Native human islet amyloid polypeptide (hIAPP) has been identified as the major component of amyloid plaques found in the pancreatic islets of Langerhans of persons affected by type 2 diabetes mellitus. Early studies of hIAPP determined that a segment of the molecule, amino acids 20–29, is responsible for its aggregation into amyloid fibrils. The present study demonstrates that the aggregation of hIAPP 20–29-Trp is a nucleation-dependent process, displaying a distinct lag time before the onset of rapid aggregation. Moreover, the lag time can be eliminated by seeding the sample of unaggregated peptide with preformed fibrils. In contrast to the expectation from the conventional model of nucleation-dependent aggregation, however, the lag time of hIAPP aggregation does not depend on peptide concentration. To explain this observation, a modified version of the standard model of nucleation-dependent aggregation is presented in which the monomeric peptide concentration is buffered by an off-aggregation-pathway formation of peptide micelles.
Native human islet amyloid polypeptide (hIAPP) is the primary component of the extracellular amyloid deposits found in the pancreas of the majority of patients with type 2 diabetes mellitus (type 2 DM) [1]. hIAPP is a 37-amino-acid peptide hormone that is normally produced in the β-cells of the islets of Langerhans in the pancreas. hIAPP is co-secreted from these cells with insulin, and functions to control hyperglycemia by restraining the rate at which dietary glucose enters the bloodstream. Though the causal relationship between the development of hIAPP amyloid and the appearance of type 2 DM is unresolved, it has been documented that sites of hIAPP amyloid deposition in the pancreas are surrounded by areas of β-cell degeneration, and hIAPP fibrils have been shown to be toxic to both human and rat islet β-cells in culture [2].
hIAPP is one of a growing number of proteins which are characterized by their abnormal deposition as highly ordered fibrillar aggregates in the tissues of individuals diagnosed with one of a variety of diseases. Some other amyloidogenic proteins and their associated diseases include the A-β peptide in Alzheimer’s disease; transthyretin in senile systemic amyloidosis; and the prion protein in various spongiform encephalopathies [3]. These proteins share little or no structural or sequence homology, yet they aggregate to form fibers which are remarkably similar in morphology and which are all characterized by intermolecular anti parallel β-sheets, oriented roughly perpendicular to the axis of the fibers, as seen with circular dichroism, nuclear magnetic resonance, and X-ray diffraction [4], [5], [6], [7], [8]. Elucidation of the relationship between various amyloid deposits and their associated disease states provides a compelling reason to study and characterize the mechanisms of amyloid formation.
Early studies comparing IAPP sequences from mammals that develop type 2 DM (humans and cats) and from those that do not develop type 2 DM (rats, hamsters, and mice) revealed important sequence differences in the peptide region spanning amino acids 20–29, as depicted in Fig. 1 [9]. Additional studies showed that this region of hIAPP is involved in the aggregation of the peptide into fibrils, and that hIAPP fragment 20–29 is capable of forming amyloid fibrils in vitro [10]. In contrast, the native sequence of rat IAPP (rIAPP) does not form amyloid in vivo and rIAPP 20–29 does not form amyloid fibrils in vitro [10].
In the experiments presented below, hIAPP 20–29-Trp (the Trp residue at the carboxy end was introduced to confer fluorescence, though this property did not serve in the present study) and rIAPP 20–29-Trp were used as models of the behavior of amyloidogenic peptides and their aggregation behavior was determined in vitro. Previous work done with fragments of both A-β and hIAPP has shown that short peptides exhibit many of the amyloidogenic characteristics of the full-length peptides and may serve as suitable model systems for in vitro amyloid formation [11], [12], [13], [14], [15], [16], [17], [18]. Using rIAPP 20–29-Trp as a control, we monitored the aggregation of hIAPP 20–29-Trp under a variety of experimental conditions. Based on the results of these experiments we propose a model for the aggregation of the hIAPP fragment which assumes a nucleation-dependent ordered aggregation of peptide molecules whose monomeric concentration in solution is fixed by the formation of peptide micellar structures. The formation of off-pathway micelles is necessary to account for our unexpected observation that hIAPP aggregation behavior is concentration independent over a wide range of peptide concentrations.
Check access to the full text by signing in through your organization.
The fragments hIAPP 20–29-Trp and rIAPP 20–29-Trp were synthesized by the Protein and Carbohydrate Structure Facility (University of Michigan, Ann Arbor), and were purified by reversed-phase high-performance liquid chromatography to a purity of >98%. The tryptophan (Trp) residue was added to the carboxyl terminus for fluorescence studies not presented in this study. Investigations by our laboratory showed that addition of the Trp residue had no effect on the peptide’s aggregation behavior, and
When solutions of hIAPP 20–29-Trp were incubated under the conditions described above, they aggregated to form amyloid as verified by the appearance of both the typical absorbance shifts and birefringence upon staining of the samples with the dye Congo red [19], [20], [21]. Fig. 2 presents the time-dependent aggregation profiles of 20 μM solutions of hIAPP 20–29-Trp and of the corresponding fragment of the rat peptide, both at pH 7, as monitored through the increase in turbidity. Several
Amyloid is a specific form of protein aggregate characterized by a typical fibrillar morphology which is created by structure-specific, ordered, molecular interactions [24], [25], [26], [27]. Other well known ordered association phenomena include protein crystallization [28], [29], viral coat protein assembly [30], sickle-cell hemoglobin polymerization [31], and microtubule formation [32]. Amyloid formation has been shown to occur via a nucleation-dependent aggregation process, as has been
This research was supported by NIA Grant #T32 AG 00114 and by the Michigan Diabetes Research and Training Center.
[1] Present address: Case Western Reserve University, Cleveland, OH, USA. Copyright © 2000 Elsevier Science B.V. All rights reserved.
