A natural hyperbranched proteoglycan inhibits IAPP amyloid fibrillation and attenuates β-cell apoptosis

Abstract

The self-association of human islet amyloid polypeptide (IAPP) into cytotoxic fibrillar assemblies that induce pancreatic β-cell apoptosis has been postulated as one of the major contributors to the development of type 2 diabetes (T2D). Hence, preventing IAPP amyloid fibrillation and β-cell apoptosis is regarded as a promising treatment for T2D. Intrigued by our previous findings that a natural amphiphilic hyperbranched proteoglycan from Ganoderma lucidum, named FYGL, can decrease plasma glucose level in vivo, we demonstrated here that FYGL efficiently inhibits IAPP fibrillation and attenuates β-cell apoptosis, using a combination of biophysical, calculation and cell culture methods, including thioflavin T (ThT) and 8-anilino-1-naphthalene sulfonate (ANS) fluorescence spectroscopy, transmission electron microscopy (TEM), atomic force microscopy (AFM), circular dichroism (CD) spectroscopy, dynamic light scattering (DLS), nuclear magnetic resonance (NMR) spectroscopy, molecular docking, cell viability tests and laser scanning confocal microscopy imaging. Moreover, a universally applicable inhibition mechanism proposed that FYGL blocks the interpeptide interaction and stabilizes IAPP structure in an α-helical conformation by forming multiple hydrogen bonds with IAPP. The findings of this study warrant FYGL to be further investigated as a potential therapeutic treatment and may provide a valuable reference for medicinal chemistry in the development of drugs against T2D as well as other amyloid diseases such as Alzheimer's disease and Huntington's disease.

---

Introduction

Type 2 diabetes (T2D), a most prevalent amyloid disease characterized by progressive insulin resistance and pancreatic β-cell apoptosis, has continued to expand in recent years due to modern lifestyle and the absence of effective treatment, posing a great threat to public health and social economy. Islet amyloid polypeptide (IAPP or amylin, Fig. S1 in the ESI†), a 37-residue polypeptide hormone co-secreted synergistically with insulin, is normally soluble and unfolded in its physiological state, functioning in controlling gastric emptying and maintaining glucose homeostasis.^1,2 However, the pathological fibrillation of IAPP has been reported to play a key role in the development of T2D by inducing β-cell membrane disruption and fragmentation.³ In T2D, IAPP self-associates into toxic oligomers and fibrils, along with its secondary structural transition from α-helix-dominant to β-sheet-rich, leading to β-cell apoptosis.⁴ Therefore, preventing IAPP cytotoxicity by impeding its aberrant fibrillation is regarded as a promising therapy against T2D.

A few inhibitors of IAPP amyloidogenicity and cytotoxicity have been reported, such as polyphenols,^5–11 graphene oxide,¹² molecular tweezers,¹³ IAPP-derived peptides¹⁴ and dendrimers,¹⁵ and much effort has been devoted to evaluate their effectiveness in the cell-like conditions.¹⁶ Despite the progress made in exploring their efficacies and mechanisms, there are no clinically approved inhibitors of islet amyloid formation for the treatment of T2D. Small-molecule polyphenols have been widely investigated for reducing IAPP amyloidogenicity, whereas their accessibility is hindered since many of these polyphenol inhibitors have poor water solubility, including curcumin,^7,8 silibinin⁹ and resveratrol.¹⁰ On the other hand, synthetic inhibitors including molecular tweezers and IAPP-derived peptides are very complicated and costly for practical use, in spite of their effective inhibition of IAPP fibrillation and cytotoxicity.

To develop a safe, effective and economical strategy for inhibiting IAPP self-assembly and attenuating its toxicity, we have successfully isolated an ingredient, named FYGL, from a traditional Chinese herb Ganoderma lucidum which has been proven beneficial in the prevention of T2D and its complications.^17,18 FYGL is a hyperbranched proteoglycan consisting of both hydrophilic and lipophilic moieties (Fig. 1).¹⁹ In addition to its potential capabilities as proteoglycan in specific recognition and communication,^20–22 its unique hyperbranched polyhydroxyl structure not only contributes to its extraordinary water solubility, but also provides multiple binding sites to form stable supramolecular complexes with IAPP, so as to block the interpeptide interaction which is critical in IAPP self-association. Our previous studies showed that FYGL decreases plasma glucose level as well as multiple plasma biochemistry indexes, such as free fatty acid, triglyceride, total cholesterol and low density lipoprotein-cholesterol, for metabolic syndrome in streptozotocin-induced diabetic rats.²³ Moreover, FYGL significantly improves the glucose homeostasis and impaired glucose tolerance in db/db mouse model of T2D, and remarkably lowers the glycosylated hemoglobin (HbAlc) level, a useful indicator in long-term diabetic control.^24–26

Fig. 1 The dominant sequence of FYGL. Ara: arabinose, Gal: galactose, Glc: glucose, Rha: rhamnose. The suffixes p and f represent pyranose and furanose respectively. Thr: threonine, Ser: serine.

Despite the beneficial behaviors of FYGL for T2D treatment, its mechanism of actions remains unclear. Herein, we aim to postulate a detailed mechanism of the protective function of FYGL on β-cells by inhibiting IAPP amyloidogenicity and cytotoxicity, using a combination of biophysical, calculation and cell culture methods including thioflavin T (ThT) and 8-anilino-1-naphthalene sulfonate (ANS) fluorescence assays, transmission electron microscopy (TEM), atomic force microscopy (AFM), circular dichroism (CD) spectroscopy, dynamic light scattering (DLS), nuclear magnetic resonance (NMR) spectroscopy, molecular docking and cell culture experiments.

Materials and methods

Materials

IAPP with an amidated C-terminus and a disulfide bridge (purity > 95%) was purchased from Chinese Peptide Company (China). Deuterated solvents for NMR use were from Cambridge Isotope Laboratories (USA). Other biophysical reagents were from Aladdin Industrial Corporation (China) unless otherwise specified. Cell culture reagents were from Gibco (Life Technologies, Canada). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) was from Sangon Biotechnology (China). DAPI and rhodamine 123 (Rh123) were from Dojindo Laboratories (Japan). Lyophilized powder of FYGL was prepared as described previously.¹⁸

Peptide preparation

IAPP was freshly dissolved at a concentration of 800 μM in 1,1,1,3,3,3-hexafluoroisopropanol (HFIP) to remove any pre-aggregates, followed by 2 min sonication and subsequently incubated at 37 °C for 4 h. For biophysical experiments, the stock solution was directly diluted to the final concentration of 16 μM with 10 mM PBS (pH 7.4) in the absence or presence of FYGL unless otherwise specified. For cell culture experiments, aliquots of the stock solution were lyophilized and dissolved by cell culture media in the absence or presence of different concentrations of FYGL swiftly before addition to cells.

ThT and ANS fluorescence assays

Amyloid fibrillation was monitored by recording ThT and ANS fluorescence intensity every 2 min for 2 h at 37 °C, with constant agitation at 60 rpm, using a Varioskan LUX 3020-Z024 microplate reader (ThermoFisher Scientific, USA). ThT fluorescence was measured with excitation at 450 nm and emission at 485 nm. ANS fluorescence was measured with excitation at 372 nm and emission at 480 nm. Background fluorescence was subtracted by using corresponding blanks without IAPP. Data are presented as mean ± SEM of four independent experiments. ThT fluorescence data were fitted using the sigmoidal logistic model.

TEM

Aliquots (10 μL) of the final mixtures from ThT measurements were spotted individually on a 300-mesh carbon-coated copper grid, and stained with 2% (w/v) uranyl acetate. The samples were viewed under a Tecnai G² 20 TWIN transmission electron microscope (FEI, USA).

AFM

Aliquots (10 μL) of the final mixtures from ThT measurements were deposited onto freshly cleaved mica, and dried in vacuum overnight. The samples were scanned using a Multimode 8 AFM instrument (Bruker, USA) in a tapping mode.

CD spectroscopy

CD analysis was performed on a Chirascan CD spectrometer (Applied Photophysics, UK) using a 1 mm path-length quartz cuvette at 25 °C. The samples were scanned over a wavelength range from 260 to 185 nm at 1.0 nm intervals. The peptide stock solution was added to the FYGL solutions to a final concentration of 16 μM immediately before each measurement. Experiments were performed three times, and the spectra of representative experiments were shown. The characteristic change of ellipticity at 217 nm corresponding to β-sheet content as a function of incubation time was plotted and fitted using the sigmoidal logistic model.

DLS

The DLS measurements were performed on a ZS90 Zetasizer Nano instrument (Malvern Instruments, UK) equipped with a 173° back scattering measurement facility. The particle size distribution of IAPP in buffer in the absence or presence of various concentrations of FYGL was monitored for up to 1 h at 25 °C in a 1 cm path-length disposable cuvette.

¹H NMR spectroscopy

HFIP-treated IAPP was lyophilized and then dissolved in deuterated DMSO to the final concentration of 100 μM in the absence or presence of 60 or 500 μg mL⁻¹ FYGL. ¹H NMR spectra were acquired at 25 °C with 128 scans on an Advance III HD 400 MHz spectrometer (Bruker BioSpin International, Switzerland). Hydrogen/deuterium (H/D) exchange was performed by adding 10% D₂O (v/v).

Molecular docking

The AutoDock Vina program was applied for molecular docking based on the calculation of binding affinity of IAPP with FYGL, and AutoDockTools 1.5.6 was employed for the visualization of the results.²⁷ The initial structure of IAPP was attained from Protein Data Bank (2KB8) and regarded as a completely rigid molecular for docking.²⁸ The dominant stereostructure of FYGL (Fig. 1) was optimized by MM2 energy minimization prior to docking studies and allowed to dock everywhere at IAPP.

Cell viability assay

The rat insulinoma (RIN5fm) cells (Shanghai Fusheng Industrial Company, China) were used as model β-cells, cultured in RPMI 1640 media supplemented with 10% FBS, 100 U mL⁻¹ penicillin and 100 μg mL⁻¹ streptomycin at 37 °C in a 5% CO₂ humidified atmosphere. Cells were seeded in 96-well plates at a density of 20 000 cells per well and incubated for 24 h. Subsequently, the cell culture media was replaced by fresh media containing IAPP in the absence or presence of different concentrations of FYGL, and incubated for another 24 h. At the end of the incubation, cells were viewed under a 37XB inverted microscope (Shanghai Yuguang Instrument, China). The cell viability was assessed by MTT assay using an ELx800 Microplate Reader (BioTek, USA), and the significance level was calculated by student's t-test.

Laser scanning confocal microscopy imaging

RIN5fm cells were planted on microscope cover glass slides placed in 12-well plates at a density of 300 000 cells per well and incubated for 24 h. Then IAPP in the absence or presence of FYGL was applied to the cells for another 24 h. Nuclei and mitochondria were stained with DAPI and Rh123, respectively. The samples were mounted and observed under a C2⁺ laser scanning confocal microscope (Nikon, Japan). Fluorescence of DAPI and Rh123 was sequentially acquired by alternating the excitation at 405 and 488 nm. Images were captured with identical parameters for comparison.

Results and discussion

FYGL inhibits IAPP amyloid fibrillation in a dose-dependent manner

To explore the inhibitory activity of FYGL on IAPP fibrillation, we followed the kinetics of 16 μM IAPP fibrillation in the absence or presence of 10, 20, 50 or 500 μg mL⁻¹ FYGL using both ThT and ANS probes. ThT and ANS are by far the most popular fluorophores used to detect β-sheet-rich fibrils in vitro, which undergo a remarkable increase in fluorescence intensity in the presence of amyloid fibrils.^29,30

The fibrillation kinetic curves monitored by ThT assay comprise of a lag phase, an exponential growth phase and an equilibrated phase (Fig. 2a). The pure IAPP displays an initial rapid increase in ThT fluorescence, with the lag phase even too short to be observed, indicating a quick formation of β-sheet-rich fibrils. With the treatment of 10 μg mL⁻¹ FYGL, the lag phase is significantly extended and the final steady-state ThT fluorescence level is substantially lower than that of IAPP alone, suggesting an efficient inhibition of β-sheet formation by FYGL in both kinetics and thermodynamics manner. As the concentration of FYGL increased to 20 μg mL⁻¹, IAPP fibrillation is further retarded and the final fluorescence level is much lower. Moreover, with the treatment of 50 or 500 μg mL⁻¹ FYGL, ThT signal remains at baseline level throughout the experiment, indicating a complete inhibition of IAPP fibrillation. Almost no fluorescence is detected for FYGL alone. It is noteworthy that the molecular weight of FYGL is about 260 kDa,²⁴ thereby the molar concentration required for the complete inhibition of IAPP fibrillation is less than 1 pM. As a control, dextran (molecular weight 70 000), a linear homo-biopolymer, shows no inhibition on IAPP fibrillation (Fig. S2 in the ESI†).

Fig. 2 Kinetics of 16 μM IAPP fibrillation in the absence or presence of 10, 20, 50 and 500 μg mL⁻¹ FYGL detected by ThT (a) and ANS (b) assay. AFM (c) and TEM (d) images of the final mixtures from ThT measurements.

In summary, an extension of lag time and a decrease in the final ThT fluorescence level were detected with the increase of FYGL concentration, indicating that FYGL impedes IAPP β-sheet fibril formation in a dose-dependent manner. To further obtain quantitative results, the kinetic curves were fitted with the sigmoidal logistic model according to eqn (1):³¹

where y is defined as the ThT signal at time t, A_max is the maximum fluorescence detected for a given sample, t₅₀ marks the time required to reach half of A_max and k_app is the apparent rate constant of fibril growth, in units of min⁻¹. The lag time t₀ was obtained from eqn (2):

The parameters of the fitted kinetic curves were summarized in Table 1. Similar results were obtained from ANS assay (Fig. 2b), whereas the overall fluorescence is lower than that of ThT, leading to relatively poor signal to noise ratio.

Table 1 Fibrillation kinetics of IAPP in the absence or presence of FYGL determined by ThT assay

FYGL (μg mL^−1)	t0 (min)	t50 (min)	kapp (min^−1)	Amax (A.U.)
a The time is too short to be determined under the experimental condition.b The time is too long to accurately calculate the parameters during the experiment.
0	–^a	3.7 ± 0.1	0.39 ± 0.02	99.0 ± 0.5
10	17.5 ± 0.8	28.1 ± 0.4	0.19 ± 0.01	53.6 ± 0.2
20	45.9 ± 2.4	59.9 ± 1.2	0.14 ± 0.01	26.7 ± 2.0
50	N^b	N^b	N^b	N^b
500	N^b	N^b	N^b	N^b

---

Theoretically, FYGL might compete with ThT or ANS probes for amyloid binding sites or quench fluorescence rather than inhibiting IAPP self-assembly.³² In this regard, we characterized the morphology of the resulting mixtures from each kinetic experiment by TEM (Fig. 2c) and AFM (Fig. 2d).³³ After 2 h incubation, the pure IAPP forms abundant tiny fibrils. While with the treatment of 10 or 20 μg mL⁻¹ FYGL, the density of tiny fibrils decreases dramatically, instead, much longer and thicker fibers or amorphous aggregates appear. This observation matches well with the theory that the propagation of existing aggregates by the addition of monomers is energetically more favorable than the nucleation of ordered fibrils,³⁴ and reveals an effective inhibition of IAPP fibrillation by FYGL. IAPP fibrillation is completely inhibited when further treated with 50 or 500 μg mL⁻¹ FYGL, presented as predominantly amorphous aggregates.

FYGL modulates IAPP secondary structural conversion

In order to further reveal IAPP conformational conversion treated by various concentrations of FYGL, we monitored the transition process using CD spectroscopy. Herein, 16 μM IAPP was incubated in the absence or presence of 10, 50 and 500 μg mL⁻¹ FYGL, and CD spectra were recorded constantly for up to 1 h (Fig. 3a–d). FYGL almost has no CD signal (Fig. S3 in the ESI†). Because of an absence of appropriate reference databases for amyloid fibrils, we were not able to obtain any ideal deconvolution data to quantitatively determine IAPP conformational distribution. Hence, the analysis here is merely qualitative.

Fig. 3 Secondary structural conversion of 16 μM IAPP incubated in the absence (a) or presence of 10 (b), 50 (c) or 500 (d) μg mL⁻¹ FYGL on CD spectroscopy. (e) The change of characteristic CD signal of IAPP at 217 nm as a function of time in the absence or presence of 10, 50 or 500 μg mL⁻¹ FYGL. (f) The variation of average size of IAPP as a function of time in the absence or presence of 10, 20 or 50 μg mL⁻¹ FYGL detected by DLS.

The initial spectrum of pure IAPP displays a characteristic α-helix-dominant conformation with a single maximum at around 189 nm and two minima at 203 and 222 nm (Fig. 3a), indicating that IAPP has been completely disaggregated prior to its incubation.¹³ The maximum near 189 nm resembles the π–π* transition, and the two minima at 203 and 222 nm resemble the π–π* and n–π* transition of α-helical structure, respectively.^35,36 With further incubation, the maximum at around 190 nm corresponding to the exciton interaction of β-sheet structure increases, and the bands at 203 and 222 nm are obscured by one single minimum at 217 nm which is assigned to the n–π* transition of β-sheet structure, indicating a continuous conformational conversion from α-helix to β-sheet.^7,36 The spectra have an isodichroic point at around 204 nm, revealing that the conformational transition occurs between the two predominant states without pronounced accumulation of intermediates. In the presence of 10 μg mL⁻¹ FYGL, the initial CD spectrum is similar to that of pure IAPP, indicating a high α-helix content, whereas the conformational transition from α-helix to β-sheet is substantially delayed. Meanwhile, the development of the minimum at 217 nm corresponding to β-sheet conformation is much slower, and the overall variation is much smaller than that of IAPP alone (Fig. 3b). With the treatment of 50 μg mL⁻¹ FYGL, the spectra exhibit no significant change throughout the monitoring course, characterized by one maximum at around 189 nm and two minima at 207 and 222 nm, indicating that IAPP maintains its α-helix-dominant structure (Fig. 3c). The spectra with 500 μg mL⁻¹ FYGL exhibit similar pattern to those with 50 μg mL⁻¹ FYGL, except for the magnified noise at around 189 nm (Fig. 3d).

Conformational transition of IAPP represented by the CD signal variation at 217 nm (Fig. 3e), a typical sign for β-sheet structure, correlates closely with the sigmoidal logistic curves as also processed from ThT assay (Fig. 2a), characterized by three steps: a lag phase, an exponential growth phase and a steady plateau.³⁷ FYGL remarkably impedes IAPP fibrillation by both extending the lag phase and reducing the β-sheet content in the final product. The pure IAPP undergoes a structural rearrangement from α-helix-dominant to β-sheet-rich over the incubation time, while 10 μg mL⁻¹ FYGL can significantly retard and inhibit this transition, and 50 μg mL⁻¹ or higher concentrations of FYGL completely suppresses this conversion by stabilizing IAPP structure in an α-helical conformation.

FYGL quickly binds with IAPP to form nonfibrillar aggregates

To elucidate the aggregation behaviors corresponding to IAPP secondary structural evolution, we conducted DLS measurements to monitor the average size variation of IAPP aggregation in the absence or presence of various concentrations of FYGL. DLS is a sensitive tool to yield qualitative real-time information about the size distribution upon protein aggregation in solution, based on the theory that an increase of the scattered intensity in DLS corresponds to an increase of the scattering particle size.³⁸ Here, 16 μM IAPP was incubated in the absence or presence of 10, 20 and 50 μg mL⁻¹ FYGL at 25 °C for DLS measurements.

The mean size of pure IAPP increases slowly throughout the experiment (Fig. 3f), suggesting a gradual formation of aggregates during IAPP self-assembly. In contrast, a rapid increase in average size is observed with the addition of 10, 20 or 50 μg mL⁻¹ FYGL, indicating a quick formation of aggregates. FYGL alone has negligible signal at the above concentrations. DLS analysis suggests that FYGL binds with IAPP swiftly at very early time to form the nonfibrillar species as characterized by TEM and AFM, preventing IAPP further misfolding into the ordered β-sheet-rich fibrils which show high fluorescence intensity.

FYGL interacts with IAPP via multiple hydrogen bonding

To explore the binding sites of IAPP with FYGL, ¹H NMR spectra of 100 μM IAPP in the absence or presence of 60 or 500 μg mL⁻¹ FYGL dissolved in deuterated DMSO were collected, and H/D exchange was carried out to determine the active protons (Fig. 4). The region within 6.85 and 9.25 ppm was chosen to analyze the IAPP resonances interfered by FYGL in order to avoid the overlap by FYGL or non-deuterated solvent peaks.

Fig. 4 NMR spectra of 100 μM IAPP in deuterated DMSO in the absence or presence of 60 or 500 μg mL⁻¹ FYGL. The top spectrum is the H/D exchange one with 10% D₂O to detect the active protons which disappear in the spectrum. The active proton signal of the imidazole ring of His18 is marked with an asterisk. The chemical shift regions of 7.35–8.00 and 6.92–7.30 ppm are highlighted in gray dashed frames.

FYGL causes noticeable changes to the ¹H NMR spectrum of IAPP, particularly to the resonances of multiple active protons. The broadening of the active hydrogen signal (δ 9.15 ppm) of His18 in the imidazole ring suggests that His18 is involved in the peptide-inhibitor interaction.^10,39 The variations of signals ranging from 7.35 to 8.00 ppm are from the active amide protons on peptide backbone, and the ones within 6.92 and 7.30 ppm are from the active amide protons on side chains of Asn, Arg and Gln.

Furthermore, we adopted molecular docking, a tool for predicting favorable binding modes based on the calculation of interaction energies, to verify the NMR data and to theoretically investigate the binding sites of IAPP with FYGL. In the most favored binding mode as Fig. 5 with the lowest binding free energy, the hydroxyl groups of FYGL interact with the side chains of Ser20 and Asn21 of IAPP via multiple hydrogen bonding. Concomitantly, the close contacts form between FYGL and Val17, His18, Ser20, Asn21, Gly24, Ala25, Ser28, Asn31 and Val32 of IAPP.

Fig. 5 The most favored binding mode of FYGL and IAPP. The close contacts are displayed by pink wireframe spheres. IAPP is represented by pink ribbons with sticks. The molecular surface of FYGL is painted cyan. The hydrogen bonds are represented by green dots. Val: valine, His: histidine, Ser: serine, Asn: asparagine, Gly: glycine, Ala: alanine.

FYGL protects RIN5fm β-cells from IAPP-induced cytotoxicity

To evaluate the protective role of FYGL on IAPP-induced β-cell apoptosis, the cell viability of RIN5fm cells incubated with IAPP in the absence or presence of FYGL was measured using MTT assay.⁴⁰ IAPP induces significant cell apoptosis in a dose-dependent manner, and the cell viability decreases to 47 ± 7% with the treatment of 16 μM IAPP for 24 h (Fig. 6a). The effect of FYGL alone on RIN5fm cell viability was investigated to determine the safe dose for use. Encouragingly, FYGL exhibits no significant effect on cell viability even at the concentration of 200 μg mL⁻¹ (Fig. S4a in the ESI†). To keep the data comparable to the above experiments, treatment with 16 μM IAPP for 24 h is adopted for all cell culture experiments. FYGL is found to remarkably attenuate IAPP-induced apoptosis dose-dependently (Fig. 6b). With the addition of 20, 50 and 200 μg mL⁻¹ FYGL to IAPP-treated β-cells, the cell viability is restored from 44 ± 2% to 49 ± 3%, 60 ± 3% and 82 ± 2%, respectively. By comparison, the linear dextran with corresponding concentrations reveals no protection against IAPP-induced apoptosis (Fig. S4b in the ESI†), which verifies the unique protective role of FYGL on β-cells.

Fig. 6 (a) Cell viability treated by 0, 0.5, 1, 5, 8 or 16 μM IAPP for 24 h. (b) Cell viability treated by 16 μM IAPP in the absence or presence of 20, 50 or 200 μg mL⁻¹ FYGL for 24 h. ^**P < 0.05 vs. negative control group, ^***P < 0.01 vs. negative control group, ^##P < 0.05 vs. IAPP-treated group, ^###P < 0.01 vs. IAPP treated group.

The morphological change of RIN5fm β-cells with IAPP in the absence or presence of FYGL or dextran was also characterized by inverted microscope (Fig. 3c and d). When incubated with 16 μM IAPP, the cells shrink and even fragment, and the cell density significantly decreases. With the addition of 20 μg mL⁻¹ FYGL, the cell density is restored, but fragmentation is still observed. Upon treatment with 50 μg mL⁻¹ FYGL, the cell density is further restored with few fragmentation. With the addition of 200 μg mL⁻¹ FYGL, almost no cell fragmentation is observed (Fig. S4c in the ESI†). In contrast, each concentration of dextran does not increase cell viability or eliminate cell fragmentation at all (Fig. S4d in the ESI†).

FYGL restores loss of MMP induced by IAPP in RIN5fm β-cells

We used a nucleic acid staining agent DAPI to locate cells and a cell-permeable cationic dye Rh123 to characterize the loss of MMP through a laser scanning confocal microscope (Fig. 7). Under normal conditions, the mitochondrial membrane is highly negative-charged, accumulating a large amount of Rh123. While in the stage of apoptosis, depolarization of MMP induces loss of Rh123 from the mitochondria and thereby a decrease in intracellular Rh123 signal.⁴¹

Fig. 7 FYGL protects RIN5fm cells from the decrease of mitochondrial membrane potential induced by IAPP. RIN5fm cells was incubated with 16 μM IAPP in the absence or presence of 20, 50 or 200 μg mL⁻¹ FYGL and fluorescently stained with DAPI and Rh123.

With the addition of 16 μM IAPP, the cell density represented by DAPI staining decreases compared to the negative control level, meanwhile, the Rh123 intensity reduces considerably, indicating that the depolarization of MMP induced by IAPP. In contrast, the reduced cell density and Rh123 intensity induced by IAPP is dose-dependently reversed with the addition of FYGL, suggesting that FYGL restores the RIN5fm β-cell activity and MMP loss.

FYGL is demonstrated capable of redirecting IAPP self-assembly towards nonamyloidogenic and nontoxic IAPP/FYGL complexes instead of toxic IAPP fibrils, and protecting β-cells from IAPP-induced apoptosis. Based on our findings and relevant studies,^40,42–44 a detailed mechanism was proposed that FYGL can associate with IAPP via multiple hydrogen bonding to form nonfibrillar complexes, which blocks the interpeptide interaction and stabilizes IAPP structure in an α-helical conformation, thereby preventing IAPP rearranging into ordered β-sheet structures and further self-associating into toxic fibrils.

Conclusion

A natural proteoglycan with amphiphilic hyperbranched structures, named FYGL, was demonstrated to efficiently inhibit IAPP fibrillation and attenuate β-cell apoptosis for type 2 diabetes treatment. The inhibition mechanism was proposed that FYGL blocks the interpeptide interaction and stabilizes IAPP structure in an α-helical conformation, along with the formation of multiple intermolecular hydrogen bonds. The findings of this study warrant FYGL to be further investigated as a potential therapeutic treatment of T2D. In addition, the structure–activity relationship of FYGL provides a valuable reference for medicinal chemistry in the development of drugs for type 2 diabetes and other amyloid diseases such as Alzheimer's disease and Huntington disease.

Acknowledgements

We thank the National Natural Science Foundation of China (No. 21374022, 81374032) and Senior Visiting Scholar Foundation of Key Laboratory of Fudan University (No. 15FGJ03) for financial support.

References

1. P. Cao, A. Abedini and D. P. Raleigh, Curr. Opin. Struct. Biol., 2013, 23, 82–89.
2. A. Abedini and A. M. Schmidt, FEBS Lett., 2013, 587, 1119–1127.
3. P. E. S. Smith, J. R. Brender and A. Ramamoorthy, J. Am. Chem. Soc., 2009, 131, 4470–4478.
4. R. Qi, Y. Luo, B. Ma, R. Nussinov and G. Wei, Biomacromolecules, 2014, 15, 122–131.
5. F. L. Palhano, J. Lee, N. P. Grimster and J. W. Kelly, J. Am. Chem. Soc., 2013, 135, 7503–7510.
6. P. Cao and D. P. Raleigh, Biochemistry, 2012, 51, 2670–2683.
7. S. Sparks, G. Liu, K. J. Robbins and N. D. Lazo, Biochem. Biophys. Res. Commun., 2012, 422, 551–555.
8. M. Daval, S. Bedrood, T. Gurlo, C. J. Huang, S. Costes, P. C. Butler and R. Langen, Amyloid, 2010, 17, 118–128.
9. B. Cheng, H. Gong, X. Li, Y. Sun, X. Zhang, H. Chen, X. Liu, L. Zheng and K. Huang, Biochem. Biophys. Res. Commun., 2012, 419, 495–499.
10. L. H. Tu, L. M. Young, A. G. Wong, A. E. Ashcroft, S. E. Radford and D. P. Raleigh, Biochemistry, 2015, 54, 666–676.
11. C. H. Chen, T. Yao, Q. Zhang, Y. M. He, L. H. Xu, M. Zheng, G. R. Zhou, Y. Zhang, H. J. Yang and P. Zhou, RSC Adv., 2016, 6, 15240–15246.
12. P. Nedumpully-Govindan, E. N. Gurzov, P. Chen, E. H. Pilkington, W. J. Stanley, S. A. Litwak, T. P. Davis, P. C. Ke and F. Ding, Phys. Chem. Chem. Phys., 2016, 18, 94–100.
13. D. H. Lopes, A. Attar, G. Nair, E. Y. Hayden, Z. Du, K. McDaniel, S. Dutt, K. Bravo-Rodriguez, S. Mittal, F. G. Klarner, C. Wang, E. Sanchez-Garcia, T. Schrader and G. Bitan, ACS Chem. Biol., 2015, 10, 1555–1569.
14. H. Wang, Z. Ridgway, P. Cao, B. Ruzsicska and D. P. Raleigh, Biochemistry, 2015, 54, 6704–6711.
15. E. N. Gurzov, B. Wang, E. H. Pilkington, P. Chen, A. Kakinen, W. J. Stanley, S. A. Litwak, E. G. Hanssen, T. P. Davis, F. Ding and P. C. Ke, Small, 2016, 12, 1615–1626.
16. M. Gao, K. Estel, J. Seeliger, R. P. Friedrich, S. Dogan, E. E. Wanker, R. Winter and S. Ebbinghaus, Phys. Chem. Chem. Phys., 2015, 17, 8338–8348.
17. R. R. Paterson, Phytochemistry, 2006, 67, 1985–2001.
18. D. Pan, L. Wang, C. Chen, B. Teng, C. Wang, Z. Xu, B. Hu and P. Zhou, Food Chem., 2012, 135, 1097–1103.
19. D. Pan, L. Wang, C. Chen, B. Hu and P. Zhou, Carbohydr. Polym., 2015, 117, 106–114.
20. R. V. Iozzo and L. Schaefer, Matrix Biol., 2015, 42, 11–55.
21. J. R. Couchman and C. A. Pataki, J. Histochem. Cytochem., 2012, 60, 885–897.
22. Y. Li, L. Wang, T. Lu, Y. Wei and F. Li, Phys. Chem. Chem. Phys., 2016, 18, 12000–12008.
23. B. S. Teng, C. D. Wang, D. Zhang, J. S. Wu, D. Pan, L. F. Pan, H. J. Yang and P. Zhou, Eur. Rev. Med. Pharmacol. Sci., 2012, 16, 166–175.
24. C. Wang, B. Teng, Y. He, J. Wu, D. Pan, L. Pan, D. Zhang, Z. Fan, H. Yang and P. Zhou, Br. J. Nutr., 2012, 108, 2014–2025.
25. D. Pan, D. Zhang, J. Wu, C. Chen, Z. Xu, H. Yang and P. Zhou, PLoS One, 2013, 8, e68332.
26. D. Pan, D. Zhang, J. Wu, C. Chen, Z. Xu, H. Yang and P. Zhou, Food Chem. Toxicol., 2014, 63, 111–118.
27. O. Trott and A. J. Olson, J. Comput. Chem., 2010, 31, 455–461.
28. S. M. Patil, S. Xu, S. R. Sheftic and A. T. Alexandrescu, J. Biol. Chem., 2009, 284, 11982–11991.
29. N. D. Younan and J. H. Viles, Biochemistry, 2015, 54, 4297–4306.
30. M. Biancalana, K. Makabe, A. Koide and S. Koide, J. Mol. Biol., 2009, 385, 1052–1063.
31. L. Nielsen, R. Khurana, A. Coats, S. Frokjaer, J. Brange, S. Vyas, V. N. Uversky and A. L. Fink, Biochemistry, 2001, 40, 6036–6046.
32. Y. Suzuki, J. R. Brender, K. Hartman, A. Ramamoorthy and E. N. G. Marsh, Biochemistry, 2012, 51, 8154–8162.
33. L. J. Ma, Y. T. Fu, L. L. Yu, X. L. Li, W. J. Zheng and T. F. Chen, RSC Adv., 2015, 5, 17405–17412.
34. J. D. Knight, J. A. Hebda and A. D. Miranker, Biochemistry, 2006, 45, 9496–9508.
35. G. Holzwart and P. Doty, J. Am. Chem. Soc., 1965, 87, 218–228.
36. N. Sreerama and R. W. Woody, in Numerical Computer Methods, Pt D, ed. L. Brand and M. L. Johnson, 2004, vol. 383, pp. 318–351.
37. C. A. De Carufel, N. Quittot, P. T. Nguyen and S. Bourgault, Angew. Chem., Int. Ed., 2015, 54, 14383–14387.
38. G. Singh and T. S. Kang, J. Phys. Chem. B, 2015, 119, 10573–10585.
39. Y. Li, W. X. Xu, Y. G. Mu and J. Z. H. Zhang, J. Chem. Phys., 2013, 139, 055102.
40. E. C. Lee, E. Ha, S. Singh, L. Legesse, S. Ahmad, E. Karnaukhova, R. P. Donaldson and A. M. Jeremic, Phys. Chem. Chem. Phys., 2013, 15, 12558–12571.
41. L. Yang, N. Li, W. Pan, Z. Yu and B. Tang, Anal. Chem., 2015, 87, 3678–3684.
42. L. E. Buchanan, E. B. Dunkelberger, H. Q. Tran, P. N. Cheng, C. C. Chiu, P. Cao, D. P. Raleigh, J. J. de Pablo, J. S. Nowick and M. T. Zanni, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 19285–19290.
43. H. H. Lee, T. S. Choi, S. J. C. Lee, J. W. Lee, J. Park, Y. H. Ko, W. J. Kim, K. Kim and H. I. Kim, Angew. Chem., Int. Ed., 2014, 53, 7461–7465.
44. P. Nedumpully-Govindan, A. Kakinen, E. H. Pilkington, T. P. Davis, P. Chun Ke and F. Ding, Sci. Rep., 2016, 6, 19463.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra23429a

---