Abstract

Semen-derived amyloid fibrils, comprising SEVI (semen-derived enhancer of viral infection) fibrils and SEM1 fibrils, could remarkably enhance HIV-1 sexual transmission and thus are potential targets for the development of an effective microbicide. Previously, we found that ADS-J1, apart from being an HIV-1 entry inhibitor, could also potently inhibit seminal amyloid fibrillization and block fibril-mediated enhancement of viral infection. However, the remodeling effects of ADS-J1 on mature seminal fibrils were unexplored. Herein, we investigated the capacity of ADS-J1 to disassemble seminal fibrils and the potential mode of action by applying several biophysical and biochemical measurements, combined with molecular dynamic (MD) simulations. We found that ADS-J1 effectively remodeled SEVI, SEM186–107 fibrils and endogenous seminal fibrils. Unlike epigallocatechin gallate (EGCG), a universal amyloid fibril breaker, ADS-J1 disaggregated SEVI fibrils into monomeric peptides, which was independent of oxidation reaction. MD simulations revealed that ADS-J1 displayed strong binding potency to the full-length PAP248–286 via electrostatic interactions, hydrophobic interactions and hydrogen bonds. ADS-J1 might initially bind to the fibrillar surface and then occupy the amyloid core, which eventually lead to fibril disassembly. Furthermore, the binding of ADS-J1 with PAP248–286 might induce conformational changes of PAP248–286. Disassembled PAP248–286 might not be favorable to re-aggregate into fibrils. ADS-J1 also exerts abilities to remodel a panel of amyloid fibrils, including Aβ1–42, hIAPP1–37 and EP2 fibrils. ADS-J1 displays promising potential to be a combination microbicide and an effective lead-product to treat amyloidogenic diseases.

Introduction

Sexual transmission, either via vaginal intercourse or via anal intercourse, is the predominant route of HIV-1 acquisition. Both oral and topical pre-exposure prophylaxes (PrEP) are promising approaches to reduce HIV-1 sexual transmission [1]. Although the oral administration of tenofovir disoproxil fumarate/emtricitabine regimen shows progress in HIV-1 prevention [2], the use of topical PrEP may be preferential over oral PrEP due to the rapid onset of protective drug levels at mucosal sites following administration and the potential reduction in systemic adverse effects [3]. The development of topical microbicide is still urgent and challenging.

Semen (SE), as an important carrier of HIV-1, plays critical roles in HIV-1 sexual transmission worldwide [4]. SE contains endogenous amyloid fibrils that could potently enhance HIV-1 infectivity and impair the anti-HIV activities of antiretroviral (ARV) drugs [5,6]. The seminal amyloid fibrils comprise proteolytic peptide fragments of prostatic acid phosphatase (PAP), a protein produced by the prostate gland. The best-characterized seminal amyloid is formed by PAP248–286 and termed SEVI (semen-derived enhancer of virus infection) [7]. Later, a second series of seminal fibrils were identified from prostate-specific antigen (PSA)-generated fragments of semenogelin (SEM1 and SEM2) and PAP85–120 [8,9] which could also facilitate HIV-1 infection. These amyloid fibrils contain a high abundance of positively charged arginine and lysine residues, which help to promote virion attachment and subsequent infection by allowing the virus to overcome the electrostatic repulsion between the negatively charged viral and cellular membranes [10].

Microbicide, simultaneously targeting virions and seminal amyloid, might be an innovative strategy to slow the global spread of HIV-1 [11,12]. Previously, our study showed that ADS-J1, an HIV-1 entry inhibitor targeting gp120 and gp41 [13,14], displayed potential abilities to counteract seminal amyloid fibril-mediated enhancement of viral infection. Due to its sulfonic groups, ADS-J1 binds to peptide PAP248–286 and inhibits SEVI fibril formation. In addition, it coats on and shields the positive charges of mature SEVI fibrils, thus antagonizing fibril-mediated enhancement of HIV-1 infection [3]. Unlike polyanions, which have failed in clinical trials, ADS-J1 does not accelerate SEVI fibril formation at low concentrations [15]. Importantly, ADS-J1 combined with ARV-based candidate microbicides displays synergistic effects against HIV-1 infection in SE, with little cytotoxicity to vaginal epithelial cells. With the anti-amyloid and anti-viral pharmaceutical effects, our results were in support of the potential application of ADS-J1 as a combination microbicide.

However, the remodeling effects of ADS-J1 on mature SEVI fibrils have been elusive. Endogenous mature fibrils are already formed and abundant in SE [6]. Thus, inhibition of the initial fibril formation would seem impractical or even futile during sexual intercourse. Although ADS-J1 could coat the amyloid surface and prevent interactions with the virus and cell surface, fibrils might also disassociate from ADS-J1 and gain chances to facilitate viral infection. A more desirable feature of ADS-J1, which could rapidly deconstruct seminal fibrils and thereby disrupt structures that promote HIV-1 infection, is highly warranted. In the present study, we intended to investigate the potential effects of ADS-J1 on remodeling seminal mature fibrils.

Materials and methods

Materials

Thioflavin T (ThT), EGCG, Triton X-100 and polybrene were purchased from Sigma (St. Louis, MO). Superoxide dismutase (SOD) and NP40 were purchased from Beyotime Biotechnology (Shanghai, China). Peptides, including PAP248–286, SEM186–107, hIAPP1–37 and EP2 peptide (>95% purity), were all synthesized by Scilight-Peptide (Beijing, China). SE samples were obtained from healthy members of laboratory with written informed consent. The protocol was approved by Human Ethics Committee of Southern Medical University, China. Seminal fluid (SE-F), representing the cell-free supernatant of SE, was collected by centrifugation for 30 min at 10 000 rpm at 4°C and stored in 1 ml aliquots at −20°C. TZM-bl cells were obtained from the Nation Institutes of Health AIDS Research and Reference Reagent Program. CCR5-tropic SF162, CXCR4-tropic NL4-3 and dual tropic infectious clone plasmids were kindly provided by Jan Münch of Ulm University, Ulm, Germany. Viruses were prepared as previously described [15]. ADS-J1 was purified by high-performance liquid chromatography (HPLC) from the crude material provided by Ciba Specialty Chemical Corp. (High Point, NC).

Fibril preparation

Lyophilized peptides of PAP248–286 and SEM186–107 were dissolved at a concentration of 2200 µM in phosphate-buffered saline (PBS) and stored at −20°C. Fibril formation was initiated by agitation at 37°C for 1–3 days at 1400 rpm with an Eppendorf Thermomixer. Fibrils of Aβ1–42 and hIAPP1–37 were prepared as previously described [16]. EP2 lyophilized peptides were dissolved in dimethyl sulfoxide and stored at −20°C. The peptides were dissolved in PBS and fibril formed immediately.

ThT staining assay

Preformed SEVI, SEM186–107, Aβ1–42, hIAPP1–37 and EP2 fibrils were agitated with or without ADS-J1 at 1400 rpm at 37°C with an Eppendorf Thermomixer. At the indicated time points, 10 µl aliquots were withdrawn from each sample for ThT staining as previously reported [15]. To assess the effects of ADS-J1 on disaggregating endogenous seminal fibrils, ADS-J1 at different concentrations or PBS was agitated with whole SE-F, and samples were withdrawn at different time points to detect fibril integrity by ThT staining.

Transmission electron microscopy (TEM)

Peptide samples were incubated in the absence or presence of ADS-J1 as described for the ThT fluorescence measurements. Samples were removed at different time points and subjected to examine their fibril shape by TEM as previously described [3].

Analysis of Tricine-SDS–PAGE

Mature SEVI fibrils were agitated with ADS-J1 or EGCG at 37°C at 1400 rpm for 12 h. Samples were centrifuged at 12 000 rpm for 15 min. The remodeling products in the supernatant were analyzed on 15% Tricine-SDS–PAGE gels. Gels were stained with Coomassie blue.

HIV-1 infection assays

TZM-bl cells (1 × 103) in 100 µl cell culture medium were seeded in 96-well flat-bottom plates the day before infection. Mature fibrils or whole SE-F were agitated in the presence or absence of ADS-J1 at various concentrations. At different time points, 60 µl samples were collected and centrifuged at 12 000 rpm for 10 min to remove the remaining free ADS-J1. The pellets were dissolved in 60 µl serum-free culture medium and mixed with 60 µl viruses (2 ng of p24 antigen). After 10 min, 100 µl of the mixtures was added to TZM-bl cells. After a 3-h incubation period, unbound viruses were removed and cells were replenished with fresh medium. Luciferase activity was measured 48-h post-infection [15]. SEVI or SEM186–107 fibrils were tested at a final concentration of 11 µM, while SE-F was used at a final dilution of 1:100.

Measurement of zeta potential and particle size

Fibrils were agitated with 2- or 4-fold excess of ADS-J1 at 1400 rpm at 37°C for 4 h. Fibrils were then centrifuged at 12 000 rpm for 10 min. The pellets were resuspended in 0.5 ml of 1 mM KCl. Samples were then loaded into a pre-rinsed folded capillary cell, and measurements were made at 25°C for all samples. At least three measurements were made for each sample. Both zeta potential and particle size were measured using the Zeta Nanosizer (Malvern, Worcestershire, U.K.).

Competition experiments

ADS-J1 (110 µM) was incubated with polybrene at different concentrations (0, 110, 220, 440 µM) at 37°C for 30 min and then the mixtures were centrifuged for 15 min at 12 000 rpm. The supernatants were agitated with SEVI. Fibrillar signal was examined by ThT staining assay as mentioned above.

To observe the hydrophobic interactions between ADS-J1 and SEVI, NP40 and Triton X-100, two neutral detergents, were agitated at 37°C at 1400 rpm with SEVI (30 µM) and ADS-J1 (60 µM). Fibril formation was detected by ThT staining method as described above.

SOD experiments

SEVI was agitated with ADS-J1 or EGCG in the presence or absence of SOD at 37°C for 4 h. The ability of ADS-J1 or EGCG to remodel SEVI fibrils in the presence of SOD was evaluated by ThT staining method.

MD simulation

Molecular dynamic (MD) was used to investigate the interaction between PAP248–286 and ADS-J1. The original structure co-ordinate for PAP248–286 was retrieved from the Protein Data Bank (PDB ID: 2L3H) [17]. The initial binding between PAP248–286 and ADS-J1 was searched by Autodock vina [18].

All MD simulations were performed 100 ns using the GROMACS 4.5.3 software package [19] together with CHARMM27 force field [20]. The system was placed in the center of a cubic box that extended 6 Å from the structure on each side. TIP3P water molecules were added together with sodium and chloride ions to offer isotonic condition and neutralize the whole system [21]. Then, the system was energy-minimized for 3000 steps by releasing all the constraints. All bond lengths were constrained by LINCS algorithm [22]. The electrostatic interaction was calculated by the particle mesh Ewald method [23]. The MD temperature was controlled at 300K and pressure was set at 1 atm. The MD simulation trajectories were saved in every 2 fs for analysis.

The most representative structure and contact map in each simulation system was gained by cluster analysis with a cutoff value of 0.3 nm. The root-mean-square deviation (RMSD) value presents the stability of two simulation systems. The electrostatic surface of PAP248-286 was generated by Pymol and the structures from the simulation trajectories were visualized using YASARA [24].

Results

Remodeling effects of ADS-J1 on SEVI and SEM186–107 fibrils

To investigate the effects of ADS-J1 on remodeling preformed seminal amyloid fibrils, we first detected its ability to remodel SEVI and SEM186–107 fibrils using ThT staining assay. In the absence of ADS-J1, 30 µM SEVI fibrils (Figure 1A) and SEM186–107 fibrils (Figure 1B), which have been agitated in PBS at 37°C for 1–3 days at 1400 rpm, remained in constant binding with ThT, a specific dye of amyloid fibrils. It indicated both fibrils were stable in the buffer within the tested period. However, in the presence of ADS-J1, the ThT fluorescence intensity of fibrils decreased by ∼50% in 4 h (Figure 1A,B). It suggested that the mature fibrils were converted into other form of species that lost the ability to bind ThT. Of note, ADS-J1 alone at tested concentrations did not influence ThT fluorescence (data not shown), indicating that ADS-J1 did not quench ThT fluorescence. Fibrillar disruption by ADS-J1 was also confirmed by TEM. Upon a 4-h incubation, TEM images revealed that the control SEVI fibrils and SEM1 fibrils remained a dense of intact and branched fibrils. In contrast, treated with 2-fold excess of ADS-J1, SEVI fibrils were converted to smaller nonfibrillar species and SEM1 fibrils were converted to shorter fibrils (Figure 1C). ADS-J1 alone was served as negative control. Particle size measurement was also applied to confirm our findings. The averaged particle sizes of SEVI and SEM1 fibrils were 3636 ± 208 nm and 2043 ± 307 nm, respectively. Treated with 4-fold excess of ADS-J1 for 4 h, the fibrils were disaggregated into smaller pieces with averaged particle sizes of 506 ± 14 nm and 543 ± 6.9 nm, respectively (Figure 1D).

Remodeling effects of ADS-J1 on SEVI and SEM186–107 fibrils.

Figure 1.
Remodeling effects of ADS-J1 on SEVI and SEM186–107 fibrils.

Fibrillar aggregates (30 µM) were exposed to different amounts of ADS-J1. Fibril integrities of SEVI (A) and SEM186–107 (B) were detected by ThT staining at the indicated time points. Values are means ± SEM (n = 3) from one of the three independent experiments, which yielded similar results. (C) Preformed SEVI fibrils and SEM186–107 fibrils (110 µM) were treated with a 2-fold excess of ADS-J1 or buffer for 4 h at 37°C. Fibrillar shapes were monitored by TEM. Images in each study group were from three independent experiments. Scale bar = 100 nm. (D) Particle sizes of SEVI and SEM186–107 fibrils agitated in the presence or absence of ADS-J1 at 37°C for 4 h were measured. Results shown are means ± SEM of triplicate measurements from one of the two independent experiments, which yielded equivalent results.

Figure 1.
Remodeling effects of ADS-J1 on SEVI and SEM186–107 fibrils.

Fibrillar aggregates (30 µM) were exposed to different amounts of ADS-J1. Fibril integrities of SEVI (A) and SEM186–107 (B) were detected by ThT staining at the indicated time points. Values are means ± SEM (n = 3) from one of the three independent experiments, which yielded similar results. (C) Preformed SEVI fibrils and SEM186–107 fibrils (110 µM) were treated with a 2-fold excess of ADS-J1 or buffer for 4 h at 37°C. Fibrillar shapes were monitored by TEM. Images in each study group were from three independent experiments. Scale bar = 100 nm. (D) Particle sizes of SEVI and SEM186–107 fibrils agitated in the presence or absence of ADS-J1 at 37°C for 4 h were measured. Results shown are means ± SEM of triplicate measurements from one of the two independent experiments, which yielded equivalent results.

Viral infection experiments were conducted to determine the activity of the disaggregating products to enhance HIV-1 infection. We found that the remodeling products showed impaired ability to facilitate HIV-1 infection. Although different fibrils enhanced various viral infections to a different extent, SEVI and SEM1 fibrils exerted stable activity to enhance HIV-1 infection within 48 h (Figure 2). ADS-J1 decreased the ability of SEVI (Figure 2A–C) and SEM1 fibrils (Figure 2D–F) to enhance HIV-1 infection in a dose- and time-dependent manner. An 8-fold molar excess of ADS-J1 could completely diminish fibril-mediated enhancement of HIV-1 infection. Taken together, our findings suggest that ADS-J1 effectively remodels SEVI and SEM1 fibrils and abrogates their infection enhancing property.

Abrogation of SEVI and SEM1 fibrils-mediated enhancement of HIV-1 infection by ADS-J1.

Figure 2.
Abrogation of SEVI and SEM1 fibrils-mediated enhancement of HIV-1 infection by ADS-J1.

SEVI at 11 µM (AC) or SEM186–107 fibrils at 11 µM (DF) were agitated with increasing concentrations of ADS-J1. At the indicated time points, samples were collected and centrifuged. The pellets were mixed with CCR5-tropic HIV-1SF162 (A and D), CXCR4-tropic HIV-1 NL4-3 (B and E) or dual tropic HIV-1 (C and F). The rate of infection was monitored by measuring the luciferase activity in TZM-bl cells. The data are means ± SEM of three independent measurements.

Figure 2.
Abrogation of SEVI and SEM1 fibrils-mediated enhancement of HIV-1 infection by ADS-J1.

SEVI at 11 µM (AC) or SEM186–107 fibrils at 11 µM (DF) were agitated with increasing concentrations of ADS-J1. At the indicated time points, samples were collected and centrifuged. The pellets were mixed with CCR5-tropic HIV-1SF162 (A and D), CXCR4-tropic HIV-1 NL4-3 (B and E) or dual tropic HIV-1 (C and F). The rate of infection was monitored by measuring the luciferase activity in TZM-bl cells. The data are means ± SEM of three independent measurements.

Remodeling effects of ADS-J1 on endogenous seminal fibrils

We then explored whether ADS-J1 could remodel endogenous fibrils in SE. We allowed fibrils to grow in SE by agitating SE-F for 8 h at 37°C, 1400 rpm, as evidenced by an increase in ThT fluorescence (data not shown). Agitated SE-F was exposed to different concentrations of ADS-J1 for 4 h. ADS-J1 could dose-dependently decrease ThT fluorescent signal in SE (Figure 3A). Accordingly, after exposure to ADS-J1, SE-F gradually lost the ability to enhance HIV-1 infection. ADS-J1 at 240 µM could completely block the enhancing property of SE-F (Figure 3B). These findings suggest that ADS-J1 could also remodel endogenous seminal amyloid fibrils.

Disaggregating effects of ADS-J1 on endogenous seminal fibrils.

Figure 3.
Disaggregating effects of ADS-J1 on endogenous seminal fibrils.

(A) Loss of ThT fluorescence after incubation of SE-F with various concentrations of ADS-J1 under constant agitation was monitored at the time point of 8 h. Values represent mean fluorescence intensities derived from triplicate measurements ± SEM. (B) SE-F was exposed to ADS-J1 at different concentrations at 37°C for 8 h. Mixtures were then centrifuged and pellets were incubated with HIV-1 SF162. Subsequently, mixtures were used to infect TZM-bl cells. Luciferase activities were determined after 48 h. Results are means ± SEM from triplicate measurements.

Figure 3.
Disaggregating effects of ADS-J1 on endogenous seminal fibrils.

(A) Loss of ThT fluorescence after incubation of SE-F with various concentrations of ADS-J1 under constant agitation was monitored at the time point of 8 h. Values represent mean fluorescence intensities derived from triplicate measurements ± SEM. (B) SE-F was exposed to ADS-J1 at different concentrations at 37°C for 8 h. Mixtures were then centrifuged and pellets were incubated with HIV-1 SF162. Subsequently, mixtures were used to infect TZM-bl cells. Luciferase activities were determined after 48 h. Results are means ± SEM from triplicate measurements.

ADS-J1 converting SEVI fibrils into monomers

We proceeded to investigate the ADS-J1's mode of action to remodel SEVI. Results from Tricine-SDS–PAGE [25,26] revealed that the free supernatant of SEVI fibrils ran with a molecular mass consistent with a PAP248–286 monomer, which indicated that SEVI fibrils contained peptide monomer (Figure 4A). When treated with increasing level of ADS-J1, the supernatant of SEVI displayed increasing intensity of PAP248–286 monomer, compared with SEVI control. Quantification of bands showed higher proportion of monomer in samples treated with ADS-J1 (Figure 4B). Compared with EGCG, a universal amyloid fibril breaker, limited smear or trapped material was seen in SEVI with EGCG. EGCG is a well-known antioxidant chemical. In our experiments, after incubation of EGCG with SEVI, the solution changed appearance from transparent to brown, suggesting EGCG underwent auto-oxidation. In the presence of SOD, a molecule that could decrease the auto-oxidation of EGCG [27], remodeling activities of EGCG were hampered by the rescue in ThT fluorescence compared with that in the absence of SOD. However, the remodeling effects of ADS-J1 were not influenced by SOD (Figure 4C).

Disaggregation of SEVI fibrils into monomers by ADS-J1.

Figure 4.
Disaggregation of SEVI fibrils into monomers by ADS-J1.

(A) PAP248-286 monomer increased after SEVI was treated with ADS-J1. SEVI was agitated with ADS-J1 or EGCG and centrifuged at 12 000 rpm for 15 min. The supernatants were electrophoresed in Tricine-SDS–PAGE. Gel was stained with Coomassie blue. Positions of molecular mass standards are shown on the left. PAP248–286 was loaded as internal control. (B) Bar chart shows that the quantification of peptide levels corresponds to the main band in the Coomassie blue staining by using Quantity One image analysis software. (C) The remodeling activities of ADS-J1 were independent of auto-oxidation. SEVI was agitated with ADS-J1 or EGCG in the presence or absence of SOD for 4 h. Mixtures were centrifuged. The remaining fibrils in the pellets were determined by ThT staining assay. Values represent mean ThT fluorescence intensity derived from triplicate measurements ± SEM. (D) The effects of ADS-J1 on remodeling SEVI were counteracted by polybrene. ADS-J1 was incubated with different concentration of polybrene at 37°C for 30 min and centrifuged. The supernatants were agitated with SEVI and fibril integrity at the time point of 4 h was assessed by ThT staining assay. Results are means ± SEM of triplicate measurements. (E) Surface charge of SEVI fibrils determined by zeta potential measurements. Fibrils were mixed with buffer or ADS-J1 at different concentrations and centrifuged for 10 min at 12 000 rpm. The pellets were resuspended in 0.5 ml of 1 mM KCl and zeta potential was measured. Values represent means ± SEM (n = 3). (F) The effects of ADS-J1 on remodeling SEVI were partially counteracted by neutral detergents. ADS-J1 was agitated with SEVI in the presence or absence of 3% Triton X-100 or 4% NP40, respectively. Fibril integrity at the time point of 4 h was assessed by ThT staining assay. Results are means ± SEM of triplicate measurements. (G) Remodeling effects of ADS-J1 on SEVI(Ala) fibrils. SEVI(Ala) fibrils (30 µM) were exposed to 60 µM of ADS-J1. Fibril integrities of SEVI(Ala) were detected by ThT staining at the indicated time points. Values are means ± SEM (n = 3) from one of the two independent experiments, which yielded similar results.

Figure 4.
Disaggregation of SEVI fibrils into monomers by ADS-J1.

(A) PAP248-286 monomer increased after SEVI was treated with ADS-J1. SEVI was agitated with ADS-J1 or EGCG and centrifuged at 12 000 rpm for 15 min. The supernatants were electrophoresed in Tricine-SDS–PAGE. Gel was stained with Coomassie blue. Positions of molecular mass standards are shown on the left. PAP248–286 was loaded as internal control. (B) Bar chart shows that the quantification of peptide levels corresponds to the main band in the Coomassie blue staining by using Quantity One image analysis software. (C) The remodeling activities of ADS-J1 were independent of auto-oxidation. SEVI was agitated with ADS-J1 or EGCG in the presence or absence of SOD for 4 h. Mixtures were centrifuged. The remaining fibrils in the pellets were determined by ThT staining assay. Values represent mean ThT fluorescence intensity derived from triplicate measurements ± SEM. (D) The effects of ADS-J1 on remodeling SEVI were counteracted by polybrene. ADS-J1 was incubated with different concentration of polybrene at 37°C for 30 min and centrifuged. The supernatants were agitated with SEVI and fibril integrity at the time point of 4 h was assessed by ThT staining assay. Results are means ± SEM of triplicate measurements. (E) Surface charge of SEVI fibrils determined by zeta potential measurements. Fibrils were mixed with buffer or ADS-J1 at different concentrations and centrifuged for 10 min at 12 000 rpm. The pellets were resuspended in 0.5 ml of 1 mM KCl and zeta potential was measured. Values represent means ± SEM (n = 3). (F) The effects of ADS-J1 on remodeling SEVI were partially counteracted by neutral detergents. ADS-J1 was agitated with SEVI in the presence or absence of 3% Triton X-100 or 4% NP40, respectively. Fibril integrity at the time point of 4 h was assessed by ThT staining assay. Results are means ± SEM of triplicate measurements. (G) Remodeling effects of ADS-J1 on SEVI(Ala) fibrils. SEVI(Ala) fibrils (30 µM) were exposed to 60 µM of ADS-J1. Fibril integrities of SEVI(Ala) were detected by ThT staining at the indicated time points. Values are means ± SEM (n = 3) from one of the two independent experiments, which yielded similar results.

Using a competitive experiment, we also found that pretreatment of ADS-J1 with polybrene, a cationic polymer, reversed the fibril remodeling properties of ADS-J1 (Figure 4D). Polybrene displayed no effects on SEVI fibrils integrity. However, pretreatment with increasing levels of polybrene, ADS-J1 gradually lost the ability to remodel SEVI fibrils. At an equimolar ratio of ADS-J1-polybrene (1:1), polybrene nearly reversed the remodeling effects of ADS-J1 on SEVI fibrils, suggesting that polybrene and fibrils compete for binding to ADS-J1. The zeta potential measurement revealed that ADS-J1 neutralized the positive surface charge of SEVI. The zeta potential of SEVI (67 µM) significantly decreased from 20.5 ± 0.6 mv to −39.4 ± 0.8 mv, as the concentration of ADS-J1 was increased from 0 µM to 268 µM (Figure 4E). Of note, we found that NP40 and Triton X-100, two neutral detergents, could both antagonize the remodeling effects of ADS-J1 on SEVI fibrils (Figure 4F). More interestingly, ADS-J1 also displayed potent capability to remodel neutral mutant SEVI(Ala) fibrils, in which all the positive residues in PAP248–286 were replaced with alanines (Figure 4G). These results show that ADS-J1 could disaggregate SEVI fibrils into monomer and, electrostatic interaction as well as hydrophobic interaction, but not oxidation function, are key determinants for ADS-J1's remodeling capability, differing from that of EGCG.

Strong binding potency of ADS-J1 to PAP248–286 revealed by MD simulation

MD simulation was applied to investigate the potential mechanism of the remodeling activities of ADS-J1 (Figure 5A) on mature SEVI fibrils. Figure 5B showed the structure of the electrostatic surface of PAP248–286 (positive charge shown in blue) with ADS-J1 molecules. PAP248–286 interacted with eight ADS-J1 molecules, which fully covered on the surface of PAP248–286. Since electrostatic interactions play vital roles in the remodeling effects of ADS-J1 on SEVI fibrils, we analyzed the ionic interactions between the S atom on ADS-J1 molecules and the N atom of the lysine side chain (or carbon atom of the guanidinium group of arginine) in PAP248–286. All eight positive charged residues in PAP248–286 were bound to ADS-J1 (Figure 5C). It showed the minimum X-S distances in the initial structure and average structure respectively for each residue.

Representative PAP248–286–ADS-J1 binding complexes and potential interactions derived from MD simulation.

Figure 5.
Representative PAP248–286–ADS-J1 binding complexes and potential interactions derived from MD simulation.

(A) Chemical structure of ADS-J1. (B) The bottom (left panel) and back (right panel) views of the structure of PAP248–286 with eight ADS-J1 molecules. PAP248–286 is shown its electrostatic potential and colored according to the charge of atoms. Positively and negatively charged zones are visualized in blue and red, respectively. ADS-J1 is shown as stuck and colored by element. (C) ADS-J1 establishes electrostatic interaction with PAP248–286 as shown by decreased X-S distances (nm) between the S atom of sulfonic groups on ADS-J1 molecules and the N atom of the lysine side chain (or carbon atom of the guanidinium group of arginine) of the average structure, compared with that of the initial structure. (D) The minimum distance of ADS-J1 interacted with PAP248–286 residues via hydrogen bond and hydrophobic interaction. (E) Sequence of PAP248–286. The N-terminal domain (ND), central domain and C-terminal domain (CD) are depicted in red, blue and green, respectively. The residues colored in purple indicate the positively charged amino acid. Black asterisks and blue asterisks represent the PAP248–286 residues interacting with ADS-J1 via hydrogen bond and hydrophobic interaction, respectively. Predicted steric-zipper hexapeptides are underlined.

Figure 5.
Representative PAP248–286–ADS-J1 binding complexes and potential interactions derived from MD simulation.

(A) Chemical structure of ADS-J1. (B) The bottom (left panel) and back (right panel) views of the structure of PAP248–286 with eight ADS-J1 molecules. PAP248–286 is shown its electrostatic potential and colored according to the charge of atoms. Positively and negatively charged zones are visualized in blue and red, respectively. ADS-J1 is shown as stuck and colored by element. (C) ADS-J1 establishes electrostatic interaction with PAP248–286 as shown by decreased X-S distances (nm) between the S atom of sulfonic groups on ADS-J1 molecules and the N atom of the lysine side chain (or carbon atom of the guanidinium group of arginine) of the average structure, compared with that of the initial structure. (D) The minimum distance of ADS-J1 interacted with PAP248–286 residues via hydrogen bond and hydrophobic interaction. (E) Sequence of PAP248–286. The N-terminal domain (ND), central domain and C-terminal domain (CD) are depicted in red, blue and green, respectively. The residues colored in purple indicate the positively charged amino acid. Black asterisks and blue asterisks represent the PAP248–286 residues interacting with ADS-J1 via hydrogen bond and hydrophobic interaction, respectively. Predicted steric-zipper hexapeptides are underlined.

Besides electrostatic interactions, PAP248–286 and ADS-J1 also formed lots of hydrophobic interactions and hydrogen bonds (Figure 5D). Most of the PAP248–286 residues contacted with ADS-J1 molecules via hydrophobic interactions. These results suggested that almost all residues in PAP248–286 (35 out of 39) were occupied by ADS-J1 molecules (Figure 5E). Electrostatic interaction, hydrogen bond and hydrophobic interaction are the main driving forces for the binding of ADS-J1 to the PAP248–286. The results from MD simulation were in agreement with those from biochemical experiments (Figure 5D,F).

Secondary structure of PAP248–286 in the presence of ADS-J1 differing from that in the absence of ADS-J1

The binding of ADS-J1 with PAP248–286 results in important conformational changes in PAP248–286. The contact maps show the distances between pairs of amino acid residues in Figure 6A. PAP248–286 alone adopts a kinked conformation, in which the C-terminus interacts with the N-terminus via hydrogen bonds leading to a U-turn in the central region. However, ADS-J1 binding converts PAP248–286 from the compact conformation to the extended one. The average distances between the N-terminal Arg257/Leu258 with Ile284/Met285/Tyr286 in C-terminal were calculated over the two MD trajectories to further explain the difference between the conformational ensembles of PAP248–286 and PAP248–286 with ADS-J1 molecules. Figure 6B,C show that the distances between Arg257/Leu258 and Ile284/Met285/Tyr286, respectively, are stable ∼0.5 nm in the absence of ADS-J1. In the presence of ADS-J1, however, the distances between Arg257–Tyr286, Leu258–Tyr286 and Arg257–Met285 become greater than 1 nm and distances between Arg257–Ile284, Leu258–Met285 and Leu258–Ile284 increase to more than 1.5 nm. The root-mean-square deviations (RMSDs) for the two systems are shown in Figure 6D. The plots reveal that both the systems reach equilibrium within 100 ns simulations and remain stable.

Induction of conformational changes of PAP248-286 by ADS-J1.

Figure 6.
Induction of conformational changes of PAP248-286 by ADS-J1.

(A) Contact map of residues in PAP248–286 in the presence (left) and absence of ADS-J1 (right). The inserts are the most representative structures of PAP248–286 obtained from the 100 ns simulation. ADS-J1 molecules are not shown. (B,C) The time-dependent changes of distance between the mass center of the Arg257–Ile284, Arg257–Met285, Arg257–Tyr286, Leu258–Ile284, Leu258–Met285, Leu258–Tyr286 in the presence (B) and absence (C) of ADS-J1. (D) Root-mean-square deviations (RMSD) plotted against time for PAP248–286 only and PAP248–286 + ADS-J1 systems, respectively.

Figure 6.
Induction of conformational changes of PAP248-286 by ADS-J1.

(A) Contact map of residues in PAP248–286 in the presence (left) and absence of ADS-J1 (right). The inserts are the most representative structures of PAP248–286 obtained from the 100 ns simulation. ADS-J1 molecules are not shown. (B,C) The time-dependent changes of distance between the mass center of the Arg257–Ile284, Arg257–Met285, Arg257–Tyr286, Leu258–Ile284, Leu258–Met285, Leu258–Tyr286 in the presence (B) and absence (C) of ADS-J1. (D) Root-mean-square deviations (RMSD) plotted against time for PAP248–286 only and PAP248–286 + ADS-J1 systems, respectively.

Remodeling effects of ADS-J1 on a wide spectrum of amyloid fibrils

Many amyloid fibrils are the pathogenic factors of several diseases, such as Aβ1–42 in Alzheimer's disease and Amylin (fibrillar form of hIAPP1–37) in type II diabetes [28,29]. In our previous study, we showed that EP2 (enhancing peptide 2) derived from HIV-1 gp120 formed amyloid fibrils and greatly enhanced HIV-1 infection [30]. We explored the remodeling effects of ADS-J1 on these amyloid fibrils. We found that ADS-J1 could effectively disassemble Aβ fibrils, amylin and EP2 fibrils (Figure 7A–C, upper panel) in a dose-dependent manner, as evidenced by the decrease in ThT fluorescence after incubation of ADS-J1 with each fibril. TEM analysis was also applied to monitor the remodeling products. It is clear that the fibrils were reduced markedly and converted into some granular aggregates after incubation with ADS-J1 for 4 h, while untreated fibrils kept the well-developed fibrillar morphology (Figure 7A–C, bottom panel). The remodeling effects were also confirmed by particle size measurement. We observed that 4-fold excess of ADS-J1 drastically disrupted three fibrils into smaller structures (Figure 7D). These findings suggest that ADS-J1 could remodel a wide range of amyloid fibrils.

Remodeling effects of ADS-J1 on a wide spectrum of amyloid fibrils.

Figure 7.
Remodeling effects of ADS-J1 on a wide spectrum of amyloid fibrils.

ADS-J1 could remodel preformed fibrils of Aβ1−42 (A), hIAPP1–37 (B) and EP2 (C) fibrils. Each of the mature fibrils was agitated in the presence or absence of ADS-J1 at molar ratio of fibril:ADS-J1 = 1:0, 1:2, 1:4, 1:8. Aliquots were withdrawn at various time points and fibril integrities were assessed using ThT. Results are means ± SEM of three replicate experiments. The fibril structures in the absence (bottom, left) and presence (bottom, right) of ADS-J1 after agitation for 4 h with molar ratios of fibrils:ADS-J1 (1:4) were visualized by TEM. Scale bar = 100 nm. (D) Particle sizes of various fibrils agitated in the presence or absence of ADS-J1 for 4 h were measured. Results shown are means ± SEM of triplicate measurements from one of the two independent experiments, which yielded similar results.

Figure 7.
Remodeling effects of ADS-J1 on a wide spectrum of amyloid fibrils.

ADS-J1 could remodel preformed fibrils of Aβ1−42 (A), hIAPP1–37 (B) and EP2 (C) fibrils. Each of the mature fibrils was agitated in the presence or absence of ADS-J1 at molar ratio of fibril:ADS-J1 = 1:0, 1:2, 1:4, 1:8. Aliquots were withdrawn at various time points and fibril integrities were assessed using ThT. Results are means ± SEM of three replicate experiments. The fibril structures in the absence (bottom, left) and presence (bottom, right) of ADS-J1 after agitation for 4 h with molar ratios of fibrils:ADS-J1 (1:4) were visualized by TEM. Scale bar = 100 nm. (D) Particle sizes of various fibrils agitated in the presence or absence of ADS-J1 for 4 h were measured. Results shown are means ± SEM of triplicate measurements from one of the two independent experiments, which yielded similar results.

Discussion

There is no effective cure to prevent HIV-1 sexual transmission topically. Therefore, therapeutic options capable of counteracting the factors involved in the sexual transmission process are actively sought [12]. In the present study, we found that ADS-J1, apart from being an HIV-1 entry inhibitor, SEVI fibril inhibitor and blocker [3,13,31], was also an effective SEVI fibril disassembler. We found that ADS-J1 could effectively disrupt seminal amyloid fibrils, including SEVI fibrils and SEM1 fibrils as evidenced by loss of ThT fluorescence, disappearance of fibrils in TEM images, reduced particle size (Figure 1), impaired ability to enhance HIV-1 infection (Figure 2) and accumulated peptide monomer in remodeled products (Figure 4A,B). ADS-J1 could also disrupt endogenous amyloid fibrils in SE and impair its ability to enhance HIV-1 infection (Figure 3). Importantly, our results showed that ADS-J1 could effectively disrupt SEVI and SEM1 fibrils within 4 h, a time scale that would be critical for prevention of HIV-1 infection (Figure 1).

ADS-J1 is also capable of disassembling multiple disease-related amyloid fibrils, such as Aβ1–42 fibrils, hIAPP1–37 fibrils, and EP2 fibrils (Figure 7). As we know, there is limited effective amyloid reversing agent lies in the exceptional stability of the amyloid fibrils, which possess high resistance to proteases, protein denaturants, temperatures up to 98°C and 2% SDS [11]. So far, the most reported fibril breakers can be divided into three categories [32]. The first class is peptide based inhibitors, which are derived from different regions of the amyloidogenic peptide sequence. The homologous fragment could bind to a complementary sequence in the full-length protein, which results in interfering with subsequent protein self-assembly and favoring to disassemble preformed fibrils [33,34]. Some drawbacks have limited the clinical use of β-sheet breaker peptides, such as the cost of preparation, hydrosolubility, and oral bioavailability. The second class is ionic agents. Similar to ADS-J1, CLR01, a Lys-specific molecular tweezer with its functional phosphate groups, has been reported to be capable of disassembling several amyloid fibrils as those also investigated in this study [3537]. Of note, 2-fold excess of ADS-J1 could effectively remodel SEVI fibrils compared with CLR01 which required 10-fold excess [35]. In addition, CLR01 fails to disassemble SEM1 fibrils and it takes several weeks to disassemble Aβ or α-synuclein [35], while ADS-J1 displays strong ability to remodel a wide range of amyloid fibrils within 4 h (Figure 7). The third class of drugs is some well-known polyphenols, including EGCG, resveratrol, curcumin, oleuropein and scyllo-inositol [32]. Our results suggest that the disassembling process of ADS-J1 on SEVI fibrils is distinct from that of EGCG, a universal fibril breaker to date. The remodeling effects of EGCG on SEVI fibrils appear to be dependent on its auto-oxidation function (Figure 4C). EGCG undergoes auto-oxidation and generates superoxide and quinones in the presence of oxygen [38]. PAP248–286 has six lysines and it is well established that quinones can react with the free amines comprising the N-terminus or the lysine side chains of proteins. It's possible that oxidized EGCG induce profound chemical modification of the peptide side chains, resulting in disruption the β-sheet alignment and fibril disassembly. It could explain why the remodeling products of SEVI fibrils by EGCG were not recognized by SDS PAGE (Figure 4A). The antioxidant function or the auto-oxidation ability of EGCG might be responsible for its ability to interfere with many amyloid fibrils and to interact with numerous protein targets involved in cancer, metabolic, neurodegenerative, inflammation and microbial diseases, etc. [39]. However, to succeed in clinical application, several problems of polyphenol are still to be solved, including solubility, stability, potential hepatotoxicity [40,41], inadequate target specificity and potential antagonizing effects in combination therapy [42,43]. ADS-J1 exerts limited oxidative capability. It might be associated with less host cytotoxicity and adverse events.

We then questioned why ADS-J1 poses excellent ability to disassemble various amyloid fibrils. We used SEVI as a model protein to elucidate the structure details of ADS-J1's binding mode. Its structure might be a key determinant since it has not only ion group and also benzene and naphthalene as hydrophobic sidewall (Figure 5A). Results from biochemical experiments (Figure 4D–F) and MD simulation (Figure 6C–E) suggest that ADS-J1 displays strong binding propensity to full-length PAP248–286 or SEVI via not only electrostatic interaction but also hydrophobic interaction and hydrogen bond. These results were similar to the observation that CLR03, a derivative of CLR01, lost the ability to disassemble amyloid fibrils because of lacking hydrophobic side chains [35]. Based on these findings, we proposed a possible mechanistic model to interpret the disassembling effects of ADS-J1 on SEVI fibrils in Figure 8. ADS-J1 might firstly attach to the surface of SEVI fibrils, possibly via the positively charged residues, which are essential for their enhancement of HIV-1 infection. Then the strong binding propensity, including hydrophobic interactions and hydrogen bonds, leads to the penetration of ADS-J1 into the inner fibrillar structure by occupying the amyloid core sequence. ADS-J1 engaged key residues, which reside in predicted steric zippers (Figure 5E) [11,44]. Steric zipper is specific residue order in a protein that has high propensity for fibrillation. If an inhibitor could interact with the steric zipper sequence, it might prevent the protein from aggregation into fibrils. ADS-J1 could interact with the predicted steric-zipper amino acid sequence in PAP248–286 (Figure 5E) [11,44]. Invaded by ADS-J1 molecules, fibrillar structures might be gradually loosened and cut into oligomers and then eventually monomers (Figure 8B).

Schematic representation for the potential mechanism by which ADS-J1 disassembling SEVI fibrils.

Figure 8.
Schematic representation for the potential mechanism by which ADS-J1 disassembling SEVI fibrils.

(A) PAP248-286 adopts a compact conformation in which the N-terminus interacts with C-terminus, leaving a deep U-turn in the central region. This conformation might be critical for the fibril nucleation and elongation. (B) ADS-J1 might firstly bind to the fibrils partially via electrostatic interaction. Then ADS-J1 molecules penetrate into the fibrils, occupying the amyloid core regions via hydrogen bonds, hydrophobic interactions and ionic interactions. Finally, fibrils were degraded into oligomers and eventually monomers. (C) The disaggregated PAP248–286 might lose the vital conformation resulting in its deficiency in re-aggregating into fibrils.

Figure 8.
Schematic representation for the potential mechanism by which ADS-J1 disassembling SEVI fibrils.

(A) PAP248-286 adopts a compact conformation in which the N-terminus interacts with C-terminus, leaving a deep U-turn in the central region. This conformation might be critical for the fibril nucleation and elongation. (B) ADS-J1 might firstly bind to the fibrils partially via electrostatic interaction. Then ADS-J1 molecules penetrate into the fibrils, occupying the amyloid core regions via hydrogen bonds, hydrophobic interactions and ionic interactions. Finally, fibrils were degraded into oligomers and eventually monomers. (C) The disaggregated PAP248–286 might lose the vital conformation resulting in its deficiency in re-aggregating into fibrils.

Furthermore, maintaining specific conformational states is reported to be vital to form β-sheet-rich amyloid fibrils [45]. As revealed by the MD images, PAP248–286 alone might adopt a compact conformation, in which the C-terminus interacts with the N-terminus leading to a deep U-turn in the central region. This unique structure might be essential to facilitate peptide aggregation (Figure 8A). However, ADS-J1 hinders the interactions between the C-terminal and the N-terminus and disturbs this potential aggregation-prone structure. Therefore, ADS-J1 could completely eliminate SEVI fibrils and yield fragments with reduced ability to form fibrils again due to the disfavoring conformations in the monomer capable of nucleation or elongation (Figure 8C).

Conclusion

Our study shows that ADS-J1 is a broad spectrum of amyloid fibril breaker. It might be a suitable option for using as a topical microbicide component since it is an effective inhibitor and breaker of seminal fibrils originating from various amyloidogenic peptides. Anti-amyloid treatment will synergize with anti-viral therapy, which appears encouraging for assessing therapies against the multifactorial nature of HIV-1 sexual transmission. In addition, the fact that ADS-J1 disaggregates various amyloid fibrils opens new perspectives for pharmaceutical applications and stimulates the synthesis of new, patentable agents towards amyloidogenic diseases.

Abbreviations

     
  • ARV

    antiretroviral

  •  
  • EGCG

    epigallocatechin gallate

  •  
  • HIV-1

    human immunodeficiency virus type I

  •  
  • HPLC

    high-performance liquid chromatography

  •  
  • MD

    molecular dynamic

  •  
  • PAP

    prostatic acid phosphatase

  •  
  • PBS

    phosphate-buffered saline

  •  
  • PrEP

    pre-exposure prophylaxis

  •  
  • PSA

    prostate-specific antigen

  •  
  • RMSF

    root-mean-square fluctuation

  •  
  • SE

    semen

  •  
  • SE-F

    seminal fluid

  •  
  • SEM

    semenogelin

  •  
  • SEVI

    semen-derived enhancer of virus infection

  •  
  • SOD

    superoxide dismutase

  •  
  • TEM

    transmission electron microscopy

  •  
  • ThT

    Thioflavin T

Author Contribution

S.T. conceived and designed the project. J.L., M.Q., W.L., and Z.L. performed experiments and analyzed the data. H.L., T.Z., and Z.Y. performed MD simulation. T.Q. and Y.Q. collected semen sample, L.L., X.Z., S.L., and S.T. discussed the results and commented on the manuscript.

Funding

This work was supported by grants from Pearl River S&T Nova Program of Guangzhou [201610010132], the Program for Outstanding Young Teacher in Guangdong Province [YQ2015039], the Natural Science Foundation of Guangdong Province [2017A030313598, 2015A030313241], and the Natural Science Foundation of China [81772194 to S.T., 81773787 to S.L., 41376162 to X.Z., and 21603095 to T.Z.]. Special Program for Applied Research on Super Computation of the NSFC-Guangdong Joint Fund (the second phase) under Grant No. U1501501.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Baeten
,
J.
and
Celum
,
C.
(
2013
)
Systemic and topical drugs for the prevention of HIV infection: antiretroviral pre-exposure prophylaxis
.
Annu. Rev. Med.
64
,
219
232
2
Nunes
,
R.
,
Sarmento
,
B.
and
das Neves
,
J.
(
2014
)
Formulation and delivery of anti-HIV rectal microbicides: advances and challenges
.
J. Control. Release
194
,
278
294
3
Xun
,
T.
,
Li
,
W.
,
Chen
,
J.
,
Yu
,
F.
,
Xu
,
W.
,
Wang
,
Q.
et al.  (
2015
)
ADS-J1 inhibits semen-derived amyloid fibril formation and blocks fibril-mediated enhancement of HIV-1 infection
.
Antimicrob. Agents Chemother.
59
,
5123
5134
4
Mujugira
,
A.
,
Coombs
,
R.W.
,
Heffron
,
R.
,
Celum
,
C.
,
Ronald
,
A.
,
Mugo
,
N.
et al.  (
2016
)
Seminal HIV-1 RNA detection in heterosexual African men initiating antiretroviral therapy
.
J. Infect. Dis.
214
,
212
215
5
Zirafi
,
O.
,
Kim
,
K.A.
,
Roan
,
N.R.
,
Kluge
,
S.F.
,
Müller
,
J.A.
,
Jiang
,
S.
et al.  (
2014
)
Semen enhances HIV infectivity and impairs the antiviral efficacy of microbicides
.
Sci. Transl. Med.
6
,
262ra157
6
Usmani
,
S.M.
,
Zirafi
,
O.
,
Müller
,
J.A.
,
Sandi-Monroy
,
N.L.
,
Yadav
,
J.K.
,
Meier
,
C.
et al.  (
2014
)
Direct visualization of HIV-enhancing endogenous amyloid fibrils in human semen
.
Nat. Commun.
5
,
3508
7
Münch
,
J.
,
Rücker
,
E.
,
Ständker
,
L.
,
Adermann
,
K.
,
Goffinet
,
C.
,
Schindler
,
M.
et al.  (
2007
)
Semen-derived amyloid fibrils drastically enhance HIV infection
.
Cell
131
,
1059
1071
8
Roan
,
N.R.
,
Müller
,
J.A.
,
Liu
,
H.
,
Chu
,
S.
,
Arnold
,
F.
,
Stürzel
,
C.M.
et al.  (
2011
)
Peptides released by physiological cleavage of semen coagulum proteins form amyloids that enhance HIV infection
.
Cell Host Microbe
10
,
541
550
9
Roan
,
N.R.
,
Liu
,
H.
,
Usmani
,
S.M.
,
Neidleman
,
J.
,
Müller
,
J.A.
,
Avila-Herrera
,
A.
et al.  (
2014
)
Liquefaction of semen generates and later degrades a conserved semenogelin peptide that enhances HIV infection
.
J. Virol.
88
,
7221
7234
10
Roan
,
N.R.
,
Münch
,
J.
,
Arhel
,
N.
,
Mothes
,
W.
,
Neidleman
,
J.
,
Kobayashi
,
A.
et al.  (
2009
)
The cationic properties of SEVI underlie its ability to enhance human immunodeficiency virus infection
.
J. Virol.
83
,
73
80
11
Castellano
,
L.M.
and
Shorter
,
J.
(
2012
)
The surprising role of amyloid fibrils in HIV infection
.
Biology
1
,
58
80
12
Sheik
,
D.A.
,
Dewhurst
,
S.
and
Yang
,
J.
(
2017
)
Natural seminal amyloids as targets for development of synthetic inhibitors of HIV transmission
.
Acc. Chem. Res.
50
,
8
13
Yu
,
F.
,
Lu
,
L.
,
Liu
,
Q.
,
Yu
,
X.
,
Wang
,
L.
,
He
,
E.
et al.  (
2014
)
ADS-J1 inhibits HIV-1 infection and membrane fusion by targeting the highly conserved pocket in the gp41 NHR-trimer
.
Biochim. Biophys. Acta
1838
,
1296
1305
14
Gonzalez-Ortega
,
E.
,
Mena
,
M.P.
,
Permanyer
,
M.
,
Ballana
,
E.
,
Clotet
,
B.
and
Este
,
J.A.
(
2010
)
ADS-J1 inhibits HIV-1 entry by interacting with gp120 and does not block fusion-active gp41 core formation
.
Antimicrob. Agents Chemother.
54
,
4487
4492
15
Tan
,
S.
,
Lu
,
L.
,
Li
,
L.
,
Liu
,
J.
,
Oksov
,
Y.
,
Lu
,
H.
et al.  (
2013
)
Polyanionic candidate microbicides accelerate the formation of semen-derived amyloid fibrils to enhance HIV-1 infection
.
PLoS ONE
8
,
e59777
16
Lopes
,
D.H.
,
Attar
,
A.
,
Nair
,
G.
,
Hayden
,
E.Y.
,
Du
,
Z.
,
McDaniel
,
K.
et al.  (
2015
)
Molecular tweezers inhibit islet amyloid polypeptide assembly and toxicity by a new mechanism
.
ACS Chem. Biol.
10
,
1555
1569
17
Nanga
,
R.P.
,
Brender
,
J.R.
,
Vivekanandan
,
S.
,
Popovych
,
N.
and
Ramamoorthy
,
A.
(
2009
)
NMR structure in a membrane environment reveals putative amyloidogenic regions of the SEVI precursor peptide PAP(248-286)
.
J. Am. Chem. Soc.
131
,
17972
9
18
Trott
,
O.
and
Olson
,
A.J.
(
2010
)
AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading
.
J. Comput. Chem.
31
,
455
461
19
Pronk
,
S.
,
Páll
,
S.
,
Schulz
,
R.
,
Larsson
,
P.
,
Bjelkmar
,
P.
,
Apostolov
,
R.
et al.  (
2013
)
GROMACS 4.5: a high-throughput and highly parallel open source molecular simulation toolkit
.
Bioinformatics
29
,
845
854
20
Brooks
,
B.R.
,
Bruccoleri
,
R.E.
,
Olafson
,
B.D.
,
States
,
D.J.
,
Swaminathan
,
S.
and
Karplus
,
M.
(
1983
)
CHARMM: a program for macromolecular energy, minimization, and dynamics calculations
.
J. Comput. Chem.
4
,
31
21
Jorgensen
,
W.L.
,
Chandrasekhar
,
J.
,
Madura
,
J.D.
,
Impey
,
R.W.
and
Klein
,
M.L.
(
1983
)
Comparison of simple potential functions for simulating liquid water
.
J. Chem. Phys.
79
,
12
22
Hess
,
B.
,
Bekker
,
H.
,
Berendesen
,
H.J.C.
and
Fraaije
,
J.G.E.M.
(
1997
)
LINCS: a linear constraint solver for molecular simulations
.
J. Comput. Chem.
18
,
10
23
Darden
,
T.
,
York
,
D.
and
Pedersen
,
L.
(
1993
)
Particle mesh Ewald: an N log(N) method for Ewald sums in large systems
.
J. Chem. Phys.
98
,
4
24
Krieger
,
E.
and
Vriend
,
G.
(
2014
)
YASARA View – molecular graphics for all devices – from smartphones to workstations
.
Bioinformatics
30
,
2981
2982
25
Bitan
,
G.
,
Lomakin
,
A.
and
Teplow
,
D.B.
(
2001
)
Amyloid β-protein oligomerization: prenucleation interactions revealed by photo-induced cross-linking of unmodified proteins
.
J. Biol. Chem.
276
,
35176
35184
26
Schägger
,
H.
(
2006
)
Tricine-SDS-PAGE
.
Nat. Protoc.
1
,
16
22
27
Palhano
,
F.L.
,
Lee
,
J.
,
Grimster
,
N.P.
and
Kelly
,
J.W.
(
2013
)
Toward the molecular mechanism(s) by which EGCG treatment remodels mature amyloid fibrils
.
J. Am. Chem. Soc.
135
,
7503
7510
28
Rushing
,
P.A.
,
Hagan
,
M.M.
,
Seeley
,
R.J.
,
Lutz
,
T.A.
and
Woods
,
S.C.
(
2000
)
Amylin: a novel action in the brain to reduce body weight
.
Endocrinology
141
,
850
853
29
Akter
,
R.
,
Cao
,
P.
,
Noor
,
H.
,
Ridgway
,
Z.
,
Tu
,
L.H.
,
Wang
,
H.
et al.  (
2016
)
Islet amyloid polypeptide: structure, function, and pathophysiology
.
J. Diabetes Res.
2016
,
1
18
30
Tan
,
S.
,
Li
,
L.
,
Lu
,
L.
,
Pan
,
C.
,
Lu
,
H.
,
Oksov
,
Y.
et al.  (
2014
)
Peptides derived from HIV-1 gp120 co-receptor binding domain form amyloid fibrils and enhance HIV-1 infection
.
FEBS Lett.
588
,
1515
1522
31
Jiang
,
S.
,
Lin
,
K.
,
Zhang
,
L.
and
Debnath
,
A.K.
(
1999
)
A screening assay for antiviral compounds targeted to the HIV-1 gp41 core structure using a conformation-specific monoclonal antibody
.
J. Virol. Methods
80
,
85
96
32
Francioso
,
A.
,
Punzi
,
P.
,
Boffi
,
A.
,
Lori
,
C.
,
Martire
,
S.
,
Giordano
,
C.
et al.  (
2015
)
β-Sheet interfering molecules acting against β-amyloid aggregation and fibrillogenesis
.
Bioorg. Med. Chem.
23
,
1671
1683
33
Li
,
H.
,
Rahimi
,
F.
and
Bitan
,
G.
(
2016
)
Modulation of amyloid β-protein (Aβ) assembly by homologous C-terminal fragments as a strategy for inhibiting aβ toxicity
.
ACS Chem. Neurosci.
7
,
845
856
34
Paul
,
A.
,
Kalita
,
S.
,
Kalita
,
S.
,
Sukumar
,
P.
and
Mandal
,
B.
(
2017
)
Disaggregation of amylin aggregate by novel conformationally restricted aminobenzoic acid containing α/β and α/γ hybrid peptidomimetics
.
Sci. Rep.
7
,
40095
35
Lump
,
E.
,
Castellano
,
L.M.
,
Meier
,
C.
,
Seeliger
,
J.
,
Erwin
,
N.
,
Sperlich
,
B.
et al.  (
2015
)
A molecular tweezer antagonizes seminal amyloids and HIV infection
.
eLife
4
,
e05397
36
Sinha
,
S.
,
Du
,
Z.
,
Maiti
,
P.
,
Klärner
,
F.G.
,
Schrader
,
T.
,
Wang
,
C.
et al.  (
2012
)
Comparison of three amyloid assembly inhibitors: the sugar scyllo-inositol, the polyphenol epigallocatechin gallate, and the molecular tweezer CLR01
.
ACS Chem. Neurosci.
3
,
451
458
37
Prabhudesai
,
S.
,
Sinha
,
S.
,
Attar
,
A.
,
Kotagiri
,
A.
,
Fitzmaurice
,
A.G.
,
Lakshmanan
,
R.
et al.  (
2012
)
A novel ‘molecular tweezer’ inhibitor of α-synuclein neurotoxicity in vitro and in vivo
.
Neurotherapeutics
9
,
464
476
38
Sang
,
S.
,
Lee
,
M.J.
,
Hou
,
Z.
,
Ho
,
C.T.
and
Yang
,
C.S.
(
2005
)
Stability of tea polyphenol (−)-epigallocatechin-3-gallate and formation of dimers and epimers under common experimental conditions
.
J. Agric. Food Chem.
53
,
9478
9484
39
Chakrawarti
,
L.
,
Agrawal
,
R.
,
Dang
,
S.
,
Gupta
,
S.
and
Gabrani
,
R.
(
2016
)
Therapeutic effects of EGCG: a patent review
.
Expert Opin. Ther. Pat.
26
,
907
916
40
Wang
,
D.
,
Taylor
,
E.W.
,
Wang
,
Y.
,
Wan
,
X.
and
Zhang
,
J.
(
2012
)
Encapsulated nanoepigallocatechin-3-gallate and elemental selenium nanoparticles as paradigms for nanochemoprevention
.
Int. J. Nanomed.
7
,
1711
1721
41
Melchardt
,
T.
,
Magnes
,
T.
,
Weiss
,
L.
,
Grundbichler
,
M.
,
Strasser
,
M.
,
Hufnagl
,
C.
et al.  (
2014
)
Liver toxicity during temozolomide chemotherapy caused by Chinese herbs
.
BMC Complement. Altern. Med.
14
,
115
42
Modernelli
,
A.
,
Naponelli
,
V.
,
Giovanna Troglio
,
M.
,
Bonacini
,
M.
,
Ramazzina
,
I.
,
Bettuzzi
,
S.
et al.  (
2015
)
EGCG antagonizes Bortezomib cytotoxicity in prostate cancer cells by an autophagic mechanism
.
Sci. Rep.
5
,
15270
43
Golden
,
E.B.
,
Lam
,
P.Y.
,
Kardosh
,
A.
,
Gaffney
,
K.J.
,
Cadenas
,
E.
,
Louie
,
S.G.
et al.  (
2009
)
Green tea polyphenols block the anticancer effects of bortezomib and other boronic acid-based proteasome inhibitors
.
Blood
113
,
5927
5937
44
Goldschmidt
,
L.
,
Teng
,
P.K.
,
Riek
,
R.
and
Eisenberg
,
D.
(
2010
)
Identifying the amylome, proteins capable of forming amyloid-like fibrils
.
Proc. Natl Acad. Sci. U.S.A.
107
,
3487
3492
45
Liu
,
G.
,
Prabhakar
,
A.
,
Aucoin
,
D.
,
Simon
,
M.
,
Sparks
,
S.
,
Robbins
,
K.J.
et al.  (
2010
)
Mechanistic studies of peptide self-assembly: transient α-helices to stable β-sheets
.
J. Am. Chem. Soc.
132
,
18223
18232

Author notes

*

These authors contributed equally to this work.