US 2017005.8007A1 (19) United States (12) Patent Application Publication (10) Pub. No.: US 2017/0058007 A1 Cox et al. (43) Pub. Date: Mar. 2, 2017

(54) SELF-ASSEMBLED BETASOLENOID (86). PCT No.: PCT/US 15/12934 PROTEIN SCAFFOLDS S 371 (c)(1), (2) Date: Jul. 14, 2016 (71) Applicant: THE REGENTS OF THE Related U.S. Application Data UNIVERSITY OF CALIFORNIA, Oakland, CA (US) (60) Provisional application No. 61/931,485, filed on Jan. 24, 2014. (72) Inventors: Daniel Cox, Oakland, CA (US); Publication Classification Gang-Yu Liu, Oakland, CA (US); (51) Int. Cl. Michael Toney, Oakland, CA (US); Xi C07K I4/435 (2006.01) Chen, Oakland, CA (US); Josh Hihath, C07K I4/45 (2006.01) Oakland, CA (US); Gergely Zimanyi, (52) U.S. Cl. Oakland, CA (US); Natha Robert CPC ...... C07K 14/43563 (2013.01); C07K 14/415 Hayre, Oakland, CA (US); Maria (2013.01) Peralta, Oakland, CA (US) (57) ABSTRACT The present invention provides amyloid fibrils comprising a (21) Appl. No.: 15/111,687 plurality of modified f solenoid protein (mEBSP) monomers. The mBSP monomers are modified to enhance self-assem (22) PCT Filed: Jan. 26, 2015 bly and are useful in a variety of applications. Patent Application Publication Mar. 2, 2017 Sheet 1 of 12 US 2017/0058007 A1

Nanoparticle binding a peptides six. r KY 3 YmXrm marrmyrmor-m-m-m-m-m-m-m-m-m-m-m-m8

Nanoparticle soxas D 's ^ s Substrate birding gi'Olips

Substrate

Fi. i. Patent Application Publication Mar. 2, 2017. Sheet 2 of 12 US 2017/0058007 A1

SY 3.

38 3. R&AF

s V8N SBAFP

ri. 2 Patent Application Publication Mar. 2, 2017 Sheet 3 of 12 US 2017/0058007 A1

fr. 3 Patent Application Publication Mar. 2, 2017. Sheet 4 of 12 US 2017/0058007 A1

$»§§§§§ *¿?????? y Patent Application Publication Mar. 2, 2017. Sheet 5 of 12 US 2017/0058007 A1

R R.

G. : Patent Application Publication Mar. 2, 2017. Sheet 6 of 12 US 2017/0058007 A1

& S.

is Patent Application Publication Mar. 2, 2017. Sheet 7 of 12 US 2017/0058007 A1

%3%), *

312 Rotatio assie ''.

G. Patent Application Publication Mar. 2, 2017 Sheet 8 of 12 US 2017/0058007 A1

40

c 2 - s : s - - 0

sess' 8 -20

-43

------I.------|------|------2O 2O 2. O 28 28O 3CO Wavelength (nm)

6 . w

Seas s

S. t

2O f 26 28 30 Wavelength (nm)

G. 8

Patent Application Publication Mar. 2, 2017. Sheet 10 of 12 US 2017/0058007 A1

3. On a 3. Rrrr. - 2.52.0 - s r it, 5. 1 2 0 ": ; , a 0. Öo Y ( x. M w stoo loo" too O 2O 400 600 fi

Patent Application Publication Mar. 2, 2017. Sheet 11 of 12 US 2017/0058007 A1

%%% Patent Application Publication Mar. 2, 2017. Sheet 12 of 12 US 2017/0058007 A1

tire is

G. 2 US 2017/005.8007 A1 Mar. 2, 2017

SELF-ASSEMBLED BETA SOLENOD designed structure is difficult as the product is at the mercy PROTEIN SCAFFOLDS of viral scaffold. Hence, precise, programmable nanometer scale ordered heterogeneity, as achieved with DNA, is not CROSS-REFERENCES TO RELATED feasible. APPLICATIONS 0007 Amyloid fibrils are self-assembled one-dimen 0001. This application claims benefit under 35 U.S.C. sional protein arrays with fi-Strands perpendicular to the S119(e) to U.S. Application No. 61/931,485, filed Jan. 24, linear axis. They arise both in unregulated self-assembly in 2014 the contents of which are incorporated herein by numerous diseases including Alzheimer's disease and type II reference. diabetes, as well as in regulated contexts in biofilm extra cellular matrices," synapse formation," and hormone res STATEMENT AS TO RIGHTS TO INVENTIONS ervoir manufacture." These fibrils have bending and twist MADE UNDER FEDERALLY SPONSORED ing persistence lengths on the micron scale,' ' which RESEARCH AND DEVELOPMENT contribute to the remarkable tensile strength of spider silk'' and the structural stability of barnacle cement." They have 0002 This work was supported by grants from the previously been used to template metallic nanowire growth, National Science Foundation (DMR-1207624 and DMR ''' and have been used to produce mechanically strong 0844115). The US Government may have certain rights in oriented films.’ this invention. 0008 Amyloid structures are remarkably robust. Gener ally, they can survive heating to the boiling point of water FIELD OF THE INVENTION 25 although there is monomer size and sequence dependence 0003. The present invention relates to amyloid fibers to this result. They are resistant to protease degradation 7 prepared from modified B solenoid protein monomers. The and UV light exposure. To date, amyloids have not been monomers are used for binding nanoparticles and other assembled to produce a significant level of transverse order, functional entities in a variety of applications. nor have they been used to template material growth other than the examples given above. There is also little system BACKGROUND OF THE INVENTION atic understanding of amyloid structure because the lack of 0004 An important goal of nanotechnology is bottom-up transverse order makes it difficult for X-ray diffraction to manufacturing of useful devices and materials via self reveal more than the generic cross f-stacking, although in assembly at room temperature in environmentally benign some instances additional scattering rings in fiber diffraction Solvents. Living systems provide numerous examples of have provided information about transverse dimensions of Such self-assembly in the guise, for example, of protein fibrils and longer periodicity repeats along the fiber axis.’ structures such as microtubules, viral capsids, bacterial S-layers, and amyloid fibrils" in which proteins grow in BRIEF SUMMARY OF THE INVENTION one-dimensional filaments with B-strands perpendicular to 0009. The present invention provides amyloid fibrils the growth axis. comprising a plurality of modified B Solenoid protein 0005. The programmable design of DNA-based nano (mBSP) monomers. The monomers may be derived from a structured scaffolds is extraordinary, allowing for the tem variety of sources, such as antifreeze proteins. The mBSP plating of ordered heterogeneous arrays of, e.g., metallic monomers are modified to enhance self-assembly, by for nanoparticles, proteins, and semiconducting wires. How example, removing an end cap that prevents amyloid aggre ever, it is plagued with technical barriers to advancement, gation. The mBSPs may also be modified to include at least including: a) difficulties in Scaling it to industrial applica one amino acid residue that promotes attachment of the fibril tions, b) high error rates of DNA replication, c) denaturation to a solid Support, a nanoparticle, a biological molecule (e.g., of DNA scaffolds/bundles at moderate temperatures (-60° an enzyme), a bacterial or eukaryotic cell (in which case the C.), and d) loss of integrity under exposure to ultraviolet scaffold can be used a matrix for tissue growth), or addi light and enzymes, e) very limited capability to carry a tional amyloid fibrils. broad range of functional groups: f) limited tenability in 0010. The invention also provides method of forming a terms of ternary and quaternary structures. nanomaterial. The methods comprise (a) contacting a plu 0006 Belcher and collaborators have used the M13 virus rality of nanoparticles with a scaffold comprising at least one as a scaffold for self-assembly of a wide variety of inorganic amyloid fibril comprising a plurality of modified B solenoid materials. Their strategy relies on modifying coat proteins protein (mBSP) monomers; and (b) fusing the nanoparticles with peptides that are selected through phage display for to form the nanomaterial. The methods may further com templating a specific material." In one example, the M13 prise the step of attaching the scaffold to a solid Support prior major coat protein was coated by a peptide with FePO to the step of contacting the plurality of nanoparticles with nanoparticle templating activity while the attachment pro the mBSP scaffold. The nature of the nanoparticles is not teins at the end of the virus were fused to a peptide known critical to the invention and can be selected based on the to adhere to carbon nanotubes.' Incubation of the virus with desired function to be achieved. iron and phosphate ions together with single-walled carbon 0011. The invention further provides scaffolds compris nanotubes generated a self-assembled working cathode. ing at least one amyloid fibril of the invention. The scaffold However, the M13 approach is limited by several factors: (a) is typically bound to a plurality of nanoparticles. viruses are large (M13 is nearly a micron in length); (b) templating sites are limited to the coat proteins, and the geometry is restricted to that provided by the virus; (c) while DEFINITIONS the viruses can order as liquid crystals, the ordering is on the (0012. The term “B-solenoid protein” (BSP) refers to micron scale; and (d) the capability to engineer or program proteins having backbones that turn helically in either a left US 2017/005.8007 A1 Mar. 2, 2017 or right-handed sense around the long axis of the protein threonine bonding, disulfide bridges, or metal mediated from the N-terminus to the C-terminus to form B-sheets, and chelation of histidine side chains can also be used. One of have regular geometric structures (triangles, rectangles, etc.) skill will recognize that by adjusting the side chain struc with 1.5-2 nm sides. The wild type (WT) BSPs are inhibited tures on different faces of mBSPs, programmable lateral from amyloid aggregation (end-to-end polymerization to assembly that can allow specific geometric arrangement of give cross B-fibrils) by natural capping features and/or the BSP scaffold can be achieved. Modifications of external structural irregularities on one or both ends. Examples of side chains of the mBSPs can be used to enable binding to non-amyloidogenic WT-BSPs that can form amyloid fibrils nanoparticles, nanoparticle templating molecules, Solid Sup upon modification include, one-sided antifreeze proteins ports, or for specific lateral self-assembly in two or three (Tenebrio molitor AFP-Protein Database (PDB) Accession dimensions. One of skill will recognize a number of modi No. 1 EZG), two-sided antifreeze (Snow Flea AFP-PDB fications that can be used to enable Such binding. 2PNE and 3BOI), rye grass AFP (PDB-3ULT), three-sided (0015 The term "amyloid fibril” refers to fibrous protein “type II left handed B-helical solenoid antifreeze proteins, aggregates that polymerize end-to-end in one-dimensional for example from the spruce budworm (PDB 1 M8N), three protein arrays. Amyloid fibrils can form naturally or they can sided bacterial enzymes (PDB 1LXA, 1FWY, 1G95, 1HV9, be produced out of intrinsically non-amyloidogenic proteins. 1J2Z, 1T3D, 1THJ, 1KGQ, 1 IMR7, 1SSM, 2WLC, 3R3R, As shown here, using a rational design concept, intrinsically 1KRV, 3EHO, 3Q1X, 3BXY, 3HJJ, 3OGZ, 4M98, 4IHH non-amyloidogenic proteins (e.g., BSPs) with natural (acyltransferases, Y-class carbonic anhydrases and cross-B structure can be transformed into proteins that homologs), three-sided motor proteins Subunits (e.g., PDB readily self-assemble into amyloid fibrils under benign con 3TVO), a three-sided “type I left handed f-helical enzyme ditions. ydcK from Salmonellae cholera (2PIG), four-sided proteins (0016. The term “mBSP scaffold” refers to a system of one (PDB 2BM6, 2W7Z, 2J8I), four-sided pentapeptide repeat or more amyloid fibrils comprising mBSP monomers, that proteins (2GOY and 3DU1), and 1XAT. One of skill will can be a platform for biomaterial-based self-assembly. recognize that the full sequence of each of these proteins is (0017. The term “antifreeze protein or AFP' refers to a available from the Protein Database. protein found in the body fluids of some poikilothermic 0013 The term “modified B solenoid protein (mBSP)” organisms, such as, Choristoneura sp. C. filmiferana or C. (also referred to as mBSP monomers) refers to genetically occidentalis, the Tenebrio molitor mealworm and plants engineered 3 solenoid proteins that allow for controlled which have the commonly known property that they reduce amyloid self-assembly. One of skill will recognize that an non-colligatively the freezing point of water. As used herein, mBSP monomer can be engineered to be any desired length “antifreeze proteins are chemically synthesized or recom and can tailored to the particular application. In a typically binantly produced polypeptides having a protein sequence embodiment, the monomer will comprise at least two beta with Substantial similarity to a naturally occurring antifreeze sheet rungs (about 30-36 residues) and more often at least protein and retaining the properties of an antifreeze poly three rungs (about 45-54 residues). The typical size of a beta peptide. In some embodiments, the modified antifreeze Strand face is about 3-6 residues, including bends the edge proteins of the invention will have altered or improved size will usually not exceed 5-8 residues, which is a range antifreeze activity and can be used for that purpose, as well. of about 2-3.2 nm. One of skill will recognize that a number 0018 Those of skill recognize that many antifreeze pro of modifications can be used to allow for self-assembly. For tein are BSPs. For example, those derived from Tenebrio, example, many BSPs include end caps that can be removed Snow Flea rye grass, and the spruce budworm. Other to allow for controlled amyloid self-assembly. Similarly, examples of antifreeze proteins useful in the present inven many BSPs include disulfides, bulges, and prolines that tion include those described in the following PDB Acces require removal to allow for controlled amyloid self-assem sions: 3VN3 B, 3VN3 A, 4DT5 B, and 4DT5 A. bly. One of skill will recognize that the three dimensional 0019. The term “nanoparticle' refers to a microscopic structure of any given BSP can be used to design an mBSP particle with at least one dimension less than 100 nm. of that desired shape. Means for modeling engineered pro Examples of nanoparticles include nanomaterial precursors, teins and characterizing their final properties are well known inorganic nanoparticles, and catalysts. The nanoparticle can to those of skill. Exemplary techniques for these procedures also be conjugated to a biomolecule (e.g., DNA, RNA, or a are described in detail below. Examples of mBSPs include protein, Such as an enzyme). The nanomaterial precursors SBAFP-m1 (SBAFP with endcap and disulfides removed), can include inorganic materials that form nanomaterials and RGAFP-m1 (RGAFP with bulges and proline removed), Such as inorganic nanocrystals. The nanoparticle can also both of which are described in more detail below. possess optimal metal binding capabilities including the 0014. The mBSPs of the invention can be functionalized ability to bind cadmium, iron, nickel, radium, uranium, in designed ways to specifically carry designated functional cobalt, lead, manganese or arsenic. The nanomaterial of the units. This includes substitution of amino acid residues with invention can comprise or consist essentially of materials side chains having desired reactivity. In some embodiments, Such as, for example, semiconducting materials, whether these residues are at the end of a nanoparticle binding doped or undoped; metallic materials; metal oxide materials, peptide, linked to the mBSP monomer. The residues can be and magnetic materials. Various oxide materials including selected to allow attachment of the mBSP or fibril to a solid silica and alumina can also be used. Nanoparticles can Support, a nanoparticle, a biological molecule (e.g., an include a metal oxide compound. The metal oxide can enzyme), a bacterial or eukaryotic cell (in which case the include a manganese oxide, a magnesium oxide, an alumi scaffold can be used a matrix for tissue growth), or addi num oxide, a silicon oxide, a Zinc oxide, a copper oxide, a tional amyloid fibrils. For example, the mBSP monomers nickel oxide, a cobalt oxide, an iron oxide, a titanium oxide, can be modified to include residues that enhance hydropho yttrium oxide, a Zirconium oxide, a niobium oxide, a ruthe bic interactions and/or salt bridging. Peptide bond chemistry, nium oxide, a rhodium oxide, a palladium oxide, a silver US 2017/005.8007 A1 Mar. 2, 2017

oxide, an indium oxide, a tin oxide, an lanthanum oxide, an 3402, respectively. Software for performing BLAST analy iridium oxide, a platinum oxide, a gold oxide, a cerium ses is publicly available through the National Center for oxide, a neodymium oxide, a praseodymium oxide, an Biotechnology Information (http://www.ncbi.nlm.nih.gov/). erbium oxide, a dysprosium oxide, a terbium oxide, a This algorithm involves first identifying high scoring Samarium oxide, a lutetium oxide, a gadolinium oxide, a sequence pairs (HSPs) by identifying short words of length ytterbium oxide, a europium oxide, a holmium oxide, a W in the query sequence, which either match or satisfy some Scandium oxide, uranium, uranium compounds, thorium or positive-valued threshold score T when aligned with a word a combination thereof. As discussed below, from these of the same length in a database sequence. T is referred to inorganic nanoparticles, a inorganic nanomaterial of the as the neighborhood word score threshold (Altschul et al. invention can be formed consisting essentially of the fused supra). These initial neighborhood word hits act as seeds for inorganic nanoparticles upon Substantial removal of the initiating searches to find longer HSPs containing them. The scaffold. word hits are then extended in both directions along each 0020. The terms “identical” or percent “identity,” in the sequence for as far as the cumulative alignment score can be context of two or more nucleic acids or polypeptide increased. Cumulative scores are calculated using, for sequences, (e.g., two mBSPs of the invention and polynucle nucleotide sequences, the parameters M (reward score for a otides that encode them) refer to two or more sequences or pair of matching residues; always >0) and N (penalty score Subsequences that are the same or have a specified percent for mismatching residues; always <0). For amino acid age of amino acid residues or nucleotides that are the same, sequences, a scoring matrix is used to calculate the cumu when compared and aligned for maximum correspondence, lative score. Extension of the word hits in each direction are as measured using one of the following sequence compari halted when: the cumulative alignment score falls off by the son algorithms or by visual inspection. quantity X from its maximum achieved value; the cumula 0021. The phrase “substantially identical,’ in the context tive score goes to Zero or below, due to the accumulation of of two nucleic acids or polypeptides of the invention, refers one or more negative-scoring residue alignments; or the end to two or more sequences or Subsequences that have at least of either sequence is reached. The BLAST algorithm param 60%, 65%, 70%, 75%, 80%, or 90-95% nucleotide or amino eters W. T. and X determine the sensitivity and speed of the acid residue identity, when compared and aligned for maxi alignment. The BLASTN program (for nucleotide mum correspondence, as measured using one of the follow sequences) uses as defaults a wordlength (W) of 11, an ing sequence comparison algorithms or by visual inspection. expectation (E) of 10, M=5, N=-4, and a comparison of both Preferably, the substantial identity exists over a region of the Strands. For amino acid sequences, the BLASTP program sequences that is at least about 50 residues in length, more uses as defaults a wordlength (W) of 3, an expectation (E) preferably over a region of at least about 100 residues, and of 10, and the BLOSUM62 scoring matrix (see Henikoff & most preferably the sequences are Substantially identical Henikoff, Proc. Natl. Acad. Sci. USA 89:10915 (1989)). over at least about 150 residues. In a most preferred embodi 0025. In addition to calculating percent sequence identity, ment, the sequences are Substantially identical over the the BLAST algorithm also performs a statistical analysis of entire length of the coding regions. the similarity between two sequences (see, e.g., Karlin & 0022. For sequence comparison, typically one sequence Altschul, Proc. Natl. Acad. Sci. USA 90:5873-5787 (1993)). acts as a reference sequence, to which test sequences are One measure of similarity provided by the BLAST algo compared. When using a sequence comparison algorithm, rithm is the smallest sum probability (P(N)), which provides test and reference sequences are input into a computer, an indication of the probability by which a match between Subsequence coordinates are designated, if necessary, and two nucleotide or amino acid sequences would occur by sequence algorithm program parameters are designated. The chance. For example, a nucleic acid is considered similar to sequence comparison algorithm then calculates the percent a reference sequence if the Smallest Sum probability in a sequence identity for the test sequence(s) relative to the comparison of the test nucleic acid to the reference nucleic reference sequence, based on the designated program acid is less than about 0.1, more preferably less than about parameters. 0.01, and most preferably less than about 0.001. 0023 Optimal alignment of sequences for comparison 0026. A further indication that two nucleic acid can be conducted, e.g., by the local homology algorithm of sequences or polypeptides of the invention are substantially Smith & Waterman, Adv. Appl. Math. 2:482 (1981), by the identical is that the polypeptide encoded by the first nucleic homology alignment algorithm of Needleman & Wunsch, J. acid is immunologically cross reactive with the polypeptide Mol. Biol. 48:443 (1970), by the search for similarity encoded by the second nucleic acid, as described below. method of Pearson & Lipman, Proc. Natl. Acad. Sci. USA Thus, a polypeptide is typically substantially identical to a 85:2444 (1988), by computerized implementations of these second polypeptide, for example, where the two peptides algorithms (GAP, BESTFIT. FASTA, and TFASTA in the differ only by conservative substitutions. Another indication Wisconsin Genetics Software Package, Genetics Computer that two nucleic acid sequences are substantially identical is Group, 575 Science Dr. Madison, Wis.), or by visual that the two molecules hybridize to each other under strin inspection (see generally, Current Protocols in Molecular gent conditions, as described below. Biology, F. M. Ausubel et al., eds. Current Protocols, a joint venture between Greene Publishing Associates, Inc. and BRIEF DESCRIPTION OF THE DRAWINGS John Wiley & Sons, Inc., (1995 Supplement) (Ausubel)). 0027 FIG. 1 is a schematic drawing of a amyloid fibril 0024 Examples of algorithms that are suitable for deter scaffold of the invention. The scaffold comprises a series of mining percent sequence identity and sequence similarity mBSP monomers bound to a solid support and each linked are the BLAST and BLAST 2.0 algorithms, which are to a nanoparticle through a nanoparticle binding peptide. described in Altschuletal. (1990).J. Mol. Biol. 215: 403-410 0028 FIG. 2 shows examples of wild type B-solenoid and Altschuel et al. (1977) Nucleic Acids Res. 25: 3389 proteins. PDB IDs are given below each. 2GOY and US 2017/005.8007 A1 Mar. 2, 2017

3DU1' are four-sided pentapeptide repeat proteins. 3BOI' of 8 days. Incubation at 37°C. leads to an increase in B-sheet is a two-sided snow flea antifreeze protein. 3ULT7 is the structure. (B) ThT fluorescence of SBAFP-m1 at 4° C. and two-sided rye grass antifreeze protein abbreviated RGAFP 37° C. compared to WTSBAFP at 37° C. WTSBAFP has herein. 2PIG7 is the three-sided “type I left handed B-he low fluorescence at 482 nm even after incubation at 37° C. lical enzyme ydcK from Salmonellae cholera. 1M8N' is (C) ThT assay comparing WT RGAFP to RGAFP-m1. WT the three-sided “type II left handed B-helical solenoid RGAFP has low fluorescence both at 4°C. and 37°C., while antifreeze protein from the spruce budworm abbreviated RGAFP-m1 has significant fluorescence at 4° C. which SBAFP herein. increases with incubation at 37° C. as expected for the 0029 FIG. 3 shows stages in the design of SBAFP-m1. growth of amyloid fibrils. For the assay, final protein con (A) The SBAFP wild type protein has a C-terminal capping centrations, except RGAFP-m1, were 5uM and ThT was 10 motif shown in red. This was removed so that the N- and uM in PBS, pH 7.4. RGAFP-m1 was added to a final C-termini form a gapless interface when brought together. concentration of 1.8 LM due to loss of protein on incubation All cysteine residues (shown in space-filling on the second at 37° C. ThT data for RGAFP-m1 were normalized for the from the left) were changed to serines, eliminating the decrease in concentration. disulfide bonds. Two monomers were fused to form a larger 0036 FIG. 10 shows AFM topography images of protein that is more manageable genetically and biochemi SBAFP-m1 and WTSBAFP. (A) SBAFP-m1 fibrils after 48 cally. Finally, the monomer-monomer interface optimized, h incubation. Inset shows fibrils at higher magnification including the addition of two Arg/Glu salt bridges. (B) (bar 200 nm). Height profiles show that the height varies Close-up of the optimized interface between two SBAFP between 1.5 and 3.0 nm. (B) AFM image of WT SBAFP. m1 proteins. The N- and C-termini form a salt bridge, as do Elongated structures and protein aggregates (indicated by the two Arg/Glu pairs placed at the interface. These are red arrows) are present. The inset shows a higher magnifi shown in ball-and-stick representation. cation image (bar for insets=50 nm). Height profiles show 0030 FIG. 4 shows (A) Sequence alignment between that the heights of fibril-like structures are 2.0-3.0 nm. (C) WTSBAFP (PDB entry 1 M8N) and the SBAFP-m1 protein Mature, three-week-old SBAFP-ml fibrils. Red arrowheads (SEQ ID NO: 1) derived from it. Only the first half of indicate lateral assembly of SBAFP-m1 fibrils. Height pro SBAFP-m1 (itself composed of two fused monomers) was files show variable (3, 8, and 10 nm) heights. The inset used in the alignment. The last 21 residues of 1 M8N were shows SBAFP-m1 fibrils in a parallel or antiparallel arrange deleted in the design of SBAFP-m1 (SEQID NO: 2) and the ment (bar-100 nm). (D) Internal structure of mature first Met are not shown in the alignment. (B) Alignment SBAFP-m1 fibril. Image indicates that mature fibrils contain between WT RGAFP and RGAFP-m1, which has a total of at least approximately four 3 nm tall individual fibrils three deletions and 13 mutations, compared to WT RGAFP bundled together. The red line corresponds to the height 0031 FIG. 5 shows stages in the design of the RGAFP profile plotted under the image. The inset shows randomly ml. (A) The N-terminal proline and amino acids causing the attached fibrils and fibril bundles of SBAFP-m1 (bar-2 um). bulge (marked in red in the left structure) were deleted. 0037 FIG. 11 shows AFM topography images of Next, amino acids at the monomer interface (marked in red RGAFP-m1 and RGAFP on poly-L-Lysine coated mica in the center structure) were mutated to optimize binding (0001) surfaces. (A) A1 umx1 um AFM image of RGAFP interactions. (B) Close up of end-to-end dimer interface of m1. The red line corresponds to the height profile under the RGAFP-n1. image. The height of RGAFP fibrils is around 2.0 nm. Tall 0032 FIG. 6 shows B-Sheet content vs. time during fibril features (indicated by red arrows) are likely aggregates of simulations. (A) SBAFP-m1 (B) RGAFP-m2. The data are RGAFP-m1 monomers. The inset shows the height profile averaged over five different runs each with random number for a single fibril (bar-40 nm). (B) 3 mx3 um AFM image seeds of the Langevin thermostat in the AMBER12 simu of WT RGAFP. The redline corresponds to the height profile lation suite. The B-content was determined by counting under the image. The inset shows a higher magnification, residues in the B-sheet region of the Ramachandran plots with individual bright features clearly shown. The red line using VMD.7 indicated by the black arrow is a single layer step of 0033 FIG. 7 shows height profile of SBAFP-m1 (gray) mica(0001) with a known height of 1 nm. and RGAFP-m2 (black) monomers. For each case, the 0038 FIG. 12 shows kinetics of SBAFP-m1 fibril for height minimum above a constraining Surface corresponds mation monitored by turbidity at 300 nm and total soluble to having a face parallel to and in contact with the Surface. protein in Solution. A reaction mixture containing 10 LM The maximum corresponds to having a line contact with an SBAFP-m1 initially on ice was placed in a 37° C. incubator edge Such that a face is perpendicular to the Surface. and shaken at 250 rpm to keep the sample mixed. At various 0034 FIG. 8 shows circular dichroism spectra of (A) times the reaction mixture was homogenized on a Vortex SBAFP-m1 in 10 mM sodium phosphate, pH 7.8 and (B) mixer, a sample removed, and turbidity measured. The RGAFP-m1 in 10 mM sodium phosphate, pH 7.8. The end-point turbidity is very close to 1 in these experiments. spectrum for SBAFP-m1 was recorded in a 1 cm path length The sample was then centrifuged to remove insoluble pro cell, while that for RGAFP-m1 was recorded in a 0.1 cm tein and the Soluble protein concentration measured. The path length cell. Both spectra show a single minimum at data are a combination of three independent experiments and ~220 nm indicative of predominantly B-sheet secondary were fitted to Eq. 1. The forward rate constant for polym structure for both proteins. erization calculated from both types of data is 14+1 M's'. 0035 FIG. 9 shows ThT fluorescence analysis of DETAILED DESCRIPTION SBAFP-m1 and RGAFP-m1. (A) ThT fluorescence assay of SBAFP-m1 as a function of urea concentration in 0.1 M 0039. The present invention provides a new approach to Tris-HCl, pH 7.8. The samples were dialyzed against amyloid design that allows programmable nanoscale struc decreasing concentrations of urea in buffer over the course tural precision for self-assembly of materials under mild US 2017/005.8007 A1 Mar. 2, 2017

conditions. The invention uses naturally occurring B-Sole 0044) The length of the fibrils can be controlled, for noid proteins (BSPs). These proteins have backbones that example through a variety of approaches including varying turn helically in either a left- or right-handed sense from the of the temperature (e.g., between 5° C. to 45° C.), by N-terminus to form B-sheets, and have regular geometric following the incubation with sonication, by the addition of structures (triangles, rectangles, etc.) with 1.5-2 nm sides. inhibitors of polymerization, or by modifying the buffer The WT proteins are inhibited from amyloid aggregation solution. For example, fibrils of several microns can be (end-to-end polymerization to give cross B-fibrils) by natural routinely produced. Alternatively, shorter fibrils (e.g., 100 capping features and/or structural distortions on one or both 200 nm) can be produced upon Sonication (panel at lower ends. The present invention describes the modifications right in FIG. 8). necessary to make linear polymers (amyloids) from these proteins, molecular simulations used to assess structural Use of Modified Beta Solenoid Proteins as Scaffolds stability and geometric properties for comparison to mea 0045. In the practice of the present invention, one skilled Surements, and the protocol for expressing and folding of the in the art can refer to technical literature for guidance on engineered proteins. As shown here, the correct monomeric how to design and synthesize a scaffold including the structures can be obtained after purification and folding, literature cited herein. For example, although the present amyloid fibrils can be produced by incubation at elevated invention relates to mBSP Scaffolds. The mBSPS described temperatures, and the kinetics of fibril formation are con above can be engineered to function as a scaffold for sistent with, though slightly faster than, other amyloid attachment and specific spatial arrangement of a number of polymerization reactions. These conclusions are Supported nanomaterials. Methods of using protein scaffolds to prepare by measurements of circular dichroism (CD), thioflavin-T nanomaterials of desired properties are known (see e.g., U.S. (ThT) fluorescence, dynamic light scattering (DLS), turbid Pat. No. 8,201,724, and US2009/0194317). A schematic of ity, and atomic force microscopy (AFM). a scaffold of the invention is shown in FIG. 1. 0046. The invention takes advantage of the ability of the Modified Beta Solenoid Proteins mBSPs of the invention to self-assemble into 1, 2, and 3 0040. The modified BSPs of the invention offer excellent dimensional scaffolds for template growth of nanoparticles. platforms for functionalization in nanotechnology without The described examples can be used in a variety of contexts, interfering with the native f3-sheet structure. For example, for example to grow photovoltaic, thermoelectric, catalytic, the large area faces together with their designable length and photocatalytic devices. can, in principle, Support nanoparticle binding peptides of 0047 One of skill will recognize that the amyloid fibrils more than one kind of nanoparticle to grow ordered hetero of the invention can be arranged in any desired, pre geneous nanoparticle arrays. Additionally, staggered place determined pattern, depending upon the particular applica ment of identical nanoparticle binding peptides can be used tion. For example, fibrils can be arranged in a repetitive to control nanoparticle aspect ratio. Even as one face is pattern, and/or in which the pattern is substantially parallel. being used for nanoparticle templating, another can be used In some embodiments, the fibrils are configured with a for binding to Surface or assuring designed lateral assembly directional order. 0048. In some embodiments binding of scaffolds to sur of the fibrils. In contrast, strategies based upon Small amy faces can be achieved via similar strategies to the templating loidogenic peptides do not immediately offer this level of discussed above. In particular, binding can be achieved by: functionalization diversity. (a) Sulfur chemistry of unoxidized cysteine or lysine side 0041. In a typical embodiment, the mBSP is modified to chains to bind to thiols decorating a prepared Surface; and enable one-dimensional growth through cross-beta Strand (b) peptide bond chemistry to link exposed lysine side (amyloid) pairing mBSPs. The exteriors and interiors of the chains to carboxyl groups decorating a prepared Surface. proteins can also be modified to enable more efficient 0049. In certain circumstances, the surface can be mica, production. Usually, the protein units are allowed to self silicon, glass, or a transparent conducting oxide, for assemble in one dimension after expressing proteins in E. example, FTO or ITO. In some embodiments, the surface coli, followed by Subsequent cell lysis, purification, dena can be poly-L-lysine coated mica (0001) surfaces. turation, and aggregation of the proteins to create the one 0050. In certain circumstances, the functionalized sub dimensional scaffolds. strate can include an aminopropylsilane functional group, a 0042. In some embodiments, at least two different mBSP carboxyethylsilane functional group, an epoxide functional monomers are designed to self-assemble in a predetermined group, or an amine functional group and a carboxylic acid order. This can be achieved by modifying the ends of the functional group, or combinations thereof. In certain cir monomers such that, for example, the N-terminus of a first cumstances, the functionalized Substrate can be positively monomer interfaces with the C-terminus of a second mono charged. mer, but not with the C-terminus of another copy of the first 0051. Using the nanoscale templating embodiments monomer. The resulting fibril comprises the two different described above allows for the generation of ordered arrays monomers in predetermined order (e.g., A-B-A-B-A-B, or of nanoscale catalysts with variable spacing controlled by A-B-C-A-B-C). the size of fused monomers and/or the end controlled linear 0043. The correct molecular mass of the amyloid fibril aggregation. Additionally, catalytic nano-structures can be can be verified through standard techniques, such as mass developed by controlling elemental identity and geometrical spectroscopy. The correct beta content can be determined arrangement of molecular catalytic moieties. through techniques such as circular dichroism. Amyloid 0052. As noted above, a wide variety of nanoparticles can aggregation can be confirmed by observing the growth of be attached to the scaffolds of the invention. The nanopar thioflavin T (ThT) fluorescence at 480 nm, according to ticles can be precursor inorganic materials that form the standard techniques. desired nanomaterial Such as inorganic nanocrystals. From US 2017/005.8007 A1 Mar. 2, 2017 these inorganic nanoparticles, inorganic nanomaterial can be precursor composition is treated to form the nanomaterial formed consisting essentially of the fused inorganic nano having the desired spatial orientation. particles upon substantial removal of the scaffold. The 0057 The step of treating the precursor to form the nanomaterial of the invention can comprise or consist essen nanomaterial will depend upon the material used. In many tially of materials such as, for example, semiconducting embodiments, a thermal treatment step is used, as is known materials, whether doped or undoped; metallic materials; in the art. The treating step may also comprise a chemical metal oxide materials, and magnetic materials. Various reduction of metal precursor salts. The scaffold can be oxide materials including silica and alumina can also be removed before or after the treating step. In general, the used. In a typical embodiment, the nanomaterials prepared reaction and the precursor materials should be compatible according to this invention conduct electricity as an electri with the scaffold. cal conductor, are semiconductive (whether inherently or via 0058. The temperatures and times for the thermal treat doping), transmit light, are magnetic, or possess some other ment step are known in the art. In general, the melting technologically useful property. Other properties of the temperatures and annealing behavior of the materials will be nanomaterials include ferroelectric properties, piezoelectric considered in selecting temperature. For example, tempera properties, converse-piezoelectric properties, and thermo tures of about 100° C. to about 1,000° C. can be used. electric properties. Thermal treatment can be used to fuse the nanoparticle 0053. In many embodiments, the nanomaterial is a semi precursors into a single structure and also to remove the conductor. Semiconductor materials are well known to those scaffold. The temperature can be selected to achieve a of skill in the art and can be, for example, alloys including desired crystalline phase which may be a low energy phase IV-IV Group (e.g., Si, Ge, Sir Ge), III-V Group binary or a high energy phase. In general, higher temperatures (e.g., (e.g., GaN. GaP), III-V Group ternary (e.g., Ga(ASP)), above about 500° C.) can be used to ensure the scaffold is II-VI Group binary (e.g., ZnS, ZnSe, CdS, CdSe, CdTe), completely removed. Lower temperatures (e.g., below about IV-VI Group binary (e.g., PbSe), transition metal oxides 300° C.) can be used to maintain the scaffold. The time of (e.g., BiTiO), and combinations thereof. the thermal treatment can be routinely determined by one of skill. Preferably, the temperature and time for thermal treat 0054. In certain circumstances, the nanomaterial precur ment can be adjusted to achieve the optimum balance for Sor can include a metal oxide compound. The metal oxide nanoparticle fusion while reducing undesired effects such as can include a manganese oxide, a magnesium oxide, an oxide formation. aluminum oxide, a silicon oxide, a Zinc oxide, a copper 0059. The scaffolds of the present invention can be used oxide, a nickel oxide, a cobalt oxide, an iron oxide, a in a variety of different commercial applications. For titanium oxide, yttrium oxide, a Zirconium oxide, a niobium example, one dimensional scaffolds can be used to produce oxide, a ruthenium oxide, a rhodium oxide, a palladium nanowires in applications requiring electrical conductivity oxide, a silver oxide, an indium oxide, a tin oxide, an or semiconductivity at the nanoscale, such as fuel cells, thin lanthanum oxide, an iridium oxide, a platinum oxide, a gold film batteries, Supercapacitors, photovoltaic devices, LEDs, oxide, a cerium oxide, a neodymium oxide, a praseodymium chemical and biological sensors, and the like. oxide, an erbium oxide, a dysprosium oxide, a terbium 0060. In the example of photovoltaic devices a multiex oxide, a Samarium oxide, a lutetium oxide, a gadolinium citon photovoltaic device with nanoparticle orientation oxide, a ytterbium oxide, a europium oxide, a holmium enabled by the templating principles of the mBSP arrays of oxide, a scandium oxide, or a combination thereof. the invention can be prepared. In these embodiments, each 0055. The nanomaterial of the invention can also be of the components of device can be precisely placed in the crystalline. The material can have one or more crystalline correct orientation with respect to the other components to domains. The crystalline phase can be either the thermody produce the device. The mBSP self-assembled scaffolds of namically favorable crystalline State or a crystalline State the invention can also be used to prepare thermoelectric which is not thermodynamically favorable but is locked in devices. By employing the end-controlled specific templat by the relative orientation of the crystalline nanoparticles ing of nanoparticles an embodiment of a thermoelectric strip before fusion. The nanoparticles can be oriented in any for heating or cooling with templating of n-type nanopar manner. For example, the crystallographic axis of the nano ticles on one side and p-type nanoparticles on the other side particles can be oriented with respect to the surface of the can be produced mBSP scaffold. The thermal treatment can be varied to 0061. The mBSP self-assembled scaffolds of the inven achieve a desired crystalline structure, or to covert poly tion can also be used for catalytic devices. It is well known crystalline structures to single crystalline structures. that certain colloidal or nanoscale minerals based upon 0056. In one embodiment, the invention provides a transition metal oxides can serve as effective catalysts for a method of making a nanomaterial using the mBSPs of the number of reactions, such as splitting of water to yield invention. In a typical embodiment, the mBSP scaffold has hydrogen under Solar illumination. Using the nanoscale a predetermined spatial orientation (e.g., one dimensional or templating of the present invention one of skill can prepare two dimensional). The scaffold comprises a plurality of ordered arrays of these nanoscale catalysts with variable binding sites along its length and/or at each end. The binding spacing controlled by the size of fused monomers and/or the sites are sites at which the desired nanoparticles are bound. end controlled linear aggregation approach described above. The binding sites can be the same or different so that one or 0062. The scaffolds of the invention can also be used to more nanomaterial precursor can be bound. For example, create arrays of specific enzymes for which (i) co-localiza different binding sites can be achieved by using different tion and (ii) immobilization can yield improved perfor monomers that self-assemble in a predetermined pattern. mance. For example, a three step enzymatic pathway on a The nanomaterial precursor is contacted the scaffold to form one dimensional scaffold can be achieved by attaching each a scaffolded precursor composition. Next, the scaffolded enzyme in the pathway to a different monomer and using end US 2017/005.8007 A1 Mar. 2, 2017

controlled linear aggregation to ensure the specific co of the helical axis. The average and standard deviations of localization of the enzymes on the scaffold. this height along the length of the monomer was then 0063. The scaffolds can also be used for adsorption of obtained. Only heavy atoms were used for the height mea atoms and molecules in environmental contexts. For SurementS. example, two- and three-dimensional mBSP scaffolds can be 0069 Protein Expression, Purification, and Folding used to nucleate specific adsorption of atoms and molecules (0070. The SBAFP-m1 and RGAFP-m1 genes in pET28a in environmental contexts for applications such as (i) get were procured from Life Technologies (Grand Island, N.Y.). tering of heavy metal ions for remediation of contaminated Proteins were expressed in E. coli BL21 (DE3) cells. For environments, and (ii) extraction of uranium and thorium SBAFP-m1, 1 L cultures were inoculated with overnight complexes from seawater for application of nuclear fuels. cultures and grown at 37° C., until ODoo reached 0.9-1.0. The scaffolds can also be engineered to template growth of Cultures were cooled on ice for 30 min after which isopropyl minerals such as calcium carbonate. This application can B-D-1-thiogalactopyranoside (IPTG) was added to a final enable such scaffolds to be added to existing cement for concentration of 1 mM. Protein expression proceeded at 30° mations as crack strengtheners. C. for 3 h. and cells were collected via centrifugation and resuspended in lysis buffer (50 mM Tris-HCl, pH 8.0, 100 Examples mMNaCl, 5 mM EDTA and 0.5% TritonX-100). Cells were 0064. The following examples are offered to illustrate, lysed by sonication and soluble and insoluble fractions but not to limit the claimed invention. separated by centrifugation. Insoluble inclusion bodies were purified by repeated sonication in lysis buffer without Triton Methods X-100 and centrifugation, total of four times. The purified inclusion bodies were resuspended in folding buffer (100 0065 Molecular Dynamics Simulations mM Tris, 50 mM glycine, pH 8.0) and were added dropwise 0066 All molecular dynamics simulations of both into denaturing buffer (100 mM Tris-HCl, 50 mM glycine, designed peptides were performed using the AMBER 12 8.5 M urea, pH 8.0) at 4° C. with stirring overnight. package at our custom built STRIDER GPU cluster at U Denatured SBAFP-m1 was purified on Fast Q Sepharose C Davis. The fl2SB parameter set was employed with a anion exchange column (GE Healthcare LifeSciences, UK). time step of 0.002 ps and fully constrained bonds to hydro The loading buffer was 50 mM Tris-HCl, 10 mM. NaCl, 8 M gen atoms. The aqueous peptide environment was simulated urea, pH 8.0 while the elution buffer was identical except for explicitly with TIP3P water at constant pressure, in a long, being supplemented with 500 mM. NaCl. Elution used a rectangular box with periodic boundary conditions (PBC). linear gradient. Purified SBAFP-m1 was concentrated using In simulating long, fibrillar multimers, a novel adaptive box Amicon centrifugal devices (EMD Millipore. Germany) algorithm was employed to economize the computation with a molecular weight cut-off of 3500 Da. Purified, (avoiding an exceedingly large Solvation geometry) all while concentrated SBAFP-m1 was refolded by stepwise dialysis maintaining a consistent solvent environment: (i) The mini out of urea using dialysis membrane with a molecular mum pairwise distance of the macromolecule (solute) with weight cut-off of 3000 Da and into 0.1 M Tris-HCl, pH 8.0 all of its periodic images was regularly recomputed, spe at 4° C. Each day the concentration of urea was decreased cifically anticipating rotational drift. (ii) When this distance by 1 M until it reached Zero. decreased below a cutoff of 15 angstrom, waters beyond this (0071. The WT SBAFP gene in pET20b was acquired distance from any solute atom were stripped; the Solute and from the Davies lab and expressed according to the pub close waters were reoriented in a new rectangular box lished protocol with minor changes. Briefly, protein was wherein the box boundaries were at least 20 angstrom from expressed in BL21 (DE3) E. coli cells in LB with IPTG any solute atom; and this box was resolvated with TIP3P induction when cell density reached 0.9 at 600 nm. Cells waters of an appropriate density for the fixed pressure. (iii) were collected via centrifugation and resuspended in a lysis The simulation was recommenced, accepting a picosecond buffer of 10 mM Tris-HCl, pH 9.0, 1 mM EDTA and 10 mM scale duration to accommodate the re-equilibration of outer 2-mercaptoethanol. Cells were disrupted by Sonication and shell waters in the new periodic box. The simulations for the IB’s denatured in lysis buffer containing 8 M urea and SBAFP-m1 and RGAFPm2 fibrils were carried out for 20 ns allowed to stir overnight at 4°C. Crude, denatured SBAFP with five different random number seeds for the Langevin was purified like SBAFP-m1 with buffers supplemented thermostat. with 10 mM 2-mercaptoethanol to keep cysteines reduced. 0067 B-Sheet content was measured from simulation Dialysis buffer for refolding was supplemented with 2% w/v. time series by using VMD7 to count the number of residues glycerol. within the typical B-sheet secondary structure region of the (0072. The RGAFP-m1 and WT RGAFP proteins were Ramachandran plot of p-up torsion angles (i.e., -180°-p<0. expressed by inoculating a 1 L culture with overnight -180°-p<-150°, and -180°-p<0°, 60°-p<180°). cultures and grown at 37° C. until ODoo reached 0.9-1.0. 0068. The height profiles of the monomers for compari Cultures were chilled in an ice-water bath for 20 min son with AFM experiments were obtained as follows. First, followed by IPTG addition to 0.5 mM. Protein expression to remove any inherent twist in the monomers, we constrain continued at 18° C. for 20 h. The cells were pelleted by Ca atoms on one side of the monomer to lie in a plane and centrifugation and resuspended in PBS, pH 7.4 (10 mM use energy minimization within the AMBER suite to relax sodium phosphate, 138 mM NaCl, and 2.7 mM KCl) fol this constrained structure. Then, each monomer was rotated lowed by freezing at -80° C. The frozen cell pellet was Such that its helical axis is aligned along the X-axis. The thawed at 37°C. and boiled for 10 minto lyse the cells. After monomers were then rotated about X-axis at 10-degree angle a 2 h. cooling period, RGAFP-m1 was sonicated extensively. intervals. At each rotation the maximum Z-coordinate height DNAsel (Worthington Biochemical Corp., Lakewood, N.J.) difference was measured for 5A thick slabs along the length was added to a final concentration of 1.6 ug/mL to the US 2017/005.8007 A1 Mar. 2, 2017

sonicated sample and incubated for 40 min followed by heat ments were made at either 4°C. or 37° C. depending on prior inactivation of the DNAsel by boiling at 100°C. for 10 min. sample treatment. A protein refractive index of 1.450 and a The sample was allowed to cool to 4°C. and dialyzed into water refractive index of 1.330 were used. Each reported PBS with 6-8 kDa dialysis tubing (Spectrum Labs, Irving, value is an average of 10 acquisitions, each lasting 300 S. Tex.). As a final purification step, the dialyzed protein was The averages and standard deviations of these 10 runs are loaded onto a Fast Q Sepharose anion exchange column (GE reported. Healthcare Lifesciences, UK) and washed with PBS pH 7.4. I0082 Atomic Force Microscopy The elution buffer was supplemented with 0.5 M NaCl. I0083 (A) Sample Preparation. Fractions containing proteins were pooled and dialyzed into 0084 Pieces of 8.0 mmx8.0 mmx0.5 mm muscovite PBS buffer before incubating samples at 37° C. for fibril mica were cut and mounted on standard microscope slides formation. using composite epoxy glue (5 Minute Epoxy, ITW Perfor 0073. The boiled cell lysate of WT RGAFP was purified mance Polymers and Fluids, FL. USA). Before protein using a nickel-NTA resin pre-equilibrated with binding deposition, the top layer(s) of mica was peeled mechanically buffer (50 mM Tris-HCl, 0.5 M NaCl, 5 mM Imidazole, pH to expose fresh (0001) surfaces. For SBAFP-m1 or wild type 7.5) in a 10 cmx1.5 cm column. The bound WT RGAFP was WTSBAFP. 20 ul of sample in Tris buffer (100 mM, pH8.0) washed with 10 column volumes of binding buffer and the were deposited on the freshly exposed mica(0001) surfaces. protein eluted with elute buffer (50 mM Tris-HCl, 0.5 M After 5 min incubation, the surfaces were washed three NaCl, 200 mM imidazole, pH 7.5). Fractions containing times with 200 ul Tris buffer (100 mM, pH 8.0) to remove protein were pooled and dialyzed into PBS using a 6-8 kDa weakly bound proteins and fibrils. The samples were imme dialysis membrane tubing. Protein concentrations for both diately imaged in the Tris buffer. Sample preparation for RGAFP-m1 and WT RGAFP were determined with the RGAFP-m1 or WT RGAFP followed similar protocols bicinchonimic acid assay (Thermo Scientific, Rockford, except for the Surface coating and imaging medium. The Ill.). freshly cleaved mica(0001) surfaces were coated with poly 0074 Amyloid Fibril Formation L-Lysine, by dropping 80 ul 0.1% (w/v) poly-L-Lysine 0075 Purified SBAFP-m1 at a concentration of 70 uM in (Sigma P8920, MW 150-300 kDa) onto the surface, incu 0.1 M Tris-HCl, pH 8.0 was transferred to an Eppendorf tube bating for 5 min, then washing with MilliO water. The and incubated at 37°C. to promote fibril formation. Purified surfaces were dried with clean air before protein deposition. RGAFP-m1 at a concentration of 98 uM in PBS pH 7.4 was The samples were imaged under ambient conditions. incubated at 37° C. in an Eppendorf tube until further I0085 (B) Imaging. analysis. I0086 AFM was performed using an MFP-3D AFM 0076. Thioflavin-T Fluorescence Assay (Asylum Research, Santa Barbara, Calif., USA). Most 0077. ThT fluorescence was measured as described." ' images were acquired using AC or tapping mode to mini ThT Stock Solutions were prepared by dissolving ~2 mg of mize perturbation and damage to Surface bound proteins and ThT (Sigma-Aldrich) in 2 mL of PBS, pH 7.4 and filtered fibrils. The typical set point was at 70%–80% of original through a 0.22 um filter. Stock concentrations were deter amplitude, and scan rate was 0.8-1 HZ. For imaging in mined using an extinction coefficient of 26,620 M' cm at aqueous media, two types of probes were used. The first was 416 nm. A working 500 LM ThT solution was prepared from a Biolever A cantilever (BL-RC 150, Olympus, Japan). Its the stock solution. For the assay, SBAFP-m1 was added to resonance frequency (f) was determined by the built in a final concentration of 5 uM and ThT to a final concentra software of MFP3D AFM. The spring constant (k) was tion of 10 uM in PBS, pH 7.4. The emission spectrum was provided by the manufacturer. Typically, f-10 kHz and k-30 recorded from 465 nm to 565 nm with excitation at 450 nm. pN/mm. The second type was an MSNL E cantilever For the assay, all protein concentrations, except for RGAFP (Brucker, USA) with f-11 kHz and k-100 pN/nm. For m1, were added to a final concentration of 5 uM and ThT to imaging under ambient (dry) conditions, AC240 cantilevers a final concentration of 10 uM in PBS, pH 7.4. RGAFP-m1 (Olympus, Japan) with f-65-75 kHz and k-2.0 N/m were was added to a final concentration of 1.8 LM due to a loss used. of protein after incubation. ThT data for RGAFP-m1 were 0087 Fibrilization Kinetics normalized for the decrease in concentration. I0088 Monomeric samples (10 uM) in 0.1 M Tris-HCl pH 0078 Circular Dichroism 8.0 were held at 4°C. in 15 mL capped plastic centrifuge 0079 Protein secondary structure was analyzed using tubes until polymerization was initiated by incubating at 37° CD. For SBAFP-m1, the spectrum is a concentration-nor C. with shaking at 250 rpm. At various times, the tube was malized combination of those for 0.02 mg/mL sample from homogenized on Vortex mixer and a sample was removed 190 nm to 200 nm and 0.2 mg/mL sample from 200 nm to for analysis by absorbance at 300 nm (turbidity). DLS, and 300 nm in 10 mM sodium phosphate, pH 7.4, in a 1 cm cell ThT fluorescence. Before ThT analysis, the sample was at 25° C. Spectra were collected at a scan rate proportional centrifuged for 5 min at 12,000xg at room temperature. The to high voltage using an OLISDSM 20 instrument. Reported soluble Supernatant and the resuspended insoluble precipi spectra are an average of 5 scans. For RGAFP-m1, spectra tate were separately analyzed with ThT as above. The were taken with 0.2 mg/mL sample in a 1 mm path length soluble protein concentration in the Supernatant was mea cell at 25°C. in 10 mM sodium phosphate, pH 7.4. sure by absorbance at 280 nm before being used in the ThT 0080 Dynamic Light Scattering assay. 0081) DLS measurements were performed using a Zeta 0089 Results and Discussion sizer NanoS (Malvern Instruments, Worcestershire, UK). (0090 Most amyloid fibrils used in the literature for Sample preparation for DLS measurements consisted of engineering purposes, e.g. templating nanowire growth.' " clarification by centrifugation at 13,000xg for 5 min. Protein are derived from naturally occurring proteins or peptides concentrations were ~ 1 mg/mL in PBS pH 7.4. Measure known to form amyloid fibrils under specific conditions. For US 2017/005.8007 A1 Mar. 2, 2017 example, nanowires have been previously grown on a prion Methods), but the expression levels are modest. Modeling variant from Saccharomyces cerevisiae known to self-as the replacement of all the cysteines by serines showed a semble' as well as on an Alzheimer's B-amyloid diphenyl good steric fit and enabled hydrogen-bonding interactions to alanine peptide.' These examples and various others in partially replace the disulfide bonds stabilizing interactions. literature’ exploit naturally occurring amyloid fibrils. An alignment of sequences Summarizing deletions and Approaches from other groups use harsh conditions (e.g., mutations between WTSBAFP and SBAFP-m1 are pre treatment with concentrated hydrochloric acid at elevated sented in FIG. 4A. Molecular dynamics simulations also temperatures for various days) to produce self-assembled showed the Cys-to-Ser mutant stable at 25° C. for 20 ns. amyloid fibrils out of intrinsically non-amyloidogenic pro Therefore, the Cys-to-Ser mutations were kept in the design. teins like lysozyme." Instead of using naturally occurring The cysteine-less design also has the advantage that future peptides already known to self-assemble or exposing a engineering efforts can employ the introduction of a unique protein to harsh conditions, we propose here a rational cysteine to enable chemically specific modification of the design concept to render intrinsically non-amyloidogenic protein. proteins with natural cross-B structure into proteins that (0097. The alterations described above left a relatively readily self-assemble into amyloid fibrils under benign con small protein of 90 amino acids. For ease of recombinant ditions. expression, purification, and functionalization of the protein, 0091 Protein Design two of the 90 amino acid monomers were genetically fused 0092 (4) Spruce Budworm Antifreeze Protein. together to give a single contiguous protein of 180 amino 0093. Here, isozyme 501 of the B-solenoid antifreeze acids in length. protein from the spruce budworm (SBAFP; PDB entry (0098 (B) Ryegrass Antifreeze Protein. 1M8N) was used to engineer one-dimensional fibrils. The (0099. As illustrated in FIG. 5, the same basic design polypeptide backbone is triangular about the long axis of the concepts used for SBAFP-m1 were applied to a ryegrass helix (FIG. 1). The structurally homologous 2PIG PDB (Lolium perenne) antifreeze left handed B-solenoid protein entry (FIG. 3) shows that the left-handed B-solenoid scaffold (RGAFP, PDB entry 3ULT) with the addition of using is tolerant to Substitutions at the apices of the triangular FoldIt to optimize the interface. In contrast to SBAFP, scaffold, and therefore likely to be robust for materials which has a triangular cross section with three B-sheet faces, engineering. RGAFP has two B-sheet faces in a rectangular arrangement 0094. There were two major considerations in the rational along the long axis of the B-helix. The structure has 8 design of the first engineered protein, termed SBAFP-m1. f-sheet rungs, each containing 14-15 amino acids per turn, The first was seamless and stable end-to-end interactions, and an exceptionally flat B-sheet on the ice-binding face. and the second was ease of biochemical handling. The first There is an additional amino acid present in the first three was addressed as follows, and as illustrated in FIG. 3. The rungs, causing a bulge at the N-terminus of the protein B-hairpin-like capping motif at the C-terminus of WT structure (FIG. 5A). These amino acids were removed to SBAFP was removed to give a clean C-terminus in the regularize the B-helix. Residues 1-4, which are missing in B-strand conformation (FIG. 3A). This required eliminating the crystal structure were excluded from the designed pro the last 21 residues. The N-terminus was modified by tein, termed RGAFP-m1. Additionally, residue 5 (Pro), removing the first six amino acids present in the crystal which might interfere with polymerization was mutated to structure of SBAFP. To avoid a heterogeneous N-terminus, Ala, which additionally facilitates a homogeneous N-termi Met-Ala were used for the first two amino acids of the nus as with the SBAFP-m1 design. SBAFP-m1 sequence. The E. coli methionine aminopepti 0100. To engineer an idealized interface, a dimer model dase has a strong preference for Small amino acids such as (FIG. 5B) was used. Residue Lys110 appeared to hinder alanine at the second position. Thus, the Met-Ala sequence interface formation and hence was mutated to ASn. As with increases the likelihood that the N-terminal amino acid is the design of SBAFP-m1, an Arg/Glu salt bridge was added homogeneously processed to Ala instead of a possible at the interface. To further strengthen the interaction mixture of Met and a different second amino acid. This between monomers at the binding interface. FoldIt opti design gives a seamless interface between the N- and mization was used as follows. The last 16 residues of the first C-termini of successive monomers, as illustrated in FIG.3B. monomer and the first 16 residues of the second monomer 0095. Two additional salt bridges were placed at the were allowed to mutate to get the best FoldIt score. This interface of the termini to increase the stability of the resulted in 10 mutations that largely increased the overall interaction between monomers. These are illustrated in FIG. charge of the proteins at the interface. Thus, a total of 13 3B. As modeled, the end-to-end interface includes three salt mutations, summarized in FIG. 4B, were incorporated in the bridge interactions: one between the N- and C-terminal design. Further testing using molecular dynamics simula ammonium and carboxylate groups, and two between the tions showed that the modeled interface was stable for 20 ns Arg/Glu side chain pairs introduced. These, along with the at 25 C. inter-monomer B-sheet hydrogen bonding, proved Sufficient 0101 Molecular Dynamics Simulations to keep the interface structure stable in MD simulations at 0102 The designed proteins were tested using MD simu 100° C. for 2 ns, and for 20 ns runs at 25° C. as described lations for 20-ns at constant pressure and temperature (25° further below. C.) using AMBER 12 suite (see Methods). Both individual 0096 WT SBAFP has a total of five disulfide bonds. monomers and longer 11-unit multimers were analyzed for These undoubtedly stabilize the folded protein, yet disul stability in MD simulations. Monomer simulations deter fides often present difficulties to high-level expression of mined the inherent stability of each design, particularly with recombinant proteins in E. coli and their Subsequent han respect to the modifications of the sequence at the termini. dling. Nevertheless, recombinant expression of WTSBAFP A concern was that new steric and electrostatic interactions in E. coli was previously achieved and reproduced here (see might lead to local instability of the B-helical motif, which US 2017/005.8007 A1 Mar. 2, 2017

could disrupt polymerization. In the case of SBAFP-m1, the 0108. The RGAFP-m1 gene in pET28a and WT RGAFP monomer simulation also probed the viability of the novel gene in pRT24a were expressed in E. coli BL21 (DE3) cells. inter-monomer interface, with fused identical 90-residue Whole cells were lysed by boiling for 10 min, releasing the segments used as proxies, where de-registry or layer sepa heat-stable RGAFP-m1 and WT RGAFP into the soluble ration of the ideal motif might have occurred. fraction. This was followed by a 2h cooling period to room 0103 Models of eleven-unit multimers were also con temperature to refold the protein and storage at 4°C. The structed to observe the behavior of ideal fibrils. Here, a heat stable properties of WT RGAFP are retained in sample often dimer interfaces could be observed simultane RGAFP-m1, as shown in FIG. S1B. A yield of ~10 mg of ously for defects or instability, along with the macroscale pure RGAFP-m1 per liter of growth medium was obtained. super-helical tendencies of a long polymeric fibril. Molecu WT RGAFP gave a yield of ~50 mg of pure protein per liter lar dynamics simulations demonstrated that both designs of growth medium. As with SBAFP, both RGAFP-m1 and were stable out to 20 ns. A measure of this is seen in FIG. WT RGAFP were incubated at 37° C. to promote fibril 6 where we plot the B-sheet content of simulated fibrils over growth. time, averaged over five different runs for the SBAFP-m1 0109 Spectroscopic Characterization fibril and the RGAFP-m2 fibril. The latter is similar in 0110 (A) Mass Spectroscopy. design to the experimentally studied RGAFP-m1. The 0111 SBAFP-m1 has a calculated molecular mass of B-sheet content is determined by counting all residues within 19,397 Da with the N-terminal Met and 19,265 Da without the B-sheet region of the Ramachandran plot (for further it. ESI-MS analysis gives a molecular weight of 19,267+4.8 details see the Methods section) within the Visual Molecular Da, corresponding to the protein without the N-terminal Dynamics program (VMD). For the idealized three-sided Met, as desired. SBAFP-m1 runs on SDS-PAGE with an structure, this yields 80% f-sheet content (twelve B-strand apparent molecular weight of 20 kDa. residues per 15 residue rung), and for the idealized two 0112 The RGAFP-m1 protein has a calculated molecular sided structure this yields 70% (ten B-strand residues per 14 mass of 11,410 Da with the N-terminal Met and 11.279 Da residue rung). The monomeric form retained all native without it. ESI-MS analysis gives a molecular weight of B-helical contacts, including those corresponding to terminal 11,280+2.8 Da, again corresponding to the mutant without sequence segments that one might expect to separate or fray. the N-terminal Met. RGAFP-m1 runs on SDS-PAGE with Simulations of polymer fibrils showed that all dimer inter an apparent molecular weight of 25 kDa. faces remained intact, and imperfections in monomer dock 0113. This MS data provides important confirmation that ing during the model construction process became annealed the sequences of the proteins produced correspond exactly to into seamless registry. those designed and tested for stability. No modifications to 0104. Another role for the simulations is to provide the proteins were detected. analyses of possible height profiles of fibrils to compare with 0114 (B) Circular Dichroism. the AFM topographic information. Note that this variation is 0115 The CD spectrum for SBAFP-m1 is presented in mainly due to the triangular and rectangular cross-sections FIG. 8A. It shows a single peak at ~220 nm, Suggesting a of the proteins. FIG. 7 shows height curves for monomers of largely 3-sheet structure. Deconvolution of the spectrum SBAFP-m1 and RGAFP-m2 found using the untwisted using the Contin Set 4' program on the online monomers. The SBAFP-m1 monomer has a mean height DichroWeb' server gave the following estimates for sec ranging from 2.25-2.55 nm, while the RGAFP-m2 monomer ondary structure content: 4% C.-helix. 64% B-sheet/turn and has a height variation between 1.6-2.6 mm. 32% random coil. Simulation gives 80% B-sheet structure, 0105 Taken together, the simulations, carried out in per FIG. 6. This may indicate somewhat less stability for the advance of the experimental fibril synthesis, portended the experimental fibrils than expected by simulations. The experimental finding that the designed B-Solenoid proteins SBAFP model has no C-helix, 89% B-sheet, 3% turns, and form stable amyloid fibrils, and allowed us to exclude 8% coil as calculated by YASARA. several intermediate designs in the process. Also, simula 0116. The CD spectrum for RGAFP-m1 is presented in tions provide important height profiles for comparison to FIG. 8B. As with SBAFP-m1, it shows a single peak at ~220 nm, indicative of a largely B-sheet structure. Deconvolution AFM data to provide corroborating evidence that the of the spectrum using CONTIN-Set 4 gave a secondary observed fibrils have the desired structure. structure content of 2% alpha helix, 63% B-sheet/turn and 01.06 Protein Expression, Purification, and Folding 35% random coil. Simulations give 72% B-sheet as shown 0107 The E. coli codon-optimized genes for both in FIG. 6. The RGAFP model has no C-helix, 88% B-sheet, SBAFP-m1 and RGAFP-m1 were procured from Life Tech and 12% coil as calculated by YASARA. nologies The SBAFP-m1 variant expressed well in E. coli 0117 The secondary structure assignments agree reason BL21 (DE3) from the plT28a vector, although it was found ably well between structural models and the CD deconvo almost completely in inclusion bodies (IBs). The protein was lutions. The average error in secondary structure assignment purified using a protocol of repeated IB washing, denatur by the CONTIN software is -5% for a structure and -10% ation in 8.5 M urea, and purification by anion exchange for B structure, although this depends on the protein and chromatography (see Methods). Purified, unfolded protein the lowest wavelength used in analysis." Additional evi was folded to its native state by stepwise urea removal via dence of extended B-sheet formation is presented below. dialysis. Generally, a yield of 30-40 mg of pure protein per 0118 (C) Thioflavin-T Fluorescence. liter of growth medium was obtained. Once the protein was 0119) Thioflavin-T is a small molecule commonly used purified and refolded, it was placed in an incubator at 37° C. for the detection of amyloid cross-B structure in peptides and to allow fibril formation. Expression of the naturally occur proteins.' " The binding mechanism is not well under ring SBAFP followed a literature procedure, with minor stood but is thought to involve either binding into “chan changes detailed in the Methods section. nels' between outward facing side chains of B-sheets and/or US 2017/005.8007 A1 Mar. 2, 2017

ThT micelle formation." Empirically, ThT fluorescence is species in Solution by DLS complements AFM imaging in significantly altered when it binds to cross-B structures: characterizing the size distribution of fibrils. compared to ThT free in Solution, the excitation maximum (0.125 For SBAFP-m1, DLS measurements performed shifts from 385 to 450 nm in the presence of B-sheet fibrils, before incubation show a species having an apparent hydro and the emission maximum changes from 445 nm to 482 dynamic diameter of 6.6+ 1.4 nm that constitutes -99.8% of nm.' This fluorescence shift has been used extensively to the sample, with the remainder consisting of minor species probe peptides for B-sheet secondary structure, primarily between 59 and 5560 nm in diameter. After incubation at 37° with amyloid fibrils such as those from AB(1-42),' insulin, C. for 24 h, the species at 6.6 nm (presumably the monomer and immunoglobulin light chain variable domain SMA." since the calculated diameter of gyration is ~4 nm for the 0120. During the refolding of SBAFP-m1 by stepwise unhydrated model) is absent. Larger species at 32-8 and dialysis against Solutions of decreasing urea concentrations, 230-43 nm in apparent diameter are present. Thus, SBAFP protein samples at each urea concentration were collected m1 may self-assemble to a minor extent at 4°C., while the and analyzed for ThT fluorescence. FIG. 9A shows the ThT monomer is absent and larger species are present after 24h fluorescence of SBAFP-m1 at urea concentrations from 8 M incubation at 37° C. urea to 0 Murea at 4°C., and SBAFP-m1 in the absence of 0.126 Dynamic light scattering results with RGAFP-m1 urea after incubation at 37° C. for 24 h. There is a gradual also show an increase in fibrils after incubation at 37° C. increase in fluorescence at 482 nm for SBAFP-m1 as the Before incubation, the sample consisted of species with urea concentration decreases; given the long time of incu apparent hydrodynamic diameters of 5.0+0.3 nm (~98.7%), bation at 4°C. (8 days total to change from 8 to O Murea) 28+4 nm (~1.1%) and 143+40 nm (-0.2%). After incubation the increase in ThT signal may arise from partial polymer at 37°C., the monomer at 5 nm is absent, and a species with ization. an apparent diameter of 268+48 nm constitutes the entire 0121 SBAFP-m1 at 4° C. gives low ThT fluorescence sample. The presence of Small amounts of large species in which is significantly above background and greater than the sample before incubation at 37° C. could either be due that for WTSBAFP as shown in FIG.9B. The larger signal to polymerization at 4° C., as with SBAFP-m1, or the from SBAFP-ml compared to WTSBAFP implies a small procedure by which RGAFP-m1 was purified and folded. degree of polymerization during the extended period (8 Unlike SBAFP-m1, RGAFP-m1 was purified and folded by days) required for stepwise dialysis at 4°C. This is further boiling followed by slow cooling to room temperature. This evidence that SBAFP-m1 folds into the predicted B-helical process gives the newly folded protein Substantial time at structure after removal of urea by dialysis. Incubation of both 4° C. and higher temperatures to undergo polymeriza SBAFP-m1 at 37° C. for 24 h significantly increases ThT tion. fluorescence at 482 nm, indicating formation of an extended (O127 AFM Imaging B-sheet species that has higher ThT binding capacity or I0128. The formation of fibers by SBAFP-m1 was ana greater effects on the spectral changes of ThT. In general, lyzed after 48 h and 3 weeks of incubation at 37° C. After incubation of amyloid-prone proteins and peptides at 48 h at 37°C., the proteins assembled into fibers, as revealed elevated temperatures is known to facilitate amyloid fibril by AFM imaging (FIG. 10A). These fibers exhibit no formation. The increase in ThT fluorescence with SBAFP branching or bundling and their lengths measure ~1-7 um. m1 after incubation at 37° C. implies that the elevated The expansion (FIG. 10A inset) allows an accurate height temperature promotes fibril formation since an identical determination of 1.5-3 nm, but predominantly between 2.5 sample held at 4°C. does not show the same large increase and 3 nm. In contrast to SBAFP-m1, WTSBAFP does not over 24 h (FIG. 9B). The kinetic studies discussed below yield well-defined fibrils under similar conditions. Instead, it confirm this. forms arch shaped assemblies (40-150 nm in length), and 0122) The RGAFP-m1 protein has an increased ThT amorphous aggregates (see red arrows in FIG. 10B). The fluorescence emission peak at 482 nm compared to WT height of the WTSBAFP monomers is 1 nm (see FIG. 10B RGAFP as shown in FIG.9C. Incubation of the WT RGAFP inset). sample at 37° C. did not change the ThT emission intensity, I0129. Extended (e.g. 3 week) incubation at 37° C. leads suggesting that WT RGAFP does not form amyloid fibrils. to bundles of fibrils in the case of SBAFP-m1, as shown in RGAFP-m1 kept at 4° C. exhibits significant ThT fluores FIG. 10C. The expanded view shows that the long axes of cence, albeit at a weaker intensity than the sample after the fibrils are parallel to each other within each bundle (FIG. incubation at 37° C.; this suggests some fibril formation at 10C inset and FIG. 10D) although the fibrils themselves, in the lower temperature. This is also evidenced by the DLS the sense of N-to-C terminal may be parallel or antiparallel. data discussed below, which shows evidence for aggregates The number of fibrils in each bundle varies from ~2-4 as in the 4°C. sample. After incubation, fluorescence at 482 nm shown on FIG. 10D. The overall height of the bundle varies increases considerably, undergoing the same pattern seen from 6-10 nm, depending on the number of component with SBAFP-m1. This indicates the formation of longer fibers. The inset of FIG. 10C shows the number of individual amyloid fibrils with B-sheet conformation. Only weak fluo SBAFP-m1 fibrils and overall diameter of a typical bundle. rescence is seen for WT RGAFP both at 4° C. and after After 3 weeks of incubation at 37° C., SBAFP-m1 also incubation at 37° C., indicating the changes made to the forms random, haystack-like aggregates along with fibrils protein template are causing the controlled aggregation. and bundles as shown in FIG. 10D. Fibril formation for both SBAFP-m1 and RGAFP-m1 is 0.130 AFM images of RGAFP-m1 samples were taken further evidenced by DLS data, discussed below. after three days incubation at 37° C. FIG. 11A shows high 0123 (D) Dynamic Light Scattering. coverage of RGAFP-m1 fibrils. These fibrils are unbranched 0.124 DLS measures the hydrodynamic size of species and very uniform in size, with height of 1.5-2.0 nm, and present in Solutions and is a non-destructive method for length of 150-400 nm. In contrast, no fibrillar structures investigation of self-assembly. Measuring the size of the were observed under the same condition using WT RGAFP. US 2017/005.8007 A1 Mar. 2, 2017

as shown in FIG. 11B. The bright features in AFM topo mentally in Solution. It is possible for anisotropic geometric graphs measure 1-10 nm tall with various lateral dimen constraints to significantly reduce k, a point we shall sions, which is consistent with monomeric and aggregates of explore in future work. RGAFP monomers. The time-dependence of fibril and I0135) In preliminary experiments, we find that polymer bundle formation is qualitatively consistent with known ization under identical conditions except that NaCl is added amyloid fibril formation kinetics. to 1 M final concentration occurs approximately 3-fold 0131 Kinetic Analysis slower. When fibrils are harvested from polymerization (0132) The kinetics of fibril formation were followed for reactions by centrifugation and resuspended in fresh buffer SBAFP-m1. DLS measurements on samples at 4°C. before at 60° C., approximately 30% of the initial turbidity is lost the initiation of polymerization at 37° C. show the monomer in 1.5 h, after which it is stable over time. Similar experi as the only species present. Within 10 min of incubation at ments show that approximately 5% of the initial turbidity is 37° C. the monomer is completely gone and replaced by lost when fibrils are resuspended in buffer containing 80% larger species. The formation of large polymers was moni ethanol in 1.5 h, after which it is stable over time. tored by turbidity at 300 nm. A time course is presented in FIG. 12. A lag phase is indicated by the data, and a simple CONCLUSIONS exponential model gives a poor fit. The data were analyzed 0.136. In conclusion, we have shown that two monomeric within the framework presented by Ferrone.’ Considering wild type antifreeze BSPs (from Spruce Budworm and the 12 rung cross-B structure of SBAFP-m1, we assumed Ryegrass) can be engineered to polymerize into amyloid that the critical nucleus size is 1 so that the initial concen fibrils as evidenced by CD, DLS. ThT fluorescence and tration of polymerization nuclei is equal to the initial con AFM imaging, with the expected height profiles observed in centration of protein. This leads to a simplified expression AFM. To our knowledge, this is the first confirmation of for the polymerization kinetics, given in Eq. 1, amyloid formation from large BSP monomers, albeit in a y=1-sech(k, "Mot) (1) synthetic context, despite the extant speculation that BSP structures should arise for known wild type amyloidogenic where k is the effective forward rate constant for polym proteins' (we refer here to large BSPs with tightly erization, Mo is the initial concentration of monomers, and packed interiors, although the Het-S fungal prion is known t is time. This equation results from a new analytic Solution to form a more open f3-solenoid amyloid structure). to the Ferrone nucleated linear polymerization equations. 0.137 The evidence provided here demonstrates the the full details of which will be presented elsewhere. By applicability of these BSPs with extraordinary geometries effective k value here, we mean that we are not explicitly for precision nanoscale applications such as templating accounting for nucleation effects Such as the need for a nanoparticles for functional devices, enzyme arrays, peptide different, rare monomer conformation to nucleate fibrils as arrays for regenerative medicine, etc. Recently, enzymatic' has been argued for PolyO aggregation. and charge transfer'activity for peptide-based amyloids has 0.133 Fitting the turbidity data to Eq. 1 (FIG. 12) gives a been demonstrated. The larger and more readily manipulated value of 14+1 M's for the rate constant, while fitting to BSPs employed here hold greater promise for tailor-made ThT time course data (not shown) gives a value of 42+7 M' nanoscale templates with precisely spatially defined func s'. These effective values for k are rapid compared to tionalization on the 2-10 nm Scale. those measured for other cross-B fibril formation reactions. 0.138. It is understood that the examples and embodi In particular, effective k values from the literature are: 0.3 ments described herein are for illustrative purposes only and M' s for insulin, 3.3 M' s for WT PrP (prion) that various modifications or changes in light thereof will be elongation in mice (assuming a concentration of PrP in Suggested to persons skilled in the art and are to be included mice brain tissues), 0.034, 0.3, and 1.4 M's for PolyO within the spirit and purview of this application and scope of polymers with 28, 36, 47 glutamines, respectively, and the appended claims. All publications, patent Database 0.14, 0.58, 1.5, 6.9 M' s for the spider silk amyloid Accessions, patents, and patent applications cited herein are peptide constructs eADF4(C2), eADF4(C4), eADF4(C8) hereby incorporated by reference in their entirety for all and eADF4(C16), respectively." While a critical nucleus for aggregation of 1 was determined for the PolyO case, it is purposes. unknown in the others, and assumed to be 2 for insulin." REFERENCES Nonetheless, it is apparent that despite the large size of the SBAFP monomers, the fact that they have preformed I0139 1. Olmsted, J. B.: Borisy, G. G. Microtubules. B-sheet structure affords much more rapid aggregation than Annu. Rev. Biochem. 1973, 42, 507-540. other amyloids, particularly in the short time limit where 0140 2. Zhang, S. G. Fabrication of Novel Biomaterials aggregation goes generically as k, Through Molecular Self-assembly. Nat. Biotechnol. 2003, 0134. The fit assumes the turbidity is proportional to the 21, 1171-1178. total mass of polymer, which is a reasonable assumption 0141 3. Sara, M.; Sleytr, U. B. Crystalline Bacterial Cell provided that the fibril lengths exceed the wavelength of Surface Layers (S-layers): From Cell Structure to Biomi light (300 nm here). This may break down in detail at short metics. Prog. Biophys. Mol. Biol. 1996, 65, 83-111. times in our experiments. We also note that our effective k, 0142. 4. Chiti, F: Dobson, C. M. Protein Misfolding, values fall many orders of magnitude below the Smolu Functional Amyloid, and Human Disease. Annu. Rev. chowski diffusion limit k, of about 10 M's', a point Biochem. 2006, 75, 333-366. noted before for polyO aggregation. In our case it is 0.143 5. Seeman, N. C. Nanomaterials Based on DNA. In unlikely that this arises from a small probability nucleation Annu. Rev. Biochem. Kornberg, R. D.; Raetz, C. R. H.; complex as argued for PolyO, since the monomer form Rothman, J. E.; Thorner, J. W., Eds. 2010; Vol. 79, pp appears highly stable on its own in simulations and experi 65-87. US 2017/005.8007 A1 Mar. 2, 2017

0144 6. Le, J. D.; Pinto, Y.: Seeman, N. C.; Musier Formation of Functionalized Nanowires by Control of Forsyth, K.; Taton, T. A.; Kiehl, R. A. DNA-templated Self-Assembly Using Multiple Modified Amyloid Pep Self-assembly of Metallic Nanocomponent Arrays on a tides. Adv. Funct. Mater. 2013, 23, 4881-4887. Surface. Nano Lett. 2004, 4, 2343-2347. (0160 22. Knowles, T. P. J.; Oppenheim, T. W.; Buell, A. (0145 7. Sacca, B.; Meyer, R.; Erkelenz, M.: Kiko, K.: K. : Chirgadze, D. Y.; Welland. M. E. Nanostructured Arndt, A.; Schroeder, H.; Rabe, K. S.: Niemeyer, C. M. Films from Hierarchical Self-assembly of Amyloidogenic Orthogonal Protein Decoration of DNA Origami. Angew. Proteins. Nat. Nanotechnol. 2010, 5, 204-207. Chem. Int. Ed. 2010, 49, 9378-9383. 0161) 23. Arora, A.; Ha, C.; Park, C. B. Insulin Amyloid 0146 8. Liu, J. F.; Uprety, B. Gyawali, S.; Woolley, A. Fibrillation at Above 100 degrees C.: New Insights into T.: Myung, N. V.; Harb, J. N. Fabrication of DNA Protein Folding Under Extreme Temperatures. Protein Templated Te and Bi2Te3 Nanowires by Galvanic Dis Sci. 2004, 13, 2429-2436. placement. Langmuir 2013, 29, 11176-11184. (0162 24. Baxa, U.: Ross, P. D.; Wickner, R. B.; Steven, 0147 9. Castro, C. E.; Kilchherr. F.; Kim, D.-N.; Shiao, A. C. The N-terminal Prion Domain of Ure2p Converts E. L.; Wauer, T.: Wortmann, P.; Bathe, M.; Dietz, H. A from an Unfolded to a Thermally Resistant Conformation Primer to Scaffolded DNA Origami. Nat. Methods 2011, Upon Filament Formation. J. Mol. Biol. 2004, 339, 259 8, 221-229. 264. 0148, 10. Mao, C. B.: Solis, D. J.; Reiss, B. D.; Kott (0163. 25. Kardos, J.; Micsonai, A.; Pal-Gabor, H.; Petrik, mann, S.T.: Sweeney, R.Y.; Hayhurst, A.; Georgiou, G.; E.; Graf. L.; Kovacs, J.; Lee, Y. H.; Naiki, H., Goto, Y. Iverson, B.: Belcher, A. M. Virus-based Toolkit for the Reversible Heat-Induced Dissociation of beta(2)-Micro Directed Synthesis of Magnetic and Semiconducting globulin Amyloid Fibrils. Biochemistry 2011, 50, 3211 Nanowires. Science 2004, 303, 213-217. 3220. 0149 11. Lee, Y. J.; Belcher, A. M. Nanostructure Design (0164. 26. Hammer, N. D.; Wang, X.; McGuffie, B. A.; of Amorphous FePO4 Facilitated by a Virus for 3 V Chapman, M. R. Amyloids: Friend or foe? Journal of Lithium Ion Battery Cathodes. J. Mater. Chem. 2011, 21, 1033-1039. Alzheimers Disease 2008, 13, 407-419. 0150 12. Barnhart. M. M.; Chapman, M. R. Curli Bio (0165 27. Ryu, J.; Park, C. B. High Stability of Self genesis and Function. Annu. Rev. Microbiol. 2006, 60, Assembled Peptide Nanowires Against Thermal, Chemi 131-147. cal, and Proteolytic Attacks. Biotechnol. Bioeng. 2010, 0151. 13. Si, K. Choi, Y.B.; White-Grindley, E.; Majum 105, 221-230. dar, A.; Kandel. E. R. Aplysia CPEB Can Form Prion-like (0166 28. Sunde, M.; Blake, C. The Structure of Amyloid Multimers in Sensory Neurons that Contribute to Long Fibrils by Electron Microscopy and X-ray Diffraction. Term Facilitation. Cell 2010, 140, 421-U179. Adv. Protein Chem. 1997, 50, 123-159. 0152 14. Maji, S. K. Perrin, M. H.; Sawaya, M. R.: (0167 29. Wille, H.; Bian, W.; McDonald, M.; Kendall, Jessberger, S.; Vadodaria, K.: Rissman, R. A. Singru. P. A.: Colby, D. W.; Bloch, L.; Ollesch, J.; Borovinskiy, A. S.: Nilsson, K. P. R.: Simon, R.; Schubert, D.; al., e. L.; Cohen. F. E.; Prusiner, S. B. Stubbs, G. Natural and Functional Amyloids As Natural Storage of Peptide Hor Synthetic Prion Structure from X-ray Fiber Diffraction. mones in Pituitary Secretory Granules. Science 2009, Proc. Natl. Acad. Sci. U.S.A 2009, 106, 16990-16995. 325, 328-332. (0168 30. Buell, A. K. Dhulesia, A.; Mossuto, M. F.; 0153. 15. Knowles, T. P. J.; Smith, J. F.; Craig, A.; Cremades, N.; Kumita, J. R.; Dumoulin, M.; Welland, M. Dobson, C. M.; Welland, M. E. Spatial Persistence of E.; Knowles. T. P. J.; Salvatella, X.; Dobson, C. M. Angular Correlations in Amyloid Fibrils. Phys. Rev. Lett. Population of Nonnative States of Lysozyme Variants 2006, 96. Drives Amyloid Fibril Formation. J. Am. Chem. Soc. 0154 16. Knowles, T. P. J.; Buehler, M. J. Nanomechan 2011, 133, 7737-7743. ics of Functional and Pathological Amyloid Materials. (0169. 31. Graether, S. P.: Kuiper, M. J.; Gagne, S. M.: Nat. Nanotechnol. 2011, 6, 469-479. Walker, V. K. Jia, Z. C. Sykes, B. D.; Davies, P. L. (O155 17. Slotta, U.: Hess, S.; Spiess. K. Stromer, T.: beta-helix structure and ice-binding properties of a hyper Serpell, L.; Scheibel, T. Spider Silk and Amyloid Fibrils: active antifreeze protein from an insect. Nature 2000, 406, A Structural Comparison. Macromol. Biosci. 2007, 7. 325-328. 183-188. (0170 32. Benbassat, A.; Bauer. K. Chang. S. Y.: 0156 18. Sullan, R. M. A.; Gunari, N.; Tanur, A. E.; Yuri. Myambo, K., Boosman, A.; Chang, S. Processing of the C.; Dickinson, G. H.; Orihuela, B., Rittschof. D.; Walker, Initiation Methionine from Proteins—Properties of the G. C. Nanoscale Structures and Mechanics of Barnacle Escherichia-Coli Methionine Aminopeptidase and its Cement. Biofouling 2009, 25, 263-275. Gene Structure. J. Bacteriol. 1987, 169, 751-757. (O157. 19. Scheibel, T.: Parthasarathy, R.: Sawicki, G.; (0171 33. Gauthier, S. Y.: Kay, C. M.: Sykes. B. D.; Lin, X. M.; Jaeger, H., Lindquist, S. L. Conducting Walker, V. K. Davies, P. L. Disulfide Bond Mapping and Nanowires Built by Controlled Self-assembly of Amyloid Structural Characterization of Spruce Budworm Anti Fibers and Selective Metal Deposition. Proc. Natl. Acad. freeze Protein. Eur, J. Biochem. 1998, 258, 445-453. Sci. U.S.A 2003, 100, 4527-4532. 0172. 34. Middleton. A. J.; Marshall, C.B.; Faucher. F.; 0158. 20. Reches, M.: Gazit, E. Casting Metal Nanowires Bar-Dolev, M.; Braslavsky, I.: Campbell, R. L.; Walker, V. Within Discrete Self-assembled Peptide Nanotubes. Sci K.; Davies, P. L. Antifreeze Protein from Freeze-Tolerant ence 2003, 300, 625-627. Grass Has a Beta-Roll Fold with an Irregularly Structured 0159. 21. Sakai, H.; Watanabe, K.: Asanomi, Y.: Ice-Binding Site. J. Mol. Biol. 2012, 416, 713-724. Kobayashi, Y.; Chuman. Y.; Shi, L. H.; Masuda. T.; (0173 35. Cooper, S.; Khatib, F.; Treuille, A.; Barbero, J.; Wyttenbach, T.; Bowers, M.T.: Uosaki. K. Sakaguchi, K. Lee, J.; Beenen. M.: Leaver-Fay, A.; Baker, D.; Popovic, US 2017/005.8007 A1 Mar. 2, 2017

Z. Players, F. Predicting Protein Structures with a Mul 0188 50. Kusumoto, Y.: Lomakin, A.; Teplow, D. B.: tiplayer Online Game. Nature 2010, 466, 756-760. Benedek, G. B. Temperature Dependence of Amyloid 0.174 36. D. A. Case, T. A. D., T. E. Cheatham, III, C. L. Beta-protein Fibrillization. Proc Natl AcadSci USA 1998, Simmerling, J. Wang, R. E. Duke, R.; Luo, R. C. W., W. 95, 12277-12282. Zhang, K. M. Merz, B. Roberts, S. Hayik, A. Roitberg, G. (0189 51. Harrison, R. S.; Sharpe, P. C.: Singh.Y.; Fairlie. Seabra: J. Swails, A. W. G. I. Kolossváry, K. F. Wong, F. D. P. Amyloid Peptides and Proteins in Review. Rev. Paesani, J. Vanicek, R. M. Wolf, J. Liu; X. Wu, S. R. B., Physiol. Biochem. Pharmacol. 2007, 159, 1-77. T. Steinbrecher, H. Gohlke. Q. Cai, X. Ye. J. Wang, M.-J. 0.190 52. Ferrone, F. Analysis of Protein Aggregation Hsieh, G., Cui, D. R. R. D. H. Mathews, M. G. Seetin, R. Kinetics. Methods Enzymol. 1999, 309, 256-274. Salomon-Ferrer, C. Sagui, V. Babin, T.; Luchko, S. G. A. (0191 53. Bhattacharyya, A. M.: Thakur, A. K.: Wetzel, Kovalenko, and P. A. Kollman AMBER 12, University of R. Polyglutamine aggregation nucleation: Thermodynam California San Francisco. 2012. ics of a highly unfavorable protein folding reaction. (0175 37. Humphrey, W.; Dalke, A.; Schulten, K. VMD: Proceedings of the National Academy of Sciences of the Visual Molecular Dynamics. J. Mol. Graphics Modell. United States of America 2005, 102, 15400-15405. 1996, 14, 33-38. (0192 54. Knowles, T. P. J.; Waudby, C. A.; Devlin, G. L.; (0176) 38. Steinle, A.; Li, P.; Morris, D. L.; Groh, V.: Cohen, S.I.A.; Aguzzi, A.; Vendruscolo, M.; Terentjev, E. Lanier, L.; Strong, R. K. Spies, T. Interactions of human M.; Welland, M. E.; Dobson, C. M. An Analytical Solu NKG2D with its ligands MICA, MICB, and homologs of tion to the Kinetics of Breakable Filament Assembly. the mouse RAE-1 protein family. Immunogenetics 2001, Science 2009, 326, 1533-1537. 52, 279-287. (0193 55. Cox, D. L.; Singh, R. R. P: Yang, S. C. Prion 0177 39. Provencher, S. W.; Glockner, J. Estimation of Disease: Exponential Growth Requires Membrane Bind Globular Protein Secondary Structure from Circular-di ing. Biophys. J. 2006, 90, L77-L79. chroism. Biochemistry 1981, 20, 33-37. 0194 56. Chen, S. M.; Ferrone. F. A.; Wetzel, R. Hun 0.178 40. Janes, R. W.; Wallace, B. A. Modern Tech tington's Disease Age-of-onset Linked to Polyglutamine niques in Circular Dichroism and Synchrotron Radiation Aggregation Nucleation. Proc Natl Acad Sci USA 2002, Circular Dichroism Spectroscopy. IOS Press: 2009. 99, 1 1884-11889. 0179 41. Whitmore, L.; Wallace, B. A. Protein Second (0195 57. Humenik, M., Magdeburg, M., and Scheibel, T. ary Structure Analyses from Circular Dichroism Spec Influence of Repeat Numbers on Self-assembly Rates of troscopy: Methods and References Databases. Biopoly Repetitive Recombinant Spider Silk Proteins. J. Struct. mers 2007, 89, 392–400. Biol. 2014, 186, 431-437. 0180. 42. Sreerama, N.; Woody, R. W. Estimation of (0196) 58. Berne, B. J. Interpretation of the Light Scat Protein Secondary Structure from Circular Dichroism tering from Long Rods J Mol Biol 1974, 89, 755-758. Spectra: Comparison of CONTIN, SELCON, and (0197) 59. Northrup, S. H.; Erickson, H. P. Kinetics of CDSSTR Methods with an Expanded Reference Set. Protein-Protein Association Explained by Brownian Anal. Biochem. 2000, 287, 252-260. Dynamics Computer-simulation. Proc Natl AcadSci USA 1992, 89, 3338-3342. 0181. 43. Greenfield, N. J. Using Circular Dichroism 0198 60. Perutz, M. F.; Finch, J. T.; Berriman. J.: Lesk, Spectra to Estimate Protein Secondary Structure. Nat. A. Amyloid Fibers are Water-filled Nanotubes. Proc Natl Protoc. 2006, 1, 2876-2890. Acad Sci USA 2002, 99, 5591-5595. 0182) 44. Vassar, P. S.: Culling, C. F. A. Fluorescent (0199 61. Govaerts, C.; Wille, H.; Prusiner, S. B.; Cohen, Stains, with Special Reference to Amyloid and Connec F. E. Evidence for Assembly of Prions with Left-handed tive Tissues. Arch. Pathol. 1959, 68, 487-498. Beta 3-helices into Trimers. Proc Natl Acad Sci USA 0183 45. Levine. H. Thioflavine-T Interaction with Syn 2004, 101, 8342-8347. thetic Alzheimers-disease Beta-amyloid Peptides—De 0200) 62. Stork, M. Giese, A.; Kretzschmar, H. A.; tection of Amyloid Aggregation in Solution. Protein Sci. Tavan, P. Molecular Dynamics Simulations Indicate a 1993, 2, 404-410. Possible Role of Parallel Beta-helices in Seeded Aggre 0184. 46. Biancalana, M.: Koide, S. Molecular Mecha gation of Poly-Gln. Biophys J 2005, 88, 2442-2451. nism of Thioflavin-T Binding to Amyloid Fibrils. Bio 0201 63. Kunes, K. C.: Clark. S. C.: Cox, D. L.; Singh, chim. Biophys. Acta, Proteins Proteomics 2010, 1804, R. R. P. Left Handed Beta-helix Models for Mammalian 1405-1412. Prion Fibrils. Prion 2008, 2, 81-90. 0185. 47. LeVine, H. Quantification of Beta-sheet Amy 0202) 64. Wasmer, C.; Lange, A.; Van Melckebeke, H.; loid Fibril Structures with Thioflavin-T. Amyloid, Prions, Siemer, A. B. Rick, R.; Meier, B. H. Amyloid Fibrils of and Other Protein Aggregates 1999, 309, 274-284. the HET-s(218-289) Prion Form a Beta Solenoid with a 0186. 48. Nielsen, L.; Khurana, R.; Coats, A.; Frokjaer, Triangular Hydrophobic Core. Science 2008, 319, 1523 S.; Brange, J.; Vyas, S.; Uversky, V. N.; Fink, A. L. Effect 1526. of Environmental Factors on the Kinetics of Insulin Fibril 0203 65. Rufo, C. M.; Moroz, Y. S.; Moroz, O. V.; Stohr, Formation: Elucidation of the Molecular Mechanism. J.; Smith, T. A.; Hu, X. Z. DeGrado, W. F.; Korendovych, Biochemistry 2001, 40, 6036-6046. I. V. Short Peptides Self-assemble to Produce Catalytic 0187 49. Ionescu-Zanetti, C.; Khurana, R.; Gillespie, J. Amyloids. Nat. R.; Petrick, J. S.; Trabachino. L. C.; Minert, L. J.; Carter, 0204 Chem. 2014, 6,303-309. S. A.; Fink, A. L. Monitoring the Assembly of Ig Light 0205 66. Ivnitski, D.; Amit, M.: Rubinov. B. Cohen chain Amyloid Fibrils by Atomic Force Microscopy. Proc Luria, R.; Ashkenasy, N.; Ashkenasy, G. Introducing Natl Acad Sci USA 1999, 96, 13175-13179. Charge Transfer Functionality into Prebiotically Relevant US 2017/005.8007 A1 Mar. 2, 2017 15

Beta-sheet Peptide Fibrils. Chem. Commun. (Cambridge. 0209 70. Pentelute, B. L.: Gates, Z. P.; Tereshko, V.: U.K.) 2014, 50, 6733-6736. Dashnau, J. L.; Vanderkooi, J. M.: Kossiakoff, A. A.; Kent, S. B. H. X-ray Structure of Snow Flea Antifreeze 0206 67. Khurana, R.; Coleman, C.; Ionescu-Zanetti, C.; Protein Determined by Racemic Crystallization of Syn Carter, S.A.: Krishna, V.; Grover. R. K.; Roy. R.: Singh. thetic Protein Enantiomers. J. Am. Chem. Soc. 2008, 130, S. Mechanism of Thioflavin-T Binding to Amyloid 9695-97.01. Fibrils. J. Struct. Biol. 2005, 151, 229-238. 0210 71. Lauersen, K. J.; Brown, A.; Middleton, A.; 0207. 68. Buchko, G. W.; Ni, S.S.; Robinson, H.; Welsh, Davies, P. L.; Walker, V. K. Expression and Character E. A.; Pakrasi, H. B.; Kennedy, M. A. Characterization of ization of an Antifreeze Protein from the Perennial Rye Two Potentially Universal Turn Motifs that Shape the Grass, Lolium Perenne. Cryobiology 2011, 62, 194-201. Repeated Five-residues Fold Crystal Structure of a 0211 72. Benach, J., Chen, Y., Vorobiev, S. M., Seeth Lumenal Pentapeptide Repeat Protein from Cyanothece araman. J., Ho, C. K., Janjua, H., Conover, K., Ma, L-C., Xiao, R., et al. Crystal Structure of ydcK from Salmonella 51142. Protein Sci. 2006, 15, 2579-2595. Cholerae at 2.38 A Resolution, http://www.pdb.org/pdb/ 0208 69. Ni, S. S.; Sheldrick, G. M.; Benning, M. M.: explore/explore.do?structureId=2PIG. Kennedy, M. A. The 2 Angstrom Resolution Crystal 0212. 73. Leinala, E. K. Davies, P. L.; Doucet, D.; Structure of HetL, a Pentapeptide Repeat Protein Tyshenko, M. G.; Walker, V. K., Jia, Z. C. A Beta-helical Involved in Regulation of Heterocyst Differentiation in Antifreeze Protein Isoform with Increased Activity— the Cyanobacterium Nostoc Sp Strain PCC 7120. J. Structural and Functional Insights. J. Biol. Chem. 2002, Struct. Biol. 2009, 165, 47-52. 277, 33349-33352.

INFORMAL SEQUENCE LISTING SEQ ID NO: 1 SBAFP-ml amino acid sequence SEQ ID NO: 2 SBAFP-ml amino acid sequence Exemplary BSPs: Chain A, Crystal Structure Of A Lumenal Pentapeptide Repeat Protein From Cyanothece Sp 51142 At 2.3 Angstrom Resolution. Tetragonal Crystal Form a. PDB : 2GOY A 1. mhhhhhhs sg lwprgsgmke taakferghm dispolgtodd dikamamvitgs sasyedvikli 61 gedfsgkslit yaciftnadlt dSnf seadilr gavfngsali gadlhgadlt nglayltsfk 121 gadltnavlt eaimmrtkfo dakitgadfs lavldvyevd klcdradgvn plktgvstres 181 lircq 1. Chain X, The 2. O Angstrom Resolution Crystal Structure Of Hetl, A Pentapeptide Repeat Protein Involved. In Heterocyst Differentiation Regulation From The Cyanobact crium Nostoc Sp. Strain Pcc 712O b. PDB: 3DU1 X 1. mgs shhhhhh ssgilviorgish mnvgeilirhy aagkrnfohi nilgeieltna slitgadl sya 61 dilrotirlgks infshtclrea dilseailwgi dlseadlyra ilreadltga klvktrleea 121 nlika slicga nilnsanlsrc llfcadilrps snortdl.gyv lltgadilsya dilraasilhha 181 nildgaklcra infgrticwgn laadils gasl qgadlsyanl esailirkanl qgadltgail 241 kdaelikgaim pdgsihd 2. Chain A, Crystal Structure Of Recombinant Human Alpha Lactalbumin c. PDB: 3BOI. A 1. mkoftkcells gllkdidgyg gialpelict mfhtsgyd to alivenneste yoglfgisnkl 61 works sqvpqs rnicci scolk flodditddi moakkildlik gidywlahka lictekleqwl 121 cekl 3. Chain B, Crystal Structure Of An Ice-Binding Protein From The Perennial Ryegrass, Lolium Perenne d. PDB:3ULT B 1. mdeqpntisg Snntvirsgsk invlagndintv isgdinns vsg snintv vsgnd ntvtgsnhviv 61 121 klaaalehhh hhh. 4. Chain A, Crystal Structure Of An Ice-Binding Protein From The Perennial Ryegrass, Lolium Perenne e. PDB:3ULT. A 1. mdeqpntisg Snntvirsgsk invlagndintv lisgdinns vsg snntwsgnd ntvtgenhw 61 Sgtnihivtdn nnnvsgndnn vsgsfhtv.sg ghntv.sgsnin tv.sgsnhws gsnkwtdaa 121 klaaalehhh hhh. 5. Chain B, Crystal Structure Of Yack From Salmonella Cholerae At 2.38 A Resolution. Northeast Structural Genomics Target Scré f. PDB: 2PIG B 1. xtkyrl segp raftycvdge kiksvillrovi avtdfndvika gtsggWvdad invil sqq.gdcw 61 iydenaxafa gteitgnari topctlynnv rigdnvwidir adisdgaris dnvtics ssv 121 reecaiygda rvlingseila igglthehad illqiydratv nhs rivhqvo lygnatitha US 2017/005.8007 A1 Mar. 2, 2017 16

- Continued

INFORMAL SEQUENCE LISTING 181 fiehraev fo falliegdkdn nv Wicccakv ygharviagt eedaiptlry issolvaehali 241 egncvlkhhv livgghaevrg gpilldorvl ieghacidge ilierqveis graaviafdd 3 O1 ntihlrg.pkv ingedritrt plvgsllehh hhhh. 6. Chain A, Crystal Structure Of Yack From Salmonella Cholerae At 2.38 A Resolution. Northeast Structural Genomics Target Scré g. PDB: 2PIG A xtkyrlsegp raftyqvdge kksvillrovi avto findvika gtsggWvdad invlsoggdcw 6 iydenaxafa gteit gnari topctlynnv rigdnvwidir adis dgaris dnvtidsssv 12 reecaiygda rvlindseila igglthehad illqiydratv nhsrivhqvo lygnatitha 18 fiehraev fo falliegdkdn nv Wicccakv ygharviagt eedaiptlry issolvaehali 24 egncvlkhhv livgghaevrg gpilldorvl ieghacidge ilierqveis graaviafdd 3 O ntihlrg.pkv ingedritrt plvgsllehh hhhh. 7. Chain A, Chorist oncura Fumiferana (Spruce Budworm) Antifreeze Protein Isoform 501 h. PDB:1M8N A dgtcvintnsq. itans qc vks tatncyidns qlvdt sictr sqys danvkk svttdcnidk 6 sqvylttctg. sqyngiyirs stttgtsisg pg.csist citi trgvatpaaa clisgc slsa 12 8. Chain B, Choristoneura Fumiferana (Spruce Budworm) Antifreeze Protein Isoform 501 i. PDB:1M8N B dgtcvintnsq. itans qc vks tatncyidns qlvdt sictr sqys danvkk svttdcnidk 6 sqvylttctg. sqyngiyirs stttgtsisg pg.csist citi trgvatpaaa clisgc slsa 12 . Chain C, Choristoneura Fumiferana (Spruce Budworm) Antifreeze Protein Isoform 501 j. PDB:1M8N C dgtcvintnsq. itans qc vks tatncyidns qlvdt sictr sqys danvkk svttdcnidk 6 sqvylttctg. sqyngiyirs stttgtsisg pg.csist citi trgvatpaaa clisgc slsa 12

10. Chain D, Choristoneura Fumiferana (Spruce Budworm) Antifreeze Protein Isoform 501 k. PDB:1M8N D dgtcvintnsq. itans qc vks tatncyidns qlvdt sictr sqys danvkk svttdcnidk 6 sqvylttctg. sqyngiyirs stttgtsisg pg.csist citi trgvatpaaa clisgc slsa 12

SEQUENCE LISTING

<16O is NUMBER OF SEO ID NOS : 15

<210s, SEQ ID NO 1 &211s LENGTH: 91 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: synthetic peptide SBAFP-m1

<4 OOs, SEQUENCE: 1 Ala Ser Arg Ile Thr Asn. Ser Glin Ile Val Lys Ser Glu Ala Thr Asn 1. 5 1O 15 Ser Asp Ile Asn. Asn. Ser Glin Lieu Val Asp Ser Ile Ser Thr Arg Ser 2O 25 3O Glin Tyr Ser Asp Ala Asn. Wall Lys Llys Ser Val Thir Thr Asp Ser Asn 35 4 O 45 Ile Asp Llys Ser Glin Val Tyr Lieu. Thir Thr Ser Thr Gly Ser Glin Tyr SO 55 6 O Asn Gly Ile Tyr Ile Arg Ser Ser Asp Thir Thr Gly Ser Glu Ile Ser 65 70 7s 8O Gly Ser Ser Ile Ser Thr Ser Arg Ile Thr Ile 85 90 US 2017/005.8007 A1 Mar. 2, 2017 17

- Continued

<210s, SEQ ID NO 2 &211s LENGTH: 111 212. TYPE: PRT <213> ORGANISM: Artificial Sequence 22 Os. FEATURE: <223> OTHER INFORMATION: synthetic peptide RGAFP-m1 <4 OOs, SEQUENCE: 2 Ala Asn Asp Ile Asp Gly Thr Asn. Asn. Glu Val Asp Gly Ser Glu Asn 1. 5 1O 15 Val Lieu Ala Gly Asn Asp Asn Thr Val Ser Gly Asp Asn. Asn. Ser Val 2O 25 3O Ser Gly Ser Asn Asn Thr Val Ser Gly Asn Asp Asn Thr Val Thr Gly 35 4 O 45 Ser Asn Met Val Val Ser Gly Thr Asn Met Ile Val Thr Asp Asn Asn SO 55 6 O Asn Asn Val Ser Gly Asn Asp Asn Asn Val Ser Gly Ser Phe Met Thr 65 70 7s 8O Val Ser Gly Gly Met Asn Thr Val Ser Gly Ser Asn Asn Thr Val Ser 85 90 95 Gly Lys Arg Met Arg Val Glin Gly Thr Asn. Asn Arg Val Thr Asp 1OO 105 11 O

<210s, SEQ ID NO 3 &211s LENGTH: 1.OO 212. TYPE PRT <213> ORGANISM: Choristoneura fumiferana

<4 OOs, SEQUENCE: 3 Asp Gly Thr Cys Val Asn Thr Asn Ser Glin Ile Thr Ala Asn Ser Glin 1. 5 1O 15 Cys Val Lys Ser Thr Ala Thr Asn. Cys Tyr Ile Asp Asn. Ser Glin Lieu. 2O 25 3O Val Asp Thir Ser Ile Cys Thr Arg Ser Glin Tyr Ser Asp Ala Asn Val 35 4 O 45 Llys Llys Ser Val Thir Thr Asp Cys Asn. Ile Asp Llys Ser Glin Val Tyr SO 55 6 O Lieu. Thir Thr Cys Thr Gly Ser Glin Tyr Asn Gly Ile Tyr Ile Arg Ser 65 70 7s 8O Ser Thir Thr Thr Gly. Thir Ser Ile Ser Gly Pro Gly Cys Ser Ile Ser 85 90 95 Thr Cys Thr Ile 1OO

<210s, SEQ ID NO 4 &211s LENGTH: 114 212. TYPE: PRT <213> ORGANISM: Lolium perenne

<4 OOs, SEQUENCE: 4 Pro Asn. Thir Ile Ser Gly Ser Asn Asn Thr Val Arg Ser Gly Ser Lys 1. 5 1O 15

Asn Val Lieu Ala Gly Asn Asp Asn Thr Val Ile Ser Gly Asp Asn. Asn 2O 25 3O

Ser Val Ser Gly Ser Asn Asn Thr Val Val Ser Gly Asn Asp Asn Thr 35 4 O 45 US 2017/005.8007 A1 Mar. 2, 2017 18

- Continued Val Thr Gly Ser Asn His Val Val Ser Gly Thr Asn His Ile Val Thr SO 55 6 O Asp Asn. Asn. Asn. Asn Val Ser Gly Asn Asp Asn. Asn. Wal Ser Gly Ser 65 70 7s 8O Phe His Thr Val Ser Gly Gly His Asn Thr Val Ser Gly Ser Asn Asn 85 90 95 Thr Val Ser Gly Ser Asn His Val Val Ser Gly Ser Asn Llys Val Val 1OO 105 11 O Thir Asp

<210s, SEQ ID NO 5 &211s LENGTH: 184 212. TYPE: PRT <213> ORGANISM: Cyanothece ATCC51142

<4 OOs, SEQUENCE: 5 Met His His His His His His Ser Ser Gly Lieu Val Pro Arg Gly Ser 1. 5 1O 15 Gly Met Lys Glu Thir Ala Ala Lys Phe Glu Arg Glin His Met Asp Ser 2O 25 3O Pro Asp Lieu. Gly Thr Asp Asp Asp Asp Lys Ala Met Ala Met Val Thr 35 4 O 45 Gly Ser Ser Ala Ser Tyr Glu Asp Wall Lys Lieu. Ile Gly Glu Asp Phe SO 55 6 O Ser Gly Lys Ser Lieu. Thir Tyr Ala Glin Phe Thr Asn Ala Asp Lieu. Thr 65 70 7s 8O Asp Ser Asn. Phe Ser Glu Ala Asp Lieu. Arg Gly Ala Val Phe Asin Gly 85 90 95 Ser Ala Lieu. Ile Gly Ala Asp Lieu. His Gly Ala Asp Lieu. Thir Asn Gly 1OO 105 11 O Lieu Ala Tyr Lieu. Thir Ser Phe Lys Gly Ala Asp Lieu. Thir Asn Ala Val 115 12 O 125 Lieu. Thr Glu Ala Ile Met Met Arg Thr Llys Phe Asp Asp Ala Lys Ile 13 O 135 14 O Thr Gly Ala Asp Phe Ser Lieu Ala Val Lieu. Asp Val Tyr Glu Val Asp 145 150 155 160 Llys Lieu. Cys Asp Arg Ala Asp Gly Val Asn Pro Llys Thr Gly Val Ser 1.65 17O 17s Thir Arg Glu Ser Lieu. Arg Cys Glin 18O

<210s, SEQ ID NO 6 &211s LENGTH: 257 212. TYPE: PRT <213> ORGANISM: Nostoc sp. PCC 7120

<4 OOs, SEQUENCE: 6 Met Gly Ser Ser His His His His His His Ser Ser Gly Lieu Val Pro 1. 5 1O 15 Arg Gly Ser His Met Asn Val Gly Glu Ile Lieu. Arg His Tyr Ala Ala 2O 25 3O

Gly Lys Arg Asin Phe Gln His Ile Asn Lieu. Glin Glu Ile Glu Lieu. Thir 35 4 O 45

Asn Ala Ser Lieu. Thr Gly Ala Asp Lieu. Ser Tyr Ala Asp Lieu. Arg Glin US 2017/005.8007 A1 Mar. 2, 2017 19

- Continued

SO 55 6 O

Thir Arg Luell Gly Ser Asn Phe Ser His Thr Cys Lell Arg Glu Ala 65 70 7s 8O

Asp Luell Ser Glu Ala Ile Lell Trp Gly Ile Asp Lieu. Ser Glu Ala Asp 85 90 95

Lell Tyr Arg Ala Ile Lell Arg Glu Ala Asp Lieu. Thir Gly Ala Luell 105 11 O

Wall Thir Arg Lell Glu Glu Ala Asn Luell Ile Llys Ala Ser Luell 115 12 O 125

Gly Ala Asn Luell Asn Ser Ala Asn Luell Ser Arg Cys Lell Luell Phe Glin 13 O 135 14 O

Ala Asp Luell Arg Pro Ser Ser Asn Glin Arg Thir Asp Lell Gly Tyr Wall 145 150 155 160

Lell Luell Thir Ala Asp Lell Ser Ala Asp Lieu. Arg Ala Ala Ser 1.65 17O

Lell His His Asn Lell Asp Gly Ala Lieu. Cys Arg Ala Asn Phe 185 19 O

Gly Arg Thir Glin Trp Gly Asn Luell Ala Ala Asp Lell Ser Gly Ala 195

Ser Luell Glin Ala Asp Lell Ser Ala ASn Lell Glu Ser Ala Ile 21 O 215

Lell Arg Asn Lell Glin Gly Ala Asp Lieu. Thir Gly Ala Ile Luell 225 23 O 235 24 O

Asp Ala Lell Gly Ala Ile Met Pro Asp Gly Ser Ile His 245 250 255 Asp

SEO ID NO 7 LENGTH: 124 TYPE : PRT ORGANISM: Artificial Sequence FEATURE: OTHER INFORMATION: synthetic peptide - chain A, recombinant human alpha lactalbumin

<4 OOs, SEQUENCE: 7

Met Lys Glin Phe Thir Glu Luell Ser Gln Lieu. Lell Asp Ile 1. 5 1O 15

Asp Gly Gly Gly Ile Ala Luell Pro Glu Lieu. Ile Thir Met Phe 25 3O

His Thir Gly Asp Thir Glin Ala Ile Wall Glu Asn Asn Glu Ser 4 O 45

Thir Glu Gly Lell Phe Glin Ile Ser Asn Llys Lieu Trp Ser SO 55 6 O

Ser Glin Wall Pro Glin Ser Arg Asn Ile Asp Ile Ser Asp 65 70 7s

Phe Luell Asp Asp Asp Ile Thir Asp Asp Ile Met Cys Ala Lys Ile 85 90 95

Lell Asp Ile Lys Gly Ile Asp Trp Luell Ala His Ala Luell 1OO 105 11 O

Thir Glu Lys Luell Glu Glin Trp Luell Glu Llys Lieu 115 12 O US 2017/005.8007 A1 Mar. 2, 2017 20

- Continued

<210s, SEQ ID NO 8 &211s LENGTH: 133 212. TYPE: PRT <213> ORGANISM: Lolium perenne <4 OOs, SEQUENCE: 8 Met Asp Glu Gln Pro Asn. Thir Ile Ser Gly Ser Asn Asn Thr Val Arg 1. 5 1O 15 Ser Gly Ser Lys Asn Val Lieu Ala Gly Asn Asp Asn Thr Val Ile Ser 2O 25 3O Gly Asp Asn. Asn. Ser Val Ser Gly Ser Asn. Asn Thr Val Val Ser Gly 35 4 O 45 Asn Asp Asn Thr Val Thr Gly Ser Asn His Val Val Ser Gly Thr Asn SO 55 6 O His Ile Val Thr Asp Asn. Asn. Asn. Asn Val Ser Gly Asn Asp Asn. Asn 65 70 7s 8O Val Ser Gly Ser Phe His Thr Val Ser Gly Gly His Asn Thr Val Ser 85 90 95 Gly Ser Asn Asn Thr Val Ser Gly Ser Asn His Val Val Ser Gly Ser 1OO 105 11 O Asn Llys Val Val Thr Asp Ala Ala Lys Lieu Ala Ala Ala Lieu. Glu. His 115 12 O 125

His His His His His 13 O

<210s, SEQ ID NO 9 &211s LENGTH: 133 212. TYPE: PRT <213> ORGANISM: Lolium perenne <4 OOs, SEQUENCE: 9 Met Asp Glu Gln Pro Asn. Thir Ile Ser Gly Ser Asn Asn Thr Val Arg 1. 5 1O 15 Ser Gly Ser Lys Asn Val Lieu Ala Gly Asn Asp Asn Thr Val Ile Ser 2O 25 3O Gly Asp Asn. Asn. Ser Val Ser Gly Ser Asn. Asn Thr Val Val Ser Gly 35 4 O 45 Asn Asp Asn Thr Val Thr Gly Ser Asn His Val Val Ser Gly Thr Asn SO 55 6 O His Ile Val Thr Asp Asn. Asn. Asn. Asn Val Ser Gly Asn Asp Asn. Asn 65 70 7s 8O Val Ser Gly Ser Phe His Thr Val Ser Gly Gly His Asn Thr Val Ser 85 90 95 Gly Ser Asn Asn Thr Val Ser Gly Ser Asn His Val Val Ser Gly Ser 1OO 105 11 O

Asn Llys Val Val Thr Asp Ala Ala Lys Lieu Ala Ala Ala Lieu. Glu. His 115 12 O 125

His His His His His 13 O

<210s, SEQ ID NO 10 &211s LENGTH: 334 212. TYPE: PRT <213s ORGANISM: Salmonella cholera 22 Os. FEATURE: <221 > NAMEAKEY: misc feature US 2017/005.8007 A1 Mar. 2, 2017 21

- Continued

<222s. LOCATION: (1) . . (1) <223> OTHER INFORMATION: Xaa can be any naturally occurring amino acid 22 Os. FEATURE: <221 > NAMEAKEY: misc feature <222s. LOCATION: (67) . . (67) <223> OTHER INFORMATION: Xaa can be any naturally occurring amino acid <4 OOs, SEQUENCE: 10 Xaa Thr Lys Tyr Arg Lieu Ser Glu Gly Pro Arg Ala Phe Thr Tyr Glin 1. 5 1O 15 Val Asp Gly Glu Lys Llys Ser Val Lieu. Lieu. Arg Glin Val Ile Ala Val 2O 25 3O Thir Asp Phe Asin Asp Wall Lys Ala Gly. Thir Ser Gly Gly Trp Val Asp 35 4 O 45 Ala Asp Asin Val Lieu. Ser Glin Glin Gly Asp Cys Trp Ile Tyr Asp Glu SO 55 6 O Asn Ala Xaa Ala Phe Ala Gly Thr Glu Ile Thr Gly Asn Ala Arg Ile 65 70 7s 8O Thr Glin Pro Cys Thr Lieu. Tyr Asn. Asn Val Arg Ile Gly Asp Asn. Wall 85 90 95 Trp Ile Asp Arg Ala Asp Ile Ser Asp Gly Ala Arg Ile Ser Asp Asn 1OO 105 11 O Val Thir Ile Glin Ser Ser Ser Val Arg Glu Glu. Cys Ala Ile Tyr Gly 115 12 O 125 Asp Ala Arg Val Lieu. ASn Gln Ser Glu Ile Lieu Ala Ile Glin Gly Lieu 13 O 135 14 O Thir His Glu. His Ala Glin Ile Lieu. Glin Ile Tyr Asp Arg Ala Thr Val 145 150 155 160 Asn His Ser Arg Ile Val His Glin Val Glin Lieu. Tyr Gly Asn Ala Thr 1.65 17O 17s Ile Thr His Ala Phe Ile Glu. His Arg Ala Glu Val Phe Asp Phe Ala 18O 185 19 O Lieu. Ile Glu Gly Asp Lys Asp Asn. Asn Val Trp Ile Cys Asp Cys Ala 195 2OO 2O5 Llys Val Tyr Gly His Ala Arg Val Ile Ala Gly Thr Glu Glu Asp Ala 21 O 215 22O Ile Pro Thr Lieu. Arg Tyr Ser Ser Glin Val Ala Glu. His Ala Lieu. Ile 225 23 O 235 24 O Glu Gly Asn. CyS Val Lieu Lys His His Val Lieu Val Gly Gly His Ala 245 250 255 Glu Val Arg Gly Gly Pro Ile Lieu. Lieu. Asp Asp Arg Val Lieu. Ile Glu 26 O 265 27 O Gly His Ala Cys Ile Glin Gly Glu Ile Lieu. Ile Glu Arg Glin Val Glu 27s 28O 285

Ile Ser Gly Arg Ala Ala Val Ile Ala Phe Asp Asp Asn. Thir Ile His 29 O 295 3 OO

Lieu. Arg Gly Pro Llys Val Ile Asin Gly Glu Asp Arg Ile Thir Arg Thr 3. OS 310 315 32O

Pro Leu Val Gly Ser Lieu Lleu. Glu. His His His His His His 3.25 330

<210s, SEQ ID NO 11 &211s LENGTH: 334 212. TYPE: PRT US 2017/005.8007 A1 Mar. 2, 2017 22

- Continued

<213s ORGANISM: Salmonella cholera 22 Os. FEATURE: <221 > NAMEAKEY: misc feature <222s. LOCATION: (1) . . (1) <223> OTHER INFORMATION: Xaa can be any naturally occurring amino acid 22 Os. FEATURE: <221 > NAMEAKEY: misc feature <222s. LOCATION: (67) . . (67) <223> OTHER INFORMATION: Xaa can be any naturally occurring amino acid <4 OOs, SEQUENCE: 11 Xaa Thr Lys Tyr Arg Lieu Ser Glu Gly Pro Arg Ala Phe Thr Tyr Glin 1. 5 1O 15 Val Asp Gly Glu Lys Llys Ser Val Lieu. Lieu. Arg Glin Val Ile Ala Val 2O 25 3O Thir Asp Phe Asin Asp Wall Lys Ala Gly. Thir Ser Gly Gly Trp Val Asp 35 4 O 45 Ala Asp Asin Val Lieu. Ser Glin Glin Gly Asp Cys Trp Ile Tyr Asp Glu SO 55 6 O Asn Ala Xaa Ala Phe Ala Gly Thr Glu Ile Thr Gly Asn Ala Arg Ile 65 70 7s 8O Thr Glin Pro Cys Thr Lieu. Tyr Asn. Asn Val Arg Ile Gly Asp Asn. Wall 85 90 95 Trp Ile Asp Arg Ala Asp Ile Ser Asp Gly Ala Arg Ile Ser Asp Asn 1OO 105 11 O Val Thir Ile Glin Ser Ser Ser Val Arg Glu Glu. Cys Ala Ile Tyr Gly 115 12 O 125 Asp Ala Arg Val Lieu. Asn Glin Ser Glu Ile Lieu Ala Ile Glin Gly Lieu 13 O 135 14 O Thir His Glu. His Ala Glin Ile Lieu. Glin Ile Tyr Asp Arg Ala Thr Val 145 150 155 160 Asn His Ser Arg Ile Val His Glin Val Glin Lieu. Tyr Gly Asn Ala Thr 1.65 17O 17s Ile Thr His Ala Phe Ile Glu. His Arg Ala Glu Val Phe Asp Phe Ala 18O 185 19 O Lieu. Ile Glu Gly Asp Lys Asp Asn. Asn Val Trp Ile Cys Asp Cys Ala 195 2OO 2O5 Llys Val Tyr Gly His Ala Arg Val Ile Ala Gly Thr Glu Glu Asp Ala 21 O 215 22O Ile Pro Thr Lieu. Arg Tyr Ser Ser Glin Val Ala Glu. His Ala Lieu. Ile 225 23 O 235 24 O Glu Gly Asn. CyS Val Lieu Lys His His Val Lieu Val Gly Gly His Ala 245 250 255 Glu Val Arg Gly Gly Pro Ile Lieu. Lieu. Asp Asp Arg Val Lieu. Ile Glu 26 O 265 27 O

Gly His Ala Cys Ile Glin Gly Glu Ile Lieu. Ile Glu Arg Glin Val Glu 27s 28O 285

Ile Ser Gly Arg Ala Ala Val Ile Ala Phe Asp Asp Asn. Thir Ile His 29 O 295 3 OO

Lieu. Arg Gly Pro Llys Val Ile Asin Gly Glu Asp Arg Ile Thir Arg Thr 3. OS 310 315 32O

Pro Leu Val Gly Ser Lieu Lleu. Glu. His His His His His His 3.25 330 US 2017/005.8007 A1 Mar. 2, 2017 23

- Continued

<210s, SEQ ID NO 12 &211s LENGTH: 121 212. TYPE: PRT <213> ORGANISM: Choristoneura fumiferana

<4 OOs, SEQUENCE: 12 Asp Gly Thr Cys Val Asn Thr Asn Ser Glin Ile Thr Ala Asn Ser Glin 1. 5 1O 15 Cys Val Lys Ser Thr Ala Thr Asn. Cys Tyr Ile Asp Asn. Ser Glin Lieu. 2O 25 3O Val Asp Thir Ser Ile Cys Thr Arg Ser Glin Tyr Ser Asp Ala Asn Val 35 4 O 45 Llys Llys Ser Val Thir Thr Asp Cys Asn. Ile Asp Llys Ser Glin Val Tyr SO 55 6 O Lieu. Thir Thr Cys Thr Gly Ser Glin Tyr Asn Gly Ile Tyr Ile Arg Ser 65 70 7s 8O Ser Thir Thr Thr Gly. Thir Ser Ile Ser Gly Pro Gly Cys Ser Ile Ser 85 90 95 Thr Cys Thr Ile Thr Arg Gly Val Ala Thr Pro Ala Ala Ala Cys Lys 1OO 105 11 O Ile Ser Gly Cys Ser Leu Ser Ala Met 115 12 O

<210s, SEQ ID NO 13 <211 LENGTH: 121 212. TYPE: PRT <213> ORGANISM: Choristoneura fumiferana

<4 OOs, SEQUENCE: 13 Asp Gly Thr Cys Val Asn Thr Asn Ser Glin Ile Thr Ala Asn Ser Glin 1. 5 1O 15 Cys Val Lys Ser Thr Ala Thr Asn. Cys Tyr Ile Asp Asn. Ser Glin Lieu. 2O 25 3O Val Asp Thir Ser Ile Cys Thr Arg Ser Glin Tyr Ser Asp Ala Asn Val 35 4 O 45 Llys Llys Ser Val Thir Thr Asp Cys Asn. Ile Asp Llys Ser Glin Val Tyr SO 55 6 O Lieu. Thir Thr Cys Thr Gly Ser Glin Tyr Asn Gly Ile Tyr Ile Arg Ser 65 70 7s 8O Ser Thir Thr Thr Gly. Thir Ser Ile Ser Gly Pro Gly Cys Ser Ile Ser 85 90 95 Thr Cys Thr Ile Thr Arg Gly Val Ala Thr Pro Ala Ala Ala Cys Lys 1OO 105 11 O Ile Ser Gly Cys Ser Leu Ser Ala Met 115 12 O

<210s, SEQ ID NO 14 &211s LENGTH: 121 212. TYPE: PRT <213> ORGANISM: Choristoneura fumiferana

<4 OOs, SEQUENCE: 14 Asp Gly Thr Cys Val Asn Thr Asn Ser Glin Ile Thr Ala Asn Ser Glin 1. 5 1O 15

Cys Val Lys Ser Thr Ala Thr Asn. Cys Tyr Ile Asp Asn. Ser Glin Lieu. 2O 25 3O US 2017/005.8007 A1 Mar. 2, 2017 24

- Continued

Wall Asp Thir Ser Ile Thr Arg Ser Glin Tyr Ser Asp Ala Asn Wall 35 4 O 45

Lys Ser Val Thr Thr Asp ASn Ile Asp Llys Ser Glin Wall SO 55 6 O

Lieu. Thir Thr Thir Gly Ser Glin Asn Ile Tyr Ile Arg Ser 65 70 8O

Ser Thir Thr Thr Gly. Thir Ser Ile Ser Gly Pro Gly Cys Ser Ile Ser 85 90 95

Thir Cys Thir Ile Thir Arg Gly Wall Ala Thr Pro Ala Ala Ala 105 11 O

Ile Ser Gly Ser Leu Ser Ala Met 115 12 O

SEO ID NO 15 LENGTH: 121 TYPE PRT ORGANISM: Choristoneura fumiferana

<4 OOs, SEQUENCE: 15

Asp Gly. Thir Cys Wall Asn Thir Asn. Ser Glin Ile Thir Ala Asn Ser Glin 1. 5 15

Wall Lys Ser Thir Ala Thir Asn Cys Ile Asp Asn Ser Gln Lieu. 25

Asp Thr Ser Ile Thr Arg Ser Glin Ser Asp Ala Asn Wall 35 4 O 45

Lys Ser Val Thr Thr Asp ASn Ile Asp Llys Ser Glin Wall SO 55 6 O

Lieu. Thir Thr Thir Gly Ser Glin Asn Ile Tyr Ile Arg Ser 65 70 8O

Ser Thir Thr Thr Gly. Thir Ser Ile Ser Gly Pro Gly Cys Ser Ile Ser 85 90 95

Thir Cys Thir Ile Thir Arg Gly Wall Ala Thr Pro Ala Ala Ala Cys 105 11 O

Ile Ser Gly Ser Leu Ser Ala Met 115 12 O

1. An amyloid fibril comprising a plurality of modified B 9. The amyloid fibril of claim 1, attached to a solid solenoid protein (mBSP) monomers. Support, a nanoparticle, a biological molecule, or a second 2. The amyloid fibril of claim 1, wherein the mBSP amyloid fibril. monomers are derived from an antifreeze protein. 10. The amyloid fibril of claim 9, wherein the nanopar 3. The amyloid fibril of claim 2, wherein the antifreeze ticles comprise a metal, a semiconductor material, a metal protein is a spruce budworm antifreeze protein. oxide, or a combination thereof. 4. The amyloid fibril of claim 3, wherein the mBSP has 11. The amyloid fibril of claim 9, wherein the biological the sequence shown in SEQ ID NO: 1. molecule is an enzyme. 5. The amyloid fibril of claim 1, wherein the antifreeze 12. A method of forming a nanomaterial, the method protein is a rye grass antifreeze protein. comprising: 6. The amyloid fibril of claim 5, wherein the mBSP has (a) coupling a plurality of nanoparticles with a scaffold the sequence shown in SEQ ID NO: 2. comprising at least one amyloid fibril comprising a 7. The amyloid fibril of any of claim 1, wherein the mBSP plurality of modified B solenoid protein (mBSP) mono is modified to remove an end cap that prevents amyloid mers; and aggregation. (b) fusing the nanoparticles to form the nanomaterial. 8. The amyloid fibril of claim 1 that is modified to include 13. The method of claim 12, further comprising the step at least one amino acid residue that promotes attachment of of attaching the scaffold to a solid Support prior to the step the fibril to a solid Support, a nanoparticle, a biological of contacting the plurality of nanoparticles with the mBSP molecule, or a second amyloid fibril. scaffold. US 2017/005.8007 A1 Mar. 2, 2017 25

14. The method of claim 12, wherein the scaffold is substantially removed prior to step (b). 15. The method of claim 12, wherein the scaffold is Substantially removed during step (b). 16. The method of claim 12, wherein the nanoparticles are crystalline. 17. The method of claim 12, wherein the nanoparticles comprise a semiconductor material, metal, a metal oxide, or a combination thereof. 18. The method of claim 12, wherein the nanomaterial is a nanowire. 19. The method of claim 12, wherein the mBSP mono mers are derived from an antifreeze protein. 20. The method of claim 19, wherein the antifreeze protein is a spruce budworm antifreeze protein. 21. The method of claim 20, wherein the mBSP has the sequence shown in SEQ ID NO: 1. 22. The method of claim 19, wherein the antifreeze protein is a rye grass antifreeze protein. 23. The method of claim 22, wherein the mBSP has the sequence shown in SEQ ID NO: 2. 24. A nanomaterial made by the method of claim 12. 25-35. (canceled)