Amyloid fibril aggregation: An insight into the

0 downloads 0 Views 1MB Size Report
Aug 29, 2017 - Results of Thioflavin T binding assay, transmission electron microscopy, and ... force microscopy (AFM) images have revealed that a large of.
Biochemical and Biophysical Research Communications xxx (2017) 1e6

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

Amyloid fibril aggregation: An insight into the underwater adhesion of barnacle cement Xingping Liu 1, Chao Liang 1, Xinkang Zhang, Jianyong Li, Jingyun Huang, Ling Zeng, Zonghuang Ye, Biru Hu*, Wenjian Wu Department of Chemistry and Biology, College of Science, National University of Defense Technology, Changsha, 410073, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 25 August 2017 Accepted 29 August 2017 Available online xxx

Barnacles robustly adhere themselves to diverse submarine substrates through a proteinaceous complex termed the “barnacle cement”. Previous studies have indicated that certain peptides derived from some barnacle cement proteins can self-assemble into amyloid fibrils. In this study, we assessed the selfassembly behavior of a full-length 19 kDa cement protein from Balanus albicostatus (Balcp19k) in different buffers. Results of Thioflavin T binding assay, transmission electron microscopy, and Fourier transform infrared spectroscopy suggested that the bacterial recombinant Balcp19k was able to aggregate into typical amyloid fibrils. The time required for the self-assembly process was close to that required for the complete curing of barnacle cement complex. Moreover, the solubility of Balcp19k amyloid deposits in guanidine hydrochloride and urea was same as that of the cured cement. These results indicated the inherent self-assembling nature of Balcp19k, implying that the amyloid fibril formation plays a critical role in barnacle cement curing procedure and its insolubility. Our results should be conducive to understanding barnacle underwater adhesion mechanisms and have implications in the development of new-generation antifouling techniques and in the designing of novel wet adhesives for biomedical and technical applications. © 2017 Published by Elsevier Inc.

Keywords: Barnacle cement Balcp19k Self-assembly Amyloid fibrils Underwater adhesion

1. Introduction Barnacles are gregarious sessile crustaceans that live in intertidal zones. They attach themselves, tenaciously and permanently, to a variety of submarine substrates. Recently, increasing attention has been focused on the underwater adhesion mechanisms of barnacles with the objective of developing new-generation antifouling techniques [1] and designing of novel wet adhesives for biomedical and technical applications [2]. The underwater adhesion of barnacles depends on a multi-protein complex called the “barnacle cement” [3]. Barnacle cement is synthesized in the cement gland and is then secreted via a canal system to the interface between the base shell of barnacle and foreign substrates [4]. At the site of adhesion, the adhesive first displaces the surfacebound water layer and then couples with both the surfaces and finally cures to achieve successful underwater attachment. Because

* Corresponding author. E-mail address: [email protected] (B. Hu). 1 These authors contributed equally to this work.

of strong intramolecular and/or intermolecular interactions, the cured cement is intrinsically insoluble; this greatly hampers the biochemical and biophysical investigations into its nature [5,6]. It has been reported that common denaturing reagents, chaotropes, surfactants, and reductants are not sufficient for rendering the barnacle cement fully soluble [7e11]. Kamino and collaborators [12e14] isolated and characterized a few proteins from the Megabalanus rosa secondary cement, by combining physical disruption of noncovalent interactions with chemical cleavage of certain covalent bonds within the cement proteins (cp). These cement proteins were named according to their apparent molecular weights revealed by SDS-PAGE; for example, some of the proteins were named as Mrcp100k, Mrcp52k, Mrcp20k, and Mrcp19k. As of date, a few curing/cross-linking mechanisms, including those that involve intermolecular disulfide bond based crosslinking [13], quinone-protein cross-linking [7], and glutamyllysine cross-linking [7,15], have been proposed. Although these hypotheses could explain certain phenomena, no conclusive evidence that confirms them has been obtained as yet. In the last few years, a view that amyloid fibrils are involved in the underwater adhesion of barnacles has become increasingly prominent. There

http://dx.doi.org/10.1016/j.bbrc.2017.08.136 0006-291X/© 2017 Published by Elsevier Inc.

Please cite this article in press as: X. Liu, et al., Amyloid fibril aggregation: An insight into the underwater adhesion of barnacle cement, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2017.08.136

2

X. Liu et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6

are three main facts indicating that amyloid fibrils are the main components of the cured barnacle cement. Firstly, it was demonstrated that b-sheet, which is the structural basis of amyloid fibrils, is the main secondary structure of cured cement. Secondly, atomic force microscopy (AFM) images have revealed that a large of portion cured cement consists of straight, smooth, and unbranched fibrils with a diameter of about 10e40 nm [16,17]. Thirdly, cured barnacle cement can be positively stained by an amyloid fibrilspecific dye, Thioflavin T (ThT). Kamino [18] has previously reported that certain peptides from Mrcp52k, which is considered to be responsible for the bulk cohesion, possess the ability to selfassemble into amyloid fibrils. However, whether a full-length cement protein can form amyloid fibrils, has not been investigated. The cp19k is a simple protein which does not contain any posttranslational modifications. It is characterized by alternative distribution of hydrophilic and charged blocks (Fig. S2) and has a bias for six amino acids, namely Ser, Thr, Ala, Gly, Val, and Lys [4,9]. Previously, recombinant Mrcp19k was demonstrated to be able to nonspecifically adsorb to different types of foreign materials. Thus, cp19k was categorized as a surface coupling protein [14]. However, it has not been conclusively shown whether cp19k is in the amyloid state when it participates in underwater adhesion. Here, we expressed a Balanus albicostatus cp19k homolog (Balcp19k) in Escherichia coli and investigated its self-assembly behavior in different buffers by ThT staining, transmission electron microscopy (TEM), and FTIR spectroscopy. Because of remarkable differences between the in vivo cement gland and in vitro sea water, the experiments were conducted in two buffers with pH and salt concentration either similar to that in the seawater or in the fluid present in the cement gland. The discrepancy in the solubility of Balcp19k formed fibrils and cured cement were also compared. The present study focused on the self-assembly property of Balcp19k. We believe that this research would aid in the understanding of mechanisms of underwater curing and adhesion of barnacle cement and would be useful for designing of new adhesives inspired by the barnacles. 2. Materials and methods 2.1. Gene cloning and vector construction The Balcp19k (GenBank, AB242295.1) gene had been already cloned into T vector as described earlier [19]. The Balcp19k gene was released from the recombinant T-Vector by digestion with EcoR

I and Not I, and then subcloned into the pET-21a vector (EMD Chemicals, USA) at the EcoR I-Not I site (Fig. 1A). Because of the presence of cysteine residues at the 5th and 39th positions (cysteine) of Balcp19k, this protein is prone to the formation of intramolecular disulfide bonds, which leads to protein multimerization (unpublished data). We, therefore, deliberately substituted these cysteine residues with serine, which is the most abundant amino acid in Balcp19k; these point mutations were performed using MutanBEST Kit (TaKaRa, USA). Finally, the sequence of the constructed plasmid was confirmed by Sanger sequencing. The detailed information is provided in the supplementary materials. 2.2. Expression and purification of recombinant Balcp19k The pET21a-Balcp19k plasmid was transformed into E. coli BL21 (DE3) component cells (Tiangen, China) and a single colony of this strain from a freshly prepared culture plate was grown overnight in 2 mL ZYM-505 culture medium containing 100 mg/mL ampicillin at 37  C, with shaking (220 rpm). The culture was transferred into 2 L ZYM-5052 auto-induction medium [20] containing 100 mg/mL ampicillin and incubated at 37  C, with shaking at 220 rpm for about 16 h. The transformed cells were harvested by centrifugation at 12,000 g for 15 min at 4  C. The pellet was gently washed twice with precooled 0.01 M phosphate buffer solution (pH 7.4) and the cells were lysed by resuspending the pellet in 5 mL lysis buffer (8 M urea, 0.1 M sodium phosphate, 0.01 M Tris-HCl, pH 8.0) per gram wet weight of the pellet. The lysate was then centrifuged at 16,000 g for 60 min at 4  C. The supernatant was collected, analyzed by SDSPAGE, and used for the purification of the recombinant protein. The supernatant and Ni-NTA slurry (QIANGEN, USA) were mixed together, incubated overnight at ambient temperature, and then loaded in an empty column. The Ni-NTA resin was subsequently washed twice with the wash buffer (8 M urea, 0.1 M sodium phosphate, 0.01 M Tris-HCl, pH 6.3). The recombinant Balcp19k protein was eluted with the elution buffer (0.01 M phosphate sodium, 0.4 M NaCl, 0.4 M imidazole, 10% (v/v) acetic acid). The eluted protein was loaded onto a preparative Resource reverse phase column (GE, USA) in batches. The column was washed with the wash fluid (0.1% (v/v) TFA in ultrapure water) and the protein was eluted with the elution fluid (0.1% (v/v) TFA, 80% (v/v) acetonitrile in ultrapure water). Finally, the purified recombinant Balcp19k was lyophilized and stored at 80  C.

Fig. 1. Vector construction, expression, and purification of Balcp19k. (A). Schematic illustration of the plasmid constructed for the expression of Balcp19k. (B) Tris-Bis SDS-PAGE analysis for the purification of recombinant Balcp19k (rBalcp19k). Lane M, protein standard; Lane WC, whole cell lysate extracted with 8 M Urea; Lane PR, purified rBalcp19k. (C) Western-blot analysis of rBalcp19k. Lane MH, protein standard with His-tag; Lane PR, purified rBalcp19k. (D) Result of MALDI-TOF analysis of rBalcp19k.

Please cite this article in press as: X. Liu, et al., Amyloid fibril aggregation: An insight into the underwater adhesion of barnacle cement, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2017.08.136

X. Liu et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6

2.3. SDS-PAGE and western blot analysis The lysate and the eluted solution were mixed with NuPAGE 4X LDS sample loading buffer (Thermo fisher scientific, USA) and boiled at 70  C for 10 min. To perform the insolubility assay, the fibril deposits of rBalcp19k were treated with 8 M urea and 6 M guanidine hydrochloride (GdnHCl). After the treatment, GdnHCl was completely removed by dialysis against 0.5% (v/v) acetic acid for about 24 h. Thereafter, all the samples, along with a prestained protein standard (Thermo fisher scientific, USA) and a Bench mark His-tag ladder (Thermo fisher scientific, USA), were loaded onto a precast 4e12% Bis-Tris gel. The electrophoresis was performed at 150 V for 35 min using MES running buffer (Thermo fisher scientific, USA). Finally, the protein was detected by staining with Coomassie brilliant blue R-250 (CBB R-250) or by western blot (Fig. 1B and C). For CBB R-250 staining, a common procedure was used and the images of gels were taken using FluorChem FCM 2 system (ProteinSample, USA). For western blot, the samples were transferred onto polyvinylidene difluoride membranes through semi-dry electrophoretic transfer at a current of 3 mA/cm2 of the membrane for 30 min using Tris-Glycine transfer buffer (25 mM Tris, 0.2 M Glycine, 20% (v/v) methanol, pH 10.0). The target protein was detected using anti-His polyclonal mouse antibody conjuncted with horse radish peroxidase (Thermo fisher scientific, USA) at a dilution of 1:10,000. After incubation with the antibody, the membrane was blocked, washed, and developed with SuperSignal West Pico Luminoi/Enhancer Solution (Thermo Scientific Scientific, USA). The membrane was imaged using a C-DiGit Blot Scanner (LI-COR, USA). 2.4. Matrix-assisted laser desorption ionization (MALDI)-time-offlight (TOF) mass spectrometry MALDI-TOF was performed at Beijing Baitaipaike Biotechnology Co. LTD. After reverse-phase chromatography, the purified rBalcp19k was subjected to molecular test using an ultrafleXtreme MALDI TOF system (Bruker, Germany). The sample preparation and experimental procedure were as per the standard protocol [21], implemented without modification. 2.5. ThT binding assay and fluorescence microscopy We utilized 0.01 M sodium phosphate buffer (pH 8.0) containing 0.6 M NaCl as the “seawater analog” [22]. Although no definite pH value and salt concentration of the fluid in the barnacle cement gland have been reported, secretory proteins usually were cumulated in vacuoles under acidic pH conditions [23]. Besides, the physiological saline concentration in marine creatures is normally less than 150 mM [24]. We thus used approximately 0.01 M sodium acetate buffer (pH 4.0) containing 0.1 M NaCl as the “cement gland fluid analog”. The lyophilized rBalcp19k was dissolved in the seawater analog as well as in the cement gland fluid analog at a concentration of 5 mg/mL. The protein solutions (100 mL) were loaded in a 96-well plate and an equal volume of ThT aqueous solution was added into the wells at a final concentration of 20 mM. In addition, the buffer solutions were subjected to the same treatment to serve as the corresponding blank control and the values obtained for it were subtracted for the background absorption. After 30 min of incubation at 25  C and shaking for 10 s at 1200 rpm on a Multiskan Spectrum fluorescence microplate reader (Thermo fisher scientific, USA), fluorescence intensities were measured at 485 nm with excitation at 430 nm. The individual data points represent the mean value of three independent measurements. After the ThT binding assay, the samples in the 96-well plate were directly imaged to obtain fluorescence micrographs using a DMI4000B

3

microscope (LEICA, German) with blue excitation in the range from 420 to 485 nm and an emission beyond 515 nm. 2.6. Transmission electron microscopy TEM was performed at the Life Science Microscopy Center at Xiangya School of Medicine, Central South China University. The sample for TEM was prepared following the standard protocol without any modification [21]. Bright-field images at different magnifications were obtained using a Tecnai G2 Spirit BioTWIN Field emission transmission electron microscope (FEI, USA), with an accelerating voltage at 120 kV. 2.7. FTIR spectrum assay The amyloid fibril sediments formed by rBalcp19k were exchanged with D2O (Sigma, USA) for 48 h to eliminate the residual H2O. Thereafter, KBr pellets were prepared for IR spectra collection. The IR spectra were collected by a Nicolet-360 transmission infrared spectrometer (Thermo fisher scientific, USA) from wavenumbers 4000 cm 1 to 400 cm1 at a resolution of 1 cm1. Each spectrum was the mean of 64 scans. The resulting spectra were processed to determine the relative fractions of different secondary structures. Firstly, self-deconvolution algorithm was applied to enhance the resolution of latent non-significant components in the Amide I; band (1700 cm 1 to 1600 cm1). We employed the Happ-Genzel aperdization with a full width at half-height (FWHH) of 13.5 cm1 and an enhancement factor of 2 defined in Nicolet Omnic version 8.2 (Thermo fisher scientific, USA) to carry out self-deconvolution. Furthermore, the deconvoluted spectra were fitted using Peakfit version 4.1.2 (Systat, USA). The detailed procedure was as follows: the baseline type was set as “Constant, D2” initially, and then the second derivative curve was obtained and smoothened at a smooth ratio of 16.2% to get rid of the noise perturbations. Thereafter, the minima of the second derivative curve were set as the initial center of latent components. The peak position, intensity, and amplitude were continuously iterated to get a minimal sum of squares due to error (SSE) value and an adjusted r-square (Adj.r2) value near 1. Once there were components whose center deviated by more than 2 cm1 from the initial position [17], the parameters were changed to conduct another fitting trial. Each component was assigned to the corresponding secondary structure based on empirical relations between absorption band and protein conformation [25]. To further evaluate the relative fraction of each conformation, the integrated area of each peak was calculated by Origin version 8.0 (OriginLab, USA). 3. Results 3.1. Expression and purification of recombinant Balcp19k The recombinant Balcp19k, with an 11-amino acid T7 tag at Nterminal and a 6-His tag at C-terminal, was successfully expressed in E. coli BL21 (DE3) cells. The purification was performed under denaturing conditions. Both Ni-NTA affinity chromatography and preparative reverse phase chromatography (PRPC) were conducted. The rBalcp19k was obtained at a high yield of about 40 mg/L of the culture and was extremely homogeneous as evidenced by the absence of any extra peak in the mass spectrum obtained in the MALDI-TOF analysis; only a single peak was predominant at 17,931.45 Da (Fig. 1D), which was identical with the theoretical weight. Results of Tris-bis SDS-PAGE (Fig. 1B) and western blotting (Fig. 1C) revealed a similar band with an apparent molecular weight of about 26 kDa, showing great discrepancy with the predicted

Please cite this article in press as: X. Liu, et al., Amyloid fibril aggregation: An insight into the underwater adhesion of barnacle cement, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2017.08.136

4

X. Liu et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6

weight (17,931.20 Da) of the coding sequence, which might be a result of the excessive basic isoelectric point (predicted value for Balcp19k is10.3). 3.2. Amyloid aggregation as monitored by ThT binding and solubility assay for the deposits Generally, the kinetic curve of ThT fluorescence used for monitoring amyloid formation consists of an initial lag-phase, followed by a sharp augmentation phase, and final plateau phase [21]. We did not observe a significant lag-phase in the experiment performed using the seawater analog. The fluorescence intensity surged at the very beginning, reaching about 57% of its final value in 200 min. After about 1400 min of incubation, the fluorescence intensity almost plateaued off, suggesting that the self-assembly was ceased (Fig. 2A). After incubation, fibrils manifested as visible opalescent deposits (Fig. S4) which were positive to ThT staining (Fig. 2B). In contrast, when incubated in cement gland fluid analog, the fluorescence intensity was low throughout, indicating extremely low self-assembly of the protein. Notably, the time (ca. 20 h) required for the self-assembly of rBalcp19k was approximating to the time [5,18,26] required by the barnacle liquid glue to cure. Furthermore, the results of SDS-PAGE (Fig. 2C) showed that the rBalcp19k amyloid fibrils could only be solubilized by 6 M GdnHCl and not by 8 M urea. 3.3. Visualization of amyloid fibrils by TEM The morphology of the rBalcp19k formed deposits was investigated using TEM. As indicated by red arrows in Fig. 3A, after 48 h of incubation in seawater analog, rBalcp19k formed the amyloid fibrils, which were straight, smooth, and unbranched, but were not as significantly twisted as the typical amyloid fibrils. The fibrils were not homogeneous in their size, being~16 ± 1e40 ± 1 nm in diameter; however, the size was comparable to that of typical amyloid fibrils [27]. Besides, a few longer and thicker fibril bundles, with diameters of ~47 ± 1e62 ± 1 nm, which were formed by selfassociation of short fibrils, were also observed (indicated by green arrows in Fig. 3B).

3.4. Evaluation of the secondary structure by FTIR The amide I band was subjected to peak fitting and a few latent components were enhanced (Fig. 4). According to Giorgia [28], an antiparallel b-sheet gives a strong peak below 1640 cm1 and a minor peak at ca.1675 cm1 whereas a parallel b-sheet has only one peak below 1640 cm1. In this work, peaks at 1622 cm1 and 1634 cm1 denoted low frequency b-sheets whereas a minor peak at 1671 cm1 indicated high frequency b-sheets. The b-sheet conformation accounted for about 48% of the entire secondary structure. All the peaks at 1661, 1678, and 1694 cm1 were deemed to derive from b-turns that accounted for about 36% of the structure and were, thus, important. The a-helix structure that generally absorbs between 1650 and 1660 cm1, was not observed. Moreover, the peak at 1648 cm1 was considered as a random coil or an orderless structure, accounting for about 17% of the secondary structure. In short, the secondary structure evaluated by FTIR indicated that b-sheet conformation is dominant in the rBalcp19k amyloid fibrils. 4. Discussion The Balcp19k was successfully expressed in E. coli and was purified by a combination of metal-affinity chromatography and PRPC. The recombinant Balcp19k was validated by SDS-PAGE, western blotting, and MADLI-TOF. Our results confirmed that rBalcp19k was self-assembled in a solution analogous to seawater (0.6 M NaCl, pH 8.0); however, no significant assembly occurred in cement gland fluid (0.1 M NaCl, pH 4.0). The micromorphology characterized by TEM, a high proportion of b-sheet secondary structure as assayed by FTIR, and positive ThT staining indicated that the assembled product were typical amyloid fibrils. Besides, the kinetic curve suggested that the time required by rBalcp19k to completely selfassemble was almost identical to that required by the liquid barnacle cement to cure. Moreover, the solubility of rBalcp19k fibrils in 8 M urea and 6 M GdnHCl was accordant with that of the cured cement. To our knowledge, this is the first report on the selfassembly of a full-length barnacle cement protein. Generally, disulfide bonds and other noncovalent interactions, including hydrogen bonds and hydrophobic and electrostatic

Fig. 2. Amyloid aggregation monitored by ThT staining and solubility assay of the deposits by SDS-PAGE. (A) Curves showing the kinetics of rBalcp19k amyloid fibril formation. (B) Fluorescence micrograph (stained with ThT) of rBalcp19k amyloid deposits in the seawater analog. (C) Solubility assay of the deposits by SDS-PAGE. Lane M, protein standard; Lane U, amyloid deposits treated with 8 M Urea; Lane G, amyloid deposits treated with 6 M GdnHCl.

Please cite this article in press as: X. Liu, et al., Amyloid fibril aggregation: An insight into the underwater adhesion of barnacle cement, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2017.08.136

X. Liu et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6

5

Fig. 3. Negative-stained TEM images of rBalcp19k amyloid fibrils. (A) Dense and non-homogeneous amyloid fibrils formed by rBalcp19k after 2 days of incubation in seawater analog. (B) The zoom-in view of the rectangular region in (A) clearly indicates the existence of nanoscale fibril bundles. The red arrows indicate the smooth and unbranched amyloid fibrils; the green arrows indicate the bundles. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

interactions are the main interactions that stabilize the amyloids [29]. Because the substitution of cysteine residues at the 5th and 39th positions with serine, to avert multimer formation, had no impact on amyloid aggregation, we conclude that rBalcp19k had been self-assembled mainly by noncovalent interactions. Aromatic interaction was considered to play an indispensable role in R1-3 peptide that is, designed according to cp52k [18]. However, the content of aromatic residues in rBalcp19k was very less (4 in 173 amino acids), indicating that aromatic interaction may not be a predominant factor. Because the sequence of Balcp19k has an alternate distribution of hydrophilic and charged segments, we propose that hydrogen bonds and electrostatic interactions are mainly responsible for the self-assembly of rBalcp19k. GdnHCl and urea are strong denaturants known to break hydrogen bonds and, thus, destroy protein conformations and facilitate protein solubility [30]. Our results showed that 6 M GdnHCl was sufficient to render the fibrils soluble, whereas 8 M urea was not, which is consistent with the solubility assay for cured cement [4,9]. Compared with another rapid-curing adhesion system in mussel, the barnacle cement takes hours or days to cure completely. It is known that the curing mechanism in mussel is dependent on quinone-based cross-linking [31]; however, no any definite cross-linking was found in barnacle cement. From the

kinetics of ThT fluorescence, we know that the self-assembly of rBalcp19k takes about 24 h, which is the approximate time required by liquid barnacle cement to cure. This time may be essential for conformational transformation of the cement proteins necessary for self-assembly. From these perspectives, we propose that underwater curing of barnacle cement might be closely related to the amyloid aggregation process. Previous studies have revealed that recombinant cp19k has the ability to adsorb to diverse foreign materials, including glass, TiO2, and SiO2, and is, therefore, considered to play a role in coupling [14]. Whether amyloid aggregation is responsible for surface binding function of cp19k is intriguing yet unknown. Amyloidbased adhesives are important natural bioadhesion systems; for example, CsgA (the major subunit of adhesive Curli fibers in E. coli) [32], harnesses hierarchically assembled amyloid nanostructures to achieve robust interfacial adhesion. This suggests that cp19k may play its surface coupling role in the amyloid state. However, it is early to conclude that amyloid formation of cp19k contributes to its surface coupling function from this research. Further studies are awaited to understand this issue. Acknowledgements We gratefully thank Dr. Ji and Dr. Song for the FTIR assay. Thanks to Dr. Wang for TEM images collection. We would like to thank Editage [www.editage.cn] for English language editing. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.bbrc.2017.08.136. Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2017.08.136. References

Fig. 4. Evaluation of the secondary structures of rBalcp19k by FTIR. Self-assembled amyloid fibrils in the seawater analog were subjected to FTIR assay using a transmission IR spectrometer.

[1] W.J. Yang, T. Cai, K.-G. Neoh, E.-T. Kang, S.L.-M. Teo, D. Rittschof, Barnacle cement as surface anchor for “clicking” of antifouling and antimicrobial polymer brushes on stainless steel, Biomacromolecules 14 (2013) 2041e2051. [2] W.J. Yang, T. Cai, K.G. Neoh, E.T. Kang, G.H. Dickinson, S.L. Teo, D. Rittschof, Biomimetic anchors for antifouling and antibacterial polymer brushes on stainless steel, Langmuir 27 (2011) 7065e7076. [3] D.A. Lacombe, Comparative study of the cement glands in some balanid

Please cite this article in press as: X. Liu, et al., Amyloid fibril aggregation: An insight into the underwater adhesion of barnacle cement, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2017.08.136

6

X. Liu et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6

barnacles (Cirripedia, balanidae), Biol. Bull. 139 (1970) 164e179. [4] K. Kamino, Barnacle underwater attachment[A], in: Biological Adhesives, Springer, 2006, pp. 145e166. [5] G. Walker, The adhesion of barnacles, J. Adhes. 12 (1981) 51e58. [6] L. Khandeparker, A.C. Anil, Underwater adhesion: the barnacle way, Int. J. Adhes. Adhes. 27 (2007) 165e172. [7] S. CR, S. JM, F. KP, E.-B. T, H. SE, L. DH, D. Z, W. C, N. SH, O. CS, Oxidase activity of the barnacle adhesive interface involves peroxide-dependant Catecholoxidase and lysyl oxidase enzymes, ACS Appl. Mater. Interfaces 9 (13) (2017) 11493e11505. [8] C.R. So, K.P. Fears, D.H. Leary, Sequence basis of barnacle cement nanostructure is defined by proteins with silk homology, Sci. Rep. (2016) 6. [9] K. Kamino, Mini-review: barnacle adhesives and adhesion, Biofouling 29 (2013) 735e749. [10] J. Otness, D. Medcalf, Chemical and physical characterization of barnacle cement, Comp. Biochem. Physiol. Part B Comp. Biochem. 43 (1972) 443e449. [11] M.J. Naldrett, The importance of sulphur cross-links and hydrophobic interactions in the polymerization of barnacle cement, J. Mar. Biol. Assoc. U. K. 73 (1993) 689e702. [12] K. Kamino, S. Odo, T. Maruyama, Cement proteins of the acorn barnacle, Megabalanus rosa, Biol. Bull. 190 (1996) 403e409. [13] K. Kamino, K. Inoue, T. Maruyama, N. Takamatsu, Harayama, Y. Shizuri, Barnacle cement proteins. Importance of disulfide bonds in their insolubility, J. Biol. Chem. 275 (2000) 6. [14] Y. Urushida, M. Nakano, S. Matsuda, N. Inoue, S. Kanai, N. Kitamura, T. Nishino, K. Kamino, Identification and functional characterization of a novel barnacle cement protein, FEBS J. 274 (2007) 4336e4346. [15] G.H. Dickinson, I.E. Vega, K.J. Wahl, B. Orihuela, V. Beyley, E.N. Rodriguez, R.K. Everett, J. Bonaventura, D. Rittschof, Barnacle cement: a polymerization model based on evolutionary concepts, J. Exp. Biol. 212 (2009) 3499e3510. [16] R.M. Sullan, N. Gunari, A.E. Tanur, Y. Chan, G.H. Dickinson, B. Orihuela, D. Rittschof, G.C. Walker, Nanoscale structures and mechanics of barnacle cement, Biofouling 25 (2009) 263e275. [17] D.E. Barlow, G.H. Dickinson, B. Orihuela, J.L. Kulp 3rd, D. Rittschof, K.J. Wahl, Characterization of the adhesive plaque of the barnacle Balanus amphitrite: amyloid-like nanofibrils are a major component, Langmuir 26 (2010) 6549e6556. [18] M. Nakano, K. Kamino, Amyloid-like conformation and interaction for the self-

assembly in barnacle underwater cement, Biochemistry 54 (2015) 826e835. [19] C. Liang, Y. Li, Z. Liu, W. Wu, B. Hu, Protein aggregation formed by recombinant cp19k homologue of Balanus albicostatus combined with an 18 kDa NTerminus encoded by pET-32a(þ) plasmid having adhesion strength comparable to several Commercial glues, PLoS One 10 (2015) e0136493. [20] F.W. Studier, Protein production by auto-induction in high density shaking cultures, Protein Expr. Purif. 41 (2005) 207. [21] M.R. Nilsson, Techniques to study amyloid fibril formation in vitro, Methods 34 (2004) 151e160. [22] M. Elimelech, W.A. Phillip, The future of seawater desalination: energy, technology, and the environment, Science 333 (2011) 712e717. [23] A.M. Power, W. Klepal, V. Zheden, J. Jonker, P. McEvilly, J. von Byern, Mechanisms of adhesion in adult barnacles[A], in: Biological Adhesive Systems, Springer, 2010, pp. 153e168. [24] H. Sumi, N. Nakajima, H. Mihara, Fibrinolysis relating substances in marine creatures, Comp. Biochem. Physiol. B Comp. Biochem. 102 (1992) 163e167. [25] D.M. Byler, H. Susi, Examination of the secondary structure of proteins by deconvolved FTIR spectra, Biopolymers 25 (1986) 469e487. [26] P. Cheung, G. Ruggieri, R. Nigrelli, A new method for obtaining barnacle cement in the liquid state for polymerization studies, Mar. Biol. 43 (1977) 157e163. [27] E. Gazit, Mechanisms of amyloid fibril self-assembly and inhibition. Model short peptides as a key research tool, FEBS J. 272 (2005) 5971e5978. €ndrich, FTIR reveals [28] G. Zandomeneghi, M.R. Krebs, M.G. McCammon, M. Fa structural differences between native b-sheet proteins and amyloid fibrils, Protein Sci. 13 (2004) 3314e3321. [29] D. Eliezer, Protein Amyloid Aggregation, Humana PressSpringer [Distributor], 2015. [30] K. Kamino, M. Nakano, S. Kanai, Significance of the conformation of building blocks in curing of barnacle underwater adhesive, FEBS J. 279 (2012) 1750e1760. [31] T. Priemel, E. Degtyar, M.N. Dean, M.J. Harrington, Rapid self-assembly of complex biomolecular architectures during mussel byssus biofabrication, Nat. Commun. 8 (2017) 14539. [32] C. Zhong, T. Gurry, A.A. Cheng, J. Downey, Z. Deng, C.M. Stultz, T.K. Lu, Strong underwater adhesives made by self-assembling multi-protein nanofibres, Nat. Nanotechnol. 9 (2014) 858e866.

Please cite this article in press as: X. Liu, et al., Amyloid fibril aggregation: An insight into the underwater adhesion of barnacle cement, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2017.08.136