Interactions between BIM protein and beta-amyloid may reveal a

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC

Article

Interactions between BIM protein and beta-amyloid may reveal a crucial missing link between Alzheimer's disease and neuronal cell death Ravit Malishev, Sukhendu Nandi, Dariusz Smilowicz, Shamchal Bakavayev, Stanislav Engel, Nir Bujanover, Roi Gazit, Nils Metzler-Nolte, and Raz Jelinek ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.9b00177 • Publication Date (Web): 29 May 2019 Downloaded from http://pubs.acs.org on May 30, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

Interactions between BIM protein and beta-amyloid may reveal a crucial missing link between Alzheimer's disease and neuronal cell death

Ravit Malishev,1 Sukhendu Nandi,2 Dariusz Śmiłowicz,2 Shamchal Bakavayev,3 Stanislav Engel,3,4 Nir Bujanover,4 Roi Gazit,4 Nils Metzler-Nolte,2 Raz Jelinek 1* 1Department

of Chemistry and Ilse Katz Institute for Nanotechnology, Ben Gurion University of

the Negev, Beer Sheva 84105, Israel. E-mail: [email protected] 2Bioinorganic

Chemistry, Ruhr-University Bochum Universitaetsstrasse 150 D - 44780 Bochum /

Germany 3Department

of Clinical Biochemistry and Pharmacology, Faculty of Health Sciences, Ben-Gurion

University of the Negev, Beer-Sheva, Israel 4National

Institute for Biotechnology in the Negev, Ben-Gurion University of the Negev, Beer-

Sheva, Israel Corresponding author: [email protected] Running title: Aβ42-BIM interactions ABSTRACT An extensive neuronal cell-death is among the pathological hallmarks of Alzheimer's disease. While neuron death is coincident with formation of plaques comprising the beta-amyloid (Aβ) peptide, a direct causative link between Aβ (or other Alzheimer's-associated proteins) and cell toxicity is yet to be found. Here we show that BH3, the primary pro-apoptotic domain of BIM - a ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

key protein in varied apoptotic cascades of which elevated levels have been found in brain cells of patients afflicted with Alzheimer's disease - interacts with the 42-residue amyloid isoform Aβ42. Remarkably, BH3 modulated the structure, fibrillation pathway, aggregate morphology, and membrane interactions of Aβ42. In particular, BH3 inhibited Aβ42 fibril-formation while simultaneously enhanced proto-fibril assembly. Furthermore, we discovered that BH3-Aβ42 interactions induced cell-death in a human neuroblastoma cell model. Overall, our data is providing a crucial mechanistic link accounting for neuronal cell-death in Alzheimer's disease patients, and the participation of both BIM and Aβ42 in the neuro-toxicity process. KEYWORDS: Beta-amyloid (Aβ); BIM; Bcl-3 homology 3 (BH3); Mitochondria; Apoptosis; Amyloid-membrane interaction INTRODUCTION Alzheimer's disease (AD) is a devastating and incurable disease, more prevalent in light of increasing human longevity, and which impose considerable emotional and economical toll. Intense research over the past decades has not resolved yet the exact molecular-mechanistic aspects of AD. Phenomenologically, AD is characterized by extensive neuronal cell-death in the brain that correlates with deposition of abundant fibrillar plaques, primarily comprising the beta-amyloid (Aβ) peptide 1, 2. Aβ is a product of sequential cleavage of amyloid β-protein precursor (APP) by the proteolytic activities of β-secretase and γ-secretase enzymes. It is yet unknown whether the fibrillar Aβ deposits actually partake in toxicity mechanisms, or whether the plaques constitute a side product and they do not inflict neuronal damage. Indeed, numerous studies have reported that Aβ oligomers and pre-fibrillar species are the actual cytotoxic agents

3, 4,

and that the primary

adverse physiological events induced by the oligomeric species occur at the cell membrane

5, 6.

Recent evidence, for example, suggests that extracellular Aβ oligomers interact with cell surfaces, leading to functional disruption of various receptors resulting in the abnormal activation of ACS Paragon Plus Environment 2

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


signaling pathways intracellularly

8, 9,

4, 7.

Notably, many studies have indicated that Aβ also accumulates

however the physiological significance of these observations is yet to be

resolved. Recent studies have reported that intracellular Aβ is specifically localized within the mitochondria 10-12. Aβ itself can be imported into mitochondria via the translocase within the outer mitochondrial membranes and is consequently found within the mitochondrial cristae in the inner mitochondrial membrane. Mitochondria are the primary cellular hosts of apoptotic proteins and as such constitute the organelle most responsible for initiating apoptotic cascades – the critical programmed cell death processes 13, 14. Intriguingly, both monomeric and oligomeric Aβ have been identified in mitochondria of neurons affected by AD with mitochondrial proteins

18;

15-17.

Moreover, Aβ was shown to interact

such interactions have been implicated in increased free radical

production 19, lower cytochrome c oxidase activity 20, inhibition of mitochondrial ATP production 21,

and overall damage to mitochondrial structures

22.

Indeed, experimental evidence suggests that

structural and functional abnormalities in mitochondria constitute intrinsic facets of AD 15. While the relationship between mitochondrial disfunctions and Aβ is still unclear, several reports have indicated elevated levels of apoptotic proteins residing primarily in mitochondria in Alzheimer's patients 23-27. Specifically, high concentrations of B-cell lymphoma (BCL-2) apoptotic proteins, including BIM, BAK, and BAD, have been reported within afflicted neuronal cells (2327). Interestingly, other studies have reported high abundance of BIM, a prominent intracellular protein participating in varied apoptotic cascades, in cells associated with other amyloidogenic pathologies including type-II diabetes 28, 29 and Huntington's disease 30-32. These observations might point to distinct relationships between Aβ and apoptosis. Indeed, association between the 1-42 isoform of Aβ (Aβ42) and BIM has been reported

18, 33, 34.

Whether these observations have

physiological or clinical significance has not been determined.

ACS Paragon Plus Environment 3


Here, we show that the Bcl-3 homology 3 (BH3) domain – the primary pro-apoptotic domain of BIM – significantly modulates structural transformations and fibrillation properties of Aβ42. Importantly, we found that interactions between BIM-BH3 and Aβ42 dramatically increased neuronal cell-toxicity, likely due to enhanced membrane interactions of BH3-induced Aβ42 prefibrillar species. Overall, our data might reveal a mechanistic link between BIM, intracellular Aβ42 accumulation, and neuronal cell-death. RESULTS Binding of Aβ42 to BIM-BH3 and structural transformations of the amyloid peptide Figure 1 depicts surface plasmon resonance (SPR) measurements, in which Aβ42 monomers were immobilized on the sensor chip and the BIM-BH3 segment was added in different concentrations. The SPR sensograms (Figure 1A) and corresponding dose-response curve (Figure 1B) indicate binding between Aβ42 and BIM-BH3, with a calculated Kd of 7.1 x 10-6 M. This binding constant is similar in magnitude to values previously reported for immobilized Aβ42 interacting with peptides and small molecules 35, 36. While Figure 1 reveals notable interactions between the pro-apoptotic BIM-BH3 peptide and Aβ42, we further examined the effect of BIM-BH3 upon Aβ42 aggregation using thioflavin-T (ThT) fluorescence (Figure 2A). ThT is a widely-used dye employed for illuminating protein aggregation 37, 38.

Specifically, ThT fluorescence increases upon association of the dye within β-strand

aggregates and fibrillar species 37. The ThT fluorescence data in Figure 2 attest to a dramatic effect of BIM-BH3 upon Aβ42 aggregation. The ThT fluorescence curve of Aβ42 alone (at a concentration of 10 μM) shows very small fluorescence increase during the measurement window (Figure 2A, black curve). This behavior may be explained by the fact that relatively small differences in A42 concentrations could result in changes of the aggregation rates; this ACS Paragon Plus Environment 4

Page 4 of 30

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


phenomenon was shown previously in the case of the amyloidogenic peptide islet amyloid polypeptide protein (IAPP).39, 40 Significantly enhanced ThT fluorescence evolution, however, was recorded upon co-incubation of the dye with BIM-BH3/Aβ42 mixtures. Notably, a direct relationship was apparent in Figure 2A between the degree of ThT fluorescence intensity and BIM-BH3 concentration. Since BIM-BH3 alone did not give rise to ThT fluorescence (data not shown), the dramatic increase in ThT fluorescence reflects the occurrence of distinct Aβ42 aggregation pathways and/or enhanced fibrillation of the peptide, that were specifically induced by BIM-BH3. The transmission electron microscopy (TEM) analysis in Figure 2B further illuminates the dramatic effects of Aβ42/BIM-BH3 interactions upon the morphology and abundance of peptide aggregates formed. The TEM images in Figure 2B were recorded after 14 hrs incubation for the respective Aβ42/BIM-BH3 mixtures analyzed in the ThT experiments (i.e. Figure 2A). Aβ42 alone produces the widely-observed elongated fibrils 41, dispersed in relatively low abundance upon the TEM grid (Figure 2B,i). Ubiquitous thicker and shorter fibrillar species appear, however, following co-incubation of Aβ42 with BIM-BH3 (Figure 2B,ii). The TEM analysis further reveals that smaller fibrillar aggregates form in solutions containing higher ratios between BIM-BH3 and Aβ42 (Figure 2B,iii-iv). The TEM results in Figure 2B complement the ThT fluorescence data (Figure 2A), as the abundant smaller Aβ42 aggregates forming upon BIM-BH3 addition provide pronounced surface area for ThT association and concomitant increased fluorescence of the dye. 37

To decipher the effects of BIM-BH3 on the structural transformations of Aβ42 in aqueous solution and the kinetics of Aβ42 aggregation we carried out circular dichroism (CD) and atomic force microscopy (AFM) experiments (Figure 3). Figure 3A presents circular dichroism (CD) spectra of Aβ42 recorded at different times after dissolving the peptide in water, with and without ACS Paragon Plus Environment 5


Page 6 of 30

co-incubation with the BIM-BH3 peptide. The time-dependent CD signatures of Aβ42 alone (Figure 3A, black spectra) reveal gradual spectra shift and appearance of the negative peak at around 215 nm and positive shoulder at 195 nm, corresponding to the random coil-to-β sheet transformation of the peptide.

42

Similar CD traces were recorded when a low concentration of

BIM-BH3 was co-incubated with Aβ42 (BH3:Aβ42 mole ratio of 1:0.5; Figure 3A, red spectra). However, significant acceleration of β-sheet assembly of Aβ42 was clearly apparent in a sample comprising 1:1 mole ratio between Aβ42 and BIM-BH3 (Figure 3A, blue spectra). Importantly, the CD spectra in Figure 3A constitute the net secondary structures of Aβ42, after subtraction of the corresponding spectra of BIM-BH3 alone (The CD traces of BIM-BH3 alone indicate a mostly random coil and a certain percentage of helical conformation, (Figure 1,SI).

43

All the signals

resulting from the BIM-BH3 were subtracted from the corresponding spectra. The dramatic BH3induced remodeling of Aβ42 structure occurred, in fact, immediately after co-dissolution of the two peptides, apparent in the spectral shift of the blue spectrum at the nominal time t=0 (Figure 3A). The CD data, attesting to BH3-promoted β-sheet formation of Aβ42 (Figure 3A), are consistent with the enhanced ThT fluorescence and TEM data in Figure 2, which demonstrated rapid formation of Aβ42 aggregates upon co-incubation with BIM-BH3. To further probe the impact of the pro-apoptotic BIM-BH3 segment upon the fibrillation process of Aβ42, we carried out microscopy experiments designed to visualize the aggregate morphologies and their time evolution (Figure 3B). Corroborating the ThT and CD experiments (Figure 2A, 3A), the atomic force microscopy (AFM) images in Figure 3B demonstrate that BIM-BH3 significantly modulated Aβ42 fibril assembly. Aβ42 alone formed small prefibrillar aggregates short time after dissolution (Figure 3B, left column, top image). The small aggregates [oligomers and protofibrils 44]

likely constitute nucleation sites for further fibril growth, yielding mature fibrils after longer

incubation times (Figure 3B, left column, middle and bottom panels). ACS Paragon Plus Environment 6

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The AFM images in Figure 3B (middle and right columns) reveal that co-incubation of BIMBH3 with Aβ42 gave rise to abundant pre-fibrillar aggregates. Moreover, the pro-apoptotic BIMBH3 domains affected Aβ42 aggregate morphologies and the extent of elongated fibril growth (Figure 3B). Specifically, ubiquitous small aggregates are apparent in the BH3-Aβ42 mixtures, at both mole ratios, after a short, 10-minute incubation (Figure 3B, top middle and right images). This result is consistent with the ThT fluorescence data in Figure 2A, pointing to rapid Aβ42 aggregation induced by the BIM-BH3 segment. Abundant, short protofibril structures are apparent in the BH3Aβ42 mixtures incubated for 24 hrs, and particularly in the mixed-peptide samples incubated for 48 hrs (Figure 3B, middle and right images, middle and bottom rows). These protofibril configurations appear larger and morphologically different than the prefibrillar species of Aβ42 alone (Figure 3B, left column). Particularly striking is the observation that the Aβ42 protofibrils were still the predominant species in the BH3-Aβ42 mixtures after 48 hours, while mature fibrils could be hardly detected (Figure 3B, bottom row, middle and right images). This result stands in contrast to the AFM results of bare Aβ42, which featured almost exclusively long, mature fibrils (e.g. bottom left panel, Figure 3B). Overall, the microscopy analyses in Figures 2B and 3B complement the CD and ThT data (Figure 2A and 3A, respectively), underscoring the pronounced effects of the proapoptotic BIMBH3 peptide upon the structural transformations, aggregation pathway, and fibril morphologies of Aβ42. Numerous studies have proposed that distinct pre-fibrillar species assembled during the aggregation process of Aβ42 are intimately linked to the biological effects and toxicity profile of the peptide, and to the pathology of Alzheimer's disease in general.

4, 45, 46

In particular, Aβ42

oligomers and small pre-fibril aggregates are believed to constitute major toxicity determinants, possibly through membrane interactions and concomitant disruption of the lipid bilayer.

47-49

Accordingly, we examined whether, and to what degree, the proapoptotic BIM-BH3 peptide affects ACS Paragon Plus Environment 7


Page 8 of 30

the functional and biological properties of Aβ42 (Figures 4-5). Importantly, the A11 antibody assay 50

confirmed the presence of abundant oligomeric Aβ42 species in the BH3-Aβ42 mixtures (Figure

2,SI). Effect of BIM-BH3 upon membrane interactions of Aβ42 To evaluate membrane interactions of Aβ42 and the BH3-Aβ42 mixtures (at different mole ratios between the two peptides) we carried out fluorescence anisotropy experiments, presented in Figure 4, utilizing lipid vesicles, which also contained the fluorescent dye diphenylhexatriene (DPH). The fluorescence anisotropy of bilayer-embedded DPH has been employed as a sensitive measure of lipid fluidity and the effect of membrane-active molecules upon lipid mobility in the bilayer. 51, 52 The fluorescence anisotropy data in Figure 4 were recorded using vesicles comprising either dioleylphosphatidylcholine (DOPC), cholesterol (chol), and sphingomyelin (Sph) – designed to mimic the plasma membrane – or vesicles prepared from DOPC and cardiolipin (CL), mimicking the mitochondrial membrane which is the main cellular compartment hosting BIM-BH3.

53, 54

Importantly, AFM experiments shown in Figure 3,SI confirmed that similar Aβ42 fibrillation pathways occurred in the presence of both types of vesicles as in buffer (i.e. Figure 3). Figure 4 reveals that, in both vesicle models, the proapoptotic BIM-BH3 peptide enhanced bilayer interactions of Aβ42. Upon addition of Aβ42 alone, the fluorescence anisotropy of DPH increased due to insertion of Aβ42 into the bilayer thereby restricting the mobility of the phospholipid acyl chains. 55 The higher Aβ42-induced anisotropy was particularly significant in the DOPC/Chol/Sph vesicles, reflecting the presumed affinity of the peptide to cholesterol-rich lipid raft domains. 56, 57 Previous studies have shown that cholesterol-enriched lipid rafts play important roles in accelerating plaque formation and membrane disruption.

58

Moreover, one has to consider the

critical micellar concentration (cmc) of phospholipid components of the vesicles produced. Specifically, previous analyses have correlated the lipids' cmc and peptide fibrils induced by free


Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


lipids in equilibrium with the assembled vesicles 59.60. In this context, the different extents of fibrils formed in the two lipid mixtures might affect the degree of Ab42-induced fluidity of the bilayers. A particularly notable result in Figure 4 is the pronounced DPH fluorescence anisotropy induced upon adding BH3-Aβ42 mixtures to the vesicles, in comparison to Aβ42 alone. BIM-induced higher anisotropy was apparent in the PC/chol/Sph vesicle model (Figure 4A, bars i and ii) and particularly in case of the mitochondrial membrane model (Figure 4B, bars i and ii). The enhanced anisotropy values indicate that BIM-BH3 promoted greater membrane internalization of Aβ42. This result is consistent with the apparent BH3-induced Aβ42 protofibril formation (e.g. Figure 3), as Aβ42 oligomers and pre-fibrillar aggregates have been shown to exhibit significant membrane interactions. 61-63 It should be noted that BIM-BH3 alone exhibited negligible effects upon bilayer fluidity (Figure 4, bars iii and iv), confirming that the enhanced bilayer interactions of Aβ42 were due to the impact of the co-added BIM-BH3 peptide. Impact of BIM-BH3 upon Aβ42-induced neuronal cell death While Figure 4 shows that the distinct effects of the proapoptotic BIM-BH3 domain upon Aβ42 structural features and fibril formation likely induced greater membrane interactions and bilayer internalization of Aβ42, a crucial question concerns the biological significance of BH3-Aβ42 interactions. Indeed, Figure 5 attests to the dramatic consequences of co-incubating BIM-BH3 and Aβ42 upon neuronal cell viability. Figure 5A depicts an PI-annexin flow cytometry apoptosis assessment utilizing the SH-SY5Y neuroblastoma cell line. 64, 65 In the absence of BIM-BH3, 15µM Aβ42 exhibited a toxic effect, reducing cell viability to around 60%, while in parallel affecting an increase in the populations of apoptotic and necrotic cells (Figure 5A, top row). Furthermore, coincubation of Aβ42 with increasing concentrations of BIM-BH3 (Aβ42/BIM-BH3 at 1:1, 1:2, and 1:3 mole ratios, respectively) induced higher toxicity, decreasing cell viability to around 40% at the ACS Paragon Plus Environment 9


highest BIM-BH3 concentration, and significantly increasing the fractions of apoptotic and necrotic cells (Figure 5A, three bottom rows).66 Importantly, the BIM-BH3 peptide alone did not produce toxic effects. The optical microscopy images in Figure 5B similarly highlight the enhanced toxic effects induced upon co-incubation of BIM-BH3 and Aβ42 with the SH-SY5Y neuroblastoma cells. While addition of Aβ42 (15 μM) to the cells induced apoptotic cell-death, reflected in the cell shape transformations to more spherical appearance (Figure 5B, middle image), co-addition of Aβ42 together with the proapoptotic BIM-BH3 peptide gave rise to considerably more extensive celldeath, accounting for the abundant spherical-like cells (Figure 5B, right image). Overall, the cell viability data in Figure 5 demonstrate significantly more pronounced cell toxicity of Aβ42 induced upon its interactions with BIM-BH3.

Discussion The structural, functional, and biological data presented in Figures 1-5 point to an intriguing and previously-undisclosed mechanism which might account for the extensive neuronal cell-death encountered in Alzheimer's disease.

Specifically, our results indicate that the proapoptotic

sequence BIM-BH3 experiences affinity and distinct binding with Aβ42 (Figure 1). Interactions between Aβ42 and BIM-BH3, which have not been previously reported, are conceivable in actual cellular environments, as BIM-BH3 is present both in the mitochondria and cytosol 67, while several reports have also localized Aβ42 in these cellular compartments 10-12, 68. A critical aspect of Aβ42 structural evolution concerns its aggregation pathways, since Aβ42 oligomers and proto-fibrillar species are believed to be major cytotoxicity and pathogenicity ACS Paragon Plus Environment 10

Page 10 of 30

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


factors 1. Indeed, the experimental data in Figure 2 demonstrate that BIM-BH3 interactions resulted in significant remodeling of Aβ42 structure and aggregation. Specifically, BIM-BH3 affected the solution conformation of Aβ42, promoting adoption of β structure by the peptide (Figure 3A), which went hand-in-hand with acceleration of aggregation manifested in the ThT fluorescence experiment (Figure 2B). These intertwined phenomena were dramatically visualized in the AFM experiment which revealed ubiquitous smaller Aβ42 fibrils upon incubation with BIM-BH3, in contrast to the longer, mature Aβ42 fibers assembled in solution without the presence of BIM-BH3. Aβ42 oligomers and protofibrils exhibit significant interactions with cellular membranes. Indeed, the fluorescence anisotropy experiments in Figure 4 complemented the spectroscopy and microscopy analyses in Figures 2-3, demonstrating that BIM-BH3 induced enhanced Aβ42membrane interactions. Notably, in two different membrane models, one mimicking the plasma membrane (Figure 4A) while another designed to mimic the mitochondrial membrane (Figure 4B), pre-incubation of Aβ42 with BIM-BH3 resulted in experimentally-significant "stiffening" of the lipid bilayer, likely reflecting insertion of membrane-active oligomeric species 69. While the experimental data in Figures 1-4 underscore the interactions between Aβ42 and BIM-BH3 and their consequences in solution and in the presence of biomimetic membrane models, a critical question one needs to address is whether such interactions actually impact cell viability and cellular processes. Importantly, flow cytometry and cell morphology analyses depicted in Figure 5 show that BH3-Aβ42 mixtures exerted higher neuronal cell toxicity compared to Aβ42 or BIM-BH3 alone, and that the toxicity effect was dependent upon BIM-BH3 concentration. These results nicely complement the biophysical analyses in Figures 1-4 and suggest that the BH3-induced greater abundance of oligomeric and pre-fibrillar Aβ42 species gave rise to cell death via enhanced membrane interactions.



Figure 6 depicts a proposed model based upon our experimental data presented in Figures 1-5, accounting for the pronounced neurotoxicity associated with BH3-Aβ42 interactions. The top part of the cartoon illustrates the structural transformation of Aβ42 monomers, produced through the enzymatic activity of APP, into membrane-active oligomers and proto-fibrils, believed to constitute major toxic determinants. The Aβ42 oligomers ultimately assemble into mature fibrils which accumulate upon the neuronal cells. The model in Figure 6 suggests, however, that co-incubation of Aβ42 with BH3, promote assembly of membrane-disruptive and cytotoxic Aβ42 oligomers, accountng for the enhanced neuronal cell death encountered in Alzheimer's disease.

Conclusions In conclusion, this study discloses an unrecognized mechanism, according to which interactions between the apoptotic BIM protein interacts and beta-amyloid may account for the devastating neurotoxic effects of Alzheimer's disease. The model reported here may open new therapeutic avenues targeting BIM-Aβ42 interactions as a possible remedy for Alzheimer's disease. Notably, elevated BIM levels have been observed in other amyloidogenic pathologies including type-II diabetes 28 and Huntington's disease 30, hinting that the phenomenon we report might be a more universal toxicity determinant in protein misfolding diseases. It is important to note, however, that AD is a complex disease likely affected by contributions from multiple molecular pathways. Thus, it should be emphasized that the precise physiological roles of BIM-Aβ42 interactions and their relationships with AD pathogenesis require further studies.

Materials and Methods ACS Paragon Plus Environment 12

Page 12 of 30

Page 13 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Materials. Aβ42 was purchased from AnaSpec (USA) in a lyophilized form at >95% purity.1,2dioleoyl-sn-glycero-3-phosphocholine (DOPC), cholesterol (ovine wool, >98%), sphingomyelin (brain, porcine) and Cardiolipin (Heart, Bovine) were purchased from Avanti Polar Lipids. Hexafluoroisopropanol (HFIP), sodium hydrosulfite and sodium phosphate monobasic were purchased from Sigma-Aldrich (Rehovot, Israel). 1,6-diphenylhexatriene (DPH) were obtained from Molecular Probes, Inc. (Eugene, Oregon). Synthesis of BIM-BH3. A Liberty microwave peptide synthesizer from CEM was used for synthesis

of

the

BH

domain

of

the

BIM

protein

(sequence

DMRPEIWIAQELRRIGDEFNAYYARR. Stepwise coupling reactions were performed with Fmoc strategy on a 0.25 mmol scale using Fmoc-Arg(Pbf)-OH Wang resin (0.5 mmol/g, Iris Biotech), enantiomerically pure Fmoc-protected amino acids, 1-hydroxybenzotriazole (HOBt), 2(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium

tetrafluoroborate

(TBTU)

and

diisopropylethylamine (DIPEA), (4:4:8 equiv.). Each amino acid was coupled using standard coupling conditions (75 °C, 24 W, 300 sec.). N-terminal deprotection of the Fmoc group during the syntheses was performed twice using 20 % piperidine in DMF (first cycle: 75 °C, 35 W, 30 sec; second cycle 75 °C, 50 W, 180 sec). The synthesis results in the linear resin-bound side chain protected Fmoc-Bim-Wang-resin. The Fmoc-Bim-Wang-resin was then completely transferred into a filter-containing syringe for Fmoc deprotection, followed by simultaneous side chain deprotection and cleavage from the resin. The Fmoc deprotection was performed twice using 20 % piperidine in DMF (2 mL). After Fmoc deprotection the resin was washed with DMF (2 × 2 mL) and DCM (2 × 2 mL), shrinked with diethyl ether and dried under vacuum for 30 min. Finally, cleavage of the Bim peptide from the resin was performed using 1 mL of cleavage mixture containing TFA/H2O/triisopropylsilane (TIS)/ 1,2-Ethanedithiol (EDT) (90 %:2.5 %:2.5 %:5 %, v/v) for 6 h at room temperature. The resin was filtered and washed with TFA (0.25 mL). Addition of cold diethyl ether/hexane mixture (1:1, v/v) yielded a white precipitate, which was repeatedly washed ACS Paragon Plus Environment 13


Page 14 of 30

with diethyl ether/hexane mixture (1:1, v/v). The crude peptide was then dissolved in acetonitrile/water (1:1, v/v) and lyophilized. The Bim peptide was purified by RP-HPLC using a RP-C18 semipreparative (Varian Dynamax, 21.4 mm x 250 mm) columns on a Varian Prostar Instrument using a gradient of acetonitrile/water containing 0.1% TFA ((40 min, 2.5 mL/min; 0 min: 5% ACN, 5 min: 5% ACN; 25 min: 95% ACN, 30 min: 95% ACN, 35 min: 5% ACN, 40 min: 5% ACN)). After collecting the desired fraction from HPLC the purity of the peptide was confirmed using a RP-C8 analytical HPLC (Varian Dynamax, 4.6 mm x 250 mm). Peptide and Sample Preparations. Aβ42 was dissolved in HFIP at a concentration of 1 mg/mL and stored at −20 °C until use to prevent aggregation. For each experiment, the solution was thawed, and the required amount was dried by evaporation for 6−7 h to remove the HFIP. The dried peptide sample was dissolved in 10 mM sodium phosphate, pH 7.4, at room temperature. Stock solutions of Bim peptide were prepared at 0.5 mM in deionized water and diluted into the Aβ42 solutions at the required concentrations. Vesicle

preparation.

Vesicles

consisting

of

DOPC/cholesterol/sphingomyelin

and

DOPC/Cardiolipin were prepared by dissolving the lipid components in chloroform/ethanol (1:1, v/v) and drying together in vacuum. Small unilamellar vesicles (SUVs; DOPC/Cho/Sph and DOPC/CL 0.67:0.25:0.08, 0.96:0.04 mole ratio, respectively) were prepared in 10 mM sodium phosphate (pH 7.4) by probe-sonication (power: 130 w, frequency: 20 kHz, at 20 % amplitude) of the aqueous lipid mixtures at room temperature for 10 min. Vesicle suspensions were allowed for 1 h at room temperature prior to usage. Surface plasmon resonance (SPR). The constant for binding of Aβ42 to BIM-BH3 peptide was determined by SPR using a ProteOn™ XPR36 instrument (Bio-Rad) as follows. Aβ42 was immobilized on the surface of the GLC chip by using the amine coupling reagents sulfo-NHS (5 mM N-hydroxysuccinimide) and EDC (20 mM 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide, ACS Paragon Plus Environment 14

Page 15 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Bio-Rad). The Aβ42 protein (2 µg) was immobilized on the chip in 10 mM sodium acetate buffer (pH 4) and run at 30 µl/sec for 300 sec. BSA (3 µg) was immobilized on the chip as a negative control. Unbound carboxyl residues were deactivated with 1 M ethanolamine HCl at pH 8.5. Each binding assay, were done using PBST (PBSx1 + 0.005% Tween 20) at 25°C. BIM peptide (the analyte) was then allowed to flow over the surface-bound at 5 different concentrations (20, 10, 5, 2.5, 1.25 and 0 µM) and a flow rate of 25 µl/min. The next step was to examine the dissociation of the proteins while allowing PBST to flow over the surface for 1800 sec. After each run, a regeneration step was performed with 50 mM NaOH at a flow rate of 100 μl/min. BSA reference channel and the BIM-BH3 zero concentration were used as background and subtracted from the BIM sensograms. Data was analyzed on the proteOn manager software version 3.01 using the equilibrium model for measuring affinity. Thioflavin T (ThT) fluorescence assay. ThT fluorescence measurements were conducted at 37 °C using 96-well path cell culture plates on a Varioskan plate reader (Thermo, Finland). Measurements were made on samples containing 10 μM Aβ42 in the absence or presence of different BIM-BH3 concentration. A 180-μL aliquot of the aggregation reaction was mixed with 20 μL of 100 μM ThT in sodium phosphate buffer, pH 7.4. The device was programmed to record fluorescence intensity every every 5 min for 14 hrs . The fluorescence intensity was measured following a 10-min incubation at λex = 440 and λem = 485 nm. Transmission Electron Microscopy (TEM). Peptide aliquots (5 μL) from samples used in the ThT experiments (after 14 h incubation) were placed on 400-mesh copper grids covered with a carbon-stabilized Formvar film. Excess solutions were removed following 2 min of incubation, and the grids were negatively stained for 30 s with a 1% uranyl acetate solution. Samples were viewed in an FEI Tecnai 12 TWIN TEM operating at 120 kV.



Circular dichroism (CD) Spectroscopy. CD spectra were recorded in the range of 190–260 nm at room temperature on a Jasco J-715 spectropolarime-ter, using 1-mm quartz cuvettes. Solutions composed of 400 μL contained 25 μM Aβ42 in the absence or presence of different BIM-BH3 peptide concentration. Spectra were recorded at time 0, 30 min and after 12 h incubation. CD signals resulting from buffer and bim peptide were subtracted from the corresponding spectra. Atomic force microscopy (AFM). Peptide samples (Aβ42 or Aβ42/BIM-BH3 mixtures solution) were scanned by AFM (Dimension 3100 SPM, Digital Instruments Veeco, NY), in tapping mode, following deposition of 5 μL peptide solution (25 μM peptide concentration) in the absence or presence of lipid vesicles (final concentration 0.3 mM) on a freshly cleaved mica substrate. After five minutes, 100 μL of 10 mM phosphate buffer, pH 7.4, was added and the sample was imaged under aqueous solution. All images were acquired using a silicon probe (AC 240, Olympus) with a spring constant of 2 N/m, frequency 70 kHz and a tip with a radius of 9 nm. Fluorescence Anisotropy. The fluorescence probes DPH was incorporated into the SUVs (DOPC/Chol/Sph and DOPC/CL 0.67:0.25:0.08 and 0.95:0.05 mole ratio, respectively) by adding the dye dissolved in THF (1 mM) to vesicles up to a final concentration of 1.25 μM. After 30 min of incubation at 30°C of DPH, fluorescence anisotropy was measured at λex=360 nm and λem=430 nm using a FL920 spectrofluorimeter (Edinburgh Co., Edinburgh, UK). Data were collected before and after the addition of freshly dissolved Aβ42, BIM-BH3, or their mixtures solution. Anisotropy values were automatically calculated by the spectrofluorimeter. software using the equation: r = (IVV – GIVH) / (IVV + 2GIVH), G = IHV / IHH, in which IVV is with excitation and emission polarizers mounted vertically; IHH corresponds to the excitation and emission polarizers mounted horizontally; IHV is the excitation polarizer horizontal and the emission polarizer vertical; IVH requires the excitation polarizer vertical and emission polarizer horizontal. Each experiment was repeated at least three times. Results are presented as means ± standard error of the mean (SEM) of seven replicates. * p

Interactions between BIM protein and beta-amyloid may reveal a

Recommend Documents