Journal of Molecular Biology
Volume 342, Issue 3, 17 September 2004, Pages 931-941
Journal home page for Journal of Molecular Biology

How Small Peptides Block and Reverse Serpin Polymerisation

https://doi.org/10.1016/j.jmb.2004.07.078Get rights and content

Many of the late-onset dementias, including Alzheimer's disease and the prion encephalopathies, arise from the aberrant aggregation of individual proteins. The serpin family of serine protease inhibitors provides a well-defined structural example of such pathological aggregation, as its mutant variants readily form long-chain polymers, resulting in diseases ranging from thrombosis to dementia. The intermolecular linkages result from the insertion of the reactive site loop of one serpin molecule into the middle strand (s4A) position of the A β-sheet of another molecule. We define here the structural requirements for small peptides to competitively bind to and block the s4A position to prevent this intermolecular linkage and polymerisation. The entry and anchoring of blocking-peptides is facilitated by the presence of a threonine which inserts into the site equivalent to P8 of s4A. But the critical requirement for small blocking-peptides is demonstrated in crystallographic structures of the complexes formed with selected tri- and tetrapeptides. These structures indicate that the binding is primarily due to the insertion of peptide hydrophobic side-chains into the P4 and P6 sites of s4A. The findings allow the rational design of synthetic blocking-peptides small enough to be suitable for mimetic design. This is demonstrated here with a tetrapeptide that preferentially blocks the polymerisation of a pathologically unstable serpin commonly present in people of European descent.

Introduction

The conformational diseases,1 which include the common dementias of Alzheimer's,2 Parkinson's3 and the spongiform encephalopathies,4 all result from aberrant β-interlinkages of an underlying protein. The serpin family of serine protease inhibitors provides the best-defined structural model of such β-linkages and of the consequent protein aggregation and disease.5, 6, 7 Mutations that affect the conformational stability of serpins such as the plasma protease inhibitors antithrombin and α1-antitrypsin (hereafter referred to as antitrypsin), lead to their polymerisation and intracellular aggregation. The accompanying loss of inhibitory activity results with antithrombin in a distinctive thrombotic disease,8 and with antitrypsin in the slow onset of the destructive lung disease, emphysema.9 But even more severe consequences arise from the concomitant cell damage due to the massive intracellular aggregation of the conformationally unstable serpins. This occurs with antitrypsin, where its accumulation within the hepatocyte leads to liver cirrhosis10 and with neuroserpin, where its intracellular accumulation is associated with a cumulative loss of neurons and the onset of an Alzheimer's-like dementia.11

The polymerisation of serpins results from an abnormality of their inhibitory mechanism. The inhibition of proteases by serpins requires the insertion of the first 14 residues of the cleaved reactive loop (P14-1) into the vacant strand 4 position of the main (A) β-sheet of the molecule.12, 13 Pathological molecular interlinkage occurs when the A-sheet aberrantly opens and allows a domain exchange with the insertion of the P8-3 portion of the loop of another molecule into the lower half of the s4A position, with sequential intermolecular linkages resulting in polymer or fibril formation14, 15, 16 (Figure 1a–c). It has been shown that such polymerisation can be blocked by competition with synthetic P14-3 or P7-2 peptides14, 17, 18, 19 (Table 1), but peptides are unsuitable agents for long-term therapy and both peptides are too large to act as a basis for mimetic drug design. Moreover, the results with different peptides seemed to be unpredictable and often surprising, with the P14-3 peptide from antithrombin acting as a better blocking agent with antitrypsin than does antitrypsin's own homologous peptide. Another indication of the unpredictability of binding came from the crystallographic study by Xue et al.,20 with the serpin plasminogen activator inhibitor, PAI-1, where a P14-10 peptide used to block inhibitory activity was unexpectedly found to also bind to the P6-2 position at the bottom of the A-sheet.

The explanation of these diverse findings developed from an observation at the commencement of this study; that a tetrapeptide WMDF (Trp-Met-Asp-Phe), unrelated to the serpins, was found to effectively block the polymerisation of both antitrypsin and antithrombin (Figure 1d). This control tetrapeptide, derived from cholecystokinin, provided a basis for the selection, by screening, of several relatively effective tetra- and tri-peptide blocking agents. The determination of the crystal structures of these small blocking peptides bound to antithrombin clarifies the specific requirements for peptide binding and also provides a structural explanation of the previously puzzling observations with natural peptides. Definition of the structural requirements for peptide binding has now enabled the elective design of a novel tetrapeptide that selectively blocks the polymerisation of the conformationally unstable variants of antitrypsin. This is demonstrated here with a priority medical target, the Z mutant of antitrypsin, commonly present in people of European descent, which causes a predisposition to emphysema and liver cirrhosis.6

Section snippets

Peptide assessment

Altogether some 40 small peptides of three to five amino acid residues randomly selected from commercial sources, together with peptides derived from serpin reactive loop sequences were screened as potential blocking agents, using different polymerogenic test-serpins. Antithrombin is converted to a polymerogenic form with the A-sheet opened by prior annealing of a P14-9 (or P14-8) peptide to the upper half of the strand 4 position.18 As shown in Figure 2a, antithrombin readily forms polymers in

Discussion

The conformational diseases such as Alzheimer's, Parkinson's and the prion encephalopathies pose a special challenge to structural biologists. They are all diseases resulting from aberrant β-strand interlinkages. Yet the end aim of research into the underlying molecular mechanisms, to develop specific therapies to prevent these intermolecular linkages, seems to be a daunting prospect. However, the insights we now have from the prototypical conformational diseases of the serpins are much more

Proteins

Human antithrombin and antitrypsin were isolated from fresh frozen human plasma as previously described.8, 29 PP4, a glycine specific cysteine protease that cleaves antitrypsin at P1021, was purchased from CN Biosciences UK, Nottingham, UK.

Peptide selection

Altogether some 40 small peptides of three to five amino acid residues were randomly selected from commercial sources and various serpin derived peptides were synthesised. The synthetic peptides Ac-SEAAASTAVVIA (antithrombin P14-3), Ac-SEAAAST (antithrombin

Acknowledgements

This study was supported by the Wellcome Trust, the NIH, the MRC, the Isaac Newton Trust, and the Alpha-1 Foundation. We thank Dr Helen Parfrey, University of Cambridge, for providing Z antitrypsin and Professor Randy J. Read for advice on crystallography.

References (40)

  • I. Bjork et al.

    Conversion of antithrombin from an inhibitor of thrombin to a substrate with reduced heparin affinity and enhanced conformational stability by binding of a tetradecapeptide corresponding to the P1 to P14 region of the putative reactive bond loop of the inhibitor

    J. Biol. Chem.

    (1992)
  • E.A. Merritt et al.

    Raster3D: photorealistic molecular graphics

    Methods Enzymol.

    (1997)
  • D.E. McRee

    XtalView/Xfit—a versatile program for manipulating atomic coordinates and electron density

    J. Struct. Biol.

    (1999)
  • R.R. Kopito et al.

    Conformational disease

    Nature Cell Biol.

    (2000)
  • D.J. Selkoe

    Alzheimer's disease: genes, proteins, and therapy

    Physiol. Rev.

    (2001)
  • K.A. Conway et al.

    Acceleration of oligomerization, not fibrillization, is a shared property of both alpha-synuclein mutations linked to early-onset Parkinson's disease: implications for pathogenesis and therapy

    Proc. Natl Acad. Sci. USA

    (2000)
  • R.W. Carrell et al.

    Alpha1-antitrypsin deficiency–a model for conformational diseases

    N. Engl. J. Med.

    (2002)
  • D.A. Lomas et al.

    Human genetics and disease: serpinopathies and the conformational dementias

    Nature Rev. Genet.

    (2002)
  • C.-B. Laurell et al.

    The electrophoretic α1-globulin pattern of serum in α1-antitrypsin deficiency

    Scand. J. Clin. Lab. Invest.

    (1963)
  • H.L. Sharp et al.

    Cirrhosis associated with alpha-1-antitrypsin deficiency: a previously unrecognised inherited disorder

    J. Lab. Clin. Med.

    (1969)
  • Cited by (77)

    • Alpha1-antitrypsin deficiency: New therapies on the horizon

      2021, Current Opinion in Pharmacology
      Citation Excerpt :

      Z Factor Ltd is another company aiming to use folding correctors to treat AATD. It relies on structural insights gained in the laboratory of Dr. James Huntington [15,16] and recently started a phase 1, double-blind, randomised, placebo-controlled, single ascending and repeat dose clinical trial (NCT04443192) recruiting healthy volunteers and individuals heterozygous for the Z-mutation (Pi∗MZ). The aims of the trial are to evaluate the safety and tolerability of the orally administered compound ZF874.

    • Alpha1-Antitrypsin: Structure and Dynamics in Health, Disease and Drug Development

      2017, Alpha-1-antitrypsin Deficiency: Biology, Diagnosis, Clinical Significance, and Emerging Therapies
    View all citing articles on Scopus
    View full text