Molecular Recognition of Structures Is Key in the Polymerization of

Subscriber access provided by Bibliothèque de l'Université Paris-Sud

Article

Molecular Recognition of Structures is Key in the Polymerization of Patterned Barnacle Adhesive Sequences Christopher R So, Elizabeth Anne Yates, Luis A. Estrella, Kenan P. Fears, Ashley M. Schenck, Catherine M. Yip, and Kathryn J. Wahl ACS Nano, Just Accepted Manuscript • Publication Date (Web): 15 Apr 2019 Downloaded from http://pubs.acs.org on April 15, 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 36 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 Nano

Molecular Recognition of Structures is Key in the Polymerization of Patterned Barnacle Adhesive Sequences Christopher R. So,1,* Elizabeth A. Yates,2,† Luis A. Estrella,1 Kenan P. Fears,1 Ashley M. Schenck,3 Catherine M. Yip,3 and Kathryn J. Wahl1

AUTHOR ADDRESSES 1Chemistry

Division, Code 6176, US Naval Research Laboratory, 4555 Overlook Ave, SW,

Washington, DC 20375-5342 USA 2US

Naval Academy Faculty sited in Code 6176, US Naval Research Laboratory, Washington,

DC, USA 3US

Naval Academy Midshipmen sited in Code 6176, US Naval Research Laboratory,

Washington, DC, USA

Corresponding Author *E-mail: [email protected]

ACS Paragon Plus Environment

1

ACS Nano 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 36

ABSTRACT The permanent adhesive produced by adult barnacles is held together by tightly folded proteins that form amyloid-like materials distinct among marine foulants. In this work, we link stretches of alternating charged and non-charged linear sequence from a family of adhesive proteins to their role in forming fibrillar nanomaterials. Using recombinant proteins and short barnacle cement derived peptides (BCPs), we find a central sequence with charged motifs of the pattern [Gly/Ser/Val/Thr/Ala -X], where X are charged amino acids, to exert specific control over timing, structure and morphology of fibril formation. While most BCPs remain dormant, the core segment demonstrates rapid polymerization as well as an ability to template other peptides with no propensity for self-assembly. Patterned charge domains assemble dormant peptides through a specific anti-parallel beta sheet structure as measured by FTIR. While charged domains favor an anti-parallel structure, BCPs without charged domains switch fibril assembly to favor simpler parallel beta sheet aggregates. In addition to activation, charged domains direct nanofibers to grow into discrete microns long fibrils similar to the natural adhesive, while segments without such domains only form short branched aggregates. The assembly of adhesive sequences through recognition of structured templates outlines a strategy used by barnacles to control physical mechanisms of underwater adhesive delivery, activation, and curing based on molecular recognition between proteins. KEYWORDS self-assembly, adhesive, peptide, amyloid, biomaterial, biomimicry, molecular recognition

Amyloid materials are used by nature to perform functions in extreme extracellular environments,1 e.g., tying bacterial colonies together,2,3 protecting fungal spores from water,4,5 and enabling


2

Page 3 of 36 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 Nano

disease-forming plaques to survive a host immune response.6 Barnacle species spanning diverse body plans have adapted amyloid-like materials to form the durable underwater bond they rely on to survive a lifetime of wave-swept seashores and extreme marine environments.7-10 Amyloids are an ideal class of materials with which to develop underwater adhesives: they can be delivered as free proteins/precursors to the bonding site, triggered to polymerize and solidify on demand, and persist underwater as highly insoluble networks that resist biological and chemical degradation.11 In the barnacle adhesive, physical mechanisms such as delivery and curing arise from specialized proteins that assemble via non-covalent interactions, a process governed by distinct patterns found in adhesive protein sequences.12 Much like protein-protein interactions responsible for sustaining living processes, delivery and curing of amyloid-forming proteins likely involve mechanisms of structuring and molecular recognition.13 Common to multiple barnacle species are a set of proteins found at the adhesive interface that span the spectrum of amino acid chemistries, charge neutral points, and molecular weights.14 Although chemically diverse, many proteins contain repetitive sequence patterns thought to form cohesive or adhesive components in the glue. Indeed, both full length proteins and segments of these proteins have been shown to assemble into fibrous materials with beta sheet structure.15-17 Studies on protein sequencing have highlighted promising candidates, however no strong sequence basis has yet been established that explains the tightly folded structure of adhesive fibers. Many repeated sequences do not span the entire length of their parent protein and also do not contain a clear pattern that relates to shaping highly ordered fibers or adhesive materials used by related crustaceans or arthropods.15 Sequence patterns that point to a core amyloid-forming protein in the barnacle adhesive have been elusive to identify because natural amyloid motifs are highly polymorphic. One reason is that


3

ACS Nano 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 4 of 36

many amyloids are misfolded globular proteins, evolved to disfavor sequence patterns that aggregate and are toxic to cells. Outside of natural sequences, the simplest conceived pattern for beta sheet structure is a binary linear sequence alternating between polar and non-polar residues, matching the intrinsic up-and-down periodicity of side chains from a fully extended backbone.18 Sheets formed from amphiphilic sequences bury their non-polar faces to stabilize larger sheet structures and fibrils, as long as the periodicity of the linear sequence registers with the alternating secondary structure of beta sheets. Materials with this pattern have an advantage of being tolerant for polymorphisms and can evolve many functions, as fibrils form regardless of the individual amino acid and its structure-forming propensity.18 Although alternating patterns have been widely adopted in wholly synthetic peptides and proteins,19, 20 few natural proteins, including amyloids, have evolved this type of alternating pattern.18, 21 Using the adhesive material from barnacles as a model, we ask whether proteins evolved to form functional amyloids might contain model betasheet forming patterns as hypothesized by the synthetic community. Deep proteomic sequencing of the dissolved barnacle glue reveals that, like in other fibrous biomaterials such as silks and elastin, compact and flexible protein segments play a key role in forming the mesh-like adhesive.12 Kamino et al. demonstrated that a 19 kD sequence with low complexity regions from the barnacle Megabalanus rosa was a component in barnacle glue and that it displayed affinity towards surfaces.22 A homologous protein sequence was identified across multiple barnacle species sharing a 20% lysine content, highlighting the possibility that the 19 kD protein is involved in surface adhesion.22 However, conserved sequences identified previously do not display any distinct patterns in the 19 kD protein similar to other biomaterials rich in beta sheet structures such as silks, disease-forming, or functional amyloids. Recently, this protein was shown to form amyloid-like fibrils in seawater conditions, suggesting that it is organized in the glue


4

Page 5 of 36 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 Nano

network and not a free protein.23 However, fibrils formed by the entire recombinant protein do not maintain the long ranged network appearance of the natural adhesive. Broader proteomic analysis of dissolved adhesive materials has uncovered 10 homologous instances of the full length 19 kD protein sequence as tandemly repeated motifs throughout at least five proteins in the glue.12 More importantly, conservation among multiple adhesive proteins from Amphibalanus amphitrite alone outlines short homologous sequences from the full length protein with distinct chemical patterns not observed in alignment of the 19 kD protein from multiple species performed by Kamino et al.22 Unlike any other cement protein studied to date, the 19 kD and 19 kD-like proteins maintain highly conserved domains alternating between short 20-residue regions of low complexity (Gly/Ser/Thr/Val/Ala residues) and regions with charged and aromatic side chains, with more than 80 such domains spanning the entire length of five proteins (Figure 1A).12 A summation of total charged amino acids per residue position confirms that conserved low complexity domains maintain little charge while variable regions account for most of the charged residues (Figures 1A and B). This analysis reveals a significant feature: charged amino acids alternate with non-charged amino acids throughout the variable regions of all 19 kD-like proteins, forming long stretches of alternating binary patterns. The unexpected conserved charge pattern strengthens previous identification of Gly/Ser/Thr/Val/Ala-X motifs throughout adhesive protein sequences, where X is a charged amino acid, and forms a clear basis for the nanostructure seen in barnacle adhesive.12 However, the role of these sequence domains in governing the molecular structure, polymerization, and morphology of assembled materials has yet to be explored. To establish a functional basis for patterned domains in the adhesive, we exploit homology among cement proteins from Amphibalanus amphitrite to design a set of peptides that contain conserved low complexity domains along with their neighboring variable charged domains. Eight


5

ACS Nano 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 6 of 36

synthetic peptides are generated from sequence patterns that span the entire length of the 19 kD homologous sequence (Figure 1C), referred to as barnacle cement peptides (BCPs). We compare two sets of peptides where one is limited to simple low complexity domains, while the other set spans additional charged domains wherein residues alternate between charged and non-charged (Figure 1C). The peptide library is screened using high throughput tools for amyloid characterization including fluorometric Thioflavin T (ThT) experiments, fluorescence microscopy, atomic force microscopy (AFM), and Fourier Transform Infrared Spectroscopy (FTIR) of formed materials. Simple peptide motifs aggregate and form fibrils, while the addition of neighboring charged patterns template microns-long fibril materials of higher order. The addition of charged domains template other sequences with no ability to form materials on their own, outlining a strategy of timed polymerization taken by the barnacle in the delivery and curing of an underwater adhesive. RESULTS Core sequences are responsible for polymerization of nanomaterials. We use the cationic benzothiazole dye Thioflavin T (ThT) which probes for beta sheet rich structures responsible for holding amyloid fibrils together.24 To verify that the amyloid-like barnacle adhesive responds to ThT, we first exposed untreated glue shavings taken directly from the barnacle Amphibalanus amphitrite to ThT.7,8 Shown in Figure 2A, natural glue shavings demonstrate enhanced ThT fluorescence response after a 15-30 minute incubation at room temperature, verifying that the natural adhesive responds to dyes sensitive to classical amyloids in agreement with previous studies highlighting the amyloid-like structure using FTIR and circular dichroism.7,8 To measure the aggregation propensity of BCPs, we survey all peptides at three concentrations: 50, 100, and 200 µM (Supplemental Figure S1) by exposing solutions to ThT over the course of up to 300


6

Page 7 of 36 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 Nano

hours. Of the BCPs, three distinct peptide sequences (BCP1C/2/2C) demonstrate the characteristic sigmoidal curvature associated with fibril formation (Figure 2D) and a positive ThT response greater than 100 a.u. (Figures 2B and C) in both simple buffer and seawater conditions. In contrast, BCPs 1/3/3C/4/4C displayed no activity over 300 h, and are referred to as dormant BCP sequences (Figure 2C). In both Tris-EDTA and artificial seawater (ASW), BCP2 shows the highest activity, followed by BCP2C and BCP1C which both contain charged domains. All three active peptides respond and undergo fibril formation in an artificial seawater environment with similar onset times as in Tris-EDTA, largely unaffected by free ions present in ASW. Lag time (Tlag) is the time at which aggregation begins, defined as the x-intercept extrapolated from the rising linear portion on the sigmoidal growth curve. Seen in Figure 2E, Tlag values were obtained using a sigmoidal model fitted by least squares regression of each experimental curve (see Methods).25 ThT fluorescence curves (Figure 2D) and estimated aggregation onset times (Figure 2E) reveal that BCP2/2C exhibit faster aggregation onset times with Tlag values within ca. 12 h in Tris-EDTA and ca. 50 h in ASW. In contrast to BCP1, BCP1C is considerably slower with a delayed aggregation time of 100+ hours. In Tris-EDTA, BCP2C noticeably had the shortest aggregation time with the lowest spread of Tlag (2 – 16h, median 6h), followed by BCP2 (4 – 53 h, median 26 h), while BCP1C exhibited the longest aggregation time (64 – 208 h, median 120 h). In ASW, BCP2C/2 revealed a large Tlag spread ranging from 2 – 102 h (median 22 h) and 9 – 73 h (median 30 h) respectively. BCP1C remained the slowest with an onset of 101 – 180 h (median 169 h). The aggregation onset of fibrillar formations follows the trend BCP2C < BCP2 < BCP1C for both Tris-EDTA and ASW with BCP1C being significantly delayed. Charged patterns confer folded, microns-long fibril morphology to simple sequences. To understand the effect of charged domains on the morphology and molecular structure of peptide


7

ACS Nano 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 8 of 36

fibrils, we characterize ThT responsive materials by AFM and FTIR (Figures 3 and 4, respectively). Clear morphological and structural differences are observed between simple peptides alone (BCP1/2) and peptides that also contain a variable charged sequence (BCP1C/2C). This is highlighted in Figure 3, where simple peptides assemble into short branched structures spanning 1-2 microns, similar to the full length protein, while their charged counterparts assemble into features that span 5+ microns. The difference between BCP2 and BCP2C is especially stark, where the charged sequence confers a matted discrete structure while the simple peptide alone clumps as aggregates from shorter subunit fibrils. Whole length 19 kD protein from Megabalanus rosa (MRCP19), containing all homologous domains, forms very short fibers that do not exceed 1 micron which is consistent with other observations.23 Since BCP1 showed no ability to assemble on its own, we introduced a single glycys mutation (G19C, mutBCP1) to promote intermolecular interactions through disulfide bonds. Indeed, like the other active peptides, a single mutation in mutBCP1 induced a classical response curve and fibril formation as seen by ThT assay and AFM respectively (Figure 3B). Fibers formed by mutBCP1 are similar to those of another simple peptide, BCP2, in that fibers remain in the 1-2 micron range, yet are 5+ microns in the presence of the neighboring charged domain as seen in BCP1C and 2C. Infrared spectroscopy of peptide nanofibers dried on CaF2 reveal clear distinctions in the secondary structure of sequences that contain patterned charge domains. Shown in Figure 4, all fibril materials formed by active peptides absorb strongly in the Amide I region, with the largest peak centered at ca. 1623 cm-1 typical of tightly folded beta-strand secondary structures. Two other prominent features range from 1655-1660 cm-1 for all peptide fibrils as well as a prominent shoulder at 1698 cm-1 for BCP1C/2C. The broad Amide I feature at ca. 1660 cm-1 is also present in amyloid beta.26 However, the distinct mode at 1698 cm-1 observed in conjunction with the


8

Page 9 of 36 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 Nano

narrow mode observed at ca. 1625 cm-1 indicates that peptides BCP1C and BCP2C form beta strands oriented in an anti-parallel fashion.26 In contrast, fibrils made from peptides without charged domains (mutBCP1 and BCP2) do not display the 1698 cm-1 mode, consistent with a parallel beta sheet structure. Quantification of the maximum peak absorbance at 1698 cm-1 as a fraction of the peak at ca. 1625 cm-1 highlights large differences in the two vibrational modes between charged and non-charged peptides. Peptides with charged domains BCP2C and BCP1C have fractions in the 0.9-1.0 range (when normalized to 1), while the absence of the feature at 1698 cm-1 causes BCP2 and mutBCP1 to exhibit low fractions between 0.15 and 0.3 (Supplemental Figure S2). These data suggest that patterned charge domains confer an anti-parallel structure to formed fibrils. Patterned charge domains activate dormant barnacle cement peptides through molecular recognition. Since charged and non-charged domains appear in a sequential order along the natural protein sequence (Figure 1), we tested the ability of organized seed structures to serve as templates in activating peptides from downstream regions. To this end, we probed the ability of BCP fibrils (BCP1C/2/2C) to polymerize dormant BCPs (BCP1/3/3C/4/4C) or accelerate kinetics of fibril formation using a cross-seeding ThT reaction in both Tris-EDTA (Figure 5A) and ASW (Figure 5B). For this, BCP seeds were formed over 7-10 days in Tris-EDTA, purified and incubated with free peptide solutions. Fibrillar BCP seeds alone have a low inherent ThT fluorescence over 300 h. Activation of dormant peptides was measured and counted only if the fluorescence exceeded the minimum response of BCP seeds alone, i.e., ≥ 100 a.u.. Interestingly, while BCP2 and mutBCP1 form fibrils as free peptides, they lack an ability to stimulate elongation and activation of dormant BCPs (Supplemental Figure S3). Seeds from peptides without charged domains have little activity, while seeds containing charged domains initiate the polymerization


9

ACS Nano 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 10 of 36

of most dormant peptides in both Tris-EDTA and ASW. The charged domain present in both BCP1C and BCP2C, as well as their anti-parallel secondary structure, adds a component of recognition for dormant peptide sequences not observed in mutBCP1 or BCP2. This robust recognition establishes BCP2C as the most self-active, as well as activity inducing, sequence of the eight peptides studied. This is exemplified in Figure 5C, where dormant peptide BCP3C is placed in a cross-seeding assay with preformed fibrils of BCP2C. While no measured onset over 300 h is observed (Figures 2C and 2D) for dormant BCP3C alone, the peptide polymerizes in the presence of BCP2C seeds in under 17 h. Furthermore, seeding by BCP2C not only activates most dormant peptides, but also accelerates the formation of amyloid fibrils in active peptides such as BCP2. The lag phase for cross-seeded reactions of active free BCP2 was < 2 h in the presence of BCP2C seeds, compared to 25-30 h for unseeded reactions (Figure 5D). Interestingly, BCP1C is observed to only activate peptides BCP1 and 4 in ASW (Figure 5B). Similar to BCP2C, BCP1C contains both a simple and charged domain that templates polymerization of the simple sequence domain (BCP1) that it carries. While BCP1 alone has no measured onset over 300 h, it readily polymerizes at 30 h in the presence of BCP1C seeds (Figure 2C and D). Thus, patterned charge domains are observed in multiple peptides to induce activation and aggregation of downstream dormant peptides, possibly due to the anti-parallel configuration of fibers. The high degree of specificity among barnacle adhesive peptides led us to then expose BCPs to human beta-amyloid 1-42 (A42), a well-known amyloid seeds organized in a parallel beta sheet structure. In these assays, no ThT response (Figures 5A and B, Supplemental Figure S3) was observed with A42 fibrils, showing that polymerization is specific to barnacle glue materials and the specific secondary structures formed by alternating charged/non-charged sequences.


10

Page 11 of 36 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 Nano

Randomization of BCP2C sequence eliminates propensity for polymerization and recognition. To test the importance of the linear sequence in both the polymerization and activation of downstream sequences, we randomize the BCP2C sequence (ranBCP2C) and monitor ThT kinetics and aggregation activity (Figure 6). Upon randomization, the N-terminal hydrophobic core sequence and corresponding amphiphilic nature were disrupted, shifting the Cterminus of the peptide to become more hydrophobic (Figure 6A and B). The grand average of hydropathy (GRAVY) value of both BCP2C and ranBCP2C was -0.7378.

Interestingly,

ranBCP2C showed no ThT activity over 300 h (Figure 6C), demonstrating that it could no longer polymerize into fibrillar structures like that of its predecessor BCP2C, and that a sequence without amphiphilic property or an alternating charge pattern can no longer undergo recognition and assembly. BCPs bind to naturally secreted barnacle adhesive. Lastly, we tested the ability for short BCP sequences to recognize naturally secreted adhesive from Amphibalanus amphitrite acorn barnacles. For this, we tested antibodies generated against adhesive proteins as well as two fluorescently labeled BCPs as probes against nitrocellulose membranes where barnacles have been resettled for three days (Figure 7A). The presence of AACP19 and AACP43, two proteins in the adhesive which display low complexity regions, were first probed to verify they had been transferred to the membrane surface. Figure 7B shows a strong response to the primary antibodies (-CP43 and -CP19), with little or no background response compared to the secondary antibody alone. Negative controls with luminol, a peroxidase activated chemiluminescent substrate, showed little cross reactivity with endogenous peroxidases known to exist in the barnacle adhesive.27,28 Both antibodies reacted strongly to regions where the organism resided, where proteins are seen distributed either localized to the barnacle periphery where new growth occurs or filled into the


11

ACS Nano 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 12 of 36

center of the settled region. We then probed a second set of membranes with BCP2C and BCP4 peptides labelled with TRITC and FITC, respectively (Figure 7C). A negative control (NC) peptide containing a GGGKDGGG sequence sensitive to oxidases was used, which showed no activity towards the glue region. Both BCP4-FITC as well as BCP2C-TRITC displayed strong fluorescence when exposed to the transferred glue region, demonstrating that there is an element of recognition for BCPs by the natural glue materials. DISCUSSION The primary sequence of BCP2C enables it to serve as a core aggregation domain, form extended fibril structures similar to barnacle glue, and activate most other dormant peptide sequences in both Tris-EDTA and ASW. Since fibril formation is largely unaffected by solution ionic strength, we first look at hydropathy trends in the BCP2C sequence (Figure 6A). Much like in amyloid beta and designed beta sheet structures,19, 29 hydrophobic sequences may play a central role in folding, activation, and self-assembly of BCP2C fibrils. The 37 residue sequence of BCP2C contains both a simple, low complexity domain (BCP2) where three aliphatic residues V7, V12, and I14 form an 11-residue hydrophobic segment (average hydropathy >3) that abruptly transitions to a charged and hydrophilic C-terminal region. Our ranBCP2C sequence disrupted the amphiphilic region, demonstrating the importance of clustered aliphatic residues in adhesive interactions. BCP1 contains only one central aliphatic valine, while BCP1C contains two leucines and an isoleucine alternating with two aspartic acids in a 5-residue stretch (Supplemental Figure S4). These three aliphatic residues may be prevented from interacting due to negative charge repulsion from neighboring aspartic acid side chains, adding to the energetic barrier for aggregation. Linus Pauling defined biological molecular recognition as a process of induced fit between


12

Page 13 of 36 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 Nano

two biomolecules occurring through reversible interactions, principally structured hydrogen bonding.30 Our findings here indicate that BCPs derived from patterned barnacle cement proteins undergo a similar mechanism to form complex nanomaterials as they first exist as free peptides and only become active in material formation when induced by an anti-parallel folded template. From the observations, we outline multiple states of aggregation which are involved in this activation process (Figure 8A) spanning monomers, nucleates and mature fibrils. Amyloid formation is classified as a nucleation-dependent reaction with a characteristic initial lag phase, an elongation/growth phase during which most fibrils form, and a final plateau where fibrillar materials are dominant among other coexisting states.31 Fully predicting the aggregation states of peptides from ThT data would require use of a descriptive chemical kinetics framework,32 survey of interactions at the equivalent supersaturation for each peptide,31 and testing the monomer reactivity to stages of aggregation during the lag phase.33 Though only a subset of these conditions were tested here, insight into participating states of aggregation during adhesive activation and curing can be outlined. Since dormant BCPs alone have no propensity to aggregate at the concentrations tested, the ability of a structured template to induce nucleation and growth is interesting. Recent findings by Knowles and Linse, et al.34,35 have revealed that secondary nucleation occurs during the lag and growth phases and may play an equal or greater role in growing the population of fibrils in solution beyond the initial primary nuclei.31 To further explore this, we observed the activation process using fluorescently labeled active and dormant BCPs (Figure 8B). Images of these materials show that dormant peptide BCP4 grows directly onto preformed BCP2C fibrils, favoring formation of heteropeptide aggregates over those comprised of BCP4 alone. Nucleation and growth of dormant peptides directly onto active templates provides evidence for a pathway where primary nucleation of dormant peptides is disfavored, instead


13

ACS Nano 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 14 of 36

favoring secondary processes via a structured template (Figure 8A). Furthermore, two pathways would lead to BCP2C existing in an active folded structure in the full-length protein: first, BCP2 being hydrophobic would induce the neighboring patterned charge domain to fold into an antiparallel configuration within the same protein. This starting structure then templates the folding of subsequent sequences into a fiber comprised of 19 kD-like proteins. Secondly, since peptides are observed to complement with the natural adhesive, homologous BCP2 domains from two proteins could condense through hydrophobic interactions to allow neighboring charged domains in BCP2C to fold and initiate activation of downstream domains in the same protein or among multiple proteins. Although BCPs are abundant in charged residues, a consistent finding is that polymerization occurs similarly in both artificial sea water and simple buffers or DI H2O (Figure 2), solutions of diverse ionic strength. This implies that while charged regions shape the structure and length of fibers, they play a minor role in initial aggregation which is instead driven by compact aliphatic side-chains and hydrogen bonds. Formation of underwater materials through simple hydrophobic aggregation would allow the organism to operate in seawater and form materials in the presence of metal ions. Although BCP2C and BCP2 polymerize rapidly, certain domains took over 100 hours, or 3-4 days, to become active even in the presence of a BCP2C template. This indicates that even in partially folded and activated states, there exist segments which delay the curing process of glue components on the timescale of real barnacle molting and growth processes.36 Alternating sequence patterns from the barnacle adhesive are uncommon among marine fouling organisms and canonical misfolded amyloids, observed in our work to have a high propensity to form beta sheet fibrils as predicted by Hecht et al.18 Though rare among natural


14

Page 15 of 36 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 Nano

marine adhesives, specialized protein fibrils are commonly used by related arthropods and crustaceans.37,38 Recently, we linked a common sequence pattern found in the barnacle adhesive to another nanofibrous adhesive protein produced by distantly related spiders, pyriform spidroin, used to attach dragline threads to surfaces.12 These materials resemble barnacle glue, as they exist as an embedded nanofibrous mesh-work that envelops a central dragline.38 Pyriform spidroin proteins do not exhibit the conventional subrepeat motifs found in spider fibroin, rather they display regions of alternating polar and non-polar amino acids.37 Two examples include the primary repeated region [AAARAQAQAEARAKAEA]x as well as the more variable region [SRTQAVTHSHAHSASHASSQASSETYAESTAHTATETHEHTSSHSQTASHSQAA]x from the primary pyriform spidroin protein PySp1. The Ser/Ala/Thr residues in the latter motif are polymorphic, similar to the observed pattern in barnacle adhesive proteins. Pauling and Corey revealed that the S-X and G-X motifs in silk fibroin proteins directly related to the ability to produce turns in shaping the threads produced by silk worms.39 In this work we demonstrate that similar alternating sequences are present in the barnacle adhesive and that they induce folding when in tandem with conserved segments that exist in at least 10 locations across multiple proteins. Spiders, moths and silk-producing crustaceans are closer relatives to barnacles than the wellstudied adhesives of bivalves such as mussels and tube worms, which bear no significant chemical or structural similarity to barnacle adhesive.40 CONCLUSION Alternating binary patterns of charge throughout homologous cement proteins from the barnacle adhesive were found to confer a distinct folded secondary structure that enables polymerization of downstream dormant cement sequences. These dormant peptides only polymerize in the presence of an anti-parallel structure, highlighting molecular recognition as a


15

ACS Nano 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 16 of 36

key mechanism in the formation of amyloid-like adhesives produced by the barnacle. This work demonstrates that structures produced by patterned cement sequences, and the progression of domain interactions, are critical in polymerizing materials that resemble the natural adhesive. Future work will involve testing how these structured materials display adhesive chemistries. Sequences identified in this work define a basic syntax used by the barnacle to fabricate its adhesive, but also adds diverse function to the growing language of materials formation through low complexity regions. METHODS AND MATERIALS Materials. The following were synthesized by Genscript USA (Piscataway, NJ): 1) ten synthetic peptides (BCP1, mutBCP1, BCP1C, BCP2, BCP2C, ranBCP2C, BCP3, BCP3C, BCP4, and BCP4C), 2) three fluorescently labelled peptides (BCP4-fluorescein-5-isothiocyanate (BCP4FITC), BCP2C-tetramethylrhodamine isothiocyanate (BCP2C-TRITC), TRITC labeled NC peptide (GGGRDGGG)), 3) full length MRCP19 recombinant protein, 4) lyophilized -amyloid 1-42 (A42) peptide, and 5) polyclonal antibodies for immunoassays: -CP43 rabbit, -CP19 rabbit and -rabbit HRP conjugated secondary antibody. Dimethyl sulfoxide (DMSO), EDTA buffer, Grenier black bottom 96-microwell plates, hexafluoroisopropanol (HFIP), phosphate buffered saline, Thioflavin T (ThT), Tris-HCl, and Tween 20 were purchased from Sigma-Aldrich (St. Louis, MO). For AFM imaging, muscovite mica (Grades V5 and V1) was purchased from Ted Pella (Redding, CA). Artificial sea salts (Instant Ocean) (Spectrum Brands, Blacksburg, VA) for all experiments were mixed with deionized (DI) water (resistivity > 18 mΩ) in 15 L carboys under constant aeration and UV sterilization to produce artificial seawater (ASW) with a concentration of 32 ppt. ASW was sterilized before use in all assays. Peptide and Protein Synthesis. Designer peptides were prepared on an automated solid-phase


16

Page 17 of 36 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 Nano

peptide synthesizer employing standard stepwise Fmoc protection and deprotection chemical procedures. Synthesis was carried out on a preloaded Fmoc-X(tBu)-Wang low-loading support resin using HBTU activation chemistry, while 20% piperidine in DMF was employed to afford the Fmoc deprotection, monitored by UV absorbance at 301 nm. Peptides were cleaved off the support and side chain deprotected by stirring the resin-bound peptide in a cocktail containing 90:5:3:2 TFA/thioanisole/EDT/anisole under an N2 atmosphere for ~6 h. Peptides were precipitated in cold ether and collected on vacuum filtration membrane, and lyophilized overnight. Purification of products was carried out by reverse-phase HPLC, where the most prominent peak was analyzed by MALDI-MS. Full length MRCP19 from Megabalanus rosa (acorn barnacle) was cloned and expressed recombinantly in the cytoplasm of E. coli. Peptide preparation. Lyophilized, synthetic barnacle derived peptides were prepared according to published protocols of -amyloid.41 In short, the peptides were treated with 200 L HFIP and sonicated (10 min) to dissolve preexisting or seed aggregates within the lyophilized stock. HFIP was evaporated off using a speedvac (Labconco, Kansas City, MO), resulting in clear peptide films. These peptide films were dissolved in varying amounts of DMSO to make a 10 mM stock solution of each BCP (with a low amount of DMSO) and sonicated (10 min). Stock solutions were stored at -80°C to prevent aggregation. To achieve the desired concentration, the stock solutions were dissolved directly into 50 mM Tris-HCl, 1 mM EDTA (Tris-EDTA) buffer or sterilized ASW with a salinity of 32 ppm. From the 10 mM stock solution of BCP1C/2/2C, 200 M seed solutions of active (fibril forming) BCPs were prepared in buffer (Tris-EDTA, ASW, or deionized water) at 32°C with orbital shaking over 7-10 days. Prepared fibrillar seeds were purified and concentrated in ~200 L desired buffer for use in seeding assays. BCP2 and BCP1C formed large aggregates which were pelleted using a MiniSpin Plus (14.1 rcf, 30-60 min) (Eppendorf), decanted, and


17

ACS Nano 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 18 of 36

reconstituted in Tris-EDTA, ASW, or deionized water accordingly to 20 M. For BCP2C, the presence of discrete fibrils inhibited pelleting and seeds were concentrated and used directly from stock concentrations. In addition, 10mM A42 stocks were prepared in DMSO using the same amyloid protocol above and seed solutions were formed over 300 hours and concentrated in TrisEDTA and ASW. Concentrated pre-formed seed BCPs and seed A42 were sonicated for 10-15 min immediately before use and ThT fluorescence of the seeds was checked prior to use in seeding assays. BCP2 underwent additional sonication (30% amplitude pulse for 5 min) with a sonic horn (Qsonica LLC, Newtown, CT) to ensure complete homogenization of BCP2 seeds. BCP1C/2/2C and mutBCP1 were used in seeding assays to activate non-fibril forming peptides in aggregation assays described below. A42 seeds were used to see if dormant BCPs had specificity to amyloidforming materials. Characterization of Peptide Polymerization. A stock solution of 1 mM ThT was prepared and a working solution of 100 µM ThT was used in each assay. Two different ThT aggregation assays were performed: 1) Free peptide BCP and 2) seeding from either BCP1C/2/2C or A42 with free peptide BCP. In the free peptide BCP assay, each BCP (BCP1/2/3/44, BCP1C/2C/3C/4C, mutBCP1, and ranBCP2C) were prepared individually and ThT fluorescence measured at various concentrations (200 µM, 100 µM, and 50 µM) of each synthetic peptide. In addition to free peptide BCP, ThT assays were also performed with full length MRCP (1X) synthesized recombinantly. In the seeded BCP assay, concentrated 100 µM seed solutions of active (fibril forming) BCP1C/2/2C and A42 were placed with dormant (50 µM) (non-fibril forming) BCP 1/3/3C/4/4C. Both ThT assays were incubated with 100 µM ThT at 32°C in 96-microwell plates sealed to prevent evaporation. ThT fluorescence was measured every 15 min for up to 300 h using a Synergy H1 Hybrid Multi-mode reader (BioTek, Winooski, VT) with excitation and emission filters set at 440


18

Page 19 of 36 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 Nano

and 480 nm, respectively with a low gain of 50, linear shaking, and top read. All ThT fluorescence assays were performed in duplicate (or more) on each plate and maximum fluorescence or normalized fluorescence was plotted in arbitrary units (a.u.). ThT assay results were verified among three or more independently performed experiments on additional 96 well plates. All kinetic data from ThT aggregation curves (free peptide BCP and seeded BCP assays) at each respective concentration were fit with a non-linear (sigmoidal) least squares regression to obtain estimated aggregation onset values/lag time (Tlag).25 𝐹0 + 𝐴

𝐹(𝑡) =

(

)

―𝑘 𝑡 ― 𝑡1

(

1+𝑒

2

)

Fitted parameters of the least squares regressions are k (elongation rate constant), A (amplitude), F0 (baseline), and T1/2 (time at half completion of aggregation). As defined by Hellstrand, et al. Tlag was calculated from the fitted parameters as:

𝑇𝑙𝑎𝑔 = 𝑇1/2 ―

2 𝑘

Once modelled, the maximum fluorescence of the least squares fit for each sample was obtained. These maximum fluorescence measurements of two or more individual samples under identical experimental conditions (i.e. same concentration, seed preparation, temperature) were averaged and reported with standard deviation. FTIR characterization. Peptide samples which formed seeds were dried onto CaF2 windows and analyzed using a single-bounce diamond ATR-IR prism in an anvil configuration (Pike technologies, Miracle accessory, Madison, WI) installed in a conventional FTIR benchtop


19

ACS Nano 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 20 of 36

spectrometer (ThermoFisher Nicolet Series, Waltham, MA). Background spectra were collected from blank regions on each seed sample and subtracted from sample spectra. Diamond prism surface was cleaned with ethanol between experiments. Nanostructure characterization by Atomic Force Microscopy. For ex situ AFM imaging, 20 L samples from identified 96 microwells of interest were spotted onto freshly cleaved muscovite mica and placed in a hood to allow for evaporation of buffer to dryness (12 – 24h). Each dry sample was washed with aliquots of deionized water and dried under a gentle stream of nitrogen. A Digital Instruments (Santa Barbara, CA) Dimension 3100 scanning probe microscope equipped with high frequency NanoSensors PPP-NCHR (NanoandMore USA, Lady’s Island, SC) probes with a 42 N/m spring constant was used to image peptide nanostructures. All imaging was carried out under tapping mode, with 512 × 512 data acquisitions at a scan speed of 0.8 Hz at room temperature under acoustic isolation. Certain samples were imaged with a Keysight 9500 AFM system (Keysight, Inc, Tempe, AZ) using the same tips. Supplier-provided software (Dimension 3100 SPM - Nanoscope, V7.3, Bruker, Santa Barbara, CA; Keysight 9500 SPM - Gwyddion 2.51, Czech Metrology Institute) was utilized for extracting quantitative data such as surface cross sections from AFM images. Barnacle Husbandry. Amphibalanus amphitrite barnacles were settled as cyprids on siliconecoated glass panels and reared at the Duke University Marine Laboratory (Beaufort, NC) as previously described.27 Panels of adult barnacles grown to 2-3 mm in diameter were shipped to the Naval Research Laboratory (Washington, DC), where they were maintained in an incubator operating at 23 C on a 12 h day/night cycle in 32 ppt ASW. The barnacles were fed Artemia spp. nauplii (Brine Shrimp Direct, Ogden, UT) three times a week, and the ASW was changed once a week during which excess algal growth was removed. Barnacles used for experiments were gently


20

Page 21 of 36 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 Nano

dislodged from the silicone-coated panels, rinsed with distilled water, and placed on alternative substrates for experiments. Immunoblotting of Barnacle Cement. Adult barnacles were placed onto a nitrocellulose membrane (0.45 µm), fed and housed in ASW at room temperature for 72 hour settlement to allow for resettlement via cement deposition. After 72 hrs., barnacles were gently peeled off (n = 3) and the resultant membrane underwent Western immunoblotting. The membrane was blocked in 5% non-fat milk dissolved in 1X PBS-T, pH 7.4, (10mM phosphate buffered saline, 0.05% Tween 20) for 1 hour, washed three times with PBS-T and probed with 1:1,000 dilution of -CP43 or -CP19 rabbit. The corresponding -rabbit HRP conjugated secondary antibody was used at 1:10,000 dilution. Membranes were developed using the a chemiluminescence kit (Pierce ECL Plus Western Blotting Substrate, 32106, Thermo Fisher Scientific, Waltham, MA) and image acquisition via Gel Doc XR+ Gel Documentation System (Bio-Rad, Hercules, CA). Activity of labeled BCP towards natural glue secretions. Adult barnacles were settled onto nitrocellulose membrane (0.45 pm) similar to immunoblotting experiments above. After 3 days, barnacles were peeled off (n = 6) and the resultant membrane was blocked in 5% non-fat milk dissolved in 1X PBS-T, pH 7.4. Blocked membranes were washed three times in PBS-T and incubated overnight in 10 M solutions of fluorescently labelled free BCPs, dormant BCP4-FITC (max excitation 490nm, max emission 525 nm) and active BCP2C-TRITC (max excitation 557 nm, max emission 576 nm). In addition, BCP2C-TRITC seeds (100 M) were formed at 32°C with orbital shaking over 7-10 days. As a negative control, a solution of a TRITC labeled peptide (GGGRDGGG) sensitive to oxidase activity was incubated with membrane deposited barnacle glue. Each incubation was repeated with multiple animals (n = 3). For fluorescence microscopy of


21

ACS Nano 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 22 of 36

fibrils, 100 µM concentrated seed solutions of labelled BCP2C-TRITC were exposed to BCP4FITC (50 µM) (non-fibril forming) in deionized H2O at room temperature. Fluorescence microscopy images were collected on a Nikon A1R+ laser scanning confocal microscope to detect free and seeded BCPs over areas with transferred adhesive. ASSOCIATED CONTENT The author(s) declare no competing financial interests. Supporting Information Additional concentration dependent ThT data, FTIR analysis, cross seeding data, and hydropathy analysis of the BCP2C sequence. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Present Addresses † Chemistry Department, U.S. Naval Academy, 572M Holloway Rd., Annapolis, MD 21402 Author Contributions C. So conceived and designed peptide sequences. C. So, K. Fears, E. Yates and L. Estrella designed and prepared experimental protocols; C. So, E. Yates, A. Schenck and C. Yip performed ThT assays; K. Fears designed and developed staining protocols and performed confocal microscopy experiments. C. So and E. Yates performed AFM, C. So performed FTIR and analyzed data. L. Estrella and C. So managed barnacle husbandry. C. So and E. Yates wrote the manuscript; L. Estrella and K. Wahl participated in manuscript preparation. C. So directed the overall project. ACKNOWLEDGMENT This work was funded by the Office of Naval Research through the Naval Research Laboratory Base Program and through the ONR Coatings Program (N0001418WX01163). E.A.Y. was


22

Page 23 of 36 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 Nano

supported through the NRL-United States Naval Academy Cooperative Program for Scientific Interchange (N0017317WR00216 and N0017318WR00229). Midshipmen A.M.S. and C.M.Y. were supported by NRL as summer research interns from USNA in Code 6176. We thank C. Taitt for discussions on testing the specificity of peptides. We also thank D. Rittschof and B. Orihuela for live barnacles. Figure 1. Derivation of barnacle cement peptide sequences from patterned charge motifs. A) Multiple sequence alignment of five barnacle cement proteins showing 10 instances of a 19 kD motif repeated along full length protein sequences. Below, plot of charged residues per sequence showing a simple and charged domain where charge is observed to alternate with non-charged residues. Peptides used in this study are boxed in orange (simple domain) and blue (charged domain). Black lines indicate segments used to generate BCPs. B) Average percentage of residues that are charged in N=25 charged domains and N=20 simple domains. C) Table showing each individual peptide sequence used in this study. Red indicates charged residues, gray indicate bulky side chains, black are compact residues. Figure 2. Sequence-dependent polymerization of dormant and active BCPs. A) Adhesive shaving from Amphibalanus amphitrite exposed to ThT (10 mM) for ca. 15 min and imaged via fluorescence microscopy. (inset) Glue shaving untreated. B) Enhanced ThT fluorescence centered at 480 nm when bound to peptide fibrils in solution. C) Synthetic barnacle cement peptides (200 M) have varied activity over 300 h. BCP 2/1C/2C are active in forming fibrils while BCP 1/3/3C/4/4C are dormant in both Tris-EDTA and ASW. D) Representative ThT curves for active BCP 1C/2/2C and dormant BCP 1/3/3C/4/4C in Tris-EDTA and ASW at 32ºC.Dormant BCP 1/3/3C/4/4C showed no activity after 300 h. E) Modelled Tlag values obtained using a non-linear (sigmoidal) least squares regression applied to 72 total experiments (x-axis, N). Red, artificial seawater, black, Tris-EDTA. Figure 3. Nano and microstructure of formed BCP fibrils from ThT screening. A) Full length protein MRCP19 forming short fibers that aggregate together. B, left) Representative ThT response curve of point mutated mutBCP1 (G19C). mutBCP1 forms short fibrils that aggregate (below). (B, right) BCP1C, containing an additional neighboring charged domain to result in fibers that are 5+ microns in length. C, left) BCP2, with only a simple domain, forming a similar fiber structure as (A) where short assemblies form larger aggregates. C, right) BCP2C fibrils, including an additional neighboring patterned charge domain that forms 5+ micron long fibrils with a discrete meshed architecture similar to the natural adhesive. Figure 4. Amide regions from ATR-FTIR of dried BCP materials on CaF2 which displayed activity by ThT. All spectra show a prominent Amide I mode at ca. 1625 cm-1 indicative of an amyloid-like material, similar to the parent barnacle glue. A) Top to bottom, BCP2 with only a simple domain maintains similar modes but does not display the prominent feature at 1698 cm-1. BCP2C spectrum showing three additional modes at 1525 cm-1 (Amide II), 1661 cm-1 and 1698 cm-1 (anti-parallel turn). mutBCP1 amide region similar to BCP2 with a main peak at 1626 cm-1


23

ACS Nano 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 24 of 36

and a shoulder at 1656 cm-1 with no modes near 1700 cm-1. Bottom, BCP1C spectrum showing similar modes as the BCP2C spectrum including the prominent shoulder at 1698 cm-1. B) Peak deconvolution of amide modes from (A) showing prominent peak at 1698 cm-1, using a five element Lorentzian fit for BCP2/2C/1C and a four element fit for mutBCP1. Figure 5. Ability of active peptide seeds to template growth of free peptides. (Top) Total fluorescence values from cross-seeding ThT assay in which preformed active BCP fibrils are incubated with free BCPs in both (A) Tris-EDTA and (B) ASW. Aβ42 pre-formed seeds show no activation of BCPs. Intrinsic fluorescence of seeds alone are indicated by a *. C - E) Representative ThT curves of seed-induced activity with free BCPs (grey circles), dormant (red circles) BCPs, and active (blue circles) BCPs. Least squares regression fit with the sigmoidal model (solid black line) are overlaid on data points. C) Dormant peptide BCP3C (red) is activated in the presence of preformed BCP2C seed fibrils in Tris-EDTA (gray). D) Preformed BCP2C seeds accelerate BCP2 aggregation onset from ca. 30 h (blue) to < 2 h (gray). E) Dormant BCP1 (red) undergoes polymerization in the presence of preformed BCP1C seed fibrils in ASW (gray). Figure 6. Hydropathy values (gray bars) of BCP2C and ranBCP2C sequences highlight importance of hydrophobic segments. Red lines represent a 5-window moving average of hydropathy using the Kyte and Doolittle scale overlaid on individual residue values with a scale of -3.5 to 1.5. A) Schematic representation of BCP2C sequence hydropathy identifies a hydrophobic segment in the simple domain, and creates an abrupt shift to hydrophilic residues in the charged domain. B) The amino acid sequence of ranBCP2C with corresponding hydropathy values showing disruption of the trend seen in BCP2C. C) Max ThT fluorescence from ranBCP2C over 300 hrs. displays no self-assembly or polymerization. Randomization of BCP2C sequence prevents aggregation by disruption of a distinctive hydrophobic core sequence. Figure 7. Binding of BCPs to newly deposited barnacle cement. A) Resettled barnacles grown on nitrocellulose membranes for 72 hours. B) Western blot analysis of membranes peeled from adhered barnacles, left panel represents a membrane incubated with HRP conjugated anti-rabbit antibodies. The center panel and right hand side panels are independent resettled barnacles blotted against -CP43 and -CP19, respectively. C) Nitrocellulose-bound barnacle cement incubated with (from left to right) FITC labeled BCP4 (positive response), TRITC labeled BCP2C (positive response) or TRITC labeled NC peptide (negative control). Observed red autofluorescence (NC, right panel) results from algal growth in regions outside of the attached barnacle. Figure 8. A) Proposed aggregation states during dormant peptide activation that involves molecular recognition of a structured template and secondary nucleation to form heteropeptide aggregates. B) Fluorescence microscopy of TRITC labelled BCP2C seeds exposed to FITC labelled BCP4 free peptides show coincident red and green fluorescence, indicating that templated growth of BCP4 occurs from the pre-existing aggregates themselves.


24

Page 25 of 36 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 Nano

REFERENCES 1. Gebbink, M. F. B. G.; Claessen, D.; Bouma, B.; Dijkhuizen, L.; Wosten, H. A. B., Amyloids - a Functional Coat for Microorganisms. Nat Rev Microbiol 2005, 3, 333-341. 2. Barnhart, M. M.; Chapman, M. R., Curli Biogenesis and Function. Annu Rev Microbiol 2006, 60, 131-147. 3. Chapman, M. R.; Robinson, L. S.; Pinkner, J. S.; Roth, R.; Heuser, J.; Hammar, M.; Normark, S.; Hultgren, S. J., Role of Escherichia coli Curli Operons in Directing Amyloid Fiber Formation. Science 2002, 295, 851-855. 4. Wosten, H. A. B.; Schuren, F. H. J.; Wessels, J. G. H., Interfacial Self-Assembly of a Hydrophobin into an Amphipathic Protein Membrane Mediates Fungal Attachment to Hydrophobic Surfaces. Embo J 1994, 13, 5848-5854. 5. Macindoe, I.; Kwan, A. H.; Ren, Q.; Morris, V. K.; Yang, W. R.; Mackay, J. P.; Sunde, M., Self-Assembly of Functional, Amphipathic Amyloid Monolayers by the Fungal Hydrophobin Eas. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, E804-E811. 6. Chiti, F.; Dobson, C. M., Protein Misfolding, Functional Amyloid, and Human Disease. Annu Rev Biochem 2006, 75, 333-366. 7. Sullan, R. M.; Gunari, N.; Tanur, A. E.; Chan, Y.; Dickinson, G. H.; Orihuela, B.; Rittschof, D.; Walker, G. C., Nanoscale Structures and Mechanics of Barnacle Cement. Biofouling 2009, 25, 263-275. 8. Barlow, D. E.; Dickinson, G. H.; Orihuela, B.; Kulp, J. L.; Rittschof, D.; Wahl, K. J., Characterization of the Adhesive Plaque of the Barnacle Balanus amphitrite: Amyloid-Like Nanofibrils Are a Major Component. Langmuir 2010, 26, 6549-6556. 9. Barlow, D. E.; Dickinson, G. H.; Orihuela, B.; Rittschof, D.; Wahl, K. J., In situ ATRFTIR Characterization of Primary Cement Interfaces of the Barnacle Balanus amphitrite. Biofouling 2009, 25, 359-366. 10. Jonker, J.-L. The Natural Adhesive of the Goose Barnacle Lepas anatifera: The Functional Morphology and Chemistry of the Adhesive Gland and an Investigation of the Adhesive Proteins. National University of Ireland, Galway, Ireland, 2013. 11. Wang, Y.; Pu, J.; An, B.; Lu, T. K.; Zhong, C., Emerging Paradigms for Synthetic Design of Functional Amyloids. J Mol Biol 2018, 430, 3720-3734. 12. So, C. R.; Fears, K. P.; Leary, D. H.; Scancella, J. M.; Wang, Z.; Liu, J. L.; Orihuela, B.; Rittschof, D.; Spillmann, C. M.; Wahl, K. J., Sequence Basis of Barnacle Cement Nanostructure Is Defined by Proteins with Silk Homology. Sci Rep 2016, 6, 36219. 13. Wang, X.; Hammer, N. D.; Chapman, M. R., The Molecular Basis of Functional Bacterial Amyloid Polymerization and Nucleation. J Biol Chem 2008, 283, 21530-21539. 14. Kamino, K., Mini-Review: Barnacle Adhesives and Adhesion. Biofouling 2013, 29, 735749. 15. Nakano, M.; Kamino, K., Amyloid-Like Conformation and Interaction for the SelfAssembly in Barnacle Underwater Cement. Biochemistry 2015, 54, 826-835. 16. Nakano, M.; Shen, J. R.; Kamino, K., Self-Assembling Peptide Inspired by a Barnacle Underwater Adhesive Protein. Biomacromolecules 2007, 8, 1830-1835. 17. So, C. R.; Liu, J.; Fears, K. P.; Leary, D. H.; Golden, J. P.; Wahl, K. J., Self-Assembly of Protein Nanofibrils Orchestrates Calcite Step Movement through Selective Nonchiral Interactions. ACS Nano 2015, 9, 5782-5791.


25

ACS Nano 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 26 of 36

18. West, M. W.; Wang, W. X.; Patterson, J.; Mancias, J. D.; Beasley, J. R.; Hecht, M. H., De Novo Amyloid Proteins from Designed Combinatorial Libraries. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 11211-11216. 19. Schneider, J. P.; Pochan, D. J.; Ozbas, B.; Rajagopal, K.; Pakstis, L.; Kretsinger, J., Responsive Hydrogels from the Intramolecular Folding and Self-Assembly of a Designed Peptide. J Am Chem Soc 2002, 124, 15030-15037. 20. Rufo, C. M.; Moroz, Y. S.; Moroz, O. V.; Stohr, J.; Smith, T. A.; Hu, X. Z.; DeGrado, W. F.; Korendovych, I. V., Short Peptides Self-Assemble to Produce Catalytic Amyloids. Nat Chem 2014, 6, 303-309. 21. Broome, B. M.; Hecht, M. H., Nature Disfavors Sequences of Alternating Polar and NonPolar Amino Acids: Implications for Amyloidogenesis. J Mol Biol 2000, 296, 961-968. 22. Urushida, Y.; Nakano, M.; Matsuda, S.; Inoue, N.; Kanai, S.; Kitamura, N.; Nishino, T.; Kamino, K., Identification and Functional Characterization of a Novel Barnacle Cement Protein. Febs J 2007, 274, 4336-4346. 23. Liu, X.; Liang, C.; Zhang, X.; Li, J.; Huang, J.; Zeng, L.; Ye, Z.; Hu, B.; Wu, W., Amyloid Fibril Aggregation: An Insight into the Underwater Adhesion of Barnacle Cement. Biochem Biophys Res Commun 2017, 493, 654-659. 24. Malmos, K. G.; Blancas-Mejia, L. M.; Weber, B.; Buchner, J.; Ramirez-Alvarado, M.; Naiki, H.; Otzen, D., Tht 101: A Primer on the Use of Thioflavin T to Investigate Amyloid Formation. Amyloid 2017, 24, 1-16. 25. Hellstrand, E.; Boland, B.; Walsh, D. M.; Linse, S., Amyloid Beta-Protein Aggregation Produces Highly Reproducible Kinetic Data and Occurs by a Two-Phase Process. ACS Chem Neurosci 2010, 1, 13-18. 26. Cerf, E.; Sarroukh, R.; Tamamizu-Kato, S.; Breydo, L.; Derclaye, S.; Dufrene, Y. F.; Narayanaswami, V.; Goormaghtigh, E.; Ruysschaert, J. M.; Raussens, V., Antiparallel Beta-Sheet: A Signature Structure of the Oligomeric Amyloid Beta-Peptide. Biochem J 2009, 421, 415-423. 27. So, C. R.; Scancella, J. M.; Fears, K. P.; Essock-Burns, T.; Haynes, S. E.; Leary, D. H.; Diana, Z.; Wang, C.; North, S.; Oh, C. S.; Wang, Z.; Orihuela, B.; Rittschof, D.; Spillmann, C. M.; Wahl, K. J., Oxidase Activity of the Barnacle Adhesive Interface Involves PeroxideDependent Catechol Oxidase and Lysyl Oxidase Enzymes. ACS Appl Mater Interfaces 2017, 9, 11493-11505. 28. Dickinson, G. H.; Vega, I. E.; Wahl, K. J.; Orihuela, B.; Beyley, V.; Rodriguez, E. N.; Everett, R. K.; Bonaventura, J.; Rittschof, D., Barnacle Cement: A Polymerization Model Based on Evolutionary Concepts. J. Exp. Biol. 2009, 212, 3499-3510. 29. Hills, R. D.; Brooks, C. L., Hydrophobic Cooperativity as a Mechanism for Amyloid Nucleation. J Mol Biol 2007, 368, 894-901. 30. Pauling, L., Molecular-Basis of Biological Specificity. Nature 1974, 248, 769-771. 31. Arosio, P.; Knowles, T. P.; Linse, S., On the Lag Phase in Amyloid Fibril Formation. Phys Chem Chem Phys 2015, 17, 7606-7618. 32. Meisl, G.; Kirkegaard, J. B.; Arosio, P.; Michaels, T. C.; Vendruscolo, M.; Dobson, C. M.; Linse, S.; Knowles, T. P., Molecular Mechanisms of Protein Aggregation from Global Fitting of Kinetic Models. Nat Protoc 2016, 11, 252-272. 33. Arosio, P.; Cukalevski, R.; Frohm, B.; Knowles, T. P.; Linse, S., Quantification of the Concentration of Abeta42 Propagons During the Lag Phase by an Amyloid Chain Reaction Assay. J Am Chem Soc 2014, 136, 219-225.


26

Page 27 of 36 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 Nano

34. Tornquist, M.; Michaels, T. C. T.; Sanagavarapu, K.; Yang, X.; Meisl, G.; Cohen, S. I. A.; Knowles, T. P. J.; Linse, S., Secondary Nucleation in Amyloid Formation. Chem Commun (Cambridge, U. K.) 2018, 54, 8667-8684. 35. Cohen, S. I.; Linse, S.; Luheshi, L. M.; Hellstrand, E.; White, D. A.; Rajah, L.; Otzen, D. E.; Vendruscolo, M.; Dobson, C. M.; Knowles, T. P., Proliferation of Amyloid-Beta42 Aggregates Occurs through a Secondary Nucleation Mechanism. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 9758-9763. 36. Burden, D. K.; Spillmann, C. M.; Everett, R. K.; Barlow, D. E.; Orihuela, B.; Deschamps, J. R.; Fears, K. P.; Rittschof, D.; Wahl, K. J., Growth and Development of the Barnacle Amphibalanus amphitrite: Time and Spatially Resolved Structure and Chemistry of the Base Plate. Biofouling 2014, 30, 799-812. 37. Blasingame, E.; Tuton-Blasingame, T.; Larkin, L.; Falick, A. M.; Zhao, L.; Fong, J.; Vaidyanathan, V.; Visperas, A.; Geurts, P.; Hu, X.; La Mattina, C.; Vierra, C., Pyriform Spidroin 1, a Novel Member of the Silk Gene Family That Anchors Dragline Silk Fibers in Attachment Discs of the Black Widow Spider, Latrodectus hesperus. J. Biol. Chem. 2009, 284, 29097-29108. 38. Wolff, J. O.; Grawe, I.; Wirth, M.; Karstedt, A.; Gorb, S. N., Spider's Super-Glue: Thread Anchors Are Composite Adhesives with Synergistic Hierarchical Organization. Soft Matter 2015, 11, 2394-2403. 39. Marsh, R. E.; Corey, R. B.; Pauling, L., An Investigation of the Structure of Silk Fibroin. Biochim Biophys Acta 1955, 16, 1-34. 40. Regier, J. C.; Shultz, J. W.; Zwick, A.; Hussey, A.; Ball, B.; Wetzer, R.; Martin, J. W.; Cunningham, C. W., Arthropod Relationships Revealed by Phylogenomic Analysis of Nuclear Protein-Coding Sequences. Nature 2010, 463, 1079-1083. 41. Stine, W. B.; Dahlgren, K. N.; Krafft, G. A.; LaDu, M. J., In vitro Characterization of Conditions for Amyloid-Beta Peptide Oligomerization and Fibrillogenesis. J. Biol. Chem. 2003, 278, 11612-11622.


27

ACS Nano 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

TOC Figure 290x145mm (268 x 268 DPI)


Page 28 of 36

Page 29 of 36 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 Nano

Figure 1. Derivation of barnacle cement peptide sequences from patterned charge motifs. A) Multiple sequence alignment of five barnacle cement proteins showing 10 instances of a 19 kD motif repeated along full length protein sequences. Below, plot of charged residues per sequence showing a simple and charged domain where charge is observed to alternate with non-charged residues. Peptides used in this study are boxed in orange (simple domain) and blue (charged domain). Black lines indicate segments used to generate BCPs. B) Average percentage of residues that are charged in N=25 charged domains and N=20 simple domains. C) Table showing each individual peptide sequence used in this study. Red indicates charged residues, gray indicate bulky side chains, black are compact residues. 195x177mm (300 x 300 DPI)


ACS Nano 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 2. Sequence-dependent polymerization of dormant and active BCPs. A) Adhesive shaving from Amphibalanus amphitrite exposed to ThT (10 mM) for ca. 15 min and imaged via fluorescence microscopy. (inset) Glue shaving untreated. B) Enhanced ThT fluorescence centered at 480 nm when bound to peptide fibrils in solution. C) Synthetic barnacle cement peptides (200 M) have varied activity over 300 h. BCP 2/1C/2C are active in forming fibrils while BCP 1/3/3C/4/4C are dormant in both Tris-EDTA and ASW. D) Representative ThT curves for active BCP 1C/2/2C and dormant BCP 1/3/3C/4/4C in Tris-EDTA and ASW at 32ºC.Dormant BCP 1/3/3C/4/4C showed no activity after 300 h. E) Modelled Tlag values obtained using a non-linear (sigmoidal) least squares regression applied to 72 total experiments (x-axis, N). Red, artificial seawater, black, Tris-EDTA. 210x190mm (300 x 300 DPI)


Page 30 of 36

Page 31 of 36 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 Nano

Figure 3. Nano and microstructure of formed BCP fibrils from ThT screening. A) Full length protein MRCP19 forming short fibers that aggregate together. B, left) Representative ThT response curve of point mutated mutBCP1 (G19C). mutBCP1 forms short fibrils that aggregate (below). (B, right) BCP1C, containing an additional neighboring charged domain to result in fibers that are 5+ microns in length. C, left) BCP2, with only a simple domain, forming a similar fiber structure as (A) where short assemblies form larger aggregates. C, right) BCP2C fibrils, including an additional neighboring patterned charge domain that forms 5+ micron long fibrils with a discrete meshed architecture similar to the natural adhesive. 254x184mm (300 x 300 DPI)


ACS Nano 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 4. Amide regions from ATR-FTIR of dried BCP materials on CaF2 which displayed activity by ThT. All spectra show a prominent Amide I mode at ca. 1625 cm-1 indicative of an amyloid-like material, similar to the parent barnacle glue. A) Top to bottom, BCP2 with only a simple domain maintains similar modes but does not display the prominent feature at 1698 cm-1. BCP2C spectrum showing three additional modes at 1525 cm-1 (Amide II), 1661 cm-1 and 1698 cm-1 (anti-parallel turn). mutBCP1 amide region similar to BCP2 with a main peak at 1626 cm-1 and a shoulder at 1656 cm-1 with no modes near 1700 cm-1. Bottom, BCP1C spectrum showing similar modes as the BCP2C spectrum including the prominent shoulder at 1698 cm-1. B) Peak deconvolution of amide modes from (A) showing prominent peak at 1698 cm-1, using a five element Lorentzian fit for BCP2/2C/1C and a four element fit for mutBCP1. 176x196mm (300 x 300 DPI)


Page 32 of 36

Page 33 of 36 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 Nano

Figure 5. Ability of active peptide seeds to template growth of free peptides. (Top) Total fluorescence values from cross-seeding ThT assay in which preformed active BCP fibrils are incubated with free BCPs in both (A) Tris-EDTA and (B) ASW. Aβ42 pre-formed seeds show no activation of BCPs. Intrinsic fluorescence of seeds alone are indicated by a *. C - E) Representative ThT curves of seed-induced activity with free BCPs (grey circles), dormant (red circles) BCPs, and active (blue circles) BCPs. Least squares regression fit with the sigmoidal model (solid black line) are overlaid on data points. C) Dormant peptide BCP3C (red) is activated in the presence of preformed BCP2C seed fibrils in Tris-EDTA (gray). D) Preformed BCP2C seeds accelerate BCP2 aggregation onset from ca. 30 h (blue) to < 2 h (gray). E) Dormant BCP1 (red) undergoes polymerization in the presence of preformed BCP1C seed fibrils in ASW (gray). 249x196mm (300 x 300 DPI)


ACS Nano 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 6. Hydropathy values (gray bars) of BCP2C and ranBCP2C sequences highlight importance of hydrophobic segments. Red lines represent a 5-window moving average of hydropathy using the Kyte and Doolittle scale overlaid on individual residue values with a scale of -3.5 to 1.5. A) Schematic representation of BCP2C sequence hydropathy identifies a hydrophobic segment in the simple domain, and creates an abrupt shift to hydrophilic residues in the charged domain. B) The amino acid sequence of ranBCP2C with corresponding hydropathy values showing disruption of the trend seen in BCP2C. C) Max ThT fluorescence from ranBCP2C over 300 hrs. displays no self-assembly or polymerization. Randomization of BCP2C sequence prevents aggregation by disruption of a distinctive hydrophobic core sequence. 246x142mm (300 x 300 DPI)


Page 34 of 36

Page 35 of 36 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 Nano

Figure 7. Binding of BCPs to newly deposited barnacle cement. A) Resettled barnacles grown on nitrocellulose membranes for 72 hours. B) Western blot analysis of membranes peeled from adhered barnacles, left panel represents a membrane incubated with HRP conjugated anti-rabbit antibodies. The center panel and right hand side panels are independent resettled barnacles blotted against -CP43 and CP19, respectively. C) Nitrocellulose-bound barnacle cement incubated with (from left to right) FITC labeled BCP4 (positive response), TRITC labeled BCP2C (positive response) or TRITC labeled NC peptide (negative control). Observed red autofluorescence (NC, right panel) results from algal growth in regions outside of the attached barnacle. 118x94mm (300 x 300 DPI)


ACS Nano 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 8. (A) Proposed aggregation states during dormant peptide activation that involves molecular recognition of a structured template and secondary nucleation to form heteropeptide aggregates. (B) Fluorescence microscopy of TRITC labelled BCP2C seeds exposed to FITC labelled BCP4 free peptides show coincident red and green fluorescence, indicating that templated growth of BCP4 occurs from the preexisting aggregates themselves. 233x161mm (300 x 300 DPI)


Page 36 of 36

Molecular Recognition of Structures Is Key in the Polymerization of

Recommend Documents