Search tips
Search criteria 


Logo of scirepAboutEditorial BoardFor AuthorsScientific Reports
Sci Rep. 2017; 7: 44041.
Published online 2017 March 13. doi:  10.1038/srep44041
PMCID: PMC5347123

The redox environment triggers conformational changes and aggregation of hIAPP in Type II Diabetes


Type II diabetes (T2D) is characterized by diminished insulin production and resistance of cells to insulin. Among others, endoplasmic reticulum (ER) stress is a principal factor contributing to T2D and induces a shift towards a more reducing cellular environment. At the same time, peripheral insulin resistance triggers the over-production of regulatory hormones such as insulin and human islet amyloid polypeptide (hIAPP). We show that the differential aggregation of reduced and oxidized hIAPP assists to maintain the redox equilibrium by restoring redox equivalents. Aggregation thus induces redox balancing which can assist initially to counteract ER stress. Failure of the protein degradation machinery might finally result in β-cell disruption and cell death. We further present a structural characterization of hIAPP in solution, demonstrating that the N-terminus of the oxidized peptide has a high propensity to form an α-helical structure which is lacking in the reduced state of hIAPP. In healthy cells, this residual structure prevents the conversion into amyloidogenic aggregates.

Type II diabetes (T2D) is the most common form of diabetes. The disease is caused by a combination of genetic and environmental factors, and results in insulin resistance and β-cell failure1. T2D is also linked to the tissue inflammation response2. The exact patho-mechanism of the disease is, however, not understood so far. In addition to insulin resistance, lipotoxicity, endoplasmic reticulum (ER) and oxidative stress, as well as amyloid deposition have been described as secondary effects of the disease2. The amyloid plaques formed within the β-cells of the pancreatic islets of Langerhans3 are composed of the 37 amino acid polypeptide human Islet Amyloid Polypeptide (hIAPP). In recent studies it has been shown that soluble oligomers of this hormone are responsible for cell toxicity4,5. Yet until now, it is not understood how hIAPP changes from a functional into a pathophysiological form. In vivo, hIAPP is soluble at concentrations of 1–4 mM6,7,8, while in vitro hIAPP aggregates at 1000 times lower concentrations. Protein aggregation is influenced by concentration and pH9. In T2D, the redox state of the ER is shifted towards more reducing conditions, resulting in protein secretion and folding deficiencies10,11,12. So far, it is unclear how this change of the redox environment affects hIAPP aggregation. We show here that the redox conditions play an important role in the aggregation of hIAPP. Furthermore, we present the first detailed structural analysis of monomeric hIAPP in aqueous buffer. We find that the disulfide bridge in hIAPPox stabilizes an α-helical structure at the N-terminus of the peptide, and thus protects the peptide from aggregation.


Assembly Kinetics of hIAPP

To better understand the influence of the redox environment on the aggregation kinetics of hIAPP, we first performed Thioflavin T (ThT) assays. In a cell, the ratio of reduced glutathione (GSH) to oxidized glutathione (GSSG) is often employed as a measure of cellular oxidative stress. In order to induce a particular oxidative state of hIAPP, we added the redox couple GSH/GSSG to the peptide solution (Fig. 1A,B). A sigmoidal fit of the ThT results yields the time at which the half-maximum fluorescence intensity has been obtained (T50). We found that the aggregation kinetics of hIAPP are affected by the oxidative environment. hIAPPred aggregates faster in comparison to hIAPPox. However, mixtures of hIAPPred and hIAPPox yield the same rate of aggregation as hIAPPred. Within the experimental error, we found the same T50 values for redox pairs that yield (100:0)%, (80:20)% or (50:50)% reduced and oxidized hIAPP. This suggests that the aggregation of hIAPP is controlled by the reduced form. The ThT results are supported by NMR experiments which show that hIAPPred has a 3-times faster aggregation kinetics in comparison to hIAPPox (Fig. 1C). To validate the biological significance of the differential aggregation behaviour of hIAPPox and hIAPPred, we determined the redox potential for the formation of the disulfide bond in hIAPP using NMR spectroscopy. 1H,15N HSQC spectra were recorded after incubating the protein with varying ratios of reduced and oxidized glutathione. A plot of the average peak intensities as a function of the [GSH]2/[GSSG] ratio is shown in Fig. 1D. As a secondary abscissa, the calculated redox potential is represented. Using the Nernst equation with the respective correction for temperature and pH13, we calculated a redox potential of (−129 ± 4) mV for hIAPP. The ER has an electrochemical potential that varies between −231 to −160 mV, and is dependent on the employed method14. Under these conditions, less than 30% of the peptide exists in the oxidized state. This theoretical redox state assumes quasi-equilibration with the redox couple GSH/GSSG.

Figure 1
(A) Thioflavin T (ThT) assay using a solution of 35 μM of hIAPP, pH 5.3 and 30 °C. Experiments were performed by adding defined amounts of GSH and GSSG to yield varying percentages of reduced and oxidized hIAPP in solution. ...

To analyze the hIAPP oxidation state in vivo, we used a transgenic mouse model (FVB/N-Tg(Ins2-IAPP)RHFSoel/J) in which the human IAPP sequence has been incorporated into the mouse genome on chromosome 15 ( We found that the morphology of the islets is dramatically changed for transgenic mice that harbor two human gene copies (TG/TG) (Fig. 1F, left) compared to healthy controls. We employed a fibril specific antibody to study whether amyloid fibrils participate in this process. We found amyloid fibrils in insulin producing β-cells in TG/TG animals (Fig. 1F). These were not present in healthy animals. These results suggest that amyloid fibrils in TG/TG mice potentially induce impaired islet morphology. To quantify the amount of reduced or oxidized hIAPP in vivo, we added N-ethylmaleimide (NEM) to TG/TG and +/+ mouse islets to yield an alkylation of free sulfhydryl groups. NEM has the ability to cross cellular membranes and thus does not require cell lysis. Modified hIAPPred-NEM yielded an increased molecular weight which could be detected by SDS-PAGE. Soluble oligomeric hIAPP could be identified using the hIAPP specific antibody A13315. In TG/TG islets, both reduced and oxidized hIAPP were found (Fig. 1E, right). As expected, no aggregates were found in the islets of the+/+control mice (Fig. 1E, right). As a reference, an aggregated sample of recombinant, oxidized hIAPP was investigated as well (Fig. 1E, left). We observed that samples that were taken at a later time point yielded a more disperse distribution of oligomeric states. An incubation time of 72 h for an oxidized hIAPP sample produced a gel band at approximately 60 kDa, which corresponds in molecular weight to the band of the hIAPP sample extracted from the diabetes mouse model islets (Fig. 1E, right, last sample).

In order to probe if the differential aggregation behavior results in a continuous conversion of hIAPPox into hIAPPred at defined redox conditions, electron microscopic (EM) images of hIAPP aggregates that are obtained in buffers with different amounts of GSH and GSSG were recorded (Fig. 2A–C). Under conditions, in which the redox equilibrium is shifted to produce a defined amount of hIAPPox (50:50 = hIAPPred:hIAPPox) (Fig. 2C), the fibril morphology resembles the morphology obtained for the pure hIAPPred (Fig. 2B). By contrast, pure hIAPPox (Fig. 2A) showed a different fibril morphology. To confirm this observation, we carried out solution-state HSQC NMR experiments of hIAPP aggregates that were dissolved in DMSO prior to the experiment. These experiments allow to probe the hIAPP redox state of the hIAPP aggregates. Again, we found that the hIAPP spectra of the samples in which the redox properties in the buffer has been adjusted to yield 100% hIAPPred or a 1:1 ratio of hIAPPox and hIAPPred are rather similar, whereas the sample containing pure hIAPPox looked significantly different (Fig. 2E). This indicates that aggregation in fact shifts the equilibrium to the reduced state of the peptide. Based on these results, we suggest a model for hIAPP aggregation in the cell that is dependent on its particular redox environment (Fig. 2F). External factors such as oxidative stress or ER stress, hyperglycemia, lipotoxity, or genetic factors have an impact on the ER redox state. A small shift of the redox conditions can result in generation of a small amount of hIAPPred. hIAPPred preferentially aggregates and is depleted from the pool of hIAPP molecules in solution. As the redox state in the cell is buffered, more hIAPPox is transformed to its reduced form which again has a higher propensity to form fibrils. Eventually, all available hIAPP is converted into its reduced state which is finally deposited in insoluble aggregates.

Figure 2
(A–C) EM images of fibrils formed in the presence of different ratios of the GSH-GSSG redox couple. In (A) hIAPP was incubated using 2 mM GSSG to produce a completely oxidized hIAPP. (B) hIAPP was incubated using 35 mM GSH yielding ...

Structural analysis of hIAPP

To study the mechanism of hIAPP aggregation, the conformation of the peptide was determined under physiological conditions. Recombinant C-terminal amidated hIAPP(1-37) was produced using the protocol published recently17. hIAPP has a strong tendency to aggregate in aqueous buffer. However, low pH (pH 5.3), temperature (4 °C) and concentration (100 μM) reduced the rate of fibril formation. Under these conditions, hIAPP(1-37) yielded a CD spectrum which contained contributions from α-helix and random coil secondary structure (Fig. 3A). Comparison of the oxidized and the reduced form of the peptide suggests that a loss of the disulfide bond induces a loss of the α-helical structure. To support this observation, we recorded solution-state NMR experiments. 1H,15N HSQC spectra of reduced and oxidized hIAPP yielded significant chemical shift differences, in particular for Cys-2 and Cys-7 as well as for residues in the vicinity of the disulfide bond (Fig. 2D). Chemical shift assignments have been obtained from 3D HNCACB experiments18. The analysis of 13 Cα/β secondary chemical shifts reveals a strong helical propensity for the N-terminal half of the oxidized hIAPP (residues 5-17), and a random coil conformation for the C-terminal half of the peptide (Fig. 3E). The same analysis for the reduced form of hIAPP showed a smaller helical propensity for the N-terminal half, suggesting that the N-terminal disulfide bond stabilizes the helical conformation. The observation that hIAPPox has a diminished tendency to aggregate stimulated a more detailed structural characterization of the peptide. For the structural analysis, 3D 13C/15N edited NOESY-HSQC data were collected for hIAPPox. The faster aggregation kinetics of hIAPPred prohibited a detailed structural characterization. The ensemble of the 20 lowest energy structures obtained after water refinement is shown in Fig. 3B,C. A summary of the structural and restraint statistics is given in Table 1. We note that no hydrogen bonding restraints were included in the structure calculations to enforce an α-helical conformation. The present solution structure of native hIAPP (oxidized and amidated) shares similarities, but also has differences when compared to other structures of amylin reported under various conditions19. A common theme in all structures is the N-terminal loop closed by the disulfide bond followed by a right-handed α-helix. In our study, the oxidation state of the peptide was confirmed by the 13Cβ secondary chemical shifts of the two cysteines (Cys-2 and Cys-7). In our analysis, the α-helix spans residues 8–17. In particular, the helical structure was confirmed by the characteristic αN(i,i+3), αN(i,i+4) and αβ(i,i+3) NOEs (Fig. 3D,G,H). We found that the histidine at position 18 is incompatible with an α-helical conformation. His-18 thus introduces a kink in line with earlier findings (Fig. 3C)20. For the C-terminal part, the NOE pattern of the 15N-edited 3D NOESY experiment was dominated by strong correlations between the respective amides and the (i−1) Hα atoms which is typical for IDPs (Intrinsically Disordered Proteins). In addition, no long-range NOEs were detected for the C-terminal amide suggesting that cross-talk to distant residues is not occuring. The structural model is supported by {1H},15N heteronuclear NOE data that showed a certain rigidity for the N-terminal half and increased flexibility towards the C-terminus of hIAPP (Fig. 3F).

Figure 3
(A) Circular dichroism (CD) spectra of reduced (red) and oxidized hIAPP (black). hIAPPox yields a negative band at 222 nm and 208 nm which is characteristic for an α-helical conformation. The reduced form shows a considerably smaller ...
Table 1
Structural statistics of hIAPP (residues 1-37).


Stabilization of the monomeric state of a protein can prevent its conversion into an aggregated state22. This paradigm is generally accepted for folded proteins, but has so far not been recognized for disordered peptides and proteins. We show here that the oxidized form of the islet hormone hIAPP is more structured in comparison to the reduced form of the peptide, and at the same time has a smaller propensity to aggregate. We find furthermore that the pancreatic islets of a T2D mouse model contain primarily aggregated hIAPP in its reduced state. We suggest that differential aggregation unleashes redox equivalents that potentially counteract oxidative stress.

Initially, it was suggested that hIAPP adopts a random coil structure23. However, more recently reports indicate that the N-terminal part of hIAPP involving residues 5–19 and to a somewhat smaller degree residues 20–22 transiently samples an α-helical structure in solution8,24. The same residues adopt a fully helical structure when bound to membranes or micelles19,24,25. In agreement with previous studies, we found that reduction of the N-terminal disulfide bond (Cys2–Cys7) in solution decreases the extent of helix formation in this region26. Miranker and coworkers observed that formation of helical structure is directly correlated with enhanced amyloid formation on a membrane surface. It was suggested that a parallel association of helices results in locally increased peptide concentration that could nucleate β-strand structure. By contrast, we find that the reduced form of hIAPP aggregates significantly faster than the oxidized peptide. Our data can be reconciled with the observation made by Miranker and coworkers by assuming that membranes can favor an α-helical structure e.g. by formation of an amphipatic helix. At the same time, an increase of the local protein density can be a very strong driving force for aggregation. In aqueous buffer, the situation is fundamentally different. The α-helical propensity at the N-terminus of hIAPP impedes conversion into β-sheet structure. Solid-state NMR studies showed that hIAPP adopts a strand–turn–strand structure in amyloid fibrils, with the N-terminal and C-terminal strands involving residues 8–17 and residues 28–37, respectively27,28. The residues in-between the disulfide bridge do not contribute to the amyloid fiber core structure. At the same time, the disulfide bridge plays a central role in the fibril assembly mechanism29: Loss of the disulfide bond substantially reduces fiber formation by secondary nucleation, and eliminates the activation step in seeded fiber growth30.

Obviously, the expression level of hIAPP has a strong impact on solubility and aggregation propensity of the peptide. Insulin and hIAPP tend to be regulated in parallel. Hyperglycemia, however, has been shown to result in disproportionate upregulation of hIAPP versus insulin, thus altering the ratio of hIAPP to insulin secreted by the β-cells31. Increased secretion of hIAPP in severe hyperglycemia has been suggested to facilitate the formation of hIAPP-derived islet amyloid deposits31. At the same time, disturbances in redox regulation, calcium homeostasis, glucose deprivation, and viral infection can lead to ER stress32. Under these conditions, the folding of proteins is slowed down, leading to an increase of unfolded proteins. Unfolded proteins cause a stress response in the ER which is known as unfolded protein response (UPR)33. ER stress is emerging as a potential cause of damage in hypoxia/ischemia, insulin resistance, and other disorders34. In T2D, pancreatic β-cells undergo cell death as a consequence of sustained ER stress35,36. We find that reduced hIAPP aggregates 3-times faster in comparison to the oxidized form of the peptide. This differential aggregation results in a continuous conversion of hIAPPox into hIAPPred in the ER. Thus, oxidation equivalents are continuously restored in the ER, a mechanism that might aid to counteract ER stress. Given the fact that hIAPP occurs at concentrations up to 4 mM in the cell37,38, hIAPP might act as a buffering system. This process works as long as the resulting aggregates are efficiently removed. Clearance can be achieved via autophagy and degradation in the lysosome. It has been shown that β-cell apoptosis is in fact correlated with autophagy and lysosomal degradation39,40,41. Inhibition of autophagy induces apoptosis, whereas stimulation of autophagy protects β-cells from hIAPP-induced toxicity. Alternatively, the endoplasmic reticulum–associated degradation (ERAD) assists to protect cells from ER stress42. It has been reported that accumulation of polyubiquitinated proteins occurs in pancreatic islets of hIAPP transgenic mice but not in mice with comparable transgenic expression of rodent-IAPP (r-IAPP). Proteasome dysfunction can mediate obesity induced ER stress and insulin resistance43. In fact, it has been found that the β-cell ERAD/ubiquitin/proteasome system of obese individuals with high BMI, that are otherwise healthy and not suffering from T2D, adapts to the increased synthetic burden of insulin and IAPP44. p62, an important factor in the autophagy/lysosomal pathway, is downregulated by overproduction of hIAPP in diabetic β-cells, resulting in induction of apoptosis39,45.

By contrast, the percentage of β-cells with ubiquitin-positive immunostaining was significantly increased in obese individuals with T2D. This may suggest that misregulation of proteasomal degradation yields an accumulation of aggregates which may eventually lead to perturbations in redox balance. In vivo, expression of other factors such as BiP, HSP70 and DnaJ which are known to interact with hIAPP can modulate the aggregation behaviour46. Similar to the small heat shock proteins such as αB-crystallin that interacts sub-stoichiometrically with Alzheimer’s disease Aβ47, HSP70 and GRP78/BiP can detect and bind misfolded hIAPP oligomers. Chaperones thus would have an influence ensuring proper redox balance.

Recently, it has been shown that the redox status of the microsomal protein folding machinery is affected in diabetes48. Streptozotocin-induced diabetes in rats results in a significant increase in the content of free protein sulfhydryl groups, as well as in the amount of free sulfhydryl groups of the major protein disulfide isomerases (PDIs), in particular the 58 kDa PDI and the 57 kDa ERp57. In this chemically induced diabetes model, hIAPP is overexpressed relative to insulin49. Our data revealed the redox potential of hIAPP to be on the order of (−129 ± 4) mV suggesting that a significant amount of hIAPP in the ER actually exists in the reduced state. A shift of the redox conditions under ER stress to a more reducing state, and an increase in the amount of produced hIAPP in T2D would immediately result in enhanced hIAPP aggregation. As a consequence of aggregation, even more redox equivalents would become available to restore the oxidizing redox environment of the ER.

Taken together, we have shows that aggregation of hIAPP can have beneficial effects on the well-being of the cell by restoring redox equivalents in the due course of aggregation. Only failure of the preoteasome degradation machinery results in β-cell disruption. It remains to be seen which molecular mechanisms in more detail induce a misfunction of proteasomal degradation in T2D.

Materials and Methods

Recombinant protein expression and purification

Human IAPP was expressed in E. coli and purified using the protocol described previously17. This protocol allows the production of an isotopically enriched peptide, which is amidated at the C-terminus and is disulfide bridged involving residues Cys-2 and Cys-7. Molecular biology reagents were obtained from Roche, New England Biolabs and from Sigma-Aldrich St. Louis, MO, USA. Isotopically labeled minimal media components were purchased from Cambridge Isotope Laboratories (CIL).

Isolation of mouse islets

hIAPP transgenic mice were purchased from Jackson Laboratory (Bar Harbor, USA) and studied between 13–18 weeks of age. Animals were killed by isoflurane overdose and islets were isolated immediately. To perfuse the pancreas, the common bile duct was clamped at the papilla of Vater and 1 mg/ml cold collagenase P solution in HBSS solution supplemented with 1% BSA was injected into the common bile duct. The pancreas was removed and incubated for 15 min at 37 °C. Digestion was stopped by addition of a cold HBSS solution supplemented with 1% BSA and centrifuged at 290 g for 2 min at 4 °C. The pellet was resuspended in the HBSS solution supplemented with 1% BSA, filtered through a metal mesh and centrifuged at 290 g for 2 min at 4 °C. The pellet was resuspended in 15% Optiprep supplemented with 3.75 mM HEPES, 12% PBS, 6.2% RPMI and layered onto 15% Optiprep supplemented with 3.75 mM HEPES, 12% PBS, 6.2% RPMI solution. The pancreatic lysate was overlaid with the HBSS solution supplemented with 1% BSA, incubated for 10 min and centrifuged at 360 g for 18 min at 4 °C. After centrifugation, islets were observed in the interface between the first and second layers and after collecting islet fraction, cell debris were filtered through a 70 μm cell strainer and islets were washed from the filter with RPMI supplemented with 11 mM glucose, 10% FBS, antibiotic-antimycotic solution. To eliminate remaining exocrine tissue, the islets were handpicked and transferred into a new culture dish.

Animals were declared to the responsible local authority (Regierung von Oberbayern). The Institutional Animal Welfare Officer (Helmholtz-Zentrum München) was consulted about the work and approved the study. All experiments were performed in accordance with relevant guidelines and regulations.

Western Blot Analysis

Extracted islets were solubilized with 30 μL PBS buffer supplemented with NEM and vortexed. Subsequently, 10 μL of SDS sample buffer without reducing agent was added. Samples were boiled for 5 min at 90 °C and separated by 12% Mini-PROTEAN-TGX stain-FreeTM protein gels, and then transferred onto a Tnas-Blot-TurboTM Mini PVDF membrane with a Bio-Rad Trans-Blot TurboTM transfer system. Membrane was blocked in Tween-Tris-buffered saline with 5% (w/v) nonfat dry milk (Roth) for 1 hour and then incubated with the primary antibody A133 in Tween-Tris-buffered saline with a ratio 1:1000 overnight at 4 °C. Reactivity was analyzed with antirabbit peroxidase-conjugated secondary antibody (Serva) and chemiluminescence detection (Bio-Rad).

hIAPP fibril antibody production

To generate a monoclonal antibody against hIAPP amyloid fibrils, synthetic human lyophilized HFIP-treated hIAPP was dissolved in 20 mM Tris buffer (pH 7.4) to a concentration of 30 μM. The solution was sonicated for 1 min and incubated for 2 d at 37 °C with shaking at 300 rpm. The resulting fibrils were used for antibody generation according to a proprietary immunization protocol of Synaptic Systems Göttingen (see also Three 8–10-week-old BALB/c female mice were subcutaneously immunized with IAPP fibrils over a period of 17 days. Cells from the knee lymph nodes were fused with the mouse myeloma cell line P3 × 63Ag8.653 (ATCC CRL-1580). The clones used in this study were re-cloned two times by limiting dilution and the immunoglobulin subclass was determined. Antibodies secreted by the hybridomas were screened for their reactivity against the immunogen using ELISA. Positive antibodies were retested by ELISA against hIAPP fibrils and monomer. Hybridomas producing antibodies with preference for IAPP fibrils were subcloned to monoclonality and further analyzed for histology and western blot applications. One of the antibodies showing high specificty for hIAPP fibrils was the antibody 91D7E8.

Immunofluorescence staining of pancreatic β-cells

Pancreata were fixed in 4% paraformaldehyde and processed as published previously50. Briefly, cryo-sections were stained with anti-insulin (Invitrogen, Life Technologies, Germany), or with anti-amyloid fibril antibody 91D7E8 employing corresponding fluorescent-labeled secondary antibodies (Invitrogen, Life Technologies, Germany). DAPI was used to visualize cell nuclei. Slides were scanned using a NanoZoomer 2.0 HT (Hamamatsu, Japan) fluorescence scanner.

NMR sample preparation

Lyophilized hIAPP(1-37) was dissolved into NMR buffer (30 mM deuterated acetic acid, pH 5.3) containing 10% D2O for locking. The samples were measured and stored at 4 °C. The formation of the intramolecular disulfide bond was confirmed by NMR. To prepare hIAPP in the oxidized state, a hIAPP solution was supplemented with 0.2 mM GSSG. To generate the reduced form of the peptide, the GSH powder was directly added to the hIAPP solution to yield a final concentration of 35 mM of reduced glutathione. To produce the hIAPPred:hIAPPox = 50:50 mixed sample, the hIAPP stock solution was added to a glutathione solution with final concentrations of 0.2 mM GSSG and 4.98 mM GSH. The final hIAPP concentration amounted to 100 μM in all cases. Afterwards, the samples were incubated at 4 °C to allow for complete oxidation or reduction.

For analysis of the oxidation state of the aggregated hIAPP samples at varying redox conditions, peptides were incubated first for two weeks at 37 °C to allow for fibril formation. After this period, samples were centrifuged for 30 min to harvest the fibrils. Pellets were dissolved in 250 μL of d6-DMSO, and subsequently sonicated for 1 hour. In order to prevent degradation of hIAPP, ice was added into the sonication bath to avoid sample heating. Afterwards, samples were directly transferred into Shigemi NMR tubes (Shigemi Inc., Allison Park, USA) for solution-state NMR measurements.

NMR Experiments

All NMR experiments were performed at 4 °C, employing Bruker Avance 500, 600, 750 MHz spectrometers, equipped with cryo-probes. The proton chemical shifts were referenced to the water resonance frequency, the 15N and 13C shifts were referenced indirectly. Backbone and side chain assignments were obtained using triple resonance experiments51. Side chain assignments were obtained from 3D-HCCH TOCSY experiments. Overall, an assignment completeness of 97% was obtained. NMR spectra were processed using the software TopSpin (Bruker) and NMRPIPE52. Spectra were analyzed using SPARKY53 and ccpNMR analysis54.

Structure Calculation

Sequential and inter-residual NOE cross-peaks that could be resolved in the 15N-edited spectrum were assigned manually for the N-terminal half and were kept fixed during structure calculations with CYANA21. Peak intensities for all cross-peaks were converted to distances using the internal function of CYANA. Dihedral angle restraints were derived from chemical shifts using the program TALOS55,56. Distance and angle restraints were used as input for the calculation of the structure on a linux-based cluster. All non-violated structures were recalculated using a water-refinement protocol.

Transmission electron microscopy (TEM)

Samples were stained with 1% uranyl acetate solution on Formvar/Carbon 300 mesh copper coated grids from Electron Microscopy Sciences (Hatfield, PA 19440, USA). 10 μL of the sample were placed on the grid for 1 min. The excess of the sample was dried using filter paper. After this process, the grid was washed by placing it three times on a drop of water for 3 s. For staining, the grid was placed on a drop of a uranyl acetate solution for 30 s. The excess of the solution was again dried using filter paper. Samples were measured immediately employing an EM 10 CR (Zeiss, Germany).

Determination of the hIAPP redox potential

100 μM hIAPPox was initially dissolved in 500 μL NMR buffer, containing 0.2 mM oxidized glutathione (GSSG). The GSH concentration was then increased stepwise to yield a final GSH concentration of 0.2, 1.0, 3.0, 5.0, 7.0, 10.0, 12.0, 15.0 and 18.0 mM, using a 100 mM GSH stock solution. 1H,15N-HSQC spectra were recorded after each addition of GSH. To determine the redox potential, the intensity of Cys-7 was determined in each spectrum. The intensities of the Cys-7 cross peak were plotted as a function of [GSH]2/[GSSG] on a logarithmic scale, and fit assuming a sigmoidal behavior. The [GSH]2/[GSSG] titration half-point was determined as 0.24 M. The redox potential for the GSH/GSSG redox pair was assumed to be −151 mV13. The Nernst equation yields then a redox potential for hIAPP of −129+/−4 mV.

Circular Dichroism Spectroscopy (CD)

CD experiments were carried out at room temperature employing a JASCO spectropolarimeter. Quartz cells with a path length of 0.1 cm were used. Spectra were recorded between 260 nm and 190 nm at 0.1 nm intervals, with a response time of 1 s. For the analysis of the data the background buffer was subtracted and the data were expressed as mean residue ellipticity θMRW = (deg cm2 dmol−1) × 10−6.

Thioflavin-T assay

Fluorescence experiments were performed using a Beckman fluorometer, using Quartz cells with a path length of 1 cm. The samples were freshly prepared with the dye ThT, and incubated in the cell while stirring at 37 °C. The solutions were prepared by adding NMR buffer (30 mM d-acetic acid, pH 5.3) to the dried peptide immediately prior to the measurements, holding the solution on ice all the time. A ThT stock solution was prepared in NMR buffer at a concentration of 25 mM. Final concentrations for the ThT experiment were 30 μM hIAPP and 10 μM ThT. The data were analyzed using the software Origin (OriginLab Corporation, Northampton, MA 01060, USA), and were fit assuming a sigmoidal behavior.

Additional Information

Accession codes: The atomic coordinates and restraint files of the oxidized and amidated hIAPP are deposited in the PDB with accession code 5MGQ, chemical shifts at BMRB, entry 34069.

How to cite this article: Camargo, D. C. R. et al. The redox environment triggers conformational changes and aggregation of hIAPP in Type II Diabetes. Sci. Rep. 7, 44041; doi: 10.1038/srep44041 (2017).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Material

Supplementary Information:


We are grateful to Gunilla Westermark for providing us with a sample of the antibody A133. Furthermore, we would like to thank Dr. Vanessa Morris and Ashish Kawale for stimulating discussions. This work was performed in the framework of the SFB-1035 (Project-B07, Deutsche Forschungsgemeinschaft, DFG, to B.R.). We acknowledge support of the Helmholtz-Gemeinschaft, and the DFG (Grants Re1435, to B.R.). We are grateful to the Center for Integrated Protein Science Munich (CIPS-M) for financial support. K.T. acknowledges support from the Ministry of Education, Youth and Sports of the Czech Republic under the project CEITEC 2020 (LQ1601). This work was supported by the Bavarian Ministry of Sciences, Research and the Arts (Bavarian Molecular Biosystems Research Network, to T.M.), the DFG (Emmy Noether program MA 5703/1-1 to T.M. and CO 291/5-1 to M.C.), and the Austrian Science Fund (FWF: P28854). C.G. gratefully acknowledges the postdoctoral fellowship program (PFP) of the Helmholtz-Zentrum-München. We thank the HELENA and the TUM graduate schools for support.


The authors declare no competing financial interests.

Author Contributions D.C.R.C. cloned, recombinantly expressed, purified and characterized hIAPP. D.C.R.C., K.T., R.S., and C.H. performed NMR experiments and analyzed the data. K.T., C.H. and C.G. carried out structure calculations. A.F., M.S. and M.H.d.A. contributed fluorescence and light microscopy experiments using islets from hIAPP transgenic mice. D.C.R.C., K.B. and M.C. performed the redox state analysis of in vivo material. A.B., C.E., H.M., and E.E.W. produced the hIAPP fibril selective antibody. D.C.R.C., M.A., G.M. and A.K.W. carried out EM experiments. D.C.R.C., K.T., T.M., M.C., E.E.W., M.H.d.A., and B.R. designed the experiments. D.C.R.C. and B.R. wrote the manuscript with contributions from all authors.


  • Amos A. F., McCarty D. J. & Zimmet P. The rising global burden of diabetes and its complications: estimates and projections to the year 2010. Diabetic Med. 14 Suppl 5, S1–S85 (1997). [PubMed]
  • Donath M. Y. & Shoelson S. E. Type 2 diabetes as an inflammatory disease. Nature Rev. Immunol. 11, 98–107 (2011). [PubMed]
  • Paulsson J. F., Andersson A., Westermark P. & Westermark G. T. 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 (2006). [PubMed]
  • Cooper G. J. S. et al. . Purification and Characterization of a Peptide from Amyloid-rich Pancreases of Type-2-Diabetic Patients. Proc. Natl Acad. Sci. USA 84, 8628–8632 (1987). [PubMed]
  • Bram Y. et al. . Apoptosis induced by islet amyloid polypeptide soluble oligomers is neutralized by diabetes-associated specific antibodies. Sci. Rep. 4 (2014). [PMC free article] [PubMed]
  • Hutton J. C. The Insulin Secretory Granule. Diabetologia 32, 271–281 (1989). [PubMed]
  • Westermark P., Li Z. C., Westermark G. T., Leckstrom A. & Steiner D. F. Effects of beta cell granule components on human islet amyloid polypeptide fibril formation. FEBS Lett. 379, 203–206 (1996). [PubMed]
  • Wiltzius J. J. W., Sievers S. A., Sawaya M. R. & Eisenberg D. Atomic structures of IAPP (amylin) fusions suggest a mechanism for fibrillation and the role of insulin in the process. Prot. Sci. 18, 1521–1530 (2009). [PubMed]
  • Jha S. et al. . pH Dependence of Amylin Fibrillization. Biochemistry 53, 300–310 (2014). [PubMed]
  • Braakman I., Helenius J. & Helenius A. Manipulating Disulphide Bond Formation and Protein Folding in the Endoplasmic Reticulum. EMBO J. 11, 1717–1722 (1992). [PubMed]
  • Cuozzo J. W. & Kaiser C. A. Competition between glutathione and protein thiols for disulphide-bond formation. Nature Cell Biol. 1, 130–135 (1999). [PubMed]
  • Nardai G., Korcsmaros T., Papp E. & Csermely P. Reduction of the endoplasmic reticulum accompanies the oxidative damage of diabetes mellitus (Reprinted from Thiol Metabolism and Redox Regulation of Cellular Functions). Biofactors 17, 259–267 (2003). [PubMed]
  • Schafer F. Q. & Buettner G. R. Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple. Free Radic. Biol. Med. 30, 1191–1212 (2001). [PubMed]
  • Birk J. et al. . Endoplasmic reticulum: reduced and oxidized glutathione revisited. J. Cell Sci. 126, 1604–1617 (2013). [PubMed]
  • Paulsson J. F. & Westermark G. T. Aberrant processing of human proislet amyloid polypeptide results in increased amyloid formation. Diabetes 54, 2117–2125 (2005). [PubMed]
  • Zimmermann J., Kuhne R., Sylvester M. & Freund C. Redox-regulated conformational changes in an SH3 domain. Biochemistry 46, 6971–6977 (2007). [PubMed]
  • Rodriguez Camargo D. C. et al. . Cloning, expression and purification of the human Islet Amyloid Polypeptide (hIAPP) from Escherichia coli. Protein Expr. Purif. 106, 49–56 (2015). [PubMed]
  • Grzesiek S. & Bax A. An efficient experiment for sequential backbone assignment of medium-sized isotopically enriched proteins. Journal of Magnetic Resonance J. Magn. Reson. 99, 201–207 (1992).
  • Nanga R. P. R., Brender J. R., Vivekanandan S. & Ramamoorthy A. Structure and membrane orientation of IAPP in its natively amidated form at physiological pH in a membrane environment. Biochim. Biophys. Acta 1808, 2337–2342 (2011). [PMC free article] [PubMed]
  • Abedini A. & Raleigh D. P. The role of His-18 in amyloid formation by human islet amyloid polypeptide. Biochemistry 44, 16284–16291 (2005). [PubMed]
  • Guntert P. Automated NMR structure calculation with CYANA. Meth. Mol. Biol. 278, 353–78 (2004). [PubMed]
  • Chiti F. & Dobson C. M. Amyloid formation by globular proteins under native conditions. Nature Chemical Biology 5, 15–22 (2009). [PubMed]
  • Kayed R. et al. . Conformational transitions of islet amyloid polypeptide (IAPP) in amyloid formation in vitro. J. Mol. Biol. 287, 781–796 (1999). [PubMed]
  • Williamson J. A. & Miranker A. D. Direct detection of transient a-helical states in islet amyloid polypeptide. Protein Science 16, 110–117 (2007). [PubMed]
  • Patil S. M., Xu S. H., Sheftic S. R. & Alexandrescu A. T. Dynamic alpha-Helix Structure of Micelle-bound Human Amylin. J. Biol. Chem. 284, 11982–11991 (2009). [PMC free article] [PubMed]
  • Williamson J. A., Loria J. P. & Miranker A. D. Helix Stabilization Precedes Aqueous and Bilayer-Catalyzed Fiber Formation in Islet Amyloid Polypeptide. J. Mol. Biol. 393, 383–396 (2009). [PMC free article] [PubMed]
  • Luca S., Yau W.-M., Leapman R. D. & Tycko R. Peptide Conformation and Supramolecular Organization in Amylin Fibrils: Constraints from Solid-State NMR. Biochemistry 46, 13505–13522 (2007). [PMC free article] [PubMed]
  • Weirich F. et al. . Structural Characterization of Fibrils from Recombinant Human Islet Amyloid Polypeptide by Solid-State NMR: The Central FGAILS Segment Is Part of the β-Sheet Core. PLoS One 11, e0161243 (2016). [PMC free article] [PubMed]
  • Koo B. W. & Miranker A. D. Contribution of the intrinsic disulfide to the assembly mechanism of islet amyloid. Prot. Sci. 14, 231–239 (2005). [PubMed]
  • Padrick S. B. & Miranker A. D. Islet amyloid: Phase partitioning and secondary nucleation are central to the mechanism of fibrillogenesis. Biochemistry 41, 4694–4703 (2002). [PubMed]
  • Obrien T. D., Westermark P. & Johnson K. H. Islet Amyloid Polypeptide and Insulin-Secretion from Isolated Perfused Pancreas of Fed, Fasted, Glucose-treated and Dexamethasone-treated Rats. Diabetes 40, 1701–1706 (1991). [PubMed]
  • Xu C. Y., Bailly-Maitre B. & Reed J. C. Endoplasmic reticulum stress: cell life and death decisions. J. Clin. Invest. 115, 2656–2664 (2005). [PMC free article] [PubMed]
  • Walter P. & Ron D. The Unfolded Protein Response: From Stress Pathway to Homeostatic Regulation. Science 334, 1081–1086 (2011). [PubMed]
  • Ozcan U. et al. . Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes. Science 306, 457–461 (2004). [PubMed]
  • Butler A. E. et al. . Beta-cell deficit and increased beta-cell apoptosis in humans with type 2 diabetes. Diabetes 52, 102–110 (2003). [PubMed]
  • Westermark P., Andersson A. & Westermark G. T. Islet Amyloid Polypeptide, Islet Amyloid and Diabetes Mellitus. Physiol. Rev. 91, 795–826 (2011). [PubMed]
  • Hutton J. C. Secretory Granules. Experientia 40, 1091–1098 (1984). [PubMed]
  • Clark A. & Moffitt J. Pancreatic Islet Amyloid and Diabetes. In Protein Misfolding, Aggregation, and Conformational Diseases. Part B: Molecular Mechanisms of Conformational Diseases, vol. 6 (eds Uversky V. N. & Fink A. L.) 199–216 (Springer, New York, 2007).
  • Rivera J. F. et al. . Human-IAPP disrupts the autophagy/lysosomal pathway in pancreatic beta-cells: protective role of p62-positive cytoplasmic inclusions. Cell Death Diff. 18, 415–426 (2011). [PMC free article] [PubMed]
  • Morita S. et al. . Autophagy protects against human islet amyloid polypeptide-associated apoptosis. J. Diabetes Investig. 2, 48–55 (2011). [PMC free article] [PubMed]
  • Lee M. S. Role of islet beta cell autophagy in the pathogenesis of diabetes. Trends Endocrinol. Metabol. 25, 620–627 (2014). [PubMed]
  • Huang C. J. et al. . Induction of endoplasmic reticulum stress-induced beta-cell apoptosis and accumulation of polyubiquitinated proteins by human islet amyloid polypeptide. Am. J. Physiol. Endocrinol. Metab. 293, E1656–E1662 (2007). [PubMed]
  • Otoda T. et al. . Proteasome Dysfunction Mediates Obesity-Induced Endoplasmic Reticulum Stress and Insulin Resistance in the Liver. Diabetes 62, 811–824 (2013). [PMC free article] [PubMed]
  • Costes S. et al. . Beta-Cell Dysfunctional ERAD/Ubiquitin/Proteasome System in Type 2 Diabetes Mediated by Islet Amyloid Polypeptide-Induced UCH-L1 Deficiency. Diabetes 60, 227–238 (2011). [PMC free article] [PubMed]
  • Rivera J. F., Costes S., Gurlo T., Glabe C. G. & Butler P. C. Autophagy defends pancreatic beta cells from human islet amyloid polypeptide-induced toxicity. J. Clin. Invest. 124, 3489–3500 (2014). [PMC free article] [PubMed]
  • Chien V. et al. . The chaperone proteins HSP70, HSP40/DnaJ and GRP78/BiP suppress misfolding and formation of beta-sheet-containing aggregates by human amylin: a potential role for defective chaperone biology in Type 2 diabetes. Biochem. J. 432, 113–121 (2010). [PubMed]
  • Mainz A. et al. . The Chaperone αB-Crystallin Deploys Different Interfaces to Capture an Amorphous and an Amyloid Client. Nature Struct. Mol. Biol. 22, 898–905 (2015). [PubMed]
  • Nardai G. et al. . Diabetic changes in the redox status of the microsomal protein folding machinery. Biochem. Biophys. Res. Comm. 334, 787–795 (2005). [PubMed]
  • Mulder H., Ahren B. & Sundler F. Islet amyloid polypeptide (amylin) and insulin are differentially expressed in chronic diabetes induced by streptozotocin in rats. Diabetologia 39, 649–657 (1996). [PubMed]
  • Franko A. et al. . Bezafibrate improves insulin sensitivity and metabolic flexibility in STZ-treated diabetic mice. Diabetes 65, 2540–2552 (2016). [PubMed]
  • Sattler M., Schleucher J. & Griesinger C. Heteronuclear multidimensional NMR experiments for the structure determination of proteins in solution employing pulsed field gradients. Prog. NMR Spect. 34, 93–158 (1999).
  • Goddard T. D. & Kneller D. G. SPARKY 3. (University of California, San Francisco, 1997).
  • Delaglio F. et al. . NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR 6, 277–293 (1995). [PubMed]
  • Vranken W. F. et al. . The CCPN Data Model for NMR Spectroscopy: Development of a Software Pipeline. Proteins 59, 687–696 (2005). [PubMed]
  • Cornilescu G., Delaglio F. & Bax A. Protein backbone angle restraints from searching a database for chemical shift and sequence homology. J. Biomol. NMR 13, 289–302 (1999). [PubMed]
  • Shen Y., Delaglio F., Cornilescu G. & Bax A. TALOS+: a hybrid method for predicting protein backbone torsion angles from NMR chemical shifts. J. Biomol. NMR 44, 213–223 (2009). [PMC free article] [PubMed]

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group