Gooptu B , Miranda E , Nobeli I , Mallya M , Purkiss A , Brown SC , Summers C , Phillips RL , Lomas DA , Barrett TE .

Medical Research Council , Wellcome Trust , G0500306 Medical Research Council , MRC_G0500306 Medical Research Council

	Fig. 1. Structure of Thr114Phe α1-antitrypsin and in silico fragment screen. (a) Crystal structure of Thr114Phe α1-antitrypsin shown in cyan with key structural features coloured. The reactive loop is depicted in red, β-sheet A in dark blue, the shutter region in orange, helix D in green, helix F in bronze and strand 1 of β-sheet C (s1C) in purple. The cavity flanking β-sheet A is indicated (ellipse). (b) Superposition of wild-type (PDB code: 1QLP,16 yellow) and Thr114Phe (cyan) α1-antitrypsin with plot of the r.m.s.d. between the main chains (red) and side chains (blue). (c) Increased stability of the native fold in the presence of the Thr114Phe mutation. TUG-PAGE of wild-type (upper panel) and Thr114Phe (middle panel) α1-antitrypsin indicates that the mutation confers increased stability of the native fold to increasing urea concentration. CD spectroscopy allows urea stability to be compared more directly. Ellipticity values measured at 222 nm with varying urea concentrations are shown for wild-type and Thr114Phe α1-antitrypsin (n = 3). Values are normalised to allow direct comparison of the loss of initial signal. Curves were fitted in Grafit 3.0 (Erithacus Software Ltd.) using an equation modelling a simple two-state denaturation as described previously.8 Unfolding measured by this method commenced at 0.7 M urea for the wild-type protein (black curve, open circles) and 1.2 M urea for Thr114Phe α1-antitrypsin (blue curve, filled circles). The transition midpoints (assuming complete unfolding at 8 M urea) were reached at similar points (3.4 M urea for wild-type α1-antitrypsin and 3.5 M for Thr114Phe α1-antitrypsin) in both profiles. (d) Electron density for the Thr114Phe mutation site (boxed) prior to restrained refinement (upper panel). The 2.0-Å structure of wild-type α1-antitrypsin (1QLP with the mutation site modelled as an alanine was used as the search model, giving the 2Fo − Fc electron density map shown in dark blue (contoured at 1 σ). Positive density for the phenylalanine residue appears in the Fo − Fc difference map shown in red (contoured at 3 σ). The same site is shown in the refined structure (lower panel). (e) The surface-accessible cavity flanking β-sheet A, shown in close up for wild-type (left panel, yellow) and in the refined structure of Thr114Phe (middle panel, cyan) α1-antitrypsin with side chains and electrostatic surfaces. The volume occupied by the Thr114Phe mutation within the cavity defines a pharmacophore target for screening of fragment compounds to mimic its effects [right panel, cyan mesh, within the wild-type α1-antitrypsin cavity (transparent surface representation)]. (f) Results of proof-of-principle fragment screen. Left panel: Ensemble of the 65 highest-scoring ligands successfully docked using the induced fit protocol. Right panel: 4-D plot showing characteristics of these ligands (results shown correspond to 903 poses). All are highly drug-like and favourable for drug development according to the Rule of 517 and the Rule of 3.18 Heat map colouring indicates ligand–target centroid proximity (≥ 5.0 Å, yellow; 2.5–5.0 Å, orange; ≤ 2.5 Å, red). Glide docking score values < − 6.8 kcal/mol correspond to micromolar binding constants, and values below − 8.2 kcal/mol are consistent with predicted submicromolar Kd. Closer proximity to the pharmacophore target site indicates increased likelihood of a docked moiety mimicking effects of the Thr114Phe mutation. Smaller ligands allow more extensive structural optimisation during drug design.19 Four ligands that formed an outlier group in terms of particularly scoring using both Glide SP and Prime scoring algorithms are indicated (box). (g) Examples of ligands with favourable characteristics following induced fit docking. The pharmacophore target is represented as a transparent cyan surface, and ‘receptor residues’ (Asn104, Thr114 and His139) are shown as optimised for each ligand. Ligand colours correspond to those of receptor site residues optimised for their binding. Non-carbon atoms are individually coloured (oxygen, red; nitrogen, dark blue; fluorine, light blue; sulfur, orange). Left panel: the five ligands top-ranked by Glide SP docking score. Middle panel: the five ligands top-ranked by proximity of ligand to the pharmacophore target. Right panel: the ‘favourable outlier group’ defined in (f).
	Fig. 2. The 3.2-Å structure of Gly117Phe α1-antitrypsin. (a) Crystal structure of Gly117Phe α1-antitrypsin (pink; β-sheet A in dark blue). The helix F–β-sheet A linker region of the superposed native wild-type protein is shown (white). Box: Initial observation of changes around the mutation site following molecular replacement and rigid-body fitting using the wild-type α1-antitrypsin search model (1QLP, shown in yellow). At this stage, difference density (shown in red mesh, contoured at 3 σ) for the mutation site and overlying F helix was calculated by excluding residues 117 and 160 (+ surrounding 3.5 Å spheres) from a simulated annealing omit map to minimise model bias. Similar difference density was seen in the F-helix region in all three copies of the molecule; the example of a single copy is shown. (b) Side-chain packing of aromatic side chains in the helix F–β-sheet A interface in wild-type (left) and Gly117Phe (right) α1-antitrypsin. Only β-sheet A (blue, stick representation) and helix F together with its linker region (gold, transparent cartoon and stick representation) are shown. Residues involved in aromatic ring interactions are coloured red, and the mutation site at Gly117 is indicated for wild-type α1-antitrypsin. (d) Increased biochemical stability of the native fold in the presence of the Gly117Phe mutation. TUG-PAGE of wild-type (upper panel) and Gly117Phe (middle panel) α1-antitrypsin indicates that the mutation confers increased stability of the native fold to increasing urea concentration. This is confirmed by CD spectroscopy (lower panel). For Gly117Phe α1-antitrypsin (red curve, filled circles), unfolding begins at 2.0 M urea (compared with 0.7 M for the wild-type protein, black curve, open circles) and the transition midpoint occurs at 4.3 M urea (compared with 3.4 M for wild-type α1-antitrypsin). Conditions are as described for Fig. 1c. Taken together, the urea unfolding studies indicate increasing stability of the native fold in the order wild-type < Thr114Phe < Gly117Phe α1-antitrypsin.
	Fig. 3. Characterisation of M, Z, Thr114Phe/Z and Gly117Phe/Z α1-antitrypsin secreted from COS-7 cells. Western blot analysis of α1-antitrypsin retained within or secreted from COS-7 cells 2 days after transient transfection. (a) Western blot analysis of samples run on SDS-PAGE with (b) luciferase transfection control for cell lysate samples [same gel as (a)]. (c) Western blot analysis of samples run on non-denaturing PAGE. Lanes 1–5, cell lysates; lanes 6–10, culture media. Lane 1, empty vector control—cell lysate; lane 2, M α1-antitrypsin—cell lysate; lane 3, Z α1-antitrypsin—cell lysate; lane 4, Thr114Phe/Z α1-antitrypsin—cell lysate; lane 5, Gly117Phe/Z α1-antitrypsin—cell lysate; lane 6, empty vector control—supernatant; lane 7, M α1-antitrypsin—supernatant; lane 8, Z α1-antitrypsin—supernatant; lane 9, Thr114Phe/Z α1-antitrypsin—supernatant; lane 10, Gly117Phe/Z α1-antitrypsin—supernatant. Arrow indicates migration of native α1-antitrypsin secreted into the culture media on non-denaturing PAGE.
	Fig. 4. Schema for formation of a partially loop-inserted M⁎ species from native α1-antitrypsin. Findings from the Thr114Phe and Gly117Phe α1-antitrypsin crystal structures are incorporated together with previous observations of requirements for loss of the s1C26 strand, remodeling of the F helix9,12,28 and destabilising of interactions involving shutter region residues.25 Sequential insertion of the reactive loop into the upper s4A position is depicted by two chimeras. These are derived from structures of native (1QLP) and latent (1IZ2) α1-antitrypsin, murine α1-antichymotrypsin (1YXA—demonstrating changes associated with opening of the P14 acceptor site) and thyroxine binding globulin (2CEO—demonstrating changes on expansion at the P12 insertion site). In all cases, the reactive loop modelled is that of α1-antitrypsin. The final image shows the α1-antitrypsin M⁎ model, generated as described in Materials and Methods, after energy minimisation with simulated annealing molecular dynamics. The effects of mutations that will facilitate this transition are shown in red whilst those that block it are shown in blue. Thus, in this scheme, Thr114Phe retards M⁎ formation by cushioning β-sheet A against expansion whilst Gly117Phe stabilises the upper turns of helix F and abolishes the steric requirement for their remodelling in response to partial loop insertion. Conversely, mutations may accelerate polymerisation by favouring partial loop insertion (e.g., Z) or by opening the lower s4A position directly (e.g., Siiyama).

Aulak, A hinge region mutation in C1-inhibitor (Ala436-->Thr) results in nonsubstrate-like behavior and in polymerization of the molecule. 1993, Pubmed

Aulak, A hinge region mutation in C1-inhibitor (Ala436-->Thr) results in nonsubstrate-like behavior and in polymerization of the molecule. 1993, Pubmed
Baglin, Crystal structures of native and thrombin-complexed heparin cofactor II reveal a multistep allosteric mechanism. 2002, Pubmed
Baugh, Human leukocyte granule elastase: rapid isolation and characterization. 1976, Pubmed
Brünger, Crystallography & NMR system: A new software suite for macromolecular structure determination. 1998, Pubmed
Cabrita, Probing the role of the F-helix in serpin stability through a single tryptophan substitution. 2002, Pubmed
Cabrita, Different conformational changes within the F-helix occur during serpin folding, polymerization, and proteinase inhibition. 2004, Pubmed
Carr, Fragment-based lead discovery: leads by design. 2005, Pubmed
Chang, Importance of the release of strand 1C to the polymerization mechanism of inhibitory serpins. 1997, Pubmed
Dafforn, A kinetic mechanism for the polymerization of alpha1-antitrypsin. 1999, Pubmed
Dunstone, Cleaved antitrypsin polymers at atomic resolution. 2000, Pubmed
Eldering, COOH-terminal substitutions in the serpin C1 inhibitor that cause loop overinsertion and subsequent multimerization. 1995, Pubmed
Elliott, Wild-type alpha 1-antitrypsin is in the canonical inhibitory conformation. 1998, Pubmed
Elliott, Topography of a 2.0 A structure of alpha1-antitrypsin reveals targets for rational drug design to prevent conformational disease. 2000, Pubmed
Emsley, Coot: model-building tools for molecular graphics. 2004, Pubmed
Fitton, Mechanisms of antithrombin polymerisation and heparin activation probed by the insertion of synthetic reactive loop peptides. 1997, Pubmed
Gooptu, Inactive conformation of the serpin alpha(1)-antichymotrypsin indicates two-stage insertion of the reactive loop: implications for inhibitory function and conformational disease. 2000, Pubmed
Guex, SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. 1997, Pubmed
Huber, Implications of the three-dimensional structure of alpha 1-antitrypsin for structure and function of serpins. 1989, Pubmed
Huntington, Structure of a serpin-protease complex shows inhibition by deformation. 2000, Pubmed
Huntington, A 2.6 A structure of a serpin polymer and implications for conformational disease. 1999, Pubmed
Im, Interactions causing the kinetic trap in serpin protein folding. 2002, Pubmed
Irwin, ZINC--a free database of commercially available compounds for virtual screening. 2005, Pubmed
James, The mechanism of alpha 1-antitrypsin polymerization probed by fluorescence spectroscopy. 1998, Pubmed
Jin, The anticoagulant activation of antithrombin by heparin. 1997, Pubmed
Laskowski, SURFNET: a program for visualizing molecular surfaces, cavities, and intermolecular interactions. 1995, Pubmed
Lipinski, Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. 2001, Pubmed
Lomas, The mechanism of Z alpha 1-antitrypsin accumulation in the liver. 1992, Pubmed
Mahadeva, 6-mer peptide selectively anneals to a pathogenic serpin conformation and blocks polymerization. Implications for the prevention of Z alpha(1)-antitrypsin-related cirrhosis. 2002, Pubmed
Mahadeva, Heteropolymerization of S, I, and Z alpha1-antitrypsin and liver cirrhosis. 1999, Pubmed , Xenbase
Mallya, Small molecules block the polymerization of Z alpha1-antitrypsin and increase the clearance of intracellular aggregates. 2007, Pubmed
Mast, Analysis of the plasma elimination kinetics and conformational stabilities of native, proteinase-complexed, and reactive site cleaved serpins: comparison of alpha 1-proteinase inhibitor, alpha 1-antichymotrypsin, antithrombin III, alpha 2-antiplasmin, angiotensinogen, and ovalbumin. 1991, Pubmed
McGowan, X-ray crystal structure of MENT: evidence for functional loop-sheet polymers in chromatin condensation. 2006, Pubmed
Miranda, Mutants of neuroserpin that cause dementia accumulate as polymers within the endoplasmic reticulum. 2004, Pubmed
Miranda, The intracellular accumulation of polymeric neuroserpin explains the severity of the dementia FENIB. 2008, Pubmed
Murshudov, Refinement of macromolecular structures by the maximum-likelihood method. 1997, Pubmed
Parfrey, Targeting a surface cavity of alpha 1-antitrypsin to prevent conformational disease. 2003, Pubmed , Xenbase
Pettersen, UCSF Chimera--a visualization system for exploratory research and analysis. 2004, Pubmed
Schechter, On the size of the active site in proteases. I. Papain. 1967, Pubmed
Sherman, Novel procedure for modeling ligand/receptor induced fit effects. 2006, Pubmed
Silverman, The serpins are an expanding superfamily of structurally similar but functionally diverse proteins. Evolution, mechanism of inhibition, novel functions, and a revised nomenclature. 2001, Pubmed
Sivasothy, Pathogenic alpha 1-antitrypsin polymers are formed by reactive loop-beta-sheet A linkage. 2000, Pubmed
Stein, What do dysfunctional serpins tell us about molecular mobility and disease? 1995, Pubmed
Tsutsui, The conformational dynamics of a metastable serpin studied by hydrogen exchange and mass spectrometry. 2006, Pubmed
Yamasaki, Crystal structure of a stable dimer reveals the molecular basis of serpin polymerization. 2008, Pubmed
Zhou, How small peptides block and reverse serpin polymerisation. 2004, Pubmed
Zhou, The serpin inhibitory mechanism is critically dependent on the length of the reactive center loop. 2001, Pubmed
Zhou, Serpin polymerization is prevented by a hydrogen bond network that is centered on his-334 and stabilized by glycerol. 2003, Pubmed