Principles governing catalytic activity of self-assembled short peptides

Subscriber access provided by University of South Dakota

Article

Principles governing catalytic activity of self-assembled short peptides Ruiheng Song, Xialian Wu, Bin Xue, Yuqin Yang, Wenmao Huang, Guixiang Zeng, Jian Wang, Wenfei Li, Yi Cao, Wei Wang, Junxia Lu, and Hao Dong J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 18 Dec 2018 Downloaded from http://pubs.acs.org on December 18, 2018

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 17 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

Journal of the American Chemical Society

Principles governing catalytic activity of self-assembled short peptides Ruiheng Songa,+, Xialian Wub,d,+, Bin Xuec, Yuqin Yanga, Wenmao Huangc, Guixiang Zenga,e, Jian Wangb, Wenfei Lic,e, Yi Caoc,e, Wei Wangc,e,*, Junxia Lub,* & Hao Donga,e,* a Kuang b School c

Yaming Honors School, Nanjing University, Nanjing 210023, China; of Life Science and Technology, ShanghaiTech University, Shanghai 201210, China;

Collaborative Innovation Center of Advanced Microstructures, National Laboratory of Solid

State Microstructure, Department of Physics, Nanjing University, Nanjing 210093, China; d University of Chinese Academy of Sciences, Beijing 100049, China; e Institute

for Brain Sciences, Nanjing University, Nanjing 210023, China.

These authors contributed equally to this work whom correspondence may be addressed. E-mail: [email protected] or [email protected] or [email protected] +

* To

Abstract Molecular self-assembly provides a chemical strategy for the synthesis of nanostructures by using the principles of nature, and peptides serve as the promising building block to construct adaptable molecular architectures. Recently, a series of heptapeptides with alternative hydrophobic and hydrophilic residues were reported to form amyloid-like structures, which are capable of catalyzing acyl ester hydrolysis with remarkable efficiency. However, it remains elusive about the atomic structures of the fibrils. What is the origin of the sequence-dependent catalytic activity? How is the ester hydrolysis catalyzed by the fibrils? In this work, the atomic structures of the aggregates were determined by using molecular modelling and further validated by solid-state NMR experiments, where the fibril with high activity adopts twisted parallel configuration within each layer, and the one with low activity is in flat antiparallel configuration. The polymorphism originates from the interactions between different regions of the building block peptides, where the delicate balance between rigidity and flexibility plays an important role. We further show that the p-nitrophenylacetate (pNPA) hydrolysis reactions catalyzed by two different fibrils follow similar mechanism, and the difference in microenvironment at the active site between the natural enzyme and the present self-assembled fibrils should account for the discrepancy in catalytic activities. The present work provides understanding of the structure and function of self-assembled fibrils formed with short-peptides at atomic level, and thus sheds new insight on designing aggregates with better functions.

1

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Introduction Self-assembly was proposed to be a facilitating bridging step between short peptides and larger structures during protein evolution.1 Among different building blocks, peptides serve as promising candidates to construct adaptable molecular architectures.2 The interplay between electrostatic, hydrogen bonding, and van der Waals forces allows peptide to form functional aggregations with variations in molecular structure.3,4 This polypeptide-based scaffold could potentially serve as a platform to fabricate controllable and tunable molecular machine with new emerging properties, mainly due to their structural diversity, high loading capacity, biocompatibility and mechanical behavior.5 A broad range of application prospects, such as targeted drug delivery, cell culturing, antimicrobial agents, biosensor devices, molecular electronics, and catalytic reaction etc. have been proposed.6 Indeed, by mimicking the active sites in nature enzyme, great success has been made in the design of bio-inspired catalytic scaffold for artificial system.7 For example, the amyloid architecture formed by short peptides provides a scaffold for enzyme-like catalysts, including catalyzing electrochemical reactions,8 promoting specific enantioselective chemical reactions,9 or enhancing ester hydrolysis.10 Aggregates formed by the building blocks of amphipathic heptapeptide with alternative hydrophobic and hydrophilic residues containing two histidine (His) residues were reported to catalyze the aryl ester hydrolysis with remarkable catalytic efficiency (Figure 1).11 The His residues on neighboring peptides were proposed to form an active site to bind the cofactor Zn2+, which mimics Zn2+-containing metalloenzymes of carbonic anhydrase that is capable of catalyzing a broad range of reactions including the interconversion between carbon dioxide and bicarbonate.12 Notably, such catalytic reaction has been extensively studied.13-27 Among different peptides, the highest catalytic efficiency come from the sequence Ac-IHIHIQI-CONH2 (abbreviated as H-sequence, where H means high activity) with the kcat/kM of 62±2 M-1 s-1, while the Q-to-R mutation at the sixth position of the peptide, the Ac-IHIHIRI-CONH2 (abbreviated as L-sequence, where L means low activity) sequence, leads to 3-fold decrease in activity (kcat/KM value of 22±8 M-1 s-1).11 Oligomer aggregated from such sequence was further found to be capable of activating oxygen through binding copper28 or prompting phosphoester hydrolysis.29 The Q-to-Y mutation at the same position, the Ac-IHIHIYI-NH2 sequence (abbreviated as Y-sequence), was found to form fibril-like structures with even higher esterase activity (reported kcat/KM values ranging from 98±3 M-1 s-1 29 to 355±15 M-1 s-1).30 Therefore, given the same number of amino acids on the peptide building block, the catalytic function of the self-aggregated structure highly depends on its sequence. However, none of the three structures has been reported, and therefore no information at atomic level is available to distinguish their differences. Further understanding of the nature of the catalytic activity possessed by the self-assembled structure depends on knowledge of their atomic structures, which is mainly hindered by a lack of accurate measurements of the dynamics with sufficient spatial and temporal resolution. Recently, the structure of a relevant catalytic amyloid formed by Ac-IHVHLQI-CONH2 sequence (abbreviated as M-sequence) was determined with solid-state NMR (ssNMR),31 which has high efficiency (kcat/KM value of 66.8 M-1 s-1) that is comparable to the H-sequence. This new artificial catalysis adopts steric zipper configuration to form stable structure.32 To be specific, the amphiphilic parallel β-sheets aggregate into stacked bilayers with alternating dry (hydrophobic 2


Page 2 of 17

Page 3 of 17 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


residues) and wet (hydrophilic residues) interfaces. The configuration of the Zn2+ binding site was also determined: both two His residues on each β-strand are involved in coordination, and each Zn2+ is coordinated by three of the four His residues from two neighboring chains, where two using nitrogen on δ position (Nδ) of imidazole ring, while the rest one using nitrogen on ε position (Nε). To achieve this configuration, the neutral (Nδ being deprotonated) and anionic (both Nδ and Nε being deprotonated) forms of His residues are equally populated.31 This structure provides great insight into the oligomeric structure and coordination geometry of the Zn2+-bound amyloid fibril based on heptapeptide that catalyzes ester hydrolysis, and may explain the Zn2+-dependent activity. However, neutral histidine involved in the His-Zn2+ interactions in nature enzyme is dominant. In contrast, according to the pKas of histidine,33 its anionic form is likely to exist at strongly basic condition, while the peptides were fibrilized at pH=8 in that work.31 It remains unclear whether the weakly basic condition used to trigger the formation of peptide fibrils is sufficient to deprotonate the histidine to the anionic form. Given the aforementioned work, there are still a couple questions remaining elusive: what are the atomic structures for the fibrils formed with H- or L- or Y-sequences? What is the origin of the sequence-dependent catalytic activity? How to improve the esterase activity within the framework of this scaffold? Attempts to address these issues may provide clues for designing fibril with higher activity. In present work, the multiple-scale computational tools were used to characterize the aggregated structure formed by H- or L- or Y-sequences, as well as the possible mechanism for the catalytic reaction of p-nitrophenylacetate (pNPA) hydrolysis, a commonly used compound for benchmarking hydrolytic enzyme, happening at the active site on the wet interface. The predicted structures of fibrils formed with either H- or L-sequences were further validated by the ssNMR experiments. Our data show that all three sequences form cross-β amyloid-like fibrils with different morphologies: the H- and Y-sequences lead to the formation of twisted morphology where beta-strands in each layer are in parallel configuration; in contrast, the L-sequence forms planar structure with anti-parallel configuration within each layer (Figure 1). The polymorphism could be attributed to the concerted interplay among hydrophobic- and hydrophilic-sidechain as well as backbone. Given the different structures, the catalytic reaction for pNPA hydrolysis follows similar reaction pathway, in which the nucleophilic attack of pNPA is the rate-limiting step, followed by the fast proton immigration. However, the fibrils formed with either H- or Y-sequence exhibit balanced rigidity and flexibility, which is better in mimicking the topology of the active site of enzyme, and therefore shows higher esterase activity. Results and Discussion (1) Configuration of peptide assembly depends on the sequence Though polymorphic diversity is a common feature in amyloid structures, and it is likely that peptides are not always uniformly aligned,31,34,35 we focused on the parallel and antiparallel configurations within each sheet to explore their stabilities. Model systems containing only two strands in either configuration in explicit water box were constructed. The free energy profile for pulling the pair of two peptides apart was characterized by using the adaptive biased force (ABF) calculations, where the pathway for association/dissociation was defined along the fibril axis perpendicular to the β-strands. For the H-sequence, the dimer formed in both parallel and antiparallel configurations are 3



more stable than those separated monomers, and the parallel configuration is ~1 kcal/mol more than stable than that in the antiparallel one (Figure 2), showing that the parallel one is energetic more favorable. Similar scenario is observed for the Y-sequence peptide, where the parallel configuration is ~5 kcal/mol more stable (Figure S1). In contrast, for L-sequence the antiparallel configuration is ~2 kcal/mol more stable than the parallel one. Seemingly, to eliminate the unfavorable electrostatic repulsion between subunits, the neighboring β-strands of L-sequence are arranged in antiparallel fashion to stabilize the structure. It should be mentioned that, though the parallel in-register arrangements has been found to be the most frequent arrangement of β-strands in the fibril, subtle variation on sequence changes the configuration from parallel to antiparallel.36 Our data demonstrates that the structural order present in the β-sheet assemblies depends on the peptide sequence, where electrostatic plays a significant role. (2) Fibril polymorphism Due to the amphiphilic nature of the sequence, we started with the initial double-layer model for the aggregates formed with H- or L- or Y-sequences.11 The parallel and antiparallel configurations within each layer were investigated, which were hereafter abbreviated as XP (X-sequence in parallel configuration) and XA (X-sequence in antiparallel configuration), where X represents H, L or Y, respectively. Following the geometry of the fibril composed of M-sequence,31 the antiparallel configuration of stacked sheets at dry interface between two layers was adopted in this work. The peptide backbone was initially fixed with sidechains fully relaxed. After equilibration with gradually released constraint, four fibril microstructures (HA, HP, LA, YP) but not the LP or YA were found to be well kept in 0.5 μs molecular dynamics (MD) simulations: the tightly packed hydrophobic core was formed by either leucine or isoleucine sidechains at the dry interface, and polar residues were exposed to water at the wet interface, which is similar to the fibril structure formed with M-sequence.31 However, apparent morphological difference was observed (Figure 1). Fibrils formed with Hor Y-sequence peptides adopted left-handedly twisted fibril-like structures, a common configuration of β-sheet in native proteins,37 while those formed with L-sequence peptides were flat, which were consistent with the Pauling-Corey model of β-sheet.38 The average angles of strand-crossing are 15.5±1.2° in HP and 9.2±1.3° in HA, while 4.6±1.7° in LA (Figure 3). It should be noted that, based on a survey of 478 experimental determined carbonic anhydrase structures from three distinct families where the two β-strands are in antiparallel configuration, these angles show a normal distribution in the range of 19.9~22.8° with the major peak at 21.5°.39 Therefore, the twisting between neighboring β-strands is a common feature in the active site of carbonic anhydrase. Presumably, the HP exactly imitates this structural characteristic to achieve the high catalytic efficiency. We further validate this hypothesis by calculating this angle in YP fibril, the more active one: the twist angle is 12.2±1.4° in YP, which resembles what we observed in the H-sequence structures. In our MD simulations, extensive inter-molecular hydrogen bonding interactions were formed between adjacent subunits, but notable differences could be observed. Specifically, HP has higher number of backbone-backbone hydrogen bonds (60.5±3.8 in HP vs. 50.0±4.8 in HA) with shorter hydrogen bonding distance than that in HA (2.83±0.01 Å in HP vs. 2.85±0.01 Å in HA), 4


Page 4 of 17

Page 5 of 17 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


suggesting higher stability within each sheet in HP than that in HA. In addition, specific interactions between sidechain were observed: ~85% of Gln residues in HP formed hydrogen bonding with adjacent Gln residue in the MD simulations, contributing further stabilization to the fibril structure. The contacts between inter-strand sidechains were proposed to be critical for the formation of stable β-sheet.40 For example, in a self-assembled structure formed with the VQIVYK sequence peptides, strong cooperativity of the hydrogen bonding between sidechain of Gln residues was found, as suggested by the short hydrogen bond length.41 In present work, the extensive hydrogen bonding between Gln residues could be another factor that the parallel arrangement is favored over the antiparallel one in the fibril formed with H-sequence peptides. LA has higher number of hydrogen bonds (68.2±4.9) than HP, though the two structures have similar hydrogen bonding distance (2.84±0.01 Å). Probably, the flat configuration of L-sequence structure accounts for its higher number of hydrogen bonds than the twisted H-sequence fibril. The formation and morphology of fibrils with H- or L-sequences were characterized with transmission electron microscopy (TEM, Figure S2) and atomic force microscopy (AFM, Figure S3). Both methods confirmed the formation of amyloid fibrils, and allowed the structure to be visualized at high resolution. The TEM images indicated that with addition of Zn2+ ions, the appearance of H-fibrils changed significantly, from curly structure in the Zn2+-free condition to short sheets with 40-60 nm width in the presence of Zn2+. The change of L-fibrils upon addition of Zn2+ ions was not so significant, compared to H-fibrils. However, L-fibrils adopted very different appearance from H-fibrils; L-fibrils were narrower than H-fibrils with width around 10-30nm. Fourier-transform infrared spectroscopy (FTIR, Figure S4) and circular dichroism spectroscopy (CD, Figure S5) measurements on both fibrils further demonstrated their discrepancy on morphology. Therefore, these experimental characterizations on structure provide preliminary evidences to support the observations obtained from in silico modelling. It should be noted that fibril-like structures may have conformational diversity.34,35 Interestingly, the off-register configuration was proposed to provide a pathway to toxic amyloid aggregates.31 Therefore mixed parallel and antiparallel configuration cannot be fully excluded. Here our modelling only focused on the in-register parallel or antiparallel configuration of the fibrils, while the off-register ones are likely the minor species. This was partially confirmed by the M-sequence fibrils,31 and was further validated by the ssNMR measurement on the present structures, as described below. (3) Energetic contributions to the different morphologies To explore the origin of difference in the morphology of the fibril-like structures, the HP in flat configuration as well as the LA in twisted configuration were constructed by restraining the inter-strand twist angle while relaxing the rest part of the system. Then the interaction energy between strands was decomposed into different components (Figure 4). In both HP and LA, the flat configuration has more favorable backbone-backbone (BB) contacts than the twisted one, which is 19.6 kcal/mol/strand more stable in HP and 32.8 kcal/mol/strand in LA. However, the sidechain interactions at the wet interface (WI) favor the twisted configuration over the flat one by 25.5 kcal/mol/strand in HP and 19.3 kcal/mol in LA. Though other interactions also contribute to the stability of the morphology (for example, the backbone-hydrophilic interactions favor the twisted configuration in HP but the flat one in LA), the aforementioned two factors are dominant. Taken together, HP prefers the twisted 5



configuration by 5.2 kcal/mol/strand with respect to the flat one, while LA in flat is 22.4 kcal/mol/strand more stable than the twisted configuration. Similar scenario was observed in literature as well.42 Therefore, the major contribution to the rigidity of the fibrils stems from the generic inter-strand hydrogen bonding network, and is further modulated by van der Waals interactions. The difference in the morphology of the fibril-like structures between the H- and L-sequence motifs could be attributed to the dedicate balance between cooperative enthalpic interactions and the contributions from entropy. (4) Balance between rigidity and flexibility in the assembly of short peptides Compared with LA, HP shows smaller degree of solvent exposure and more twisted structure, as shown by the correlation between solvent-accessible surface area (SASA) and twist angle between neighboring strands (Figure 5). Given the similar size of polar residues at the wet interface, it is expected that HP has larger amount of buried surface than other constructs, mainly because of its tighter packing of hydrophobic residues between sheets at the dry interface. Seemingly, the steric zipper structure in HP enhances the propensity for fibril formation, and increases the rigidity of the structure. On the other hand, both SASA and twist angle in HP have relatively broad range, indicating that the HP fibril has the adaptability to adjust its morphology in response to the environment, and therefore showing some flexibility. Indeed, twisting along an in-plane axis that is perpendicular to the strand orientation was identified as one of the two dominant modes of flexibility for β-sheet, based on an extensive survey on 3516 representative structures from protein database bank.43 Our data demonstrates that the twisting of the fibril is associated with its solvent exposure (Figure 5). Similar conclusion was obtained for YP (Figure S6), as well as the fibril aggregated by the C-terminal truncated α-synuclein with residues 109–140 removed.44 Therefore, the delicate balance between flexibility and rigidity is achieved in HP (or YP), which is probably the reason for its high catalytic activity. (5) Solid-state NMR characterization of the fibrils The detailed structures of the self-assembled fibrils from two different sequences were then determined by ssNMR spectroscopy. By mixing peptides with 15NH-I3 and 13CO-I5 label in a 1:1 ratio, the inter-strand distance between 15NH-I3 and 13CO-I5 could be measured. The rotational-echo double-resonance (REDOR) technique45 was used to measure inter-nuclear distances between specific 13C-15N spin pairs. As shown in Figure 6a, the experimentally determined 13CO intensity ratio S/S0 as a function of mixing time tells the inter-strand 15NH-I3 and 13CO-I5 distance to be 40.5 Å for L fibrils, indicating the formation of LA. The distance was obtained by fitting data with a simulated REDOR experiment run by the software SIMPSON.46 This inter-nuclear separation is in good agreement with ~4.1 Å distance predicted from our MD simulations for LA. In contrast, H-sequence fibrils showed no significant decay in the REDOR signal, indicating a more than 6 Å distance between 15NH-I3 and 13CO-I5. Therefore, the H-sequence fibrils do not adopt the antiparallel configuration, and is likely to be arranged in the parallel one. In order to confirm the parallel configuration, further experiment was carried out on H-sequence fibrils with only 13CO-I5 labelled. The inter-strand 13C-13C dipole-dipole couplings were determined using PITHIRDS-CT experiment47 performed at 293K with magic-angle spinning (MAS) of 20.00 kHz 6


Page 6 of 17

Page 7 of 17 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


(Figure 6b). The inter-strand

13C-13C

distance between two neighboring I5 residues in HP was

measured to be 4.80.2 Å, which was obtained by fitting data with a simulated PITHIRDS-CT experiment run by the software SIMPSON.46 This distance is in line with the calculated 4.87±0.22 Å obtained from our in silico simulations for HP. Therefore, ssNMR experiments confirmed that HP and LA are the preferred configurations for the H- and L- sequence fibrils, respectively. (6) Zinc coordination geometry It should be noted that, we failed to obtain the histidine sample with sidechain-isotope-labelled. Therefore the conformation of the imidazole ring for coordination cannot be determined experimentally. Instead, exhaustive sampling of rotamer space was carried out by rotating the χ1 and χ2 angles of histidine sidechain based on the well-defined architectures. More details could be found in the Supplementary Information. According to the survey of carbonic anhydrase experimental structures, the three histidine residues at the active site used one Nδ and two Nε positions to bound Zn ions. Such configuration was identified to be stable in HP, LA and YP structures as well, and was well kept in 200 ns MD simulations. Because one of the functions of these fibrils is to mimic the catalytic role of carbonic anhydrase, such a Nε-Nε-Nδ configuration was adopted to study the catalytic reaction pathway by using ab initio calculations, as described below, though other configurations cannot be excluded. Presumably, the capacity to accommodate various metal ions at the active site makes the fibril-like structure a robust catalyst for different reactions.11,48 Therefore, we explored the coordination of either Zn2+ (complex 1 discussed in the following session) or Cu2+ at the active site with the Nε-Nε-Nδ configuration of His residues. With an additional water/hydroxide ion bound to the metal ion, a tetrahedral coordination was formed, which resembles the structure in the active-site cavity of carbonic anhydrase.12 This coordination sphere could be partially validated by electron paramagnetic resonance (EPR) spectrum of Cu2+ bound to the H-sequence fibrils, which shows gII = 2.27 and AII = 167 G, a characteristic of 3N1O coordination environment of Cu2+.28 Geometry optimized structures containing different metal ions could be well aligned, with the root-mean-square deviation (RMSD) of only 0.66 Å (Figure S7). Seemingly, the binding pocket is pretty rigid to well accommodate metal ions which favor catalysis of different reactions, as further described below. (7) Reaction pathway for the catalytic reaction Based on the scaffold obtained, we further explored the catalytic mechanism. Representative structures were selected from the trajectory generated with MD simulations, where the two neighboring strands has a twist angle of 16° for HP and 4° for LA, mimicking the twisted and flat morphology of the fibrils. These two strands of peptide, metal ion, water molecules nearby and pNPA ligand were used to build a cluster model in each system. As shown in Figure 7, the reaction catalyzed by HP is initiated by the proton migration from the Zn2+-bound water in complex 1 to the deprotonated nitrogen atom on the fourth histidine sidechain, a distal histidine that is indirectly involved in coordination through a two-molecule water chain. Through TS1/2, this proton transfer step results in a hydroxide bound to the metal center (complex 2), which is endothermic by 4.31 kcal/mol (G0) with the energy barrier (G0) of 11.03 kcal/mol. Then, the pNPA substrate approaches the negatively charged hydroxide ligand 7



in complex 2 to form an intermediate complex 3. Subsequently, the hydroxide group nucleophilic attacks the carbon atom of the carbonyl group of pNPA to form a complex 4 through TS3/4. The G0 of this step is 22.15 kcal/mol. In this process, the C1-O5 bond distance of the NPA moiety is significantly elongated from 1.37 Å in 3 to 1.54 Å in 4. Consequently, the C1-O5 bond cleavage occurs through TS4/5 to form a complex 5, which produces an acetic acid moiety and a p-nitrophenoxide anion. This process is significantly exothermic by 13.64 kcal/mol and occurs with G0 value of 16.35 kcal/mol. Successively, the proton transfers from the hydroxide group of the acetic acid moiety to the p-nitrophenoxide anion through TS5/6, with the G0 and G0 values of barrierless and -13.57 kcal/mol, respectively. In this step the product p-nitrophenol is produced and is released from the active site (complex 6). In the last step, the proton transfer from the distal imidazole moiety to the acetoxyl group on Zn2+ through TS6/7 to form the acetic acid (in complex 7) which is another product. This step occurs with the G0 and G0 values of -5.70 and 4.81 kcal/mol, respectively. Overall, the rate-determining step of the whole reaction is the nucleophilic attack step. To explore the effect of twisting between neighboring strands, two systems based on HP with either increased (21°, HP21) or decreased (14°, HP14) twist angle were manually constructed. The initial complex 1 and the TS3/4 with the highest free energy along the original reaction coordinate were optimized. The free energy differences between the two are 25.55 kcal/mol in HP14 and 28.25 kcal/mol in HP21, respectively. Both two shows higher free energy barrier than the abovementioned structure with the optimal 16° twist angle. Seemingly an optimal configuration induced by the proper twist angle is critical for its catalytic activity. For the LA system, the reaction proceeds through a similar pathway as mentioned above, and the rate-determining step of this LA mediated reaction is also the nucleophilic attack step, which occurs with the G0 value of 22.68 kcal/mol, ~0.5 kcal/mol higher than that of 22.15 kcal/mol in HP. In contrast, calculations based on YP structure shows a much lower free energy difference between the complex 1 and TS3/4 (17.44 kcal/mol). Though the accuracy of density functional theory (DFT) calculations is 2~3 kcal/mol, the calculated difference in free energy barrier height of the two reactions is in line with the experimental observations that the H-sequence fibrils are 2.8 times more active than the L-sequence ones, while the Y-sequence fibrils are even more active than the H-sequence fibrils. The active-site cavity of carbonic anhydrase contains a variety of residues: besides the three catalytic histidines and a water molecule that coordinate to the Zn2+, a well-defined hydrogen bonding network was formed with His64, Glu106, Thr199, etc (PDB entry: 2CBA). In the present system (taking HP as an example), other than the three histidines coordinated to Zn2+, a fourth histidine is found to have active role in accepting the excessive proton, which is likely to mimic the role of His64 in natural enzyme. However, other functional groups in natural enzyme are missing in the present construct. Consequently, the natural carbonic anhydrase (kcat/kM = 2550 M-1 s-1)49 for the hydrolysis of pNPA was found to be tens of times more efficient than that of HP (kcat/kM = 62±2 M-1 s-1) 11 and YP (kcat/KM values ranging from 98±3 M-1 s-1 29 to 355±15 M-1 s-1),30 respectively. Though different elementary steps in the catalytic cycle may affect the efficiency of the catalyst, which should be considered during rational design,50 the difference between microenvironment at the active site between natural enzyme and the present self-assembled fibrils should account for the major difference in catalytic activity. Therefore incorporating proper residue in the sequence to mimic the environment in natural enzyme is likely to be a 8


Page 8 of 17

Page 9 of 17 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


working strategy to further improve the catalytic activity of functional fibrils. Conclusion In this work, we determined the atomic structures of self-assembled fibrils formed by different peptide sequences by using combined molecular modelling and ssNMR measurements. These aggregates had been reported to be active in the catalytic reaction of pNPA hydrolysis. Our data shows that the backbone-backbone interactions between neighboring β-strands are dominant for fibril stability, and the interplay between different regions of the building blocks cooperatively determines its morphology and thus affects its catalytic activity. The higher catalytic activity of fibrils is a consequence of balanced rigidity and flexibility in the assembly of short peptides. We further explored the mechanism for the hydrolysis reaction with DFT calculations, and found that the rate-determining step of the whole reaction is the nucleophilic attack process. The difference in local environment at the active site between natural enzyme and the self-assembled fibrils studied in this work should account for the discrepancy in catalytic activity. The present work provides understanding about the structure and function of self-assembled fibrils formed with short-peptides at atomic level, and thus sheds new insight on designing aggregates with better functions. Corresponding Author [email protected] or [email protected] or [email protected] Notes The authors declare no competing financial interest. ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (Grant Nos. 21503107 and 21773115 to H.D.), the Natural Science Foundation of JiangSu Province (Grant No. SBK2015041570), the “Jiangsu Specially-Appointed Professor” program (to H.D.), the “Fundamental Research Funds for the Central Universities” (021514380012 and 021514380014 to H.D.), and the “1000 Young Talents” Program (to J. L.). Parts of the calculations were performed using computational resources on an IBM Blade cluster system from the High Performance Computing Center (HPCC) of Nanjing University and the Shenzhen Supercomputer Center. The TEM images were obtained at National Center for Protein Science Shanghai. NMR experiments were performed at NMR core facility of School of Life Science and Technology at ShanghaiTech University. We thank Mr. Tianyu Du for suggestion, and we also appreciate Dr. Kent Thurber and Robert Tycko from National Institute of Health, USA for providing PITHIRDs solid-state NMR pulse program.

Supporting Information Available: The system setup, computational details, representative structures of HP, LA and YP, list of PDB used for survey, and experimental details. This information is available free of charge via the Internet at http://pubs.acs.org.

9



REFERENCE (1) Maury, C. P. J. Amyloid and the origin of life: self-replicating catalytic amyloids as prebiotic informational and protometabolic entities. Cell. Mol. Life Sci. 2018, 75, 1499. (2) Ulijn, R. V.; Smith, A. M. Designing peptide based nanomaterials. Chem. Soc. Rev. 2008, 37, 664. (3) Whitesides, G. M.; Mathias, J. P.; Seto, C. T. Molecular self-assembly and nanochemistry: a chemical strategy for the synthesis of nanostructures. Science 1991, 254, 1312. (4) Li, Y.; Cao, Y. The physical chemistry for the self-assembly of peptide hydrogels. Chinese J. Polym. Sci. 2018, 1. (5) Lakshmanan, A.; Zhang, S.; Hauser, C. A. Short self-assembling peptides as building blocks for modern nanodevices. Trends Biotechnol. 2012, 30, 155. (6) Habibi, N.; Kamaly, N.; Memic, A.; Shafiee, H. Self-assembled peptide-based nanostructures: smart nanomaterials toward targeted drug delivery. Nano Today 2016, 11, 41. (7) Zozulia, O.; Dolan, M.; Korendovych, I. Catalytic peptide assemblies. Chem. Soc. Rev. 2018, 47, 3621. (8) Jang, H.-S.; Lee, J.-H.; Park, Y.-S.; Kim, Y.-O.; Park, J.; Yang, T.-Y.; Jin, K.; Lee, J.; Park, S.; You, J. M. Tyrosine-mediated two-dimensional peptide assembly and its role as a bio-inspired catalytic scaffold. Nat. Commun. 2014, 5, 3665. (9) Omosun, T. O.; Hsieh, M.-C.; Childers, W. S.; Das, D.; Mehta, A. K.; Anthony, N. R.; Pan, T.; Grover, M. A.; Berland, K. M.; Lynn, D. G. Catalytic diversity in self-propagating peptide assemblies. Nat. Chem. 2017, 9, 805. (10) Zhang, C.; Xue, X.; Luo, Q.; Li, Y.; Yang, K.; Zhuang, X.; Jiang, Y.; Zhang, J.; Liu, J.; Zou, G. Self-assembled peptide nanofibers designed as biological enzymes for catalyzing ester hydrolysis. Acs Nano 2014, 8, 11715. (11) Rufo, C. M.; Moroz, Y. S.; Moroz, O. V.; Stohr, J.; Smith, T. A.; Hu, X.; DeGrado, W. F.; Korendovych, I. V. Short peptides self-assemble to produce catalytic amyloids. Nat. Chem. 2014, 6, 303. (12) Alterio, V.; Di Fiore, A.; D’Ambrosio, K.; Supuran, C. T.; De Simone, G. Multiple Binding Modes of Inhibitors to Carbonic Anhydrases: How to Design Specific Drugs Targeting 15 Different Isoforms? Chem. Rev. 2012, 112, 4421. (13) Supuran, C. T. How many carbonic anhydrase inhibition mechanisms exist? J. Enzyme Inhib. Med. Chem. 2016, 31, 345. (14) Riccardi, D.; Yang, S.; Cui, Q. Proton transfer function of carbonic anhydrase: Insights from QM/MM simulations. BBA Proteins Proteom. 2010, 1804, 342. (15) Riccardi, D.; Cui, Q. pKa Analysis for the Zinc-Bound Water in Human Carbonic Anhydrase II: Benchmark for “Multiscale” QM/MM Simulations and Mechanistic Implications. J. Phys. Chem. A 2007, 111, 5703. (16) Riccardi, D.; König, P.; Guo, H.; Cui, Q. Proton transfer in carbonic anhydrase is controlled by electrostatics rather than the orientation of the acceptor. Biochemistry 2008, 47, 2369. (17) Smedarchina, Z.; Siebrand, W.; Fernández-Ramos, A.; Cui, Q. Kinetic isotope effects for concerted multiple proton transfer: a direct dynamics study of an active-site model of carbonic anhydrase II. J. Am. Chem. Soc. 2003, 125, 243. (18) Cui, Q.; Karplus, M. Is a “proton wire” concerted or stepwise? A model study of proton 10


Page 10 of 17

Page 11 of 17 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


transfer in carbonic anhydrase. J. Phys. Chem. B 2003, 107, 1071. (19) Bottoni, A.; Lanza, C. Z.; Miscione, G. P.; Spinelli, D. New Model for a Theoretical Density Functional Theory Investigation of the Mechanism of the Carbonic Anhydrase: How Does the Internal Bicarbonate Rearrangement Occur? J. Am. Chem. Soc. 2004, 126, 1542. (20) Shimahara, H.; Yoshida, T.; Shibata, Y.; Shimizu, M.; Kyogoku, Y.; Sakiyama, F.; Nakazawa, T.; Tate, S.-i.; Ohki, S.-y.; Kato, T. Tautomerism of histidine 64 associated with proton-transfer in catalysis of carbonic anhydrase. J. Biol. Chem. 2007, 282, 9646. (21) Sjöblom, B.; Polentarutti, M.; Djinović-Carugo, K. Structural study of X-ray induced activation of carbonic anhydrase. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 10609. (22) Miscione, G. P.; Stenta, M.; Spinelli, D.; Anders, E.; Bottoni, A. New computational evidence for the catalytic mechanism of carbonic anhydrase. Theor. Chem. Acc. 2007, 118, 193. (23) Sattler, W.; Parkin, G. Structural characterization of zinc bicarbonate compounds relevant to the mechanism of action of carbonic anhydrase. Chem. Sci. 2012, 3, 2015. (24) Piazzetta, P.; Marino, T.; Russo, N. Promiscuous ability of human carbonic anhydrase: QM and QM/MM investigation of carbon dioxide and carbodiimide hydration. Inorg. Chem. 2014, 53, 3488. (25) Jiao, D.; Rempe, S. B. Combined Density Functional Theory (DFT) and Continuum Calculations of pKa in Carbonic Anhydrase. Biochemistry 2012, 51, 5979. (26) Hakkim, V.; Subramanian, V. Role of second coordination sphere amino acid residues on the proton transfer mechanism of human carbonic anhydrase II (HCA II). J. Phys. Chem. A 2010, 114, 7952. (27) Dong, H. A computational study towards understanding the mechanism of phosphodiester cleavage by two mononuclear Zn (ii) complexes. Phys. Chem. Chem. Phys. 2010, 12, 10434. (28) Makhlynets, O. V.; Gosavi, P. M.; Korendovych, I. V. Short Self-Assembling Peptides Are Able to Bind to Copper and Activate Oxygen. Angew. Chem. 2016, 55, 9017. (29) Lengyel, Z.; Rufo, C. M.; Moroz, Y. S.; Makhlynets, O. V.; Korendovych, I. V. Copper-Containing Catalytic Amyloids Promote Phosphoester Hydrolysis and Tandem Reactions. ACS Catal. 2017, 8, 59. (30) Al-Garawi, Z.; McIntosh, B.; Neill-Hall, D.; Hatimy, A.; Sweet, S.; Bagley, M.; Serpell, L. The amyloid architecture provides a scaffold for enzyme-like catalysts. Nanoscale 2017, 9, 10773. (31) Lee, M.; Wang, T.; Makhlynets, O. V.; Wu, Y.; Polizzi, N. F.; Wu, H.; Gosavi, P. M.; Stoehr, J.; Korendovych, I. V.; DeGrado, W. F.; Hong, M. Zinc-binding structure of a catalytic amyloid from solid-state NMR. Proc. Natl. Acad. Sci. U.S.A. 2017, 114, 6191. (32) Zhang, S.; Holmes, T.; Lockshin, C.; Rich, A. Spontaneous assembly of a self-complementary oligopeptide to form a stable macroscopic membrane. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 3334. (33) Edgcomb, S. P.; Murphy, K. P. Variability in the pKa of histidine side-chains correlates with burial within proteins. Proteins: Struct., Funct., Bioinf. 2002, 49, 1. (34) Tycko, R. Physical and structural basis for polymorphism in amyloid fibrils. Protein Sci. 2014, 23, 1528. (35) Al-Garawi, Z. S.; Morris, K. L.; Marshall, K. E.; Eichler, J.; Serpell, L. C. The diversity and utility of amyloid fibrils formed by short amyloidogenic peptides. Interface Focus 2017, 7, 20170027. (36) Do, T. D.; LaPointe, N. E.; Sangwan, S.; Teplow, D. B.; Feinstein, S. C.; Sawaya, M. R.; Eisenberg, D. S.; Bowers, M. T. Factors That Drive Peptide Assembly from Native to Amyloid Structures: Experimental and Theoretical Analysis of [Leu-5]-Enkephalin Mutants. J. Phys. Chem. B 2014, 118, 7247. 11



Page 12 of 17

(37) Chothia, C. Conformation of twisted β-pleated sheets in proteins. J. Mol. Biol. 1973, 75, 295. (38) Pauling, L.; Corey, R. B. The pleated sheet, a new layer configuration of polypeptide chains. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 251. (39) Hakansson, K.; Carlsson, M.; Svensson, L. A.; Liljas, A. Structure of native and apo carbonic anhydrase II and structure of some of its anion-ligand complexes. J. Mol. Biol. 1992, 227, 1192. (40) Gsponer, J.; Haberthür, U.; Caflisch, A. The role of side-chain interactions in the early steps of aggregation: Molecular dynamics simulations of an amyloid-forming peptide from the yeast prion Sup35. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 5154. (41) Plumley, J. A.; Dannenberg, J. The importance of hydrogen bonding between the glutamine side chains to the formation of amyloid VQIVYK parallel β-sheets: An ONIOM DFT/AM1 Study. J. Am. Chem. Soc. 2010, 132, 1758. (42) Knowles, T. P.; Fitzpatrick, A. W.; Meehan, S.; Mott, H. R.; Vendruscolo, M.; Dobson, C. M.; Welland, M. E. Role of intermolecular forces in defining material properties of protein nanofibrils. Science 2007, 318, 1900. (43) Emberly, E. G.; Mukhopadhyay, R.; Tang, C.; Wingreen, N. S. Flexibility of β ‐ sheets: Principal component analysis of database protein structures. Proteins 2004, 55, 91. (44) Iyer, A.; Roeters, S. J.; Kogan, V.; Woutersen, S.; Claessens, M. M. A. E.; Subramaniam, V. C-terminal truncated α-synuclein fibrils contain strongly twisted β-sheets. J. Am. Chem. Soc. 2017, 139, 15392. (45) Gullion, T.; Schaefer, J. Rotational-echo double-resonance NMR. J. Magn. Reson. (1969) 1989, 81, 196. (46) Bak, M.; Rasmussen, J. T.; Nielsen, N. C. SIMPSON: a general simulation program for solid-state NMR spectroscopy. J. Magn. Reson. 2011, 213, 366. (47) Tycko, R. Symmetry-based constant-time homonuclear dipolar recoupling in solid state NMR. J. Chem. Phys. 2007, 126, 064506. (48) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.; Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J. V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams; Ding, F.; Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.; Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Throssell, K.; Montgomery Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J. J.; Brothers, E. N.; Kudin, K. N.; Staroverov, V. N.; Keith, T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J.

Wallingford, CT,

2016. (49) Verpoorte, J. A.; Mehta, S.; Edsall, J. T. Esterase activities of human carbonic anhydrases B and C. J. Biol. Chem. 1967, 242, 4221. (50) Korendovych, I. V.; DeGrado, W. F. Catalytic efficiency of designed catalytic proteins. Curr. Opin. Struct. Biol. 2014, 27, 113.

12


Page 13 of 17 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


Figures

Figure 1. Building block peptide and the aggregated structures. Top panel: the high activity sequence peptide with amphiphilic nature; bottom panel: flat (left) and twisted (right) fibril morphologies. Space-filling mode shows the tight packing at the dry interface in the aggregates. Inset shows the cartoon representative of the aggregates.

Figure 2. Free energy profiles of the dissociation of two β-strands formed with H- (left) or L-sequence (right) peptides. Both the parallel (red line) and the antiparallel (blue line) configurations were explored. The inset shows the initial structures of the β-sheet peptide dimers in the parallel (left) and antiparallel (right) configurations and the arrows indicate the direction for dissociation. The reaction coordinate was defined as the distance between the center of mass of two strands.

13



Figure 3. The distribution of twisting angles between two neighboring β-strands in HP (red), LA (blue) and natural carbonic anhydrase (green). The angle was defined as the angles between two principal axes of neighboring strands. Trajectories of the last 200 ns simulations of HP or LA were used. For natural enzyme, a survey of 478 structures deposited in protein database bank (PDB) was carried out. The distribution was normalized for comparison.

Figure 4. Contribution from different segments towards the stability of twisted and flat morphologies in HP and LA. The interaction energies between different segments of backbone (BB), the wet interface (WI) and the dry interface (DI) were calculated, and the energy difference, ΔE, between the twisted and the flat configuration of each segment were used to represent the contribution to the relative stability of the structure.

14


Page 14 of 17

Page 15 of 17 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


Figure 5. Correlation between solvent exposure (represented by SASA) and the structure twisting for HP and LA.

Figure 6. Solid-state NMR REDOR (left) and PITHIRDS-CT (right) measurements of fibril inter-strand distance. The inter-molecular 13C–15N magnetic dipole–dipole couplings were measured for both H- and L-fibrils with 15NH-I3 and 13CO-I5 labelled peptides mixed in a 1:1 ratio in the specific fibril samples (left). The inter-molecular 13C–13C magnetic dipole–dipole couplings in 13CO-I5 labeled H fibrils were also determined (right). Comparison with numerical simulations confirms L-fibrils form antiparallel β-sheet structure with intermolecular distances of ~4.0 Å between 15NH-I3 and 13CO-I5, while H-fibrils form in-register parallel β-sheet structure with intermolecular distances of ~4.8 Å for 13CO-I5.

15



Figure 7. Catalytic reaction mechanism and the optimized structures along the reaction pathway. The β-strand is represented in cartoon mode, the sidechain of His residue is shown in stick mode, waters and pNPA are shown in CPK mode. Key region of the transition state in each elementary reaction is cycled with red dashed-line. Some key atoms are labelled on complex 3. The calculated free energy differences between 1 and TS3/4 for HP (red), LA (blue), HP14 (pink), HP21 (black), and YP (green) are labelled in the free energy profile.

16


Page 16 of 17

Page 17 of 17 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


Table of contents graphic

17


Principles governing catalytic activity of self-assembled short peptides

Recommend Documents