Abstract
Islet amyloid polypeptide (IAPP) forms cytotoxic oligomers and amyloid fibrils in islets in type 2 diabetes (T2DM). The causal factors for amyloid formation are largely unknown. Mechanisms of molecular folding and assembly of human IAPP (hIAPP) into β-sheets, oligomers and fibrils have been assessed by detailed biophysical studies of hIAPP and non-fibrillogenic, rodent IAPP (rIAPP); cytotoxicity is associated with the early phases (oligomers/multimers) of fibrillogenesis. Interaction with synthetic membranes promotes β-sheet assembly possibly via a transient α-helical molecular conformation. Cellular hIAPP cytotoxicity can be activated from intracellular or extracellular sites. In transgenic rodents overexpressing hIAPP, intracellular pro-apoptotic signals can be generated at different points in β-cell protein synthesis. Increased cellular trafficking of proIAPP, failure of the unfolded protein response (UPR) or excess trafficking of misfolded peptide via the degradation pathways can induce apoptosis; these data indicate that defects in intracellular handling of hIAPP can induce cytotoxicity. However, there is no evidence for IAPP overexpression in T2DM. Extracellular amyloidosis is directly related to the degree of β-cell apoptosis in islets in T2DM. IAPP fragments, fibrils and multimers interact with membranes causing disruption in vivo and in vitro. These findings support a role for extracellular IAPP in β-sheet conformation in cytotoxicity. Inhibitors of fibrillogenesis are useful tools to determine the aberrant mechanisms that result in hIAPP molecular refolding and islet amyloidosis. However, currently, their role as therapeutic agents remains uncertain.
Introduction
Two aspects of the amyloidogenic process are significant for diabetes. Firstly the role of small assemblies (so-called ‘oligomers’) of islet amyloid polypeptide (IAPP, also known as amylin), which can have a cytotoxic role in the islet (Westermark et al. 2011) and secondly, the degree of replacement of islet cells by the invasive amyloid deposits in type 2 diabetes (T2DM) (Clark et al. 1988, Jurgens et al. 2011). These processes are likely to be linked in T2DM since the process of amyloidogenesis damages cells whatever the mechanism of the IAPP-induced toxic insult. Cell death and toxicity in relation to diabetes is a more apparent and testable end point in animal models than is possible in humans; but it must be remembered that studies in vitro and in created animal models in vivo are unlikely to model exactly the slow (possibly over decades) process of islet amyloid deposition in T2DM. Moreover, islet amyloidosis and substantial islet cell cytotoxicity are not universal features of all patients with T2DM and are likely to be only one of the many factors that contribute to T2DM in humans.
Amyloid fibrils are macromolecular assemblies in an elongated fibrillary form which are deposited in different tissues of the body in different disease states. The common feature of the amyloid diseases is that the component molecules have become refolded from their normal conformation as a result of genetic mutations or disease conditions (Glenner 1980, Chiti & Dobson 2006). Amyloid fibrils share several common structural features despite the wide diversity of proteins and polypeptides that form amyloid in vivo or in vitro. The intermolecular cross-β conformation in which β-strands run perpendicular to the long axis of the fibril and interstrand hydrogen bonds are parallel to the fibril axis is a common feature as is the long and unbranched nature of the fibrils (Riek & Eisenberg 2016). The term ‘oligomer’ describes assemblies of small numbers of β-strands/molecules (dimers-multimers) at the beginning of the process of amyloidogenesis (Glabe 2006). The protein molecules are non-covalently linked in the polymeric (fibril) state by backbone hydrogen bonding and side-chain/side-chain interactions. Thus, amyloid formation differs from most synthetic polymeric molecules, which involve covalently linked monomers.
The biophysical properties of IAPP and proIAPP have much in common with the Alzheimer peptide Aβ, which is the major component of cerebral amyloid plaques in Alzheimer’s disease (AD). The major difference between amyloidosis of these two peptides is that, although increased concentration/decreased clearance of an alternative cleavage product of the larger Alzheimer precursor protein (APP) results in the amyloid plaques in AD (Selkoe 2011), islet amyloid is formed from the normal sequence of the peptide in type 2 diabetes. Whether there is decreased clearance of proIAPP/IAPP from the islet spaces in islet amyloidosis is yet to be determined.
The structure of IAPP and fibril assembly
Early in the study of IAPP amyloidogenesis it was observed that differences in the sequence of IAPP within positions 20–29 in rodents and humans correlated with the lack of islet amyloid in rats and mice (Westermark et al. 1990); human IAPP (hIAPP) and rodent IAPP (rIAPP) differ in six amino acids in this region (Fig. 1). The most important of these differences are the three proline residues in rIAPP at positions 25, 28 and 29 and the His-18 to Arg substitution. Proline disrupts the hydrogen bonding and intermolecular β-sheet formation, and this could potentially explain the lack of islet amyloid in rodent models of diabetes (Betsholtz et al. 1989). Replacement of His-18 in hIAPP with Arg in rIAPP ensures that the side chain will be positively charged at all physiologically relevant pH values and decreases the tendency to aggregate and form amyloid (Abedini & Raleigh 2005, Jha et al. 2014). Work with small fragments of rIAPP and hIAPP have shown that other regions of the molecule can themselves aggregate to form β-sheet-rich amyloid structures and thus could interact in the assembly process of hIAPP 1–37. It is now known that the peptide fragments hIAPP 30–37, 8–20 and 10–19 also form amyloid fibrils (Nilsson & Raleigh 1999, Jaikaran et al. 2001, Mazor et al. 2002, Gilead & Gazit 2008). The sequence IAPP 30–37 is identical in rodent and human IAPP (Fig. 1), suggesting that this region of the rodent sequence has the potential to form β-sheets, but the adjacent proline-rich region in rIAPP and the additional charge at position 18 prevents amyloid formation by the full-length rodent polypeptide. Studies with variants of hIAPP have shown that proline substitutions outside the 20–29 region can abolish amyloid formation by hIAPP, and replacement of Asn-14 or Asn-21 can have drastic effects (Abedini & Raleigh 2006, Koo et al. 2008). Conversely, replacement of residues Arg-18, Leu-23 and Val-26 in rIAPP by the corresponding amino acids found in hIAPP leads to a weakly amyloidogenic polypeptide even though it still contains the three proline substitutions found in rodent IAPP (Green et al. 2003). Collectively, these studies and the studies of IAPP-derived peptide fragments show that factors other than the sequence within residues 20–29 play an important role in dictating the amyloidogenecity of hIAPP; none-the-less, the sequence within the 20–29 segment is an important amyloidogenic determinate.
Amino acid sequence of proIAPP in different species: ProIAPP is 67 amino acids long in humans, Macaques, cats and dogs but 70 amino acids in rodents and guinea pigs. The propeptide is cleaved by the enzymes PC2 and PC1/3 and further amidated at the C-terminus by peptidylglycine α-amidating monooxygenase (PAM). This results in the mature 37 amino acid IAPP. The conservation of the amino acid components between the different species is shown in the lower panel.
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
Native hIAPP is a positively charged, hydrophobic peptide due to the amidated C-terminus and the lack of acidic residues. In cells and in the circulation hIAPP is considered to be in a so-called ‘intrinsically disordered’ state (Williamson & Miranker 2007). This term refers to the fact that the IAPP molecule does not adopt a rigid globular conformation, but rather samples an ensemble of rapidly converting, partial and less structured conformations. The question is how and when does the conformation change to create cytotoxic refolded molecular structures? There have been suggestions from nuclear magnetic resonance (NMR), crystallography studies of IAPP-maltose-binding protein fusions and circular dichroism (CD) studies that, prior to the refolding into a β-conformation (the crucial unknown condition which leads to islet amyloidosis), the molecule adopts a transient α-helical conformation (Williamson & Miranker 2007, Apostolidou et al. 2008, Wiltzius et al. 2008, Abedini & Raleigh 2009). Lipids and interactions with membranes can promote the formation of a partial helical state (Knight et al. 2006, Apostolidou et al. 2008, Williamson et al. 2009). It is now considered that this helical confirmation facilitates the conversion of hIAPP to β-conformation and subsequent intermolecular association.
Several structural models have been proposed for hIAPP amyloid fibres formed in vitro and they share common feature, although they differ in the details of the assemblies (Luca et al. 2007, Wiltzius et al. 2008). Each monomer of IAPP forms two β-strands connected by a loop/hairpin region. Monomers stack on top of each other such that the backbone to backbone hydrogen bonds are located between different molecules rather than within a single polypeptide. The fibre is made up of two symmetrical related U-shaped stacks of monomers oriented such that polypeptides in adjacent stacks are oriented antiparallel to each other (Fig. 2). The region IAPP 20–29, although now known not to be the only sequence to regulate fibril formation, is likely to be important for the initial refolding of the molecule into a hairpin shape with turn/loop in the region of residues 18–27 (Westermark et al. 1990, Luca et al. 2007, Wiltzius et al. 2008, Akter et al. 2016) (Fig. 2). Different hypotheses concerning the intramolecular stacking of the side chains between two adjacent monomers have been proposed. One model proposes that two monomers in adjacent stacks pack via the side chains of residues 26–32; this forms the building block of the oligomer (Fig. 2). The disulphide bridge between residues 2 and 7 at the N-terminal end of the peptide is not part of the β-sheet assembly in this model and remain largely exterior to the longitudinal axis of the protofibril (Fig. 2) (Luca et al. 2007, Wiltzius et al. 2008). Alternatively, the turn/loop connecting the two β-strands within each monomer has been proposed to occur between residues 18 and 22; association of the two monomers has been postulated to occur via a steric zipper between the regions residues 20–27 and residues 29–33 (Wiltzius et al. 2008).
Structural models of the hIAPP protofibril. (A) A β-bend in regions 18–27 results in a horseshoe shaped structure with the N-and C-termini (N-term, C-term) on the same face. There are no intra peptide backbone hydrogen bonds in the structure; instead the backbone hydrogen bonding occurs between adjacent monomers in one of the horseshoe shaped stacks. (B) The two stacks are aligned so that two molecules lie antiparallel in the same plane and interact via their sidechains. Protofibrils have a super helical twist, although the β-sheets are still relatively flat by the standards of globular proteins. Protofibrils further assemble into fibrils containing two or more protofibrils (Wiltzius et al. 2008).
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
The naturally occurring mutation Ser20Gly (S20G) found in a small number of Asian diabetic patients (Sakagashira et al. 1996, Lee et al. 2001) has been shown to enhance amyloidosis and cytotoxicity in hIAPP S20G transgenic mice compared to hIAPP-native sequence transgenic mice (Meier et al. 2016) and to accelerate fibril formation in vitro (Sakagashira et al. 2000, Ma et al. 2001, Cao et al. 2012, Krotee et al. 2017). It has been suggested that this mutation creates different intermolecular energetics at this site, which affects β-sheet assembly (Krotee et al. 2017). However, the real question is which part or parts of the molecule and which assemblies of multimers act as the toxic species to instigate cell death.
Amyloid proteins/peptides assemble into amyloid fibrils with a distinct time course, which represents different stages in the assembly processes (Fig. 3) (Chiti & Dobson 2006). Amyloid formation typically proceeds through a so-called ‘lag phase’, during which little or no detectable amyloid is present. This leads to a growth phase and culminates in a plateau phase (saturation phase) in which amyloid fibrils are in equilibrium with residual soluble peptide. These stages can be identified in vitro by analysis of molecular conformations and β-sheet assemblies and are typically monitored in vitro using the small dye thioflavin-T (Abedini et al. 2016). It is presumed that similar processes would occur in vivo but with different time courses depending upon the cellular conditions. The first or lag phase starts with monomeric peptide and, if the conditions are favourable, refolding of monomers and assembly of oligomers; oligomers then assemble into protofibrils/protofilaments, which are components of mature amyloid fibrils (Sunde & Blake 1997) (Fig. 3). The growth phase represents the elongation of the protofibrils and assembly of the fibrils (Fig. 3). Eventually, the vast majority of the amyloidogenic molecules are incorporated into amyloid fibrils and no further activity is present; this is the saturation phase (Fig. 3). In vivo, this last theoretical situation would never occur since the β-cells continue to secrete IAPP and new molecules would be added to existing deposits in the extracellular space or create new sites of amyloid formation in the islets. It has been suggested from in vitro experiments that there is little or no cytotoxicity of the mature amyloid fibrils from the saturation phase (Abedini et al. 2016). It is notable that synthetic hIAPP in fibrillar form when added to cultured cells or islets is not cytotoxic to cells in culture. Extending this in vitro finding to the in vivo situation, it suggests that the extensive ‘mature’ IAPP amyloid deposits in the extracellular space containing many millions of molecules in fibrillar form are only minimally cytotoxic; newly refolded IAPP monomers/oligomers that are not incorporated into the extensive fibrillary assembly could then be the cytotoxic species.
Schematic diagram of amyloid fibril formation from hIAPP. During the initial phase (lag phase) monomeric hIAPP forms oligomers and adopts conformations which differ from the monomeric state. The growth phase involves elongation of protofibrils and assembly into fibrils. A saturation phase, (also called the plateau phase) occurs with synthetic hIAPP in vitro when most of the monomers have become assembled into mature fibrils.
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
During the lag phase, the peptide structure appears largely amorphous by transmission electron microscopy (EM) and soluble assemblies ranging from monomers to hexamers have been detected by photochemically induced cross-linking and by ion mobility mass spectrometry (Young et al. 2014, Abedini et al. 2016). At this stage, unlike material present in the elongation phase, there is little evidence of measureable well-developed β-sheet structure and the assemblies could be easily digested with proteolytic enzymes (Young et al. 2014, Abedini et al. 2016). These preparations from the lag phase were cytotoxic, whereas samples taken later from the reaction, from the elongation and saturation phases had less or no cytotoxic properties. Parallel investigations with rIAPP and hIAPP peptides with amino acid substitutions indicated that although oligomers were formed, they had limited or no cytotoxic properties (Young et al. 2014, Abedini et al. 2016).
Cytotoxicity, peptide fragments and membrane interactions
Experimentally, cytotoxicity in vitro is determined by addition of different fragments of peptides to cells (usually β-cells) in culture. Alternatively, peptide interactions with model membranes or liposomes in vitro can provide data on how peptides would combine with cell surfaces. However, many of the early observations on hIAPP peptides in vitro were hampered by the tendency of synthetic hIAPP and its fragments to rapidly form amyloidogenic assemblies in vitro. Thus, careful handling of the peptide is required to ensure that correct conformations are being investigated at that moment in time (Higham et al. 2000b ). This is a more difficult scenario with cells in culture where hIAPP fibril formation is a dynamic and rapid process; many of the early reported studies were not able to differentiate between the IAPP conformation during the early phases of the oligomeric assembly (when the peptide was first added to culture) and the presence of mature fibrils (at a later time point when the cells died). Measurement of β-sheet structures by thioflavin-T fluorescent assay is not suitable to probe the initial assemblies of IAPP. In addition, most analytical methods such as circular dichroism (CD) or Fourier transform infrared spectroscopy (FTIR) of peptides undergoing molecular structural changes indicates only the overall nature of the intermediates of fibril assembly in the sample. However, nuclear magnetic resonance (NMR) and ion mobility spectrometry–mass spectrometry (IMS–MS) can detect small molecular intermediates (Young et al. 2014) whilst newly developed two-dimensional IR methods can provide residue-specific information about the development of β-sheet structure (Kamata et al. 2014). These methods provide a dynamic view of the fibril formation with appropriate time resolution within the life time of the process (Abedini et al. 2016).
It is now apparent that many different IAPP fragments including components of rIAPP and mutated sequences will form oligomers. However, not all of these are cytotoxic (Abedini et al. 2016) as measured by immediate responses (not the terminal point of cell death) such as upregulation of proinflammatory markers. In general, the molecular species that have been identified to interact with the cells and elicit a response were those taken from the lag phase and low order oligomeric assemblies with little β-sheet component (Fig. 3); the samples containing the mature fibrils in β-sheet conformation (Fig. 3) were less cytotoxic (Abedini et al. 2016). This corresponds to results from other methods of analysis of the early phases of fibrillogenesis (Young et al. 2014). However, other similar studies have suggested that the toxic species include hIAPP fragments that are already in β-sheet conformation (Krotee et al. 2017).
Although it has proved to be exceptionally difficult to correlate in vitro studies with cell biology, fibrillogenesis of many amyloidogenic proteins is enhanced at a synthetic membrane interfaces and this may be linked to an increase in permeability (Fig. 4A). The membrane-induced self-assembly pathway of any amyloidogenic protein, including IAPP, is yet to be determined and a multitude of conflicting mechanisms have been proposed to explain their membrane activity. The confusion is likely due, in part, to the wide variety of model membrane systems employed for in vitro studies. Most existing model membranes are incomplete mimics of the relevant biological membranes; they all lack asymmetry, most lack cholesterol and they often have high levels of anionic lipids. It is not clear how studies with these simplified model membranes translate to the situation in vivo. Human IAPP is believed to initially interact with membranes though the N-terminal region (Knight et al. 2006), but the role of individual residues is not understood. It is not known if features that modulate amyloid formation in solution and in vitro also do so in the environment of a biological membrane. Not only is the structure of the lipid bilayer and relative concentrations of phospholipids and cholesterol important but also the presence of many membrane-associated proteins; these include membrane receptors, ion channels and proteins interacting with extracellular matrix. Intragranular IAPP will come into close proximity with all of these components on exocytosis of the secretory granule.
(A) Schematic representation of the influence of membranes on the rate of hIAPP fibril formation. Synthetic membrane liposomes (blue line) reduce the lag time and accelerate the growth phase compared to fibrillogenesis in the absence of membrane (red line). (B) Diagrammatic representation of potential processes of membrane interactions of hIAPP. (a) hIAPP (red) hydrophobic, monomeric hIAPP adopts a transient helical conformation at the membrane surface which becomes β-sheet leading to oligomer assembly at the membrane. These multimers interact with the lipid bilayer to form a pore. (b) multimers assembled at the membrane bind to components of the outer leaflet of the bilayer and destroy the structure. (c) multimers assembled at the membrane continue to extend into protofibrils and fibril growth occurs with part of the fibril inserted into the membrane and membrane damage.
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
It has been shown that interaction of monomeric human IAPP with synthetic phospholipid membranes or bilayers promotes a transient conformational change to a molecular α-helical structure at the membrane interface with the helices lying parallel to the membrane surface (Knight et al. 2006, Apostolidou et al. 2008). Synthetic membranes that contain a significant fraction of anionic lipids significantly accelerate human IAPP amyloid formation in vitro, with more highly charged membranes having a larger effect (Zhang et al. 2017). Most studies of hIAPP–membrane interactions have made use of simple model membranes comprising pure anionic lipids, such as phosphatidylglycerol (PG) or phosphatidylserine (POPS) or mixtures of anionic lipids with zwitterionic lipids, such as 1-palmitoyl-2-oleoyl-phosphatidylcholine (POPC). The content of anionic lipid typically ranges from 25 to 100 mol% of the total phospholipids.
Standard model membranes employed in biophysical studies fail to capture the biology of membrane disruption in vivo. The β-cell plasma membrane contains a low percentage of anionic lipids, (estimates range from 2 to 13%) and cholesterol and is asymmetric. The outer leaflet is enriched in the neutral lipids and contains a very low percentage of anionic lipids, while the inner leaflet is enriched in phosphatidyl-ethanolamine (PE) and the anionic lipids, POPS and phosphatidylinositol (PI) (Trikha & Jeremic 2011, Abedini et al. 2016). Thus, extracellular IAPP will face a membrane with a very low percentage of anionic lipids. Even the inner leaflet contains a lower percentage of anionic lipids than is employed in most model membrane studies. Cholesterol is important for the uptake of hIAPP in vivo, (Trikha & Jeremic 2011) but the majority of biophysical studies of human IAPP membrane interactions have used vesicles that lack cholesterol. Some biophysical studies that included cholesterol also employed high concentrations of anionic lipids and these can mask the effect of the sterol. Higher concentrations of cholesterol in model membranes decrease the rate of fibril formation at the surface, whereas increased POPS leads to increased rate of fibrillogenesis (Zhang et al. 2017).
In vitro, hIAPP binding to synthetic liposomes induces both membrane leakage (an in vitro marker of potential cell damage) and cellular toxicity (Kumar et al. 2015). However, synthetic rIAPP as well as other hIAPP variants (which are unable to adopt β-sheet conformation and do not form fibrils nor induce cytotoxicity) induce low leakage with less efficiency than hIAPP in vitro, although higher concentrations of rIAPP can induce vesicle leakage that is comparable to that induced by toxic hIAPP even though rIAPP is not toxic under the conditions studied (Cao et al. 2013a,b). The mechanism of IAPP-mediated synthetic membrane leakage is controversial. IAPP-induced membrane damage has been proposed to result from a range of different mechanisms some of which can be related to biological systems. It was suggested that hIAPP caused changes in membrane curvature and pore formation of synthetic membranes (Engel et al. 2008, Engel 2009) and increased membrane conductance has been detected in cells (Mirzabekov et al. 1996, Kayed et al. 2004, Casas et al. 2007, 2008, Engel et al. 2008) (Fig. 4Ba). However, it is unclear if these pores are persistent or transient. Other studies have argued for a carpeting, detergent like mechanism removing some parts of the membrane (Fig. 4Bb) (Lee et al. 2012). There is also evidence for association of membrane damage with the growth of amyloid fibrils at the membrane surface (Engel et al. 2008). Changes in the composition of the model membranes, including incorporation of cholesterol demonstrated that hIAPP dye leakage from liposomes was more sensitive to membrane composition than amyloid fibril formation at the surface (Zhang et al. 2017). However, the time course of sustained leakage and fibril formation are comparable which is consistent with fibrillogenesis at the membrane surface (Engel 2009, Brender et al. 2011, Cao et al. 2013b). Taken together these studies suggest that hIAPP fibrils developing at the surface of the cell might exert cytotoxicity and that the membrane composition of β-cells (or other islet cell types) might be critical in conferring susceptibility to external attack by multimeric hIAPP or hIAPP fibrils.
A live cell approach to membrane interactions has been made using radioimaging techniques. MIN-6, clonal β-cells and a lipophilic dye, laurdan has been used to indicate changes in membrane fluidity. The data suggest that in the early stages of fibrillogenesis, hIAPP partitioned into the membrane and later mature fibrils can induce cell blebbing and death (Pilkington et al. 2016). Experiments on cells treated with various preparations of soluble or partially insoluble preparations of IAPP, identified potential membrane channels as sites for action (Mirzabekov et al. 1996, Kayed et al. 2004, Casas et al. 2008, Engel et al. 2008). These oligomeric forms of human, but not rIAPP, were proposed to act via the transient receptor channel (TRPV4) in β-cell membranes to increase influx of calcium (Casas et al. 2008) and to downregulate the intracellular proteasome degradation pathway (Casas et al. 2007). Clustering of hIAPP at the plasma membrane and internalisation of the peptide has been noted in INS-1 β-cells cultured with hIAPP (Trikha & Jeremic 2011) suggesting that external interaction of the peptide with the membrane could lead to internalisation of fibrils. This is turn could lead to activation of cell death pathways from within the cell.
Where does the cytotoxicity originate in vivo? Extracellular or intracellular?
The question of the origin of the insult that results in cell death has been a long debated point. In vivo, does the insult arise from within the cell (as a defect in the pathway of a secreted peptide) or is cell death triggered from the secreted peptide accumulating at an extracellular site and interacting with the cell membrane or a receptor? It is likely that there is more than one mechanism of toxicity and different mechanisms may be more or less effective depending upon the cellular conditions. It is notable that in vivo, it is primarily the islet β-cells (from which the peptide is secreted) and not the adjacent α- or δ-cells that are affected by the cytotoxicity. Alpha cells also have higher survival rates when exogenous synthetic hIAPP is applied to intact islet cells (Law et al. 2010). This suggests that the β-cells (and/or their plasma membranes) are more susceptible to the effect of hIAPP compared to α-cells, and it is not just the process of IAPP peptide secretion from β-cells, which differentiates the cell susceptibility.
Intracellular IAPP and fibrils
Most amyloidogenic proteins are secreted proteins/peptides, which have undergone considerable post-translation modification, including protein folding within their cells of origin. Some are misfolded during their original production (often as a result of a genetic mutation) and others become misfolded at later stages (Chen et al. 2015). ProIAPP undergoes post-translational processing in islet β-cells and aberrant reactions at any one of the many events in this pathway could result in conversion to β-conformation and aggregation. This in turn can trigger cytotoxicity and cell death (Fig. 5).
Diagram of intracellular mechanisms contributing to cytotoxicity of hIAPP in β-cells. With normal IAPP expression (left hand diagram) proIAPP (red) synthesised in the endoplasmic reticulum (ER) is transported to the Golgi through the cytoplasm. If aberrant folding or transcription/translation occurs, misfolded molecules are targeted (UPR) by ubiquitination for degradation via the proteasome. proIAPP is packaged in the Golgi, together with insulin and processing enzymes, into secretory granules. Maturation and post-translational processing of the peptide occurs in the granule where prohIAPP/hIAPP is largely stabilised by interaction with insulin. Some granules are targeted for destruction via the lysosomal system (crinophagy) and fuse with lysosomes. Granule components, including amino acids and lipids, arising from this degradation process are recycled back for re-synthesis. Human IAPP is partially resistant to lysosomal proteases and is retained in the lysosomal storage compartment (lipofuscin bodies). Overexpression (right hand side) of prohIAPP affects trafficking through the ER, overloading the UPR and creating ER stress which can lead to cell death. High concentrations of prohIAPP will aggregate in the ER or cytoplasm forming fibrils. These are removed by autophagy and targeted to the lysosomal system. Overload of this protective mechanism can result in apoptosis. The high concentrations of prohIAPP/hIAPP can result in fibril formation in secretory granules. On release from the cell by exocytosis of the granule, hIAPP can interact with the membrane causing cytotoxicity or be deposited as fibrils in the extracellular space.
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
ProIAPP in the endoplasmic reticulum
In the β-cell, transcribed propeptides/proteins (e.g. proIAPP, proinsulin) undergo molecular folding in the endoplasmic reticulum (ER); disulphide bridges are formed and peptides, which adopt well-folded structures, undergo correct folding into their near final conformation. At this point there is a strict process of quality control. If there is overproduction of the peptide or an aberrant sequence is translated/transcribed, the misfolded or excess peptide/proteins are targeted for destruction via the UPR. This is a collective name for several signalling pathways which are designed to operate continuously to reduce the effects of abnormal ER protein accumulation on cellular viability (Chen et al. 2015). These processes include detection of overexpression of normal proteins and/or misfolded proteins, signalling for disposal of this excess via the proteasome or lysosomal systems and cellular survival when this regulatory system becomes overburdened. Aberrantly misfolded peptides/proteins are targeted for ubiquitination and move to the cytosol for degradation of the misfolded material via the proteosome. However, this normal process has a limited capacity; if the UPR is overloaded, this can result in ER stress that can have disastrous consequences for the viability of the cells if it is not alleviated by removal of the insult or following activation of other protective systems. Under normal situations, proIAPP expression is regulated through the IAPP gene promoter; this promoter, although glucose responsive, is less sensitive than the insulin promoter (Zhu et al. 2017), and normal secretion of IAPP is 1–10% of that of insulin. Similarly, production of proIAPP/IAPP co-ordinately changes with that of insulin in rodents and humans in relation to stimulation with glucose and arginine and in T2DM diabetes (Butler et al. 1990, Kahn et al. 1990, 1998, Alam et al. 1992, MacNamara et al. 2000). Transgenic rodents expressing hIAPP gene are useful models. However, in many of these rodent models, many copies of the transgene are expressed; up to 74 copies have been identified in one model (Zhang et al. 2014). Since these transgenes are under control of the insulin gene promoter, they over-respond to glucose compared to rIAPP in normal wild-type mice; overproduction of the transgenic peptide was as much as ×2 in transgenic rats (Matveyenko & Butler 2006) and ×5 in transgenic mice (Fox et al. 1993). Cytoplasmic IAPP fibrils are frequent in circumscribed compartments in the cytoplasm of transgenic animals and in human islets in vitro (Paulsson et al. 2006) (Fig. 6A) but not seen in cells of humans with T2DM. It is unclear if these fibrils are situated in the ER or in other cellular compartments.
IAPP and proIAPP fibrils in overexpressing hIAPP transgenic mouse and human insulinoma cells viewed by electron microscopy. (A) A bundle of fibrils (f), immunogold labelled for IAPP in cytoplasm of a β-cell of a hIAPP transgenic mouse. (B) Arrays of fibrils immunoreactive for IAPP in the cytoplasm of a cell from a human insulinoma. i, insulin granules; amy, IAPP amyloid fibrils; m, mitochondrion; n, nucleus. Scale bars A, 500 nm; B, 1.0 µm.
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
ProIAPP is also fibrillogenic (Exley et al. 2010). Increased concentrations of human proIAPP peptide in the ER have been shown to activate the UPR and target the newly transcribed proIAPP to pathways of degradation. At the same time there is activation of other ER stress responses (Westermark et al. 2011, Chen et al. 2015) which, if not checked, can lead to cell death. Experiments in transgenic mice have suggested that overexpression of hIAPP, but not the rodent peptide, generates ER stress and activation of the ubiquitin pathway targeting degradation (Huang et al. 2007a) suggesting that it is hIAPP but not just the volume of peptide in the ER that is the primary factor. Similar conclusions were drawn from observations on islets from type 2 diabetic subjects in vitro although there was no evidence that hIAPP affected proteasome activity or that IAPP aggregates were the cause of these effects (Costes et al. 2011). The disulphide bridge in proIAPP (and proinsulin) is also formed at this stage by the enzyme protein disulphide isomerase (PDI) which allows correct folding of the peptides. However, it has been shown in hIAPP transgenic mouse islets that overexpression of this enzyme does not reduce hIAPP fibril assembly suggesting that cysteine bridge formation is unlikely to be a critical step for fibril assembly (Montane et al. 2016).
The human model of overexpression of human ProIAPP/IAPP occurs in sporadic insulinomas. These are adenomas, which are usually non-proliferative consisting of cellular β-cell clusters in the pancreas about the size of a large islet. Cells in these tumours have unregulated secretion of insulin/proinsulin (Guettier et al. 2013); this results in severe and debilitating hypoglycaemia for the patient. Surgical removal of the adenoma is the usual form of treatment. These β-cells also have unregulated secretion of IAPP/proIAPP (Toshimori et al. 1991) and many contain extracellular and intracellular IAPP amyloid (Toshimori et al. 1991, Clark 1992, O’Brien et al. 1994a). Although most of the IAPP amyloid is extracellular in these tumours, intracellular accumulations of fibrillar IAPP is found in some cells, often in extensive arrays as seen in transgenic mice (Paulsson et al. 2006) (Fig. 6B). Intracellular fibrils were immunoreactive for proIAPP (Paulsson et al. 2006) suggesting that refolding of proIAPP and aggregation of fibrils inside the cell (probably in the ER) had occurred. The cytotoxic nature of the intracellular fibrils in tumour cells cannot be properly assessed as these cells have a different pattern of turnover and lifespan compared to normal islet cells.
Autophagy and IAPP/ProIAPP in the cytoplasm
Autophagy (‘self eating’) is a normal and important process of the cell to remove damaged/dysfunctional organelles and cytoplasmic material, which have the potential to be cytotoxic. In the β-cell, this housekeeping activity is also responsible for maintaining the secretory granule population (Halban 1991) and for removal of damaged mitochondria (Ashrafi & Schwarz 2013) and other foreign material in the cytoplasm. Autophagy is also involved in removal of ER stress related material (Bartolome et al. 2014). Observations in hIAPP transgenic mouse islets indicated that intracellular thioflavin S-positive material (amyloid fibrils) was bound to p62 (a scaffolding protein essential for targeting proteins for autophagy), suggesting that fibrillary forms of IAPP are recognised by the autophagic system for degradation; in addition, β-cell content of hIAPP was increased in mice deficient in the autophagic machinery indicating that degradation via autophagy usually regulates IAPP (Westermark et al. 1990, Meier et al. 2014, Rivera et al. 2014). Cells containing IAPP fibrils also showed evidence of apoptosis suggesting that either the rescue process in the cell has become overwhelmed or that the proIAPP fibrils are cytotoxic within the cytoplasm (Paulsson et al. 2006). Although fibrils have been identified in these experiments, the molecular structure of proIAPP/IAPP in the cytoplasm and its mechanism of cytotoxicity is a matter for debate (Zraika et al. 2010).
It has been proposed that prefibrillar, oligomeric hIAPP, ‘toxic oligomers’ are present in the cytoplasm in overexpressing cells. An antibody-recognising oligomeric IAPP has identified cytoplasmic material in β-cells of hIAPP transgenic mice and patients with type 2 diabetes; this pathology of cytoplasmic ‘toxic oligomers’ has been proposed to form an unifying hypothesis for loss of insulin-secreting β-cells in type 2 diabetes (Ritzel et al. 2007, Zhao et al. 2009, Gurlo et al. 2010). However, all of these findings are based on the binding of a so-called ‘conformational antibody’ which does not recognise an antigenic amino acid sequence but a conformational shape of the molecule (Glabe 2008). These antibodies are useful for identifying small assemblies of IAPP in vitro, but their use in biological material and tissue sections is less well defined since specificity could be compromised by recognition of a large variety of molecules with β-conformation and their applicability to IAPP detection has been questioned (Zraika et al. 2010). Whilst there may be some evidence for cytotoxicity of visible extracellular fibrils or clearly identified, intracellular molecular aggregates in hIAPP producing cells in vivo, it is unclear that less visible, intracellular ‘toxic oligomers’ could be a major insult in human islets and cells where no islet amyloid is visible: the lack of good reagents for conclusively identifying hIAPP oligomers hinders such studies. It seems unlikely that minute assemblies of cytoplasmic hIAPP in β-cells which otherwise appear normal and continue to secrete insulin, could be a major causative factor for type 2 diabetes. This metabolic condition involves many factors both in islets and in the periphery which contribute to the dysregulated insulin secretion; loss of β-cells and islet amyloid formation are likely to be contributory rather than causative factors for the disease (Clark & Nilsson 2004).
ProIAPP/IAPP in granules
In the β-cell, translated proIAPP together with proinsulin is transferred to the Golgi apparatus to be packaged into secretory granules (Fig. 5). The granules also contain the prohormone convertase enzymes PC1 (also known as PC3) and PC2 and the amidating enzyme peptidylglycine-alpha-amidating mono-oxygenase (PAM) that are responsible for post-translation modification of proIAPP and proinsulin (Fig. 1) (Hutton 1994, Higham et al. 2000a, Wang et al. 2001b, Marzban et al. 2004). Proinsulin processing is defective in T2DM leading to an increased secretion of intact proinsulin and processing intermediates (Ward et al. 1987, Temple et al. 1992, MacNamara et al. 2000). hIAPP overexpressing β-cells lacking PC1/3 and PC2, exhibit increased proIAPP associated with increased islet amyloid deposition and islet cell death compared to cells possessing the processing enzymes (Marzban et al. 2006). Islets from hIAPP transgenic mice lacking PC2 were less viable when transplanted than islets from hIAPP mice with PC2 suggesting that proIAPP is a more toxic form of the molecule than the mature peptide (Courtade et al. 2017). Just as with human insulinomas, it is likely that intracellular fibrils are formed from the abundant proIAPP in this condition (Fig. 5).
Secretory peptides are in a very high concentration in the granule; insulin/proinsulin granule concentrations are estimated to be ~100 mM and IAPP concentrations 1–10 mM in man (Kanatsuka et al. 1989). Granules are at pH 3–5 (Hutton 1994) and the concentration of zinc is estimated to be 20 mM (Hutton et al. 1983). It is therefore surprising that, at this relatively high concentration, proIAPP or IAPP do not become fibrillar in the granule under normal conditions. Fibril-like structures are present in granules of some animal models overexpressing hIAPP (Paulsson et al. 2006) (Fig. 7A). However, under normal conditions it is likely that the granule peptides also have influences on each other. Interaction in vitro of insulin, c-peptide, proinsulin and zinc with hIAPP have been investigated and shown to have variable effects on IAPP fibrillogenesis; high concentrations of zinc accelerate, but lower concentrations reduce fibril formation (Westermark et al. 1996, Brender et al. 2010). Human proinsulin and insulin formed heteromolecular complexes with hIAPP and prevented fibrillogenesis (Jaikaran et al. 2004, Larson & Miranker 2004, Knight et al. 2008, Wiltzius et al. 2009). Thus it seems that under conditions of normal hIAPP gene expression, molecular buffering capacity in secretory granules where peptide concentrations are high, are effective in preventing β-sheet formation of hIAPP or human proIAPP.
IAPP in insulin secretory granules of a hIAPP transgenic mouse and in a lipofuscin body of a non-diabetic human subject. (A) Immunogold labelling with an antibody against hIAPP with N-terminal of proIAPP intact demonstrates fibrillar IAPP/proIAPP in mature insulin granules (i) (with halos) (arrow) and in immature granules (without halos) (arrowhead) suggesting that fibrils are formed at this site from unprocessed proIAPP. Scale bar 200 nm. (B) Immunoreactivity for IAPP in a lipofuscin body (arrows) of a non-diabetic subject. m, mitochondrion; i, insulin secretory granule. Scale bar 200 nm.
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
Secretory granules surplus to requirements in the cell are removed and degraded to constituent components via the lysosomal system by a type of autophagy known as crinophagy (Schnell et al. 1988). The finding of high concentrations of immunoreactive IAPP in human beta-cell lysosomes and lipofuscin bodies (Clark et al. 1989) (Fig. 7B) indicates that this is a normal pathway for degradation of granule peptides including IAPP. Lipofuscin bodies are intracellular storage organelles for material in the lysosomal system and since they accumulate with age, also markers for β-cell ageing (Cnop et al. 2010); cellular material which has not been degraded by lysosomal proteases and lipases accumulates in these bodies. The finding of hIAPP in lipofuscin bodies in human cells but not rIAPP in mouse or rat cells suggests that proteolysis of hIAPP is inefficient compared to rIAPP resulting in accumulation over time in man. Although it has been proposed that autophagy is impaired in β-cells of patients with T2DM (Westermark et al. 1990, Meier et al. 2014, Rivera et al. 2014), there is little evidence for an increase of autophagic bodies containing hIAPP in cells from diabetic patients (Cnop et al. 2010).
Thus, in the cell, at various stages of post-translational modifications or passage through the secretory granule mechanisms, there is the potential for excess hIAPP to oligomerise and form fibrils. This could result in activation of cell death signals and apoptosis. However, how this relates to IAPP in T2DM remains a critical question. At present there is little evidence for overproduction of IAPP in diabetes (Butler et al. 1990, Sanke et al. 1991, Eriksson et al. 1992, Kahn et al. 1998, Jorgensen et al. 2000). In addition, currently there is no compelling data to suggest that IAPP-related cellular damage could be a pathogenic factor for onset of the disease in man. Moreover, IAPP is not in the list of 39 susceptibility loci for T2DM (Gaulton et al. 2015, Prasad & Groop 2015).
Extracellular IAPP and fibrils
IAPP is co-secreted with insulin from islet β-cells into the extracellular space in islets by exocytosis of secretory granules (Rorsman & Renstrom 2003). It is assumed that, under normal conditions, IAPP in the secretory granule is in a natively unstructured molecular form and would be released on granule exocytosis in that form. Secreted peptides cross the extracellular islet space to the islet capillaries and enter the circulation. At the point of exocytosis, the concentrations of IAPP and insulin fall and the ambient pH rises to 7.0 (Rorsman & Renstrom 2003, Obermuller et al. 2005). Thus the space immediately adjacent to the plasma membrane will transiently contain a high IAPP and insulin concentration together with the other contents of the granule. This change in environment and loss of molecular buffers is likely to have impact on the potential for IAPP to aggregate. The mechanism of, and trigger for oligomerisation in the extracellular space is largely unknown and may be initiated by several different factors. The process of exocytosis involves membrane fusion between the granule and the plasma membrane. SNARE proteins in the membranes are involved in the fusion of the granule membrane with the cell membrane and release of the granule contents (Rorsman & Renstrom 2003). Due to the structural properties of the SNARE transmembrane domains the membrane properties (e.g. viscosity) are likely to be modified during exocytosis and could promote the interaction between hIAPP and the membrane. This could potentially initiate the process of fibrillogenesis at the point of exocytosis (Han et al. 2017, Hastoy et al. 2017). In all animal and cell models and human T2DM, IAPP fibrils are found in invaginations of the β-cell membrane. This was first shown in 1973 in islets and islet macrophages from a subject with T2DM (Westermark 1973). This type of membrane interaction is characteristic of both biosynthetic IAPP (from secreted peptide) in human and transgenic hIAPP mouse islets (Westermark et al. 2011) and synthetic exogenous IAPP (added to the media) (Janson et al. 1999) (Fig. 8A, B and C). This suggests that oligomeric hIAPP interacts with sites on the cell membrane. These characteristic membrane invaginations are present in animal, human and cellular models including islets from patients with T2DM (Westermark 1973, de Koning et al. 1994a, Westermark et al. 2011) (Fig. 8A). At this point, the IAPP fibrils can be visualised arranged in parallel orientation and in close contact with the membrane (Westermark 1973, de Koning et al. 1994b, Janson et al. 1999) (Fig. 8B and C). In contrast, these invaginations are not present in α-cells adjacent to amyloid deposits (Fig. 8A). The association of the fibrils with the membrane is more random and irregular in cells to which synthetic peptide has been added (Janson et al. 1999, Casas et al. 2008). Normally, following exocytosis, the granule membrane is quickly recovered by endocytosis involving clathrin on the cytoplasmic side (Ceridono et al. 2011). Under normal circumstances, these ‘omega-shaped’ (as seen in EM images) endocytotic secretory vesicles are so quickly recaptured from the cell membrane that they are not apparent even in highly secreting cells. However, the presence of clathrin coated endocytoses is markedly increased in areas of β-cell membrane adjacent to extracellular IAPP fibrils in islets of transgenic mice expressing hIAPP (Fig. 8B) or in cultured normal human islets. This leads to the intriguing possibility that the extracellular IAPP fibrils/oligomers may be arresting complete granule membrane recovery, possibly by retaining an attachment to the membrane or stimulating endocytosis. Such interference with the normal fluidity of the membrane by extracellular fibrils could compromise cellular function and could lead to cytotoxicity. The sections of the membrane influenced by the fibrils are likely to be dysfunction in terms of further granule release or receptor-mediated signalling. This in turn, could affect cellular function and possibly cell viability.
Extracellular IAPP fibrils in plasma membrane invaginations of β-cells seen with electron microscopy. (A) Islet with extracellular amyloid immunogold labelled for IAPP from a Type 2 diabetic subject. The amyloid (amy) is situated between the capillary (c) and the β-cell (β) and in deep invaginations of the β-cell membrane (arrows). Adjacent alpha cells (α) although in close contact with the amyloid (arrowhead) do not have membrane invaginations or show nuclear (n) signs of apoptosis. Scale bar 1.0 µm. (B) Extracellular amyloid fibrils immunogold labelled for IAPP (amy) from an islet of an hIAPP transgenic mouse. The fibrils lie in invaginations of the β-cell plasma membrane (arrows). Some of these membrane invaginations show thickening characteristic of clathrin coating suggesting an arrested process of membrane recovery/endocytosis (e.g. red arrow). m, mitochondrion. Scale bar 200 nm. (C) β-cell of an islet isolated from an hIAPP transgenic mouse and cultured with 10 mM glucose. Extensive IAPP immunogold positive fibrils (amy) lie in very deep invaginations of the cell. i, insulin granules; m, mitochondrion. Scale bar 500nm.
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
The extracellular surface of the β-cell plasma membrane is coated with glycosylated proteins including heparan sulphate proteoglycans (HSPGs) which is a known ligand for IAPP (Westermark 1973, Westermark & Grimelius 1973, Meng & Raleigh 2011) and also a component of islet amyloid deposits (Castillo et al. 1998). Enzymatic removal of HSPG (Oskarsson et al. 2015, Potter et al. 2015) or blockade of HSPG binding sites (Hull et al. 2007) reduced islet amyloid formation in hIAPP transgenic mouse or human islets in vitro; this suggests that the proteoglycans are a binding site for IAPP on the outside of the cell. If HSPGs are a primary anchoring site in the extracellular space for monomeric secreted hIAPP then hIAPP/HSPG interactions could lead to a high local concentration of hIAPP and promote aggregation (Westermark 1973).
The role of the extracellular amyloid deposits and extracellular oligomers in cell death has been challenged; studies on transgenic mice expressing hIAPP showed that β-cell death could occur in the absence of visible extracellular islet amyloidosis (Huang et al. 2007a, Haataja et al. 2008). From this observation, it was concluded that cell death pathways were generated as a result of ER stress in cells overexpressing hIAPP (but not in cells overexpressing rIAPP) rather than IAPP fibril formation outside the cells (Huang et al. 2007a). In humans, one model of massive overexpression of IAPP is the condition of anti-insulin receptor antibodies resulting in severe insulin resistance. This is coupled with overexpression of insulin and extensive extracellular islet amyloid (O’Brien et al. 1994b). However, the extent of cell death has not been quantified in this condition.
In contrast, the importance of extracellular hIAPP has been demonstrated in a different hIAPP transgenic mouse model with a lower degree of hIAPP expression where there was a close association of the degree of extracellular amyloid deposition and loss of islet cells suggesting that toxicity was related to extracellular islet amyloidosis (Wang et al. 2001a). In human post-mortem samples, the degree of cell death was directly associated with the amount of extracellular islet amyloid (Jurgens et al. 2011) implicating the fibrils (or the process of forming the fibrils) in cytotoxicity. Since there is little evidence that IAPP is overexpressed in T2DM (Jorgensen et al. 2000), it is more likely that IAPP secreted from the cell becomes misfolded outside the cell and activates cell death pathways by interaction with the cell membrane or membrane-bound receptors. Once the misfolded hIAPP is attached to the membrane (or other sites adjacent to the membrane), it could form a nidus for further secreted peptide to bind and aggregate thus creating extracellular fibrils (Jarrett & Lansbury 1993). It is likely that the early misfolded peptide (which is probably the cytotoxic species of IAPP) converts to amyloid fibrils.
Initially, islet amyloid deposits in islets in type 2 diabetic patients and spontaneously diabetic macaque monkeys are located in extracellular, pericapillary regions of the islet and specifically between the two basement membranes (one surrounding the islet cells and one adjacent to the capillary) (de Koning et al. 1993) (Fig. 9A). Although in material taken at the time of autopsy, some of these islet deposits are not adjacent to islet cells undergoing apoptosis in that snapshot of time, it is unclear if cells have died previously. This site of initial fibril formation also suggests that accumulation of the peptide might arise as a result of poor clearance of the peptide into the capillaries and increased concentration in this restricted space. In biological systems, fibrils forming in human or hIAPP transgenic mouse islets in vitro are found more frequently between islet cells and in the central regions of islets than on the well-perfused cell surface (Potter et al. 2010). IAPP-immunoreactive material (some fibrillary) accumulates on the basal surface of cells grown on a matrix (Fig. 9B) but not on the exposed cell surface. This all suggests that restricted diffusion of the peptide from a more confined space (and increased concentration) could play a part in fibrillogenesis as proposed for Aβ in AD (Selkoe 2011). This effect could explain the finding of amyloid in human and hIAPP transgenic mouse islets when transplanted into sites which are poorly perfused, such as islets in the liver (Westermark et al. 2012) or under the kidney capsule or when encapsulated (Udayasankar et al. 2009, Bohman & Westermark 2012, Westermark et al. 2012; this amyloid compromises their viability and function (Potter et al. 2010).
Localisation of IAPP amyloid fibrils in an islet of diabetic macaque monkey and in hIAPP expressing β-cells cultured on a matrix. (A) Extracellular amyloid fibrils in humans and all animal models are initially deposited in vivo between islet cells and islet capillaries. Gold labelled IAPP fibrils (amy) are situated between the two basement membranes (arrows) in the islet; one adjacent to the islet capillary (cap) and a second adjacent to the β-cell. i, insulin granule. Scale bar 1.0 µm. (B) Electron micrograph of hIAPP expressing β-cells cultured on a filter culture insert. Section taken through the filter (filter) shows the base of the cells with invaginations containing fibrils gold labelled for IAPP (amy). Some of these invaginations (arrows) have a thickened membrane suggesting clathrin coating. m, mitochondrion; n, nucleus; i, insulin granule. Scale bar 200 nm.
Citation: Journal of Molecular Endocrinology 59, 3; 10.1530/JME-17-0105
Cellular pathways of cytotoxicity
It is now evident that the small oligomeric hIAPP species induce cytotoxicity whether the amyloid peptide is exogenously applied (Lorenzo et al. 1994, Abedini et al. 2016) (either synthetic hIAPP or peptide fragments) or secreted from the cell. Induction of cell death from intracellular events and from extracellular application appears to follow a final common pathway of apoptosis rather than necrosis. This pathway involves a series of events in the cell, which finally results in chromatin condensation, nuclear fragmentation and total loss of cellular function.
Extrinsic apoptotic pathways can be initiated on the outside of the cell largely via Fas/Fas ligand binding to the cell membrane Fas receptor. This in turn initiates activation of caspase 8 and subsequently cleavage of caspase 3 in the apoptotic cascade. Exogenous application of hIAPP has been shown to activate this pathway and blockade/deletion of the Fas receptor or inhibition of caspase 3 activation reduced hIAPP-induced cytotoxicity (Zhang et al. 2003, 2008, Law et al. 2010, Park et al. 2017). Further evidence for an inflammatory-type of response to hIAPP has been shown by production of proinflammatory cytokines, including interleukin1-α (Masters et al. 2010) and interleukin1-β (IL1-β) (Westwell-Roper et al. 2011). This response was observed in islets in vitro with exogenous application of synthetic hIAPP and in cultured islets from transgenic mice expressing hIAPP (Westwell-Roper et al. 2011); blocking the IL1-β receptor reduced production of IAPP-induced cytokines suggesting that this was the mode of action of the cytotoxic hIAPP. Moreover, a link has been shown between hIAPP-induced cytokine production and upregulation of Fas; blocking experiments demonstrated that IL1-β released from the cells activated apoptosis via Fas/Fas ligand binding (Park et al. 2017). The molecular complex known as the inflammasome has been implicated in release of cytokines in hIAPP-induced cell death (Masters et al. 2010).
The question remains whether this pathway exists in T2DM. Increased β-cell apoptosis is a hallmark of islets in type 2 diabetes (Butler et al. 2003). However, the inflammatory pathway is better represented in islet cell death in autoimmune type 1 diabetes than in T2DM. Increased expression of IL-1β has been observed in islets in T2DM (Boni-Schnetzler et al. 2008) and associated with glucotoxicity (Maedler et al. 2002). These data suggest that inflammation could be a factor in the pathogenesis of type 2 diabetes (Donath & Shoelson 2011). Macrophages also release cytokines and an increased density of tissue macrophages in islets in type 2 diabetes has been proposed (Richardson et al. 2009). Quantitative analysis of macrophage content of human pancreas from type 2 diabetic subjects has suggested that, at least in some patients, islet amyloid is associated with increased numbers of islet macrophages (Kamata et al. 2014). However, specific analysis of macrophage content of amyloid-containing and amyloid-free islets in spontaneously diabetic monkeys showed no increased macrophage density associated with islet amyloidosis although macrophages containing immunoreactive IAPP material were identified in islets (de Koning et al. 1998). Observations in hIAPP transgenic mice have indicated that islet macrophages are associated with increased inflammatory signals in amyloid-containing islets (Meier et al. 2014). Thus, tissue macrophages may play both scavenger and inflammatory roles in this pathology.
Activation of intrinsic (intracellular) mechanisms of apoptosis does not involve cell membrane receptor signalling. In addition, the cell has the potential to suppress some intracellular pro-apoptotic signalling by activation of anti-apoptotic mechanisms. Anti-apoptotic restoration of cellular homeostasis operates largely via the Bcl-2 family of proteins (Taylor et al. 2008). The activity of this family of proteins balances the pro- and anti-apoptotic signalling cascades at different steps of protein processing in the cell (Gross & Katz 2017).
This balancing mechanism operates in the ER. The UPR is the first step in protein production that can lead to apoptosis. If there is an excess production of misfolded protein such as hIAPP, a pro-apoptotic pathway is initiated involving activation of PERK (protein kinase like ER kinase) and transcription factor CHOP (C/EBP homologous protein) and related proteins (Gross & Katz 2017). This pathway is also activated in so-called ‘ER stress’ responses and can develop as a result of other cellular disorders- notably gluco- or lipotoxicity in β-cells and has been demonstrated in relation to hIAPP cytotoxicity (Huang et al. 2007b). CHOP activated proteins promote apoptosis via inhibition of protective Bcl-2 but the balance of pro- and anti-apoptotic proteins is delicate, can be reversed quickly and can be modulated by external stimuli. The involvement of CHOP in hIAPP-induced cell death and onset of diabetes has been demonstrated in transgenic mice overexpressing hIAPP (Huang et al. 2007b, Gurlo et al. 2016).
Inhibitors of IAPP fibrillogenesis: a suitable treatment in type 2 diabetes?
Since the discovery of the peptide 30 years ago, there has been considerable interest in development of inhibitors of IAPP amyloid formation for potential treatment of type 2 diabetes. This has run in parallel (but somewhat behind) ideas for treatment of AD (Westermark 1972). Some most successful inhibitors of in vitro fibrillogenesis are small peptide molecules that interact with hIAPP and modulate the peptide aggregation process. Many of these have been developed and tested both on aggregation and cytotoxicity in vitro and in some cases administered to hIAPP transgenic animals with some degrees of successful inhibition of amyloidosis (Westermark & Grimelius 1973, Scrocchi et al. 2002, Potter et al. 2009, Wang et al. 2014, Wijesekara et al. 2015, Sivanesam et al. 2016). Investigating the mechanism of action of some of these inhibitory compounds has been useful in providing information on the mechanisms of refolding and fibrillogenesis (Cohen et al. 2006, Young et al. 2014, 2015). However, although many of these compounds have been tested in mice, not all of them would be suitable for clinical use. Alternative inhibitors which could be administered to humans have included some aromatic compounds including resevatrol (a component of red wine) (Porat et al. 2004, Evers et al. 2009, Mishra et al. 2009) and some polyphenols including (−)-Epigallocatechin 3-Gallate (EGCG) (a natural component of green tea); inhibition of fibrillogenesis was accompanied by disaggregation of fibrils and formation of non-toxic oligomers (Meng et al. 2010, Engel et al. 2012). Other inhibitors effective in vitro are tetracycline, congo red, ruthenium red, and organic designed ‘foldamers’ (Aitken et al. 2003, Kumar & Miranker 2013, Zhu et al. 2017). Therapeutic drugs used to treat hyperglycaemia have also been tested as inhibitors including thiazolidenediones, metformin and sulphonylureas with varying degrees of success (Fortin & Benoit-Biancamano 2015).
It is evident that in hIAPP transgenic animals and in spontaneously diabetic cats and monkeys, the degree of amyloidosis is directly related to the onset of hyperglycaemia since all diabetic animals have varying but quite severe islet amyloidosis (de Koning et al. 1993, Wang et al. 2001a, Guardado-Mendoza et al. 2009, Zhang et al. 2014). However, the islet amyloid load of islets in most patients with T2DM is not high and unlike animal models of islet amyloid and diabetes, the degree of amyloidosis is not related to onset or duration of diabetes. Islet amyloidosis is one of the few visible pathological changes in islets in T2DM (Ehrlich & Ratner 1961, Westermark 1972). The prevalence of islet amyloid (the percentage of islets in a pancreas affected by amyloid) has been assessed as being between <1% and 90% in patients with T2DM; a human pancreas has been estimated to have approximately 1000 islets. However, in more than 40% of patients with T2DM <25% of islets are affected (Clark & Moffitt 2007, Jurgens et al. 2011). Moreover, the patients with low prevalence of islet amyloidosis had <10% of the islet space filled with amyloid (amyloid severity) (Jurgens et al. 2011). Thus, in many patients with established T2DM, sometimes of many years’ duration, little amyloidosis is visible; low degrees of islet amyloid are associated with low percentages of β-cell loss (Jurgens et al. 2011). This would argue against the hypothesis that IAPP and ER stress were a major feature of β-cell death which was in turn a major aetiological factor for onset of T2DM. A more likely scenario is that functional changes in islet cells are associated with T2DM onset and that IAPP and amyloidosis are a result of these alterations. However, it is known that the distribution of islet amyloidosis in the pancreas is variable and it has been proposed that not all islets respond to a glucose challenge at any one time (Zhu et al. 2016). It may therefore be important to identify which islets are amyloid-containing (even if only a few are affected) in relation to any functional changes. The real questions are firstly, does this level of islet pathology require treatment and secondly, how would the efficacy of this treatment be assessed?
Current therapies for T2DM are directed towards optimising the release of insulin and glucagon from islet cells which are largely desensitised to glucose; the end points of this type of therapy are improved glycaemia (improved haemoglobin A1c) and better glucose homeostasis. These factors are critical for long term minimisation of the devastating complications of T2DM such as retinopathy, kidney disease and amputations. Whilst islet amyloid and the associated β-cell destruction will contribute to eventual reduction of insulin secretion, inhibitors for IAPP fibrillogenesis may not be the initial therapy of choice for T2DM unless it can be proved that IAPP and its cytotoxicity has a definite role in the pathogenesis of T2DM. Currently it seems more relevant to focus IAPP research on and working towards understanding the cellular and biochemical factors that promote misfolding/aggregation of IAPP in T2DM than developing anti-amyloid clinical therapies.
In addition, since the role of amyloidosis in onset and progression of T2DM is unclear, the development and assessment of an islet amyloid inhibitor would require relevant end points such as evidence of reduction of islet amyloidosis. Currently, it is not possible to determine which patients have islet amyloid or its rate of deposition. The search for treatments for AD and the progression of cerebral plaque formation have been accompanied by development of methods quantify cerebral amyloid plaques by imaging. Positron Emission Tomography scanning in combination with markers for amyloid and tau (Pittsburg Compound, PiB) (Klunk et al. 2004) has allowed identification of cerebral plaques and efficacy of treatment regimens. However, at the present levels of sensitivity, this type of analysis would not be successful for detection of a few islets with amyloid in vivo.
In conclusion, hIAPP, if overexpressed, can exert its toxic effects at many intracellular stages of protein production and degradation in β-cells. In T2DM, it is more likely that secreted IAPP misfolds outside the cells and the early stages of β-sheet assembly – the small oligomeric molecular species – are responsible for inducing apoptosis and contribute to fibril formation. These small molecular assemblies of hIAPP are the misfolded assassins of the β-cell in T2DM.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
Funding
This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.
Acknowledgements
All human material was obtained with appropriate Ethical Permissions and animal material was obtained from animals maintained under US or UK guidelines and regulations for animal care. A C thanks the many students and members of her team over the years who have made contributions to this story and whose work (and some unpublished data) is represented in this review; they include Eelco de Koning, Sophie Chargé, Emma Jaikaran, Claire Higham, Diane MacArthur, Rashpal Bhogal, Michael Badman, Jenni Moffitt, Britta Laube, Cathy Sloan, Lynne Scott and Chris Partridge. Much of this research work was financially supported by Diabetes UK and the Wellcome Trust. B H acknowledges support from the Wellcome Trust (095101/Z/10/Z and 200837/Z/16/Z) and the Medical Research Council UK (MR/L020149/1). D P R thanks the many members of his group who have contributed to our work on IAPP over the years and is grateful to Profs. Andisheh Abedini, Martin Zanni, Ann Marie Schmidt and Steven Kahn for continued long term collaborations and numerous helpful discussions and acknowledges the Wellcome Trust for funding.
References
Abedini A, Plesner A, Cao P, Ridgway Z, Zhang J, Tu LH, Middleton CT, Chao B, Sartori DJ & Meng F et al. 2016 Time-resolved studies define the nature of toxic IAPP intermediates, providing insight for anti-amyloidosis therapeutics. eLife 5 e12977. (doi:10.7554/eLife.12977)
Abedini A & Raleigh DP 2005 The role of His-18 in amyloid formation by human islet amyloid polypeptide. Biochemistry 44 16284–16291. (doi:10.1021/bbib51432v)
Abedini A & Raleigh DP 2006 Destabilization of human IAPP amyloid fibrils by proline mutations outside of the putative amyloidogenic domain: is there a critical amyloidogenic domain in human IAPP? Journal of Molecular Biology 355 274–281.
Abedini A & Raleigh DP 2009 A role for helical intermediates in amyloid formation by natively unfolded polypeptides? Physical Biology 6 015005. (doi:10.1088/1478-3975/6/1/015005)
Aitken JF, Loomes KM, Konarkowska B & Cooper GJ 2003 Suppression by polycyclic compounds of the conversion of human amylin into insoluble amyloid. Biochemcal Journal 374 779–784. (doi:10.1042/bj20030422)
Akter R, Cao P, Noor H, Ridgway Z, Tu LH, Wang H, Wong AG, Zhang X, Abedini A & Schmidt AM et al. 2016 Islet amyloid polypeptide: structure, function, and pathophysiology. Journal of Diabetes Research 2016 2798269. (doi:10.1155/2016/2798269)
Alam T, Chen L, Ogawa A, Leffert JD, Unger RH & Luskey KL 1992 Coordinate regulation of amylin and insulin expression in response to hypoglycemia and fasting. Diabetes 41 508–514. (doi:10.2337/diab.41.4.508)
Apostolidou M, Jayasinghe SA & Langen R 2008 Structure of alpha-helical membrane-bound human islet amyloid polypeptide and its implications for membrane-mediated misfolding. Journal of Biological Chemistry 283 17205–17210. (doi:10.1074/jbc.M801383200)
Ashrafi G & Schwarz TL 2013 The pathways of mitophagy for quality control and clearance of mitochondria. Cell Death and Differentiation 20 31–42. (doi:10.1038/cdd.2012.81)
Bartolome A, Kimura-Koyanagi M, Asahara S, Guillen C, Inoue H, Teruyama K, Shimizu S, Kanno A, Garcia-Aguilar A & Koike M et al. 2014 Pancreatic beta-cell failure mediated by mTORC1 hyperactivity and autophagic impairment. Diabetes 63 2996–3008. (doi:10.2337/db13-0970)
Betsholtz C, Christmansson L, Engstrom U, Rorsman F, Svensson V, Johnson KH & Westermark P 1989 Sequence divergence in a specific region of islet amyloid polypeptide (IAPP) explains differences in islet amyloid formation between species. FEBS Letters 251 261–264. (doi:10.1016/0014-5793(89)81467-X)
Bohman S & Westermark GT 2012 Extensive amyloid formation in transplanted microencapsulated mouse and human islets. Amyloid 19 87–93. (doi:10.3109/13506129.2012.679988)
Boni-Schnetzler M, Thorne J, Parnaud G, Marselli L, Ehses JA, Kerr-Conte J, Pattou F, Halban PA, Weir GC & Donath MY 2008 Increased interleukin (IL)-1beta messenger ribonucleic acid expression in beta-cells of individuals with type 2 diabetes and regulation of IL-1beta in human islets by glucose and autostimulation. Journal of Clinical Endocrinology and Metabolism 93 4065–4074. (doi:10.1210/jc.2008-0396)
Brender JR, Hartman K, Nanga RP, Popovych N, de la Salud Bea R, Vivekanandan S, Marsh EN & Ramamoorthy A 2010 Role of zinc in human islet amyloid polypeptide aggregation. Journal of the American Chemical Society 132 8973–8983. (doi:10.1021/ja1007867)
Brender JR, Lee EL, Hartman K, Wong PT, Ramamoorthy A, Steel DG & Gafni A 2011 Biphasic effects of insulin on islet amyloid polypeptide membrane disruption. Biophysical Journal 100 685–692. (doi:10.1016/j.bpj.2010.09.070)
Butler AE, Janson J, Bonner-Weir S, Ritzel R, Rizza RA & Butler PC 2003 Beta-cell deficit and increased beta-cell apoptosis in humans with type 2 diabetes. Diabetes 52 102–110. (doi:10.2337/diabetes.52.1.102)
Butler PC, Chou J, Carter WB, Wang YN, Bu BH, Chang D, Chang JK & Rizza RA 1990 Effects of meal ingestion on plasma amylin concentration in NIDDM and nondiabetic humans. Diabetes 39 752–756. (doi:10.2337/diab.39.6.752)
Cao P, Abedini A & Raleigh DP 2013a Aggregation of islet amyloid polypeptide: from physical chemistry to cell biology. Current Opinion in Structural Biology 23 82–89. (doi:10.1016/j.sbi.2012.11.003)
Cao P, Abedini A, Wang H, Tu LH, Zhang X, Schmidt AM & Raleigh DP 2013b Islet amyloid polypeptide toxicity and membrane interactions. PNAS 110 19279–19284. (doi:10.1073/pnas.1305517110)
Cao P, Tu LH, Abedini A, Levsh O, Akter R, Patsalo V, Schmidt AM & Raleigh DP 2012 Sensitivity of amyloid formation by human islet amyloid polypeptide to mutations at residue 20. Journal of Molecular Biology 421 282–295. (doi:10.1016/j.jmb.2011.12.032)
Casas S, Gomis R, Gribble FM, Altirriba J, Knuutila S & Novials A 2007 Impairment of the ubiquitin-proteasome pathway is a downstream endoplasmic reticulum stress response induced by extracellular human islet amyloid polypeptide and contributes to pancreatic beta-cell apoptosis. Diabetes 56 2284–2294. (doi:10.2337/db07-0178)
Casas S, Novials A, Reimann F, Gomis R & Gribble FM 2008 Calcium elevation in mouse pancreatic beta cells evoked by extracellular human islet amyloid polypeptide involves activation of the mechanosensitive ion channel TRPV4. Diabetologia 51 2252–2262. (doi:10.1007/s00125-008-1111-z)
Castillo GM, Cummings JA, Yang W, Judge ME, Sheardown MJ, Rimvall K, Hansen JB & Snow AD 1998 Sulfate content and specific glycosaminoglycan backbone of perlecan are critical for perlecan’s enhancement of islet amyloid polypeptide (amylin) fibril formation. Diabetes 47 612–620. (doi:10.2337/diabetes.47.4.612)
Ceridono M, Ory S, Momboisse F, Chasserot-Golaz S, Houy S, Calco V, Haeberle AM, Demais V, Bailly Y & Bader MF et al. 2011 Selective recapture of secretory granule components after full collapse exocytosis in neuroendocrine chromaffin cells. Traffic 12 72–88. (doi:10.1111/j.1600-0854.2010.01125.x)
Chen JJ, Genereux JC & Wiseman RL 2015 Endoplasmic reticulum quality control and systemic amyloid disease: impacting protein stability from the inside out. IUBMB Life 67 404–413. (doi:10.1002/iub.1386)
Chiti F & Dobson CM 2006 Protein misfolding, functional amyloid, and human disease. Annual Review of Biochemistry 75 333–366. (doi:10.1146/annurev.biochem.75.101304.123901)
Clark A 1992 Islet amyloid: an enigma of type 2 diabetes. Diabetes/Metabolism Research and Reviews 8 117–132. (doi:10.1002/dmr.5610080204)
Clark A, Edwards CA, Ostle LR, Sutton R, Rothbard JB, Morris JF & Turner RC 1989 Localisation of islet amyloid peptide in lipofuscin bodies and secretory granules of human B-cells and in islets of type-2 diabetic subjects. Cell and Tissue Research 257 179–185.
Clark A & Moffitt J 2007 Pancreatic Amyloid and Diabetes. In Protein Misfolding, Aggregation and Conformational Diseases, pp 199–216. Eds Uversky VN, Fibk AL. New York, NY, USA: Springer.
Clark A & Nilsson MR 2004 Islet amyloid: a complication of islet dysfunction or an aetiological factor in Type 2 diabetes? Diabetologia 47 157–169. (doi:10.1007/s00125-003-1304-4)
Clark A, Wells CA, Buley ID, Cruickshank JK, Vanhegan RI, Matthews DR, Cooper GJ, Holman RR & Turner RC 1988 Islet amyloid, increased A-cells, reduced B-cells and exocrine fibrosis: quantitative changes in the pancreas in type 2 diabetes. Diabetes Research 9 151–159.
Cnop M, Hughes SJ, Igoillo-Esteve M, Hoppa MB, Sayyed F, van de Laar L, Gunter JH, de Koning EJ, Walls GV & Gray DW et al. 2010 The long lifespan and low turnover of human islet beta cells estimated by mathematical modelling of lipofuscin accumulation. Diabetologia 53 321–330. (doi:10.1007/s00125-009-1562-x)
Cohen T, Frydman-Marom A, Rechter M & Gazit E 2006 Inhibition of amyloid fibril formation and cytotoxicity by hydroxyindole derivatives. Biochemistry 45 4727–4735. (doi:10.1021/bbib51525c)
Costes S, Huang CJ, Gurlo T, Daval M, Matveyenko AV, Rizza RA, Butler AE & Butler PC 2011 beta-Cell dysfunctional ERAD/ubiquitin/proteasome system in type 2 diabetes mediated by islet amyloid polypeptide-induced UCH-L1 deficiency. Diabetes 60 227–238. (doi:10.2337/db10-0522)
Courtade JA, Wang EY, Yen P, Dai DL, Soukhatcheva G, Orban PC & Verchere CB 2017 Loss of prohormone convertase 2 promotes beta cell dysfunction in a rodent transplant model expressing human pro-islet amyloid polypeptide. Diabetologia 60 453–463. (doi:10.1007/s00125-016-4174-2)
de Koning EJ, Bodkin NL, Hansen BC & Clark A 1993 Diabetes mellitus in Macaca mulatta monkeys is characterised by islet amyloidosis and reduction in beta-cell population. Diabetologia 36 378–384. (doi:10.1007/BF00402271)
de Koning EJ, Hoppener JW, Verbeek JS, Oosterwijk C, van Hulst KL, Baker CA, Lips CJ, Morris JF & Clark A 1994a Human islet amyloid polypeptide accumulates at similar sites in islets of transgenic mice and humans. Diabetes 43 640–644.
de Koning EJ, Morris ER, Hofhuis FM, Posthuma G, Hoppener JW, Morris JF, Capel PJ, Clark A & Verbeek JS 1994b Intra- and extracellular amyloid fibrils are formed in cultured pancreatic islets of transgenic mice expressing human islet amyloid polypeptide. PNAS 91 8467–8471.
de Koning EJ, van den Brand JJ, Mott VL, Charge SB, Hansen BC, Bodkin NL, Morris JF & Clark A 1998 Macrophages and pancreatic islet amyloidosis. Amyloid 5 247–254. (doi:10.3109/13506129809007297)
Donath MY & Shoelson SE 2011 Type 2 diabetes as an inflammatory disease. Nature Reviews Immunology 11 98–107. (doi:10.1038/nri2925)
Ehrlich JC & Ratner IM 1961 Amyloidosis of the islets of Langerhans. A restudy of islet hyalin in diabetic and non-diabetic individuals. American Journal of Pathology 38 49–59.
Engel MF 2009 Membrane permeabilization by islet amyloid polypeptide. Chemistry and Physics of Lipids 160 1–10. (doi:10.1016/j.chemphyslip.2009.03.008)
Engel MF, Khemtemourian L, Kleijer CC, Meeldijk HJ, Jacobs J, Verkleij AJ, de Kruijff B, Killian JA & Hoppener JW 2008 Membrane damage by human islet amyloid polypeptide through fibril growth at the membrane. PNAS 105 6033–6038. (doi:10.1073/pnas.0708354105)
Engel MF, vandenAkker CC, Schleeger M, Velikov KP, Koenderink GH & Bonn M 2012 The polyphenol EGCG inhibits amyloid formation less efficiently at phospholipid interfaces than in bulk solution. Journal of the American Chemical Society 134 14781–14788. (doi:10.1021/ja3031664)
Eriksson J, Nakazato M, Miyazato M, Shiomi K, Matsukura S & Groop L 1992 Islet amyloid polypeptide plasma concentrations in individuals at increased risk of developing type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 35 291–293. (doi:10.1007/BF00400933)
Evers F, Jeworrek C, Tiemeyer S, Weise K, Sellin D, Paulus M, Struth B, Tolan M & Winter R 2009 Elucidating the mechanism of lipid membrane-induced IAPP fibrillogenesis and its inhibition by the red wine compound resveratrol: a synchrotron X-ray reflectivity study. Journal of the American Chemical Society 131 9516–9521. (doi:10.1021/ja8097417)
Exley C, House E, Patel T, Wu L & Fraser PE 2010 Human pro-islet amyloid polypeptide (ProIAPP(1–48)) forms amyloid fibrils and amyloid spherulites in vitro. Journal of Inorganic Biochemistry 104 1125–1129. (doi:10.1016/j.jinorgbio.2010.05.012)
Fortin JS & Benoit-Biancamano MO 2015 In vitro evaluation of hypoglycemic agents to target human islet amyloid polypeptide: a key protein involved in amyloid deposition and beta-cell loss. Canadian Journal of Diabetes 39 373–382. (doi:10.1016/j.jcjd.2015.01.291)
Fox N, Schrementi J, Nishi M, Ohagi S, Chan SJ, Heisserman JA, Westermark GT, Leckstrom A, Westermark P & Steiner DF 1993 Human islet amyloid polypeptide transgenic mice as a model of non-insulin-dependent diabetes mellitus (NIDDM). FEBS Letters 323 40–44. (doi:10.1016/0014-5793(93)81444-5)
Gaulton KJ, Ferreira T, Lee Y, Raimondo A, Magi R, Reschen ME, Mahajan A, Locke A, Rayner NW & Robertson N et al. 2015 Genetic fine mapping and genomic annotation defines causal mechanisms at type 2 diabetes susceptibility loci. Nature Genetics 47 1415–1425. (doi:10.1038/ng.3437)
Gilead S & Gazit E 2008 The role of the 14–20 domain of the islet amyloid polypeptide in amyloid formation. Experimental Diabetes Research 2008 256954.
Glabe CG 2006 Common mechanisms of amyloid oligomer pathogenesis in degenerative disease. Neurobiology of Aging 27 570–575. (doi:10.1016/j.neurobiolaging.2005.04.017)
Glabe CG 2008 Structural classification of toxic amyloid oligomers. Journal of Biological Chemistry 283 29639–29643. (doi:10.1074/jbc.R800016200)
Glenner GG 1980 Amyloid deposits and amyloidosis. The beta-fibrilloses (first of two parts). New England Journal of Medicine 302 1283–1292.
Green J, Goldsbury C, Mini T, Sunderji S, Frey P, Kistler J, Cooper G & Aebi U 2003 Full-length rat amylin forms fibrils following substitution of single residues from human amylin. Journal of Molecular Biology 326 1147–1156. (doi:10.1016/S0022-2836(02)01377-3)
Gross A & Katz SG 2017 Non-apoptotic functions of BCL-2 family proteins. Cell Death and Differentiation 8 1348–1358. (doi:10.1038/cdd.2017.22)
Guardado-Mendoza R, Davalli AM, Chavez AO, Hubbard GB, Dick EJ, Majluf-Cruz A, Tene-Perez CE, Goldschmidt L, Hart J & Perego C et al. 2009 Pancreatic islet amyloidosis, beta-cell apoptosis, and alpha-cell proliferation are determinants of islet remodeling in type-2 diabetic baboons. PNAS 106 13992–13997. (doi:10.1073/pnas.0906471106)
Guettier JM, Lungu A, Goodling A, Cochran C & Gorden P 2013 The role of proinsulin and insulin in the diagnosis of insulinoma: a critical evaluation of the Endocrine Society clinical practice guideline. Journal of Clinical Endocrinology and Metabolism 98 4752–4758. (doi:10.1210/jc.2013-2182)
Gurlo T, Rivera JF, Butler AE, Cory M, Hoang J, Costes S & Butler PC 2016 CHOP contributes to, but is not the only mediator of, IAPP induced beta-cell apoptosis. Molecular Endocrinology 30 446–454. (doi:10.1210/me.2015-1255)
Gurlo T, Ryazantsev S, Huang CJ, Yeh MW, Reber HA, Hines OJ, O’Brien TD, Glabe CG & Butler PC 2010 Evidence for proteotoxicity in beta cells in type 2 diabetes: toxic islet amyloid polypeptide oligomers form intracellularly in the secretory pathway. American Journal of Pathology 176 861–869. (doi:10.2353/ajpath.2010.090532)
Haataja L, Gurlo T, Huang CJ & Butler PC 2008 Islet amyloid in type 2 diabetes, and the toxic oligomer hypothesis. Endocrine Reviews 29 303–316. (doi:10.1210/er.2007-0037)
Halban PA 1991 Structural domains and molecular lifestyles of insulin and its precursors in the pancreatic beta cell. Diabetologia 34 767–778. (doi:10.1007/BF00408349)
Han J, Pluhackova K & Bockmann RA 2017 The multifaceted role of SNARE proteins in membrane fusion. Frontiers in Physiology 8 5. (doi:10.3389/fphys.2017.00005)
Hastoy B, Scotti PA, Milochau A, Fezoua-Boubegtiten Z, Rodas J, Megret R, Desbat B, Laguerre M, Castano S & Perrais D et al. 2017 A central small amino acid in the VAMP2 transmembrane domain regulates the fusion pore in exocytosis. Scientific Reports 7 2835. (doi:10.1038/s41598-017-03013-3)
Higham CE, Hull RL, Lawrie L, Shennan KI, Morris JF, Birch NP, Docherty K & Clark A 2000a Processing of synthetic pro-islet amyloid polypeptide (proIAPP) ‘amylin’ by recombinant prohormone convertase enzymes, PC2 and PC3, in vitro. European Journal of Biochemistry 267 4998–5004. (doi:10.1046/j.1432-1327.2000.01548.x)
Higham CE, Jaikaran ET, Fraser PE, Gross M & Clark A 2000b Preparation of synthetic human islet amyloid polypeptide (IAPP) in a stable conformation to enable study of conversion to amyloid-like fibrils. FEBS Letters 470 55–60. (doi:10.1016/S0014-5793(00)01287-4)
Huang CJ, Haataja L, Gurlo T, Butler AE, Wu X, Soeller WC & Butler PC 2007a Induction of endoplasmic reticulum stress-induced beta-cell apoptosis and accumulation of polyubiquitinated proteins by human islet amyloid polypeptide. American Journal of Physiology: Endocrinology and Metabolism 293 E1656–E1662. (doi:10.1152/ajpendo.00318.2007)
Huang CJ, Lin CY, Haataja L, Gurlo T, Butler AE, Rizza RA & Butler PC 2007b High expression rates of human islet amyloid polypeptide induce endoplasmic reticulum stress mediated beta-cell apoptosis, a characteristic of humans with type 2 but not type 1 diabetes. Diabetes 56 2016–2027. (doi:10.2337/db07-0197)
Hull RL, Zraika S, Udayasankar J, Kisilevsky R, Szarek WA, Wight TN & Kahn SE 2007 Inhibition of glycosaminoglycan synthesis and protein glycosylation with WAS-406 and azaserine result in reduced islet amyloid formation in vitro. American Journal of Physiology: Cell Physiology 293 C1586–C1593. (doi:10.1152/ajpcell.00208.2007)
Hutton JC 1994 Insulin secretory granule biogenesis and the proinsulin-processing endopeptidases. Diabetologia 37 (Supplement 2) S48–S56. (doi:10.1007/BF00400826)
Hutton JC, Penn EJ & Peshavaria M 1983 Low-molecular-weight constituents of isolated insulin-secretory granules. Bivalent cations, adenine nucleotides and inorganic phosphate. Biochemical Journal 210 297–305. (doi:10.1042/bj2100297)
Jaikaran ET, Higham CE, Serpell LC, Zurdo J, Gross M, Clark A & Fraser PE 2001 Identification of a novel human islet amyloid polypeptide beta-sheet domain and factors influencing fibrillogenesis. Journal of Molecular Biology 308 515–525. (doi:10.1006/jmbi.2001.4593)
Jaikaran ET, Nilsson MR & Clark A 2004 Pancreatic beta-cell granule peptides form heteromolecular complexes which inhibit islet amyloid polypeptide fibril formation. Biochemical Journal 377 709–716. (doi:10.1042/bj20030852)
Janson J, Ashley RH, Harrison D, McIntyre S & Butler PC 1999 The mechanism of islet amyloid polypeptide toxicity is membrane disruption by intermediate-sized toxic amyloid particles. Diabetes 48 491–498. (doi:10.2337/diabetes.48.3.491)
Jarrett JT & Lansbury PT Jr 1993 Seeding ‘one-dimensional crystallization’ of amyloid: a pathogenic mechanism in Alzheimer’s disease and scrapie? Cell 73 1055–1058.
Jha S, Snell JM, Sheftic SR, Patil SM, Daniels SB, Kolling FW & Alexandrescu AT 2014 pH dependence of amylin fibrillization. Biochemistry 53 300–310. (doi:10.1021/bi401164k)
Jorgensen JO, Rosenfalck AM, Fisker S, Nyholm B, Fineman MS, Schmitz O, Madsbad S, Holst JJ & Christiansen JS 2000 Circulating levels of incretin hormones and amylin in the fasting state and after oral glucose in GH-deficient patients before and after GH replacement: a placebo-controlled study. European Journal of Endocrinology 143 593–599. (doi:10.1530/eje.0.1430593)
Jurgens CA, Toukatly MN, Fligner CL, Udayasankar J, Subramanian SL, Zraika S, Aston-Mourney K, Carr DB, Westermark P & Westermark GT et al. 2011 beta-Cell loss and beta-cell apoptosis in human type 2 diabetes are related to islet amyloid deposition. American Journal of Pathology 178 2632–2640. (doi:10.1016/j.ajpath.2011.02.036)
Kahn SE, D’Alessio DA, Schwartz MW, Fujimoto WY, Ensinck JW, Taborsky GJ Jr & Porte D Jr 1990 Evidence of cosecretion of islet amyloid polypeptide and insulin by beta-cells. Diabetes 39 634–638. (doi:10.2337/diab.39.5.634)
Kahn SE, Verchere CB, Andrikopoulos S, Asberry PJ, Leonetti DL, Wahl PW, Boyko EJ, Schwartz RS, Newell-Morris L & Fujimoto WY 1998 Reduced amylin release is a characteristic of impaired glucose tolerance and type 2 diabetes in Japanese Americans. Diabetes 47 640–645. (doi:10.2337/diabetes.47.4.640)
Kamata K, Mizukami H, Inaba W, Tsuboi K, Tateishi Y, Yoshida T & Yagihashi S 2014 Islet amyloid with macrophage migration correlates with augmented beta-cell deficits in type 2 diabetic patients. Amyloid 21 191–201. (doi:10.3109/13506129.2014.937857)
Kanatsuka A, Makino H, Ohsawa H, Tokuyama Y, Yamaguchi T, Yoshida S & Adachi M 1989 Secretion of islet amyloid polypeptide in response to glucose. FEBS Letters 259 199–201. (doi:10.1016/0014-5793(89)81527-3)
Kayed R, Sokolov Y, Edmonds B, McIntire TM, Milton SC, Hall JE & Glabe CG 2004 Permeabilization of lipid bilayers is a common conformation-dependent activity of soluble amyloid oligomers in protein misfolding diseases. Journal of Biological Chemistry 279 46363–46366. (doi:10.1074/jbc.C400260200)
Klunk WE, Engler H, Nordberg A, Wang Y, Blomqvist G, Holt DP, Bergstrom M, Savitcheva I, Huang GF & Estrada S et al. 2004 Imaging brain amyloid in Alzheimer’s disease with Pittsburgh Compound-B. Annals of Neurology 55 306–319. (doi:10.1002/ana.20009)
Knight JD, Hebda JA & Miranker AD 2006 Conserved and cooperative assembly of membrane-bound alpha-helical states of islet amyloid polypeptide. Biochemistry 45 9496–9508. (doi:10.1021/bbib60579z)
Knight JD, Williamson JA & Miranker AD 2008 Interaction of membrane-bound islet amyloid polypeptide with soluble and crystalline insulin. Protein Science 17 1850–1856.
Koo BW, Hebda JA & Miranker AD 2008 Amide inequivalence in the fibrillar assembly of islet amyloid polypeptide. Protein Engineering, Design and Selection 21 147–154. (doi:10.1093/protein/gzm076)
Krotee P, Rodriguez JA, Sawaya MR, Cascio D, Reyes FE, Shi D, Hattne J, Nannenga BL, Oskarsson ME & Philipp S et al. 2017 Atomic structures of fibrillar segments of hIAPP suggest tightly mated beta-sheets are important for cytotoxicity. eLife 6 e19273. (doi:10.7554/eLife.19273)
Kumar S & Miranker AD 2013 A foldamer approach to targeting membrane bound helical states of islet amyloid polypeptide. Chemical Communications 49 4749–4751. (doi:10.1039/c3cc41452c)
Kumar S, Schlamadinger DE, Brown MA, Dunn JM, Mercado B, Hebda JA, Saraogi I, Rhoades E, Hamilton AD & Miranker AD 2015 Islet amyloid-induced cell death and bilayer integrity loss share a molecular origin targetable with oligopyridylamide-based alpha-helical mimetics. Chemistry and Biology 22 369–378. (doi:10.1016/j.chembiol.2015.01.006)
Larson JL & Miranker AD 2004 The mechanism of insulin action on islet amyloid polypeptide fiber formation. Journal of Molecular Biology 335 221–231. (doi:10.1016/j.jmb.2003.10.045)
Law E, Lu S, Kieffer TJ, Warnock GL, Ao Z, Woo M & Marzban L 2010 Differences between amyloid toxicity in alpha and beta cells in human and mouse islets and the role of caspase-3. Diabetologia 53 1415–1427. (doi:10.1007/s00125-010-1717-9)
Lee CC, Sun Y & Huang HW 2012 How type II diabetes-related islet amyloid polypeptide damages lipid bilayers. Biophysical Journal 102 1059–1068. (doi:10.1016/j.bpj.2012.01.039)
Lee SC, Hashim Y, Li JK, Ko GT, Critchley JA, Cockram CS & Chan JC 2001 The islet amyloid polypeptide (amylin) gene S20G mutation in Chinese subjects: evidence for associations with type 2 diabetes and cholesterol levels. Clinical Endocrinology 54 541–546. (doi:10.1046/j.1365-2265.2001.01244.x)
Lorenzo A, Razzaboni B, Weir GC & Yankner BA 1994 Pancreatic islet cell toxicity of amylin associated with type-2 diabetes mellitus. Nature 368 756–760. (doi:10.1038/368756a0)
Luca S, Yau WM, Leapman R & Tycko R 2007 Peptide conformation and supramolecular organization in amylin fibrils: constraints from solid-state NMR. Biochemistry 46 13505–13522. (doi:10.1021/bi701427q)
Ma Z, Westermark GT, Sakagashira S, Sanke T, Gustavsson A, Sakamoto H, Engstrom U, Nanjo K & Westermark P 2001 Enhanced in vitro production of amyloid-like fibrils from mutant (S20G) islet amyloid polypeptide. Amyloid 8 242–249. (doi:10.3109/13506120108993820)
MacNamara CM, Barrow BA, Manley SE, Levy JC, Clark A & Turner RC 2000 Parallel changes of proinsulin and islet amyloid polypeptide in glucose intolerance. Diabetes Research and Clinical Practice 50 117–126. (doi:10.1016/S0168-8227(00)00188-1)
Maedler K, Sergeev P, Ris F, Oberholzer J, Joller-Jemelka HI, Spinas GA, Kaiser N, Halban PA & Donath MY 2002 Glucose-induced beta cell production of IL-1beta contributes to glucotoxicity in human pancreatic islets. Journal of Clinical Investigation 110 851–860. (doi:10.1172/JCI200215318)
Marzban L, Rhodes CJ, Steiner DF, Haataja L, Halban PA & Verchere CB 2006 Impaired NH2-terminal processing of human proislet amyloid polypeptide by the prohormone convertase PC2 leads to amyloid formation and cell death. Diabetes 55 2192–2201. (doi:10.2337/db05-1566)
Marzban L, Trigo-Gonzalez G, Zhu X, Rhodes CJ, Halban PA, Steiner DF & Verchere CB 2004 Role of beta-cell prohormone convertase (PC)1/3 in processing of pro-islet amyloid polypeptide. Diabetes 53 141–148. (doi:10.2337/diabetes.53.1.141)
Masters SL, Dunne A, Subramanian SL, Hull RL, Tannahill GM, Sharp FA, Becker C, Franchi L, Yoshihara E & Chen Z et al. 2010 Activation of the NLRP3 inflammasome by islet amyloid polypeptide provides a mechanism for enhanced IL-1beta in type 2 diabetes. Nature Immunology 11 897–904. (doi:10.1038/ni.1935)
Matveyenko AV & Butler PC 2006 Islet amyloid polypeptide (IAPP) transgenic rodents as models for type 2 diabetes. ILAR Journal 47 225–233. (doi:10.1093/ilar.47.3.225)
Mazor Y, Gilead S, Benhar I & Gazit E 2002 Identification and characterization of a novel molecular-recognition and self-assembly domain within the islet amyloid polypeptide. Journal of Molecular Biology 322 1013–1024. (doi:10.1016/S0022-2836(02)00887-2)
Meier DT, Entrup L, Templin AT, Hogan MF, Mellati M, Zraika S, Hull RL & Kahn SE 2016 The S20G substitution in hIAPP is more amyloidogenic and cytotoxic than wild-type hIAPP in mouse islets. Diabetologia 59 2166–2171. (doi:10.1007/s00125-016-4045-x)
Meier DT, Morcos M, Samarasekera T, Zraika S, Hull RL & Kahn SE 2014 Islet amyloid formation is an important determinant for inducing islet inflammation in high-fat-fed human IAPP transgenic mice. Diabetologia 57 1884–1888. (doi:10.1007/s00125-014-3304-y)
Meng F, Abedini A, Plesner A, Verchere CB & Raleigh DP 2010 The flavanol (−)-epigallocatechin 3-gallate inhibits amyloid formation by islet amyloid polypeptide, disaggregates amyloid fibrils, and protects cultured cells against IAPP-induced toxicity. Biochemistry 49 8127–8133. (doi:10.1021/bi100939a)
Meng F & Raleigh DP 2011 Inhibition of glycosaminoglycan-mediated amyloid formation by islet amyloid polypeptide and proIAPP processing intermediates. Journal of Molecular Biology 406 491–502. (doi:10.1016/j.jmb.2010.12.028)
Mirzabekov TA, Lin MC & Kagan BL 1996 Pore formation by the cytotoxic islet amyloid peptide amylin. Journal of Biological Chemistry 271 1988–1992.
Mishra R, Sellin D, Radovan D, Gohlke A & Winter R 2009 Inhibiting islet amyloid polypeptide fibril formation by the red wine compound resveratrol. Chembiochem 10 445–449. (doi:10.1002/cbic.200800762)
Montane J, de Pablo S, Obach M, Cadavez L, Castano C, Alcarraz-Vizan G, Visa M, Rodriguez-Comas J, Parrizas M & Servitja JM et al. 2016 Protein disulfide isomerase ameliorates beta-cell dysfunction in pancreatic islets overexpressing human islet amyloid polypeptide. Molecular and Cellular Endocrinology 420 57–65. (doi:10.1016/j.mce.2015.11.018)
Nilsson MR & Raleigh DP 1999 Analysis of amylin cleavage products provides new insights into the amyloidogenic region of human amylin. Journal of Molecular Biology 294 1375–1385. (doi:10.1006/jmbi.1999.3286)
O’Brien TD, Butler AE, Roche PC, Johnson KH & Butler PC 1994a Islet amyloid polypeptide in human insulinomas. Evidence for intracellular amyloidogenesis. Diabetes 43 329–336.
O’Brien TD, Rizza RA, Carney JA & Butler PC 1994b Islet amyloidosis in a patient with chronic massive insulin resistance due to antiinsulin receptor antibodies. Journal of Clinical Endocrinology and Metabolism 79 290–292.
Obermuller S, Lindqvist A, Karanauskaite J, Galvanovskis J, Rorsman P & Barg S 2005 Selective nucleotide-release from dense-core granules in insulin-secreting cells. Journal of Cell Science 118 4271–4282. (doi:10.1242/jcs.02549)
Oskarsson ME, Singh K, Wang J, Vlodavsky I, Li JP & Westermark GT 2015 Heparan sulfate proteoglycans are important for islet amyloid formation and islet amyloid polypeptide-induced apoptosis. Journal of Biological Chemistry 290 15121–15132. (doi:10.1074/jbc.M114.631697)
Park YJ, Warnock GL, Ao Z, Safikhan N, Meloche M, Asadi A, Kieffer TJ & Marzban L 2017 Dual role of interleukin-1beta in islet amyloid formation and its beta-cell toxicity: implications for type 2 diabetes and islet transplantation. Diabetes, Obesity and Metabolism 5 682–694. (doi:10.1111/dom.12873)
Paulsson JF, Andersson A, Westermark P & Westermark GT 2006 Intracellular amyloid-like deposits contain unprocessed pro-islet amyloid polypeptide (proIAPP) in beta cells of transgenic mice overexpressing the gene for human IAPP and transplanted human islets. Diabetologia 49 1237–1246. (doi:10.1007/s00125-006-0206-7)
Pilkington EH, Gurzov EN, Kakinen A, Litwak SA, Stanley WJ, Davis TP & Ke PC 2016 Pancreatic beta-cell membrane fluidity and toxicity induced by human islet amyloid polypeptide species. Scientific Reports 6 21274. (doi:10.1038/srep21274)
Porat Y, Mazor Y, Efrat S & Gazit E 2004 Inhibition of islet amyloid polypeptide fibril formation: a potential role for heteroaromatic interactions. Biochemistry 43 14454–14462. (doi:10.1021/bbib48582a)
Potter KJ, Abedini A, Marek P, Klimek AM, Butterworth S, Driscoll M, Baker R, Nilsson MR, Warnock GL & Oberholzer J et al. 2010 Islet amyloid deposition limits the viability of human islet grafts but not porcine islet grafts. PNAS 107 4305–4310. (doi:10.1073/pnas.0909024107)
Potter KJ, Scrocchi LA, Warnock GL, Ao Z, Younker MA, Rosenberg L, Lipsett M, Verchere CB & Fraser PE 2009 Amyloid inhibitors enhance survival of cultured human islets. Biochimica et Biophysica Acta 1790 566–574. (doi:10.1016/j.bbagen.2009.02.013)
Potter KJ, Werner I, Denroche HC, Montane J, Plesner A, Chen Y, Lei D, Soukhatcheva G, Warnock GL & Oberholzer J et al. 2015 Amyloid formation in human islets is enhanced by heparin and inhibited by heparinase. American Journal of Transplantation 15 1519–1530. (doi:10.1111/ajt.13134)
Prasad RB & Groop L 2015 Genetics of type 2 diabetes-pitfalls and possibilities. Genes 6 87–123. (doi:10.3390/genes6010087)
Richardson SJ, Willcox A, Bone AJ, Foulis AK & Morgan NG 2009 Islet-associated macrophages in type 2 diabetes. Diabetologia 52 1686–1688. (doi:10.1007/s00125-009-1410-z)
Riek R & Eisenberg DS 2016 The activities of amyloids from a structural perspective. Nature 539 227–235. (doi:10.1038/nature20416)
Ritzel RA, Meier JJ, Lin CY, Veldhuis JD & Butler PC 2007 Human islet amyloid polypeptide oligomers disrupt cell coupling, induce apoptosis, and impair insulin secretion in isolated human islets. Diabetes 56 65–71. (doi:10.2337/db06-0734)
Rivera JF, Costes S, Gurlo T, Glabe CG & Butler PC 2014 Autophagy defends pancreatic beta cells from human islet amyloid polypeptide-induced toxicity. Journal of Clinical Investigation 124 3489–3500. (doi:10.1172/JCI71981)
Rorsman P & Renstrom E 2003 Insulin granule dynamics in pancreatic beta cells. Diabetologia 46 1029–1045. (doi:10.1007/s00125-003-1153-1)
Sakagashira S, Hiddinga HJ, Tateishi K, Sanke T, Hanabusa T, Nanjo K & Eberhardt NL 2000 S20G mutant amylin exhibits increased in vitro amyloidogenicity and increased intracellular cytotoxicity compared to wild-type amylin. American Journal of Pathology 157 2101–2109. (doi:10.1016/S0002-9440(10)64848-1)
Sakagashira S, Sanke T, Hanabusa T, Shimomura H, Ohagi S, Kumagaye KY, Nakajima K & Nanjo K 1996 Missense mutation of amylin gene (S20G) in Japanese NIDDM patients. Diabetes 45 1279–1281. (doi:10.2337/diab.45.9.1279)
Sanke T, Hanabusa T, Nakano Y, Oki C, Okai K, Nishimura S, Kondo M & Nanjo K 1991 Plasma islet amyloid polypeptide (Amylin) levels and their responses to oral glucose in type 2 (non-insulin-dependent) diabetic patients. Diabetologia 34 129–132. (doi:10.1007/BF00500385)
Schnell AH, Swenne I & Borg LA 1988 Lysosomes and pancreatic islet function. A quantitative estimation of crinophagy in the mouse pancreatic B-cell. Cell and Tissue Research 252 9–15.
Scrocchi LA, Chen Y, Waschuk S, Wang F, Cheung S, Darabie AA, McLaurin J & Fraser PE 2002 Design of peptide-based inhibitors of human islet amyloid polypeptide fibrillogenesis. Journal of Molecular Biology 318 697–706. (doi:10.1016/S0022-2836(02)00164-X)
Selkoe DJ 2011 Resolving controversies on the path to Alzheimer’s therapeutics. Nature Medicine 17 1060–1065. (doi:10.1038/nm.2460)
Sivanesam K, Shu I, Huggins KN, Tatarek-Nossol M, Kapurniotu A & Andersen NH 2016 Peptide Inhibitors of the amyloidogenesis of IAPP: verification of the hairpin-binding geometry hypothesis. FEBS Letters 590 2575–2583. (doi:10.1002/1873-3468.12261)
Sunde M & Blake C 1997 The structure of amyloid fibrils by electron microscopy and X-ray diffraction. Advances in Protein Chemistry 50 123–159.
Taylor RC, Cullen SP & Martin SJ 2008 Apoptosis: controlled demolition at the cellular level. Nature Reviews Molecular Cell Biology 9 231–241. (doi:10.1038/nrm2312)
Temple R, Clark PM & Hales CN 1992 Measurement of insulin secretion in type 2 diabetes: problems and pitfalls. Diabetic Medicine 9 503–512. (doi:10.1111/j.1464-5491.1992.tb01830.x)
Toshimori H, Narita R, Nakazato M, Asai J, Mitsukawa T, Kangawa K, Matsuo H, Takahashi K & Matsukura S 1991 Islet amyloid polypeptide in insulinoma and in the islets of the pancreas of non-diabetic and diabetic subjects. Virchows Archivs 418 411–417. (doi:10.1007/BF01605927)
Trikha S & Jeremic AM 2011 Clustering and internalization of toxic amylin oligomers in pancreatic cells require plasma membrane cholesterol. Journal of Biological Chemistry 286 36086–36097. (doi:10.1074/jbc.M111.240762)
Udayasankar J, Kodama K, Hull RL, Zraika S, Aston-Mourney K, Subramanian SL, Tong J, Faulenbach MV, Vidal J & Kahn SE 2009 Amyloid formation results in recurrence of hyperglycaemia following transplantation of human IAPP transgenic mouse islets. Diabetologia 52 145–153. (doi:10.1007/s00125-008-1185-7)
Wang F, Hull RL, Vidal J, Cnop M & Kahn SE 2001a Islet amyloid develops diffusely throughout the pancreas before becoming severe and replacing endocrine cells. Diabetes 50 2514–2520. (doi:10.2337/diabetes.50.11.2514)
Wang J, Xu J, Finnerty J, Furuta M, Steiner DF & Verchere CB 2001b The prohormone convertase enzyme 2 (PC2) is essential for processing pro-islet amyloid polypeptide at the NH2-terminal cleavage site. Diabetes 50 534–539. (doi:10.2337/diabetes.50.3.534)
Wang H, Abedini A, Ruzsicska B & Raleigh DP 2014 Rationally designed, nontoxic, nonamyloidogenic analogues of human islet amyloid polypeptide with improved solubility. Biochemistry 53 5876–5884. (doi:10.1021/bi500592p)
Ward WK, LaCava EC, Paquette TL, Beard JC, Wallum BJ & Porte D Jr 1987 Disproportionate elevation of immunoreactive proinsulin in type 2 (non-insulin-dependent) diabetes mellitus and in experimental insulin resistance. Diabetologia 30 698–702. (doi:10.1007/BF00296991)
Westermark GT, Davalli AM, Secchi A, Folli F, Kin T, Toso C, Shapiro AM, Korsgren O, Tufveson G & Andersson A et al. 2012 Further evidence for amyloid deposition in clinical pancreatic islet grafts. Transplantation 93 219–223. (doi:10.1097/TP.0b013e31823e46ef)
Westermark P 1972 Quantitative studies on amyloid in the islets of Langerhans. Upsala Journal of Medical Sciences 77 91–94. (doi:10.1517/03009734000000014)
Westermark P 1973 Fine structure of islets of Langerhans in insular amyloidosis. Virchows Archivs 359 1–18. (doi:10.1007/BF00549079)
Westermark P, Andersson A & Westermark GT 2011 Islet amyloid polypeptide, islet amyloid, and diabetes mellitus. Physiological Reviews 91 795–826. (doi:10.1152/physrev.00042.2009)
Westermark P, Engstrom U, Johnson KH, Westermark GT & Betsholtz C 1990 Islet amyloid polypeptide: pinpointing amino acid residues linked to amyloid fibril formation. PNAS 87 5036–5040. (doi:10.1073/pnas.87.13.5036)
Westermark P & Grimelius L 1973 The pancreatic islet cells in insular amyloidosis in human diabetic and non-diabetic adults. Acta Pathologica et Microbiologica Scandinavica: Section A 81 291–300.
Westermark P, Li ZC, Westermark GT, Leckstrom A & Steiner DF 1996 Effects of beta cell granule components on human islet amyloid polypeptide fibril formation. FEBS Letters 379 203–206. (doi:10.1016/0014-5793(95)01512-4)
Westwell-Roper C, Dai DL, Soukhatcheva G, Potter KJ, van Rooijen N, Ehses JA & Verchere CB 2011 IL-1 blockade attenuates islet amyloid polypeptide-induced proinflammatory cytokine release and pancreatic islet graft dysfunction. Journal of Immunology 187 2755–2765. (doi:10.4049/jimmunol.1002854)
Wijesekara N, Ahrens R, Wu L, Ha K, Liu Y, Wheeler MB & Fraser PE 2015 Islet amyloid inhibitors improve glucose homeostasis in a transgenic mouse model of type 2 diabetes. Diabetes, Obesity and Metabolism 17 1003–1006. (doi:10.1111/dom.12529)
Williamson JA, Loria JP & Miranker AD 2009 Helix stabilization precedes aqueous and bilayer-catalyzed fiber formation in islet amyloid polypeptide. Journal of Molecular Biology 393 383–396. (doi:10.1016/j.jmb.2009.07.077)
Williamson