Amyloid fibril formation from full-length and fragments of amylin

Amyloiddeposits of fibrillar human amylin (hA) in the pancreas may be a causative factor in type-2 diabetes. A detailed comparison of in vitro fibril formation by full-length hA(1-37) versus fragments of this peptide-hA(8-37) and hA(20-29)-is presented. Circular dichroism spectroscopy revealed that fibril formation was accompanied by a conformational change: random coil to beta-sheet/alpha-helical structure. Fibril morphologies were visualized by electron microscopy and displayed a remarkable diversity. hA(20-29) formed flat ribbons consisting of numerous 3. 6-nm-wide protofibrils. In contrast, hA(1-37) and hA(8-37) formed polymorphic higher order fibrils by lateral association and/or coiling together of 5.0-nm-wide protofibril subunits. For full-length hA(1-37), the predominant fibril type contained three protofibrils and for hA(8-37), the predominant type contained two protofibrils. Polymerization was also monitored with the thioflavin-T binding assay, which revealed different kinetics of assembly for hA(1-37) and hA(8-37) fibrils. hA(20-29) fibrils did not bind thioflavin-T. Together the results demonstrate that the N-terminal region of the hA peptide influences the relative frequencies of the various higher order fibril types and thereby the overall kinetics of fibril formation. Furthermore, while residues 20-29 contribute to the fibrils' beta-sheet core, the flanking C- and N-terminal regions of the hA peptide determine the interactions involved in the formation of higher order coiled polymorphic superstructures.     

The parallel superpleated beta-structure as a model for amyloid fibrils of human amylin

Human amylin is a 37 amino acid residue peptide hormone whose fibrillogenesis has been correlated with type 2 diabetes. These fibrils are rope-like bundles of several 5nm diameter protofilaments. Here, we propose, as a model for the protofilament, a variant of the parallel superpleated beta-structure previously derived for amyloid filaments of the yeast prion Ure2p. In the amylin model, individual polypeptides from residues 9 to 37 have a planar S-shaped fold with three beta-strands. These serpentines are stacked in register, with a 0.47 nm axial rise and a small rotational twist per step, generating an array of three parallel beta-sheets in cross-beta conformation. The interior, the two "bays" sandwiched between adjacent sheets, are occupied by non-polar and by polar/uncharged residues that are predicted to form H-bonded ladders, similar to those found in beta-helical proteins. The N-terminal peptide containing a disulfide bond occupies an extraneous peripheral position in the protofilament. The left-handed twist of the beta-sheets is shown to underlie left-handed coiling of amylin protofilaments in fibrils. The model is consistent with current biophysical, biochemical and genetic data and, in particular, affords a plausible explanation for why rodent amylin does not form fibrils.     

Atomic force microscopy reveals defects within mica supported lipid bilayers induced by the amyloidogenic human amylin peptide

To date, over 20 peptides or proteins have been identified that can form amyloid fibrils in the body and are thought to cause disease. The mechanism by which amyloid peptides cause the cytotoxicity observed and disease is not understood. However, one of the major hypotheses is that amyloid peptides cause membrane perturbation. Hence, we have studied the interaction between lipid bilayers and the 37 amino acid residue polypeptide amylin, which is the primary constituent of the pancreatic amyloid associated with type 2 diabetes. Using a dye release assay we confirmed that the amyloidogenic human amylin peptide causes membrane disruption; however, time-lapse atomic force microscopy revealed that this did not occur by the formation of defined pores. On the contrary, the peptide induced the formation of small defects spreading over the lipid surface. We also found that rat amylin, which has 84% identity with human amylin but cannot form amyloid fibrils, could also induce similar lesions to supported lipid bilayers. The effect, however, for rat amylin but not human amylin, was inhibited under high ionic conditions. These data provide an alternative theory to pore formation, and how amyloid peptides may cause membrane disruption and possibly cytotoxicity.     

A partially structured region of a largely unstructured protein, Plasmodium falciparum merozoite surface protein 2 (MSP2), forms amyloid-like fibrils.

Merozoite surface protein 2 (MSP2) from the human malaria parasite Plasmodium falciparum is expressed as a GPI-anchored protein on the merozoite surface. It has been implicated in the process of erythrocyte invasion and is a leading vaccine candidate. MSP2 is an intrinsically unstructured protein (IUP), and recombinant MSP2 forms amyloid-like fibrils upon storage. We have examined synthetic peptides corresponding to sequences in the conserved N-terminal region of MSP2 for the presence of local structure and the ability to form fibrils related to those formed by full-length MSP2. In a 25-residue peptide corresponding to the entire N-terminal region of mature MSP2, structures calculated from NMR data show the presence of nascent helical and turn-like structures. An 8-residue peptide from the central region of the N-terminal domain (residues 8-15) also formed a turn-like structure. Both peptides formed fibrils that were similar but not identical to the amyloid-like fibrils formed by full-length MSP2. Notably, the fibrils formed by the peptides bound both Congo Red and Thioflavin T, whereas the fibrils formed by full-length MSP2 bound only Congo Red. The propensity of peptides from the N-terminal conserved region of MSP2 to form amyloid-like fibrils makes it likely that this region contributes to fibril formation by the full-length protein. Thus, in contrast to the more common pathway of amyloid formation by structured proteins, which proceeds via partially unfolded intermediates that then undergo beta-aggregation, MSP2 is an example of a largely unstructured protein with at least one small structured region that has an important role in fibril formation.     

Receptor-mediated toxicity of human amylin fragment aggregated by short- and long-term incubations with copper ions

Human amylin (hA1–37) is a polypeptide hormone secreted in conjunction with insulin from the pancreatic β-cells involved in the pathogenesis of type 2 diabetes mellitus (T2DM). The shorter fragment hA17–29 than full-length peptide is capable to form amyloids "in vitro". Here, we monitored the time course of hA17–29 β-amyloid fibril and oligomer formation [without and with copper(II)], cellular toxicity of different amyloid aggregates, and involvement of specific receptors (receptor for advanced glycation end-products, RAGE; low-affinity nerve growth factor receptor, p75-NGFR) in aggregate toxicity. Fibril and oligomer formation of hA17–29 incubated at 37 °C for 0, 48, and 120 h, without or with copper(II), were measured by the thioflavin T fluorescence assay and ELISA, respectively. Toxicity of hA17–29 aggregates and effects of anti-RAGE and anti-p75-NGFR antibodies were evaluated on neuroblastoma SH-SY5Y viability. Fluorescence assay of hA17–29 indicates an initial slow rate of soluble fibril formation (48 h), followed by a slower rate of insoluble aggregate formation (120 h). The highest quantity of oligomers was recorded when hA17–29 was pre-aggregated for 48 h in the presence of copper(II) showing also the maximal cell toxicity (?44% of cell viability, p < 0.01 compared to controls). Anti-RAGE or anti-p75-NGFR antibodies almost abolished cell toxicity of hA17–29 aggregates. These results indicate that copper(II) influences the aggregation process and hA17–29 toxicities are especially attributable to oligomeric aggregates. hA17–29 aggregate toxicity seems to be mediated by RAGE and p75-NGFR receptors which might be potential targets for new drugs in T2DM treatment.     

Backbone-modified amylin derivatives: implications for amyloid inhibitor design and as template for self-assembling bionanomaterials.

This report reviews our approach to the design, synthesis and structural/morphological analysis of backbone-modified amylin(20-29) derivatives. Depending on the position in the peptide backbone and the type of amide bond isostere/modification, the amylin(20-29) peptides behave either as inhibitors of amyloid fibril formation, which are able to retard amyloid formation of native amylin(20-29), or as templates for the formation of self-assembled supramolecular structures. Molecular fine-tuning of the hydrogen-bond accepting/donating properties allows the control over the morphology of the supramolecular aggregation motifs such as helical ribbons and tapes, ribbons progressing to closed peptide nanotubes, (twisted) lamellar sheets or amyloid fibrils.     

Atomic structures of IAPP (amylin) fusions suggest a mechanism for fibrillation and the role of insulin in the process

Islet Amyloid Polypeptide (IAPP or amylin) is a peptide hormone produced and stored in the β‐islet cells of the pancreas along with insulin. IAPP readily forms amyloid fibrils in vitro, and the deposition of fibrillar IAPP has been correlated with the pathology of type II diabetes. The mechanism of the conversion that IAPP undergoes from soluble to fibrillar forms has been unclear. By chaperoning IAPP through fusion to maltose binding protein, we find that IAPP can adopt a α‐helical structure at residues 8–18 and 22–27 and that molecules of IAPP dimerize. Mutational analysis suggests that this dimerization is on the pathway to fibrillation. The structure suggests how IAPP may heterodimerize with insulin, which we confirmed by protein crosslinking. Taken together, these experiments suggest the helical dimerization of IAPP accelerates fibril formation and that insulin impedes fibrillation by blocking the IAPP dimerization interface.     

The amylin peptide implicated in type 2 diabetes stimulates copper-mediated carbonyl group and ascorbate radical formation

Human amylin (hA), which is toxic to islet β-cells, can self-generate H₂O₂, and this process is greatly enhanced in the presence of Cu(II) ions. Here we show that carbonyl groups, a marker of oxidative modification, were formed in hA incubated in the presence of Cu(II) ions or Cu(II) ions plus H₂O₂, but not in the presence of H₂O₂ alone. Furthermore, under similar conditions (i.e., in the presence of both Cu(II) ions and H₂O₂), hA also stimulated ascorbate radical formation. The same observations concerning carbonyl group formation were made when the histidine residue (at position 18) in hA was replaced by alanine, indicating that this residue does not play a key role. In complete contrast to hA, rodent amylin, which is nontoxic, does not generate H₂O₂, and binds Cu(II) ions only weakly, showed none of these properties. We conclude that the hA-Cu(II)/Cu(I) complex is redox active, with electron donation from the peptide reducing the oxidation state of the copper ions. The complex is capable of forming H₂O₂ from O₂ and can also generate ^•OH via Fenton chemistry. These redox properties of hA can explain its ability to stimulate copper-mediated carbonyl group and ascorbate radical formation. The formation of reactive oxygen species from hA in this way could hold the key to a better understanding of the damaging consequences of amyloid formation within the pancreatic islets of patients with type 2 diabetes mellitus.     

Human amylin oligomer growth and fibril elongation define two distinct phases in amyloid formation

Human amylin (hA), a 37-amino-acid polypeptide, is one of a number of peptides with the ability to form amyloid fibrils and cause disease. It is the main constituent of the pancreatic amyloid deposits associated with type 2 diabetes. Increasing interest in early assembly intermediates rather than the mature fibrils as the cytotoxic agent has led to this study in which the smallest hA oligomers have been captured by atomic force microscopy. These are 2.3 +/- 1.9 nm in height, 23 +/- 14 nm in length, and consist of an estimated 16 hA molecules. Oligomers first grow to a height of about 6 nm before they begin to significantly elongate into fibrils. Congo red inhibits elongation but not the growth in height of hA oligomers. Two distinct phases have thus been identified in hA fibrillogenesis: lateral growth of oligomers followed by longitudinal growth into mature fibrils. These observations suggest that mature fibrils are assembled directly via longitudinal growth of full-width oligomers, making assembly by lateral association of protofibrils appear less likely.     

Identification of minimal peptide sequences in the (8-20) domain of human islet amyloid polypeptide involved in fibrillogenesis

We have examined a series of overlapping peptide fragments from the 8-20 region of human islet amyloid polypeptide (IAPP) with the objective of defining the smallest fibril-forming domain. Peptide fragments corresponding to LANFLV (residues 12-17) and FLVHSS (residues 15-20) were strong enhancers of beta-sheet transition and fibril formation. Negative stain electron microscopy illustrated the ability of these peptide fragments to form fibrils independently when incubated alone in solution. Circular dichroism analysis revealed that when full-length human IAPP was incubated in the presence of these two fragments, fibrillogenesis was accelerated. While the two fragments, LANFLV and FLVHSS, were able to enhance the recruitment of additional IAPP molecules during fibril formation, the "seeding" activity of these peptides had no effect on altering IAPP-induced cytotoxcity as determined by cell culture studies. Therefore, this study has identified two internal IAPP peptide fragments within the 8-20 domain that may have a role in enhancing the folding and aggregation of human IAPP. These fragments are the smallest sequences identified, within the 8-20 region of hIAPP, that can independently form fibrils, and that can interact with IAPP to assemble into fibrils with characteristics similar as those formed by human IAPP alone.     

Destabilization of human IAPP amyloid fibrils by proline mutations outside of the putative amyloidogenic domain: is there a critical amyloidogenic domain in human IAPP?

Islet amyloid polypeptide (IAPP; amylin) is responsible for amyloid formation in type-2 diabetes. Not all organisms form islet amyloid, and amyloid formation correlates strongly with variations in primary sequence. Studies of human and rodent IAPP have pointed to the amino acid residues 20-29 region as the important amyloid-modulating sequence. The rat 20-29 sequence contains three proline residues and does not form amyloid, while the human sequence contains no proline and readily forms amyloid. This has led to the view that the 20-29 region constitutes a critical amyloidogenic domain that dictates the properties of the entire sequence. The different behavior of human and rat IAPP could be due to differences in the 20-29 region or due simply to the fact that multiple proline residues destabilize amyloid fibrils. We tested how critical the 20-29 region is by studying a variant identical with the human peptide in this segment but with three proline residues outside this region. We designed a variant of the amyloidogenic 8-37 region of human IAPP (hIAPP(8-37) 3xP) with proline substitutions at positions 17, 19 and 30. Compared to the wild-type, the 3xP variant was much easier to synthesize and had dramatically greater solubility. Fourier transform infra red spectroscopy, transmission electron microscopy, Congo red staining and thioflavin-T binding indicate that this variant has a reduced tendency to form beta-sheet structure and forms deposits with much less structural order than the wild-type. Far-UV CD studies show that the small amount of beta-sheet structure developed by hIAPP(8-37) 3xP after long periods of incubation dissociates readily into random-coil structure upon dilution into Tris buffer. The observation that proline substitutions outside the putative core domain effectively abolish amyloid formation indicates that models of IAPP aggregation must consider contributions from other regions.     

Amyloidogenic Mutation Promotes Fibril Formation of the N-terminal Apolipoprotein A-I on Lipid Membranes

The N-terminal amino acid 1–83 fragment of apolipoprotein A-I (apoA-I) has a strong propensity to form amyloid fibrils at physiological neutral pH. Because apoA-I has an ability to bind to lipid membranes, we examined the effects of the lipid environment on fibril-forming properties of the N-terminal fragment of apoA-I variants. Thioflavin T fluorescence assay as well as fluorescence and transmission microscopies revealed that upon lipid binding, fibril formation by apoA-I 1–83 is strongly inhibited, whereas the G26R mutant still retains the ability to form fibrils. Such distinct effects of lipid binding on fibril formation were also observed for the amyloidogenic prone region-containing peptides, apoA-I 8–33 and 8–33/G26R. This amyloidogenic region shifts from random coil to α-helical structure upon lipid binding. The G26R mutation appears to prevent this helix transition because lower helical propensity and more solvent-exposed conformation of the G26R variant upon lipid binding were observed in the apoA-I 1–83 fragment and 8–33 peptide. With a partially α-helical conformation induced by the presence of 2,2,2-trifluoroethanol, fibril formation by apoA-I 1–83 was strongly inhibited, whereas the G26R variant can form amyloid fibrils. These findings suggest a new possible pathway for amyloid fibril formation by the N-terminal fragment of apoA-I variants: the amyloidogenic mutations partially destabilize the α-helical structure formed upon association with lipid membranes, resulting in physiologically relevant conformations that allow fibril formation.     

Copper(II) inhibits the formation of amylin amyloid in vitro

The amyloidogenic peptide amylin is found associated with pancreatic islet beta-cells and is implicated in the aetiology of type-2 diabetes mellitus. We have used fluorimetry and transmission electron microscopy to investigate in vitro the influence of Al(III), Fe(III), Zn(II) and Cu(II) on amylin amyloid formation under near-physiological conditions. Cu(II) at 10.0 microM inhibited amylin of 0.4 and 2.0 microM from forming amyloid fibrils while the same concentration of either Al(III) or Zn(II) promoted the formation of beta-pleated sheet structures. If amylin amyloid is cytotoxic to beta-cells then Cu(II) should protect against the degeneration of the islets in type-2 diabetes mellitus.     

Design of peptide-based inhibitors of human islet amyloid polypeptide fibrillogenesis

Human islet amyloid polypeptide (IAPP) is the major component of amyloid deposits found in the pancreas of over 90% of all cases of type-2 diabetes. We have generated a series of overlapping hexapeptides to target an amyloidogenic region of IAPP (residues 20-29) and examined their effects on fibril assembly. Peptide fragments corresponding to SNNFGA (residues 20-25) and GAILSST (residues 24-29) were strong inhibitors of the beta-sheet transition and amyloid aggregation. Circular dichroism indicated that even at 1:1 molar ratios, these peptides maintained full-length IAPP (1-37) in a largely random coil conformation. Negative stain electron microscopy revealed that co-incubation of these peptides with IAPP resulted in the formation of only semi-fibrous aggregates and loss of the typical high density and morphology of IAPP fibrils. This inhibitory activity, particularly for the SNNFGA sequence, also correlated with a reduction in IAPP-induced cytotoxicity as determined by cell culture studies. In contrast, the peptide NFGAIL (residues 22-27) enhanced IAPP fibril formation. Conversion to the amyloidogenic beta-sheet was immediate and the accompanying fibrils were more dense and complex than IAPP alone. The remaining peptide fragments either had no detectable effects or were only weakly inhibitory. Specificity of peptide activity was illustrated by the fragments, SSNNFG and AILSST. These differed from the most active inhibitors by only a single amino acid residue but delayed the random-to-beta conformational change only when used at higher molar ratios. This study has identified internal IAPP peptide fragments which can regulate fibrillogenesis and may be of therapeutic use for the treatment of type-2 diabetes.     

Effect of Proline Mutations on the Monomer Conformations of Amylin

The formation of human islet amyloid polypeptide (hIAPP) is implicated in the loss of pancreatic β-cells in type II diabetes. Rat amylin, which differs from human amylin at six residues, does not lead to formation of amyloid fibrils. Pramlintide is a synthetic analog of human amylin that shares three proline substitutions with rat amylin. Pramlintide has a much smaller propensity to form amyloid aggregates and has been widely prescribed in amylin replacement treatment. It is known that the three prolines attenuate β-sheet formation. However, the detailed effects of these proline substitutions on full-length hIAPP remain poorly understood. In this work, we use molecular simulations and bias-exchange metadynamics to investigate the effect of proline substitutions on the conformation of the hIAPP monomer. Our results demonstrate that hIAPP can adopt various β-sheet conformations, some of which have been reported in experiments. The proline substitutions perturb the formation of long β-sheets and reduce their stability. More importantly, we find that all three proline substitutions of pramlintide are required to inhibit β conformations and stabilize the α-helical conformation. Fewer substitutions do not have a significant inhibiting effect.     

Identification and characterization of a novel molecular-recognition and self-assembly domain within the islet amyloid polypeptide

The islet amyloid polypeptide (hIAPP) is a 37 amino acid residue polypeptide that was found to accumulate as amyloid fibrils in the pancreas of individuals with type II diabetes. Previous studies identified various fragments of hIAPP that can form amyloid fibrils in vitro (e.g. hIAPP(8-20), hIAPP(23-27), and hIAPP(30-37)). However, no comparative and systematic information was available on the role of these structural domains (or others) in the process of molecular recognition that mediates fibrillization, in the context of the full-length polypeptide. To systematically map and compare potential recognition domains, we studied the ability of hIAPP to interact with an array of 28 membrane-spotted overlapping peptides that span the entire sequence of hIAPP (i.e. hIAPP(1-10), hIAPP(2-11...), hIAPP(28-37)). Our study clearly identified a major domain of molecular recognition within hIAPP, as the polypeptide was found to bind with high affinity to a defined linear group of peptides ranging from hIAPP(7-16) to hIAPP(12-21). The maximal binding of the full-length polypeptide was to the hIAPP(11-20) peptide fragment (with the sequence RLANFLVHSS). In order to define the minimal fragment, within this apparent recognition motif, that is capable of self-association and thus may serve as the core molecular recognition motif, we examined the ability of truncated analogs of the recognition sequence to self-assemble into amyloid fibrils. The shortest active fragments capable of self-assembly were found to be the pentapeptides FLVHS and NFLVH. The apparent role of this motif in the process of hIAPP self-assembly is consistent with the profile of the hIAAP-binding distribution to the peptide array. The identification of such short recognition motifs is extremely useful in the attempts to develop means to block amyloid fibril formation by hIAPP. It is worth mentioning that this is only the second time in which peptides as short as a pentapeptide were shown to form amyloid fibrils (the other pentapeptide is FGAIL).     

Effects of sequential proline substitutions on amyloid formation by human amylin20-29

Amylin, also known as islet amyloid polypeptide (IAPP), is the major protein component of the fibril deposits found in the pancreas of individuals with type II diabetes. The central region of amylin, residues 20-29, has been implicated as a key determinate of amyloid formation. To establish which positions are most important for amyloid formation, the wild-type sequence of the 20-29 fragment and a set of 10 variants have been synthesized in which a proline was placed at each position. Proline is energetically unfavorable in the extended cross-beta structure found in amyloid. If a particular position is critical for amyloid formation, then substitution with a proline should inhibit amyloid formation. A proline substitution at any position inhibited aggregation and amyloid formation. Substitution of Asn22, Gly24, and residues 26-28 had the largest effect. Fourier transform infrared (FTIR) spectroscopy showed little secondary structure in these peptides, and transmission electron microscopy (TEM) showed mostly amorphous material. The peptides were much more soluble than the wild-type sequence, and no birefringence was observed with Congo Red staining. Proline substitutions at the N (residues 20 and 21) and C termini showed the least effect. These peptides showed the classic fibril morphology, a significant amount of beta-sheet structure, and exhibited green birefringence when stained with Congo Red. The results indicate that residues 22, 24, and 26-28 play a key role in formation of amyloid by amylin. Positions 23 and 25 also appear to be important, but may be less critical than positions 22, 24, and 26-28.     

Effect of Proline Mutations on the Monomer Conformations of Amylin

The formation of human islet amyloid polypeptide (hIAPP) is implicated in the loss of pancreatic β-cells in type II diabetes. Rat amylin, which differs from human amylin at six residues, does not lead to formation of amyloid fibrils. Pramlintide is a synthetic analog of human amylin that shares three proline substitutions with rat amylin. Pramlintide has a much smaller propensity to form amyloid aggregates and has been widely prescribed in amylin replacement treatment. It is known that the three prolines attenuate β-sheet formation. However, the detailed effects of these proline substitutions on full-length hIAPP remain poorly understood. In this work, we use molecular simulations and bias-exchange metadynamics to investigate the effect of proline substitutions on the conformation of the hIAPP monomer. Our results demonstrate that hIAPP can adopt various β-sheet conformations, some of which have been reported in experiments. The proline substitutions perturb the formation of long β-sheets and reduce their stability. More importantly, we find that all three proline substitutions of pramlintide are required to inhibit β conformations and stabilize the α-helical conformation. Fewer substitutions do not have a significant inhibiting effect.     

Identification of key residues involved in fibril formation by the conserved N-terminal region of Plasmodium falciparum merozoite surface protein 2 (MSP2)

Merozoite surface protein 2 (MSP2) from the human malaria parasite Plasmodium falciparum is expressed as a GPI-anchored protein on the merozoite surface. MSP2 is assumed to have a role in erythrocyte invasion and is a leading vaccine candidate. Recombinant MSP2 forms amyloid-like fibrils upon storage, as do peptides corresponding to sequences in the conserved N-terminal region, which constitutes the structural core of fibrils formed by full-length MSP2. We have investigated the roles of individual residues in fibril formation and local ordered structure in two peptides, a recombinant 25-residue peptide corresponding to the entire N-terminal domain of mature MSP2 and an 8-residue peptide from the central region of this domain (residues 8–15). Both peptides formed fibrils that were similar to amyloid-like fibrils formed by full-length MSP2. Phe11 and Ile12 have important roles both in stabilising local structure in these peptides and promoting fibril formation; the F11A and I12A mutants of MSP2_8–15 were essentially unstructured in solution and fibril formation at pH 7.4 and 4.7 was markedly retarded. The T10A mutant showed intermediate behaviour, having a less well defined structure than wild-type and slower fibril formation at pH 7.4. The mutation of Phe11 and Ile12 in MSP2_1–25 significantly retarded but did not abolish fibril formation, indicating that these residues also play a key role in fibril formation by the entire N-terminal conserved region. These mutations had little effect on the aggregation of full-length MSP2, however, suggesting that regions outside the conserved N-terminus have unanticipated importance for fibril formation in the full-length protein.     

Role of amino acid hydrophobicity, aromaticity, and molecular volume on IAPP(20-29) amyloid self-assembly

Aromatic amino acids strongly promote cross-β amyloid formation; whether the amyloidogenicity of aromatic residues is due to high hydrophobicity and β-sheet propensity or formation of stabilizing π-π interactions has been debated. To clarify the role of aromatic residues on amyloid formation, the islet amyloid polypeptide 20-29 fragment [IAPP(20-29)], which contains a single aromatic residue (Phe 23), was adopted as a model. The side chain of residue 23 does not self-associate in cross-β fibrils of IAPP(20-29) (Nielsen et al., Angew Chem Int Ed 2009;48:2118-2121), allowing investigation of the amyloidogenicity of aromatic amino acids in a context where direct π-π interactions do not occur. We prepared variants of IAPP(20-29) in which Tyr, Leu, Phe, pentafluorophenylalanine (F5-Phe), Trp, cyclohexylalanine (Cha), α-naphthylalanine (1-Nap), or β-naphthylalanine (2-Nap) (in order of increasing peptide hydrophobicity) were incorporated at position 23 (SNNXGAILSS-NH2), and the kinetic and thermodynamic effects of these mutations on cross-β self-assembly were assessed. The Tyr, Leu, and Trp 23 variants failed to readily self-assemble at concentrations up to 1.5 mM, while the Cha 23 mutant fibrillized with attenuated kinetics and similar thermodynamic stability relative to the wild-type Phe 23 peptide. Conversely, the F5-Phe, 1-Nap, and 2-Nap 23 variants self-assembled at enhanced rates, forming fibrils with greater thermodynamic stability than the wild-type peptide. These results indicate that the high amyloidogenicity of aromatic amino acids is a function of hydrophobicity, β-sheet propensity, and planar geometry and not the ability to form stabilizing or directing π-π bonds.