy chain (CD98hc, Slc3a2) affected the development of colitis in an experimental animal model. Transferring CD98hc-deficient CD4⁺CD25⁻ T cells into Rag2^−/− mice did not cause colitis accompanied by increasing Foxp3⁺ inducible regulatory T cells. By comparison, CD98hc-deficient naturally occurring regulatory T cells (nTregs) had a decreased capability to suppress colitis induced by CD4⁺CD25⁻ T cells, although CD98hc-deficient mice did not have a defect in the development of nTregs. Blocking CD98hc with an anti-CD98 blocking antibody prevented the development of colitis. Our results indicate that CD98hc regulates the expansion of autoimmune CD4⁺ T cells in addition to controlling nTregs functions, which suggests the CD98hc as an important target molecule for establishing strategies for treating colitis.