The aggregation potential of human amylin determines its cytotoxicity towards islet β‐cells

Abstract

Human amylin (hA) is a small protein cosecreted with insulin from pancreatic islet b-cells upon stimulation by glucose or other chemical signals [1,2]. Wild-type hA, also known as insulinoma amyloid peptide, insulinoma amyloid polypeptide (IAPP) or islet amyloid polypeptide [3], demonstrates a strong in vitro tendency to aggregate into fibrils [4,5], which is dependent on specific residues in different regions of the molecule, notably, the amyloidogenic region of amino acids 20-29 [6]. It is the main protein in the amyloid aggregates that are frequently present in the islets of human subjects with type 2 diabetes mellitus (T2DM) [7,8], as well as in diabetic cats and primates [9]. It has been reported that aggregated hA might contribute to the loss of insulin-producing pancreatic b-cells cells in T2DM [10,11]. The following lines of evidence have associated hA with the pancreatic pathology of T2DM: (a) amyloid is often concentrated near areas of islet b-cell degeneration in humans or primates with T2DM [10,[12][13][14]; (b) synthetic hA is toxic to pancreatic b-cells in vitro [11,[15][16][17]; (c) spontaneous loss of b-cells has been reported in a line of hA-transgenic mice [18]; and (d) toxicity of hA towards cultured cells correlates with its ability to form fibrils -for example, rat amylin does not form fibrils and nor does it evoke b-cell apoptosis in vitro [11,17,19]. Human amylin is a small fibrillogenic protein that is the major constituent of pancreatic islet amyloid, which occurs in most subjects with type 2 diabetes. There is evidence that it can elicit in vitro apoptosis in islet b-cells, but the physical properties that underpin its cytotoxicity have not been clearly elucidated. Here we employed electron microscopy, thioflavin T fluorescence and CD spectroscopy to analyze amylin preparations whose cytotoxic potential was established by live-dead assay in cultured b-cells. Highly toxic amylin contained few preformed fibrils and initially showed little b-sheet content, but underwent marked time-dependent aggregation and b-conformer formation following dissolution. By contrast, low-toxicity amylin contained abundant preformed fibrils, and demonstrated high initial b-sheet content but little propensity to aggregate further once dissolved. Thus, mature amylin fibrils are not toxic to b-cells, and aggregates of fibrils such as occur in pancreatic islet amyloid in vivo are unlikely to contribute to b-cell loss. Rather, the toxic molecular species is likely to comprise soluble oligomers with significant b-sheet content. Attempts to find ways of protecting b-cells from amylin-mediated death might profitably focus on preventing the conformational change from random coil to b-sheet.Abbreviations AM, acetomethoxyl ester; EthD, ethidium homodimer; hA, human amylin; HFIP, hexafluoroisopropanol; IAPP, insulinoma amyloid polypeptide; JNK1, Jun NH 2 -terminal kinase 1; KRB, Krebs-Ringer bicarbonate buffer; RINm5F, rat insulinoma m5F; T2DM, type 2 diabetes mellitus; TEM, transmission electron microscopy; ThT, thioflav...

Amyloid Fiber Formation and Membrane Disruption are Separate Processes Localized in Two Distinct Regions of IAPP, the Type-2-Diabetes-Related Peptide

“…Previous studies have shown that the N-terminal 1-19 region, rather than the amyloidogenic 20-29 region, is primarily responsible for the interaction of the IAPP peptide with membranes. Liposome leakage experiments presented in this study confirm that the pathological membrane disrupting activity of the full-length hIAPP is also shared by hIAPP [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] . The hIAPP 1-19 fragment at a low concentration of peptide induces membrane disruption to a near identical extent as the full-length peptide.…”

“…The hIAPP [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] and hIAPP [20][21][22][23][24][25][26][27][28][29] peptides were synthesized using standard FMOC (9-fluornylmethoxycarbonyl) methods. A PAL-PEG resin was used to provide an amidated C-terminus.…”

“…The hIAPP 20-29 peptide was synthesized as described above but with both an amidated C-terminus and an acetylated N-terminus. After synthesis, the crude hIAPP [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] peptide was subjected to air oxidation to form the disulfide bond. The identity of the hIAPP [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] and hIAPP [20][21][22][23][24][25][26][27][28][29] sequences and formation of the disulfide bond for hIAPP [1][2][3][4]…”

“…Lyophilized hIAPP and hIAPP [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] were dissolved in HFIP for 1 h to obtain a 4 mg/mL stock solution. Aliquots of the peptide stock solution were lyophilized again and then resolubilized in either 10 mM sodium phosphate buffer, pH 7.3 or 10 mM sodium phosphate buffer with 150 mM NaF, pH 7.3.…”

“…It has previously been shown through fluorescence anisotropy, infrared reflectionabsorption spectroscopy, and lipid monolayer expansion experiments that the N-terminal part of IAPP is the primary membrane binding site of IAPP.11,14,15 Indeed, the 1 -19 fragment of human IAPP (hIAPP [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] ) inserts into lipid membranes with a higher efficiency than full-length IAPP.11 We report here that the N-terminal region of the peptide (residues 1 -19) can disrupt synthetic lipid vesicles to a similar extent as the full-length IAPP peptide without forming amyloid fibers and may therefore be an important model for the study of membrane disruption by IAPP and other amyloid proteins.…”