Journal of Molecular Biology
Volume 341, Issue 5, 27 August 2004, Pages 1175-1187
Journal home page for Journal of Molecular Biology

Phospholipid Catalysis of Diabetic Amyloid Assembly

https://doi.org/10.1016/j.jmb.2004.06.086Get rights and content

Islet amyloid polypeptide (IAPP) is a 37-residue hormone that forms cytotoxic amyloid fibers in the endocrine pancreas of patients with type II diabetes (NIDDM). A potential origin for cytotoxicity is disruption of lipid membranes by IAPP as has been observed in vitro. The cause of amyloid formation during NIDDM is not known, nor is the mechanism by which membrane disruption occurs in vitro. Here, we use kinetic studies in conjunction with assessments of lipid binding and electron microscopy to investigate the interactions of IAPP with phospholipid bilayers and the morphological effects of membranes on IAPP fibers. Fibrillogenesis of IAPP is catalyzed by synthetic and human tissue-derived phospholipids, leading to >tenfold increases in the rate of fibrillogenesis. The molecular basis of this phenomenon includes a strong dependence on the concentration and charge density of the membrane. IAPP binds to lipid membranes of mixed anionic (DOPG) and zwitterionic (DOPC) content. The transition for binding occurs over a physiologically relevant range of anionic content. Membrane binding by IAPP occurs on timescales that are short compared to fibrillogenesis and results in assembly into preamyloid states via ordered interactions at the N but not C terminus of the protein. These assemblies lead both to gross morphological changes in liposomes and to alterations in the appearance of early fibers when compared to liposome-free fibril formation. Intact bilayer surfaces are regenerated upon dissociation of fibers from the membrane surface. These findings offer a structural mechanism of membrane destabilization and suggest that changes in lipid metabolism could induce IAPP fiber formation in NIDDM.

Introduction

Amyloid formation is a critical component of the pathology of many degenerative diseases, including Alzheimer's and type II diabetes (NIDDM).1., 2. For each disease, a unique protein undergoes changes in conformation resulting in polymerization to a fibrous structure. Amyloid fibers are characterized by a cross-β structure in which β strands are organized orthogonal to the long axis of an unbranched fiber.3 Biochemically, these fibers are characterized by extreme apparent resistance to chemical and proteolytic degradation. As a result, in vivo deposits of amyloid are not readily cleared by cellular processes. Fiber formation is destructive either through macroscopic obstruction, e.g. in light chain amyloidosis,4 or by leading to cellular dysfunction and death, e.g. in NIDDM.1., 5.

NIDDM is a disease that afflicts more than 150 million individuals worldwide. The pathology of NIDDM includes the presence of amyloid deposits in the islets of Langerhans. The main component of these deposits, called islet amyloid polypeptide (IAPP) or amylin, is a 37-residue hormone (Figure 1) that is normally stored with insulin in vesicles and secreted by islet β-cells. Islet amyloidosis in patients with NIDDM is associated with a reduced mass of insulin-producing β-cells.6 In addition, oligomeric intermediates of fiber formation have been shown to be cytotoxic to cultured β-cells.7., 8. Fiber formation is therefore an important factor in the development of β-cell failure. Since IAPP isolated from amyloid fibers in diabetic patients is wild-type, the obvious candidates for agents that might affect fiber formation are the other granule components. The most abundant protein in the granule is insulin, which is secreted in ∼100-fold molar excess over IAPP. In vitro, however, substoichiometric amounts of insulin readily inhibit IAPP fiber formation.9 Therefore, fiber formation in vivo must involve an accelerant and/or isolation of IAPP from interaction with this inhibitor.

Several mechanisms for induction of β-cell death in NIDDM have been proposed.10 One possibility is that IAPP forms cation-permeable pores, which disrupt membranes and lead to cell death.11., 12. Indeed, preamyloid conformations of IAPP are able to induce membrane destabilization and apoptosis in healthy cultured islet cells.7 An alternative hypothesis is that elevated glucose and lipid levels associated with early NIDDM overstimulate metabolic and signaling pathways, resulting in β-cell failure and apoptosis.13., 14. These two hypotheses need not be mutually exclusive. It is reasonable to suggest that alterations in metabolism accompanying NIDDM onset provide the stimulus for IAPP fiber formation. Immunostaining studies of IAPP in healthy human β-cells show localization of the protein in a narrow region of the granule between a central core of insulin and the lipid bilayer.15 Furthermore, in transgenic mice which express human IAPP, fiber formation requires high fat intake or genetic alteration of lipid metabolism.16., 17. It is therefore plausible that alterations in lipid metabolism accompanying obesity and NIDDM provide for changes of the bilayer environment which lead to pathological interactions with IAPP.

We have previously characterized the fiber formation pathway of IAPP upon its in vitro dilution into physiological buffer from a stock solution in organic solvent.9., 18., 19. IAPP fibrillogenesis kinetics generally follow a sigmoidal profile; an initial lag phase of relatively slow fiber nucleation precedes a rapid elongation phase during which the remainder of the soluble protein is converted into fiber. One central result from our studies is that the rate of fiber formation is exceptionally sensitive to the residual presence of the cosolvent hexafluoroisopropanol (HFIP). For example, at 4% HFIP, fiber formation occurs rapidly following a lag phase of a few minutes. In contrast, at 2% HFIP, the midpoint of fiber formation is reached after several hours.19 Fluorinated alcohols have both hydrophobic and polar moieties and are therefore capable of providing a variety of interactions that can affect conformation and assembly processes.19., 20. A similar variety of interactions can be provided in vivo by phospholipid membranes. Phospholipids can stabilize polar and electrostatic interactions through their head groups and hydrophobic interactions in the transmembrane region.21., 22. Here, we use an established lipid binding assay and kinetic studies coupled with electron microscopy to elucidate the mechanism of IAPP fibrillogenesis in the presence of synthetic and tissue-derived lipid membranes.

Section snippets

Results

In this study, we investigate the mechanism of IAPP fiber formation under conditions approximating those found in the NIDDM β-cell. First, we establish an in vitro system amenable to kinetic studies in the presence of lipid bilayers. We then assess the effects of patient-derived and synthetic lipid bilayers on IAPP fibrillogenesis. Our analyses focus on the kinetic consequences of varying lipid headgroup and protein:lipid ratio, explicit measurement of protein–lipid binding, and the

Discussion

IAPP is a peptide hormone, packaged in proximity to the lipid bilayer of a secretory granule. Its conversion to amyloid is central to the progression of NIDDM. The following in vitro observations offer insight into the mechanism of lipid-mediated fibrillogenesis: (i) IAPP fiber formation is dramatically accelerated by negatively charged phospholipid bilayers, including those derived from diabetic pancreas. (ii) Lipid binding by IAPP is rapid compared to the timescale of fiber formation. (iii)

Materials

Synthetic lipids were obtained from Avanti (Alabaster, AL) dissolved in chloroform; DMSO was from J.T. Baker or Sigma; Thioflavin T (ThT) was from Acros; heparin and salmon sperm DNA were from Sigma. Human IAPP was synthesized using standard TBOC methods and purified by the W.M. Keck facility (New Haven, CT). TAMRA-IAPP, rIAPP, IAPP8-37, and IAPPK1E were synthesized using Fmoc methods and cyclized and purified in house.

Isolation of tissue phospholipids and liposome preparation

Pancreatic tissue was obtained post-mortem and flash-frozen from a

Acknowledgements

The authors thank Drs V. Unger and D. Chester for assistance with EM and helpful discussions, Dr D. Engelman for helpful discussion, Dr F. Sigworth for use of fluorimeter, and Drs S. Jaswal and C. Atreya for critical reading of this manuscript. This work was supported by a grant from the National Institutes of Health (DK54899). J.D.K. is supported by a National Science Foundation Graduate Research Fellowship.

References (60)

  • D.J. Watson et al.

    Heparin-binding properties of the amyloidogenic peptides Abeta and amylin. Dependence on aggregation state and inhibition by Congo red

    J. Biol. Chem.

    (1997)
  • J. Green et al.

    Full-length rat amylin forms fibrils following substitution of single residues from human amylin

    J. Mol. Biol.

    (2003)
  • S. Potter-Perigo et al.

    Proteoglycans synthesized and secreted by pancreatic islet beta-cells bind amylin

    Arch. Biochem. Biophys.

    (2003)
  • L.R. Brown et al.

    High resolution nuclear magnetic resonance studies of the conformation and orientation of melittin bound to a lipid–water interface

    Biophys. J.

    (1982)
  • C.J. Morgan et al.

    Kidney dialysis-associated amyloidosis: a molecular role for copper in fiber formation

    J. Mol. Biol.

    (2001)
  • J. Folch et al.

    A simple method for the isolation and purification of total lipides from animal tissues

    J. Biol. Chem.

    (1957)
  • M.J. Newman et al.

    Solubilization and reconstitution of the lactose transport system from Escherichia coli

    J. Biol. Chem.

    (1980)
  • H. Schagger et al.

    Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa

    Anal. Biochem.

    (1987)
  • A.M. Batenburg et al.

    Interaction of melittin with negatively charged phospholipids: consequences for lipid organization

    FEBS Letters

    (1987)
  • B. Caughey et al.

    Protofibrils, pores, fibrils, and neurodegeneration: separating the responsible protein aggregates from the innocent bystanders

    Annu. Rev. Neurosci.

    (2003)
  • M. Sunde et al.

    From the globular to the fibrous state: protein structure and structural conversion in amyloid formation

    Q. Rev. Biophys.

    (1998)
  • M.B. Pepys

    Pathogenesis, diagnosis and treatment of systemic amyloidosis

    Philos. Trans. R. Soc. Lond. B Biol. Sci.

    (2001)
  • J.W. Hoppener et al.

    Islet amyloid and type 2 diabetes mellitus

    N. Engl. J. Med.

    (2000)
  • S.E. Kahn et al.

    Islet amyloid: a long-recognized but underappreciated pathological feature of type 2 diabetes

    Diabetes

    (1999)
  • J. Janson et al.

    The mechanism of islet amyloid polypeptide toxicity is membrane disruption by intermediate-sized toxic amyloid particles

    Diabetes

    (1999)
  • A. Lorenzo et al.

    Pancreatic islet cell toxicity of amylin associated with type-2 diabetes mellitus

    Nature

    (1994)
  • D. Porte et al.

    Beta-cell dysfunction and failure in type 2 diabetes: potential mechanisms

    Diabetes

    (2001)
  • M. Anguiano et al.

    Protofibrillar islet amyloid polypeptide permeabilizes synthetic vesicles by a pore-like mechanism that may be relevant to type II diabetes

    Biochemistry

    (2002)
  • M. Prentki et al.

    Malonyl-CoA signaling, lipid partitioning, and glucolipotoxicity: role in beta-cell adaptation and failure in the etiology of diabetes

    Diabetes

    (2002)
  • M. Shimabukuro et al.

    Fatty acid-induced beta cell apoptosis: a link between obesity and diabetes

    Proc. Natl Acad. Sci. USA

    (1998)
  • Cited by (314)

    • Preclinical toxicity assessment of anionic nanoliposomes

      2024, Advanced Industrial and Engineering Polymer Research
    View all citing articles on Scopus
    View full text