Scholarly article on topic 'Protocol for sortase-mediated construction of DNA–protein hybrids and functional nanostructures'

Protocol for sortase-mediated construction of DNA–protein hybrids and functional nanostructures Academic research paper on "Chemical sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Methods
OECD Field of science
Keywords
{"DNA–protein hybrids/chimeras" / "Molecular self-assembly" / "Sortase coupling" / Bioconjugation / "DNA origami" / Site-directed}

Abstract of research paper on Chemical sciences, author of scientific article — Mounir A. Koussa, Marcos Sotomayor, Wesley P. Wong

Abstract Recent methods in DNA nanotechnology are enabling the creation of intricate nanostructures through the use of programmable, bottom-up self-assembly. However, structures consisting only of DNA are limited in their ability to act on other biomolecules. Proteins, on the other hand, perform a variety of functions on biological materials, but directed control of the self-assembly process remains a challenge. While DNA–protein hybrids have the potential to provide the best-of-both-worlds, they can be difficult to create as many of the conventional techniques for linking proteins to DNA render proteins dysfunctional. We present here a sortase-based protocol for covalently coupling proteins to DNA with minimal disturbance to protein function. To accomplish this we have developed a two-step process. First, a small synthetic peptide is bioorthogonally and covalently coupled to a DNA oligo using click chemistry. Next, the DNA–peptide chimera is covalently linked to a protein of interest under protein-compatible conditions using the enzyme sortase. Our protocol allows for the simple coupling and purification of a functional DNA–protein hybrid. We use this technique to form oligos bearing cadherin-23 and protocadherin-15 protein fragments. Upon incorporation into a linear M13 scaffold, these protein–DNA hybrids serve as the gate to a binary nanoswitch. The outlined protocol is reliable and modular, facilitating the construction of libraries of oligos and proteins that can be combined to form functional DNA–protein nanostructures. These structures will enable a new class of functional nanostructures, which could be used for therapeutic and industrial processes.

Academic research paper on topic "Protocol for sortase-mediated construction of DNA–protein hybrids and functional nanostructures"

ELSEVIER

Contents lists available at ScienceDirect

Methods

journal homepage: www.elsevier.com/locate/ymeth

Protocol for sortase-mediated construction of DNA-protein hybrids and functional nanostructures

CrossMark

Mounir A. Koussaa, Marcos Sotomayorb, Wesley P. Wong

c,d,e,*

a Program in Neuroscience, Department of Neurobiology, Harvard Medical School, Boston, MA, United States b Department of Chemistry and Biochemistry, The Ohio State University, Columbus, OH, United States cProgram in Cellular and Molecular Medicine, Boston Children's Hospital, Boston, MA, United States d Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA, United States e Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA, United States

ARTICLE INFO

Article history:

Available online 22 February 2014 Keywords:

DNA-protein hybrids/chimeras

Molecular self-assembly

Sortase coupling

Bioconjugation

DNA origami

Site-directed

ABSTRACT

Recent methods in DNA nanotechnology are enabling the creation of intricate nanostructures through the use of programmable, bottom-up self-assembly. However, structures consisting only of DNA are limited in their ability to act on other biomolecules. Proteins, on the other hand, perform a variety of functions on biological materials, but directed control of the self-assembly process remains a challenge. While DNAprotein hybrids have the potential to provide the best-of-both-worlds, they can be difficult to create as many of the conventional techniques for linking proteins to DNA render proteins dysfunctional. We present here a sortase-based protocol for covalently coupling proteins to DNA with minimal disturbance to protein function. To accomplish this we have developed a two-step process. First, a small synthetic peptide is bioorthogonally and covalently coupled to a DNA oligo using click chemistry. Next, the DNA-pep-tide chimera is covalently linked to a protein of interest under protein-compatible conditions using the enzyme sortase. Our protocol allows for the simple coupling and purification of a functional DNA-protein hybrid. We use this technique to form oligos bearing cadherin-23 and protocadherin-15 protein fragments. Upon incorporation into a linear M13 scaffold, these protein-DNA hybrids serve as the gate to a binary nanoswitch. The outlined protocol is reliable and modular, facilitating the construction of libraries of oligos and proteins that can be combined to form functional DNA-protein nanostructures. These structures will enable a new class of functional nanostructures, which could be used for therapeutic and industrial processes.

© 2014 Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://

creativecommons.org/licenses/by-nc-nd/3.0/).

1. Introduction

DNA is emerging as the platform of choice for the bottom-up self-assembly of intricate, biocompatible nanostructures. As techniques for the programmed patterning of these structures have advanced, it is now possible to create complex assemblies, ranging from two dimensional shapes to three-dimensional nano-robots [1-3]. Structures built using only DNA, although intricate, are limited in their ability to interact with biological substrates. Proteins, on the other hand, are the molecular machines of biology, and have a wide range of functions including chemically modifying, binding to, and applying forces to other molecules. However, to build intricate machines using DNA-protein hybrids, which harness the

* Corresponding author at: 3 Blackfan Circle, 3rd Floor, Boston, MA 02115, United States.

E-mail address: wong@idi.harvard.edu (W.P. Wong).

advantages of both materials, one needs to attach the proteins in a targetable and biocompatible way [4,5].

One approach is to attach the proteins to the oligonucleotides (oligos) that bind to designated locations on the DNA origami scaffold. The challenge, however, is to attach these proteins to oligos while preserving protein function. Existing chemistries, such as disulfide and SMCC (sulfosuccinimidyl-4-(N-maleimidometh-yl)cyclohexane-1-carboxylate) linkage [6], are sometimes effective, but they react with functional groups ubiquitous in biology. This reactivity hinders both the ability to link to a desired location on the protein, and to control the number of linkers bound to each molecule (stoichiometry), as one biomolecule may contain many reactive species. To overcome this issue, bioorthogonal techniques, such as copper-free click-chemistry, have been developed [7]. Although these techniques do in fact overcome the issues of regi-oselectivity and stoichiometry, they still suffer from the problem common to all of these techniques: reaction conditions are suboptimal for many proteins, be it due to long incubation periods at

http://dx.doi.org/10.1016/j.ymeth.2014.02.020 1046-2023/© 2014 Published by Elsevier Inc.

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

room temperature, extreme oxidizing/reducing conditions, or non-physiological pH. These methods also fail when working with non-thermostable proteins that have a tendency to aggregate and/ or precipitate out of solution under the requisite conditions. Additionally, purification can be difficult due to the challenge of easily distinguishing products from reactants—as we will see, our technique solves this problem by directly marking successful reactions by the simultaneous removal of an affinity label. Finally, each protein has specific considerations with regards to its reactivity and structural stability. Thus protocols optimized for one protein may not be suitable for another protein.

Here, we present a protocol for covalent coupling developed to meet the challenges outlined above. In order to preserve protein function, protein-DNA coupling is performed under physiological conditions. In this two-step process, a small synthetic peptide is first coupled to a DNA oligo. Next, utilizing the enzyme sortase [8,9], the protein is coupled to the DNA-peptide chimera under physiological conditions. This strategy frontloads all of the protein-incompatible chemistry so that it is performed on an oligo and a synthetic peptide, which are far more tolerant of non-physiological conditions. The use of sortase also facilitates purification, as coupling of the oligo is accompanied by the removal of an affinity tag. Our protocol, detailed below, allows for the use of commercially available purification resins to yield the product of interest free of any side products and reac-tants. This technique can be used to generate libraries of oligo nucleotides and proteins such that any proteins in the library can be easily and reliably attached anywhere along the DNA origami scaffold.

To demonstrate the utility of this technique we build the simplest DNA-protein nanomachine, a binary DNA-nanoswitch [6], gated by the interacting pair of cadherin-23 and protocadherin-15 protein fragments (CDH23 and PCDH15) [10,11]. This self-assembled mechanical switch changes state to report the formation or rupture of biomolecular bonds: in this case, the switch is closed when CDH23 is bound to PCDH15 and open otherwise. The CDH23 fragment is covalently linked to an oligo that hybridizes one-third of the way in from one edge of the DNA scaffold (M13). In addition, the PCDH15 fragment is linked to an oligo that hybridizes one-third of the way in from the other edge of the scaffold (Fig. 1). The result is a nanoswitch with an end-to-end length of 3 im when the proteins are not interacting, and an end-to-end length of 2 im when the proteins are interacting. These two states are resolvable via gel electrophoresis as described in [6]. In vivo some variants of sortase, such as sortase-A are used to anchor proteins to the cell surface by covalently linking proteins to the pepti-doglycan on the cell walls of bacteria. In this in vitro system sortase covalently links the N-terminus of one protein to a location near (within ~100 amino acids of) the C-terminus of another protein. Sortase recognizes an N-terminal GGG and a C-terminal LPX1TGX2, where X1 can be any amino acid, and X2 can be any string of amino acids of length 1-99. Sortase then facilitates the transposition of the glycine residues in the two proteins resulting in a covalent linkage between the two proteins and the release of GX2 (Fig. 2).

Maximillian Popp et al. in 2007 [9] first described using sortase to selectively attach fluorescent markers to a protein of interest. Chen et al. in 2011 [8] evolved a sortase variant with 140-fold increased activity, lowering coupling times from hours to minutes. Sortase has also been used to link peptide nucleic acids (PNA) to peptides [12], to label proteins N-terminally [13], C-terminally, and in loops [14]. Additionally sortase has been used in combination with click chemistry to make unnatural N-N- and C-C- linked protein chimeras [15].

Our goal was to use sortase technology to generate DNAprotein hybrids for self-assembled nanostructures. While synthetic PNA oligos offer the ability to append amino acids directly to a string of nucleic acids, they are less soluble than DNA oligos, and

Fig. 1. A method for the formation of the binary DNA-nanoswitch. Linearized M13 single-stranded DNA (dark blue), complimentary oligonucleotides (golden-rods), CDH23 EC1+2 (cyan), and PCDH15 EC1 + 2 (magenta). (A->B) Annealing of functionalized and non-functionalized oligo nucleotides to the M13 ssDNA scaffold (see Figs. 3-6 for functionalization of oligos). (B->C) Sortase is used to link an LPETG-containing CDH23 fragment (2.4.4) to the Gly-Gly-Gly-modified oligo (the other oligo is protected by having the N-terminus blocked by a Flag-TEV sequence) (2.1). (C->D) After successful coupling, the TEV protease is used to deprotect the second Gly-Gly-Gly-oligo, thereby priming it for sortase coupling (2.4.4 e). (D->E) sortase is then used again to attach an LPETG-containing PCDH15 fragment (2.4.4 k). (E<->F) Upon binding of CDH23 to PCDH15 the DNA-nanoswitch is closed. In the paper we also outline a method for attaching thermo-stable proteins to a DNA origami scaffold. For thermo-stable proteins, one can perform the sortase-based coupling reactions on the individual oligos separately prior to thermally annealing the oligos to the M13 scaffold, thus obviating the need for the protecting Flag-tag (2.3).

harder to synthesize. Thus, we wanted instead to develop a convenient way to couple proteins to DNA oligos which are more readily-available. To do this we needed a means of attaching a peptide to an oligo. Click chemistry was chosen to allow for the process to be both bioorthogonal and efficient.

2. Methods

Here we present four protocols describing: (2.1) the formation of a DNA-oligo bearing a sortase-compatible GGG-peptide, (2.2)

Fig. 2. (A). Sortase coupling schematic: The sortase enzyme catalyzes the formation of a covalent bond between two proteins by recognizing, and then coupling two specific peptide sequences. Note that the recognition sequences do not first align, but rather, sortase binds to the LPXiTGX2 sequence first, as described below. (B). Reaction Diagram: Sortase first recognizes a C-terminal LPX,TGX2 sequence, in which X, can be any one amino acid, and X2 can be any string of 1-99 amino acids. Sortase then transposes the N-terminal-most glycine from the Gly-Gly-Gly sequence with the glycine from the LPX,TGX2 sequence (red box), resulting in a peptide bond between the two proteins and the release of GX2. (G and T refer to the amino acids Glycine and Threonine respectively.)

the sortase-catalyzed coupling of a protein to the DNA-peptide chimera, and (2.3/2.4) the integration of DNA-protein hybrids into self-assembling nanostructures for thermostable/non-thermostable proteins. The oligos we were interested in functionalizing for our application were both 60 bp oligos herein referred to as oligo 1 and oligo 2. We ordered oligo 1 with a 3'-azide and oligo 2 was ordered with a 5'-azide (IDT custom oligo). The following peptide was synthesized by NeoBioLab: (N->C) Flag-TEV-GGG-Pra (DYKDDDDK-ENLYFQ-GGG-Pra) where Pra is the unnatural amino acid propargylglycine. This can be incorporated into a synthetic peptide to provide an alkyne, the complimentary click reagent. Additionally, to facilitate purification, a Flag-tag was added to the N-terminus. As the sortase requires the GGG to be on a free N-ter-minus, a tobacco etch virus cleavage site (TEV) was inserted to allow for removal of the Flag-tag.

2.1. Protocol for the formation of oligonucleotides with sortase-compatible GGG peptide

2.1.1. Preparation of reagents

d. Prepare a 264.2 g/L (0.21 M) aqueous ascorbic acid stock. Note: The ascorbic acid serves to reduce the Cu(II) (blue) to the catalytically active Cu(I) (green).

2.1.2. Click-coupling of the peptide to the oligo (Fig. 3)

a. Combine the following in a 250 iL DNA-lowbind tube: 35 iL of the azide-oligo, 30 iL of the peptide, 12 iL of the ascorbic acid, and 8.5 iL of CuSO4. Note: CO2(g) is produced from the ammonium bicarbonate. Also, although the peptide is insoluble at neutral pH, it is soluble under both the slightly basic ammonium bicarbonate conditions and the acidic ascorbic acid conditions.

b. Allow the reaction to sit for 2 h at room temperature to ensure completion.

c. Note: some of the Cu will be reduced to Cu(0) metal, which will precipitate out.

Fig. 3. Click coupling of peptide to oligo. The Flag-TEV-GGG-Pra (DYKDDDDK-ENLYFQ-GGG-Pra) peptide is attached to an azide oligo (golden-rod) through copper(I)-catalyzed click chemistry.

a. Solubilize the peptide to 1 mg/ml (0.5 mM) in nuclease-free water. Note: The propargylglycine reduces solubility of the peptide and a small amount of ammonium bicarbonate can be added to solubilize it.

b. Solubilize the oligo at 100 iM in nuclease-free water.

c. Prepare a 94.2 g/L (59 mM) aqueous CuSO4 stock. Note: Anhydrous CuSO4 should be used.

2.1.3. Purification of the peptide-oligo chimera

a. Removal of uncoupled peptide (select either Method 1 or Method 2) (Fig. 4).

Method 1: Neutralization

1. Neutralizing the solution via the addition of 250 iL of TBS (50 mM TrisHCl pH 7.6, 300 mM NaCl) will cause uncoupled peptide to precipitate.

2. The copper metal and precipitated peptide can be pelleted by centrifugation at 16,000g for 5 min. The supernatant will contain coupled and uncoupled oligo and excess Cu(I) can be dialyzed out using a 6-8 kDa membrane (Mini GeBA-flex-tube, T070-6).

Method 2: QiaQuick nucleotide removal kit (Qiagen) (The Qia-gen kit works by precipitating out DNA, passing it over a membrane, and washing away any soluble products such as unconjugated peptides, small molecules, soluble copper etc.).

1. Following the kit protocol will remove the copper metal, Cu(I), Cu(II), and the uncoupled peptide.

2. The protocol should be followed as instructed by the manufacturer, but the wash step should be repeated a second time.

3. Perform the elution with 200 iL of TBS.

2.1.4. Removal of uncoupled oligo (Fig. 5)

a. Wash 1 ml of Anti-Flag M2 magnetic beads (Sigma-Aldrich, M8823) three times with 1 ml of TBS.

b. Apply the product of the Qiagen purification column and allow to bind for 1-2h rotating at room temperature.

c. Wash the beads at least 4 times with 500 il of TBS being sure to agitate the beads to remove any uncoupled oligo. Note: A wide boar pipette is recommended when agitating the beads.

d. Elute with 1 ml of 0.1 mg/ml (Sigma) Flag peptide in eTBS (50 mM TrisHCl pH 7.6, 150 mM NaCl). Allow 1 h rotating at room temperature for elution.

e. A second elution can be performed, but >85% will be recovered in the first elution.

2.1.5. TEV-cleavage of the Flag-tag (Fig. 6)

a. Add 2 iL of 2 mg/ml TEV protease (Sigma-Aldrich, T4455) to each ml of eluted product.

b. Incubate in a 30 oC water bath over night.

c. Running a 4-20% gradient poly-acrylamide gel with the binding, wash, and elution (cleaved and uncleaved) superna-tants reveals that the product has been successfully purified of uncoupled oligo (The 60 bp band is eliminated with successive washes) and cleaved (after cleavage the band shifts back near 60 bp).

The final product will herein be referred to as GGG-oligo. Additionally the TEV and cleaved peptide can be removed by repeating the Qiagen nucleotide removal kit as described above. This step, however, has not proven to be necessary.

2.2. Protocol for sortase coupling LPETG-tagged proteins to GGG-oligonucleotides

2.2.1. Preparation of reagents

a. The GGG-oligo should be at a concentration of iM as judged by band intensity on the Polyacrylamide gel. The gel was stained with SYBR-Gold (Invitrogen, S11494), imaged using a GE Typhoon FLA-9500, and analyzed using ImageJ (NIH, 1.46r).

b. Two proteins, protein1 and 2, are coupled to two different oligos, oligo1 and 2. In this example protein 1 is a fragment of the CDH23 protein containing two (out of 27) extracellular repeats. Likewise, protein 2 is a PCDH15 fragment containing two (out of 11) extracellular repeats. Both proteins were modified by appending the LPETG amino-acid sequence between the protein fragment C-terminus and a His-tag. Protein fragments were produced as described in [11,16].

c. Proteins should be used at a minimum concentration of 0.1 mM. CDH23-LPETG and PCDH15-LPETG stocks were at 2.5 and 2.7 mg/ml respectively in TBS + 5 mM CaCl2 (~0.1 mM). CDH23-LPETG and PCDH15-LPETG were fragments, each containing two extracellular domains used in a crystallographic study [11]. The proteins were modified by appending LPETG between the His-tag and the C-termi-nus of the protein.

d. Sortase stock was at 1.5 mg/ml in TBS + 10%-glycerol. An evolved variant of sortase-A from Staphylococcus aureus [8] was expressed and purified from Escherichia coli (plasmids and expression protocol provided by Brent Dorr in David Liu's lab [8]).

e. Sortase Reaction buffer consisted of the following: 300 mM TrisHCl pH 7.5, 5 mM MgCl2, 5 mM CaCl2, and 150 mM NaCl.

Flag-TEV-GGG e © :n=N=N^ C=CH Cu(l) Flag-TEV-GGG- C=CH + Cu(l) Flag-TEV-GGG V-N T> \ ' 1 Mill

Flag-TEV-GGGx T 1 1 II 1 1 -A, rr 0 © :n=n=n

N II 1 1 1 1 1

Qiagen Nucleotide Removal Kit Qiagen Nucleotide Removal Kit

Fig. 4. Removal of Cu and uncoupled peptide. Running the reaction product through a Qiagen nucleotide removal kit allows for the isolation of the coupled and uncoupled oligo. Excess peptide and copper will remain on the membrane and/or be removed in the wash steps.

Fig. 5. Removal of uncoupled oligo. Binding the product of the Qiagen nucleotide removal kit to Anti-Flag beads allows one to wash away uncoupled oligo. Elution with Flag-peptide releases the oligo from the affinity matrix.

TEV Protease

Flag-TEV-GGG,,

Purification Cieavag

W1 W2 W3 W4 W5 W6

Fig. 6. TEV cleavage of Flag-tag. The TEV protease (brown) can be applied to cleave the Flag-tag resulting in an oligo with a sortase-compatible GGG-peptide. The left-most lane is a Bio-Rad 20 bp Molecular Ruler.

2.2.2. Sortase coupling of protein1 to oligo1 (Fig. 7)

a. Mix the following in a 250 iL mini GEBAflex-tube: 140 iL of 2 iM GGG-oligol, 40 iL of 0.1 mM proteinl-LPETG-HHHHHH, 5 iL of sortase, 65 iL of sortase reaction Buffer. Note: The protein is added in large excess to drive coupling to completion with respect to the oligo. Place the GEBAflex tube in 1L of sortase reaction buffer and allow the reaction to go for 1hr at RT or 4-5 h at 4 °C. The dialysis column will allow any Flag peptide to dialyze out and will remove the sortase reaction byproduct, G-HHHHHH, which can compete with the oligo.

b. Fig. 8 shows SDS-PAGE analysis indicating that the protein oligo chimera is only formed when all components are present.

2.2.3. Purification of protein-oligo chimera (Fig. 9)

b. If your protein buffer contains Ca2+, wash 1 ml of magnetic Anti-His beads (GenScript, L00275) 2 times with 1 ml of TBS. If, however, your protein buffer does not contain Ca2+ PBS can be used for these washes and for those in step c. This is to remove any phosphate from the storage solution to prevent calcium-phosphate crystal formation. Omitting this step will result in large losses in later steps.

c. Wash 3 more times with 1 ml of TBS + 5 mM CaCl2.

d. Apply the product of the sortase reaction and allow 2 h rotating at 4 °C for binding.

e. The supernatant contains the DNA-protein hybrid, free of any other proteins. In our molecular system we obtain a yield of ~65% with respect to the amount of starting oligo.

2.3. Protocol for hybridization of DNA-protein hybrid to scaffold (thermostable proteins)

a. The TEV, sortase, protein 1, and protein 2 all bear His tags. The sortase reaction, however, selectively cleaves the His tag off of the final product. Thus passing the product over anti-His magnetic beads will remove these reactants leaving the final product in the supernatant. Note: anti-His beads should be used rather than Ni-NTA. The Ni-NTA beads tend to bind the oligos quite strongly and very high salt is required to remove them.

If the protein used can withstand being heated to 40 oC, the oligos can be hybridized to the scaffold, in this case linearized M13 [6], by adding the oligos in a one-to-one ratio to the scaffold, then ramping the temperature from 40 °C to 20 °C at half a degree per minute in a thermocycler [6].

One can anneal all unfunctionalized oligos from 95 °C in 0.5 degree steps to 20 °C. The functionalized oligos can be added during this run by pausing the thermocycler once it hits 40 °C, adding the

Fig. 7. Sortase-catalyzed production of protein-DNA hybrids. An LPETG-containing protein, in this case CDH23 EC1 + 2, is coupled to the Gly-Gly-Gly-oligo via sortase. Sortase transposes the two glycine residues highlighted by the red box, resulting in the formation of a peptide bond between the protein and the Gly-Gly-Gly-oligo (we note that the recognition sequences do not first align, but rather, sortase binds to the LPX1TGX2 sequence first, as described in Fig. 2(B)). In this case the sequence after the C-terminal-most glycine on the protein was a 6xHis-tag. Sortase cleaves and releases this affinity tag, replacing it with the Gly-Gly-Gly-oligo. The procedure is carried out in a dialysis membrane so the Flag-TEV and G-His-tag peptides are dialyzed out.

Sortase + + + - - + + - + + +

CDH23 + - - + - + - + - + -

PCDH15 - + - - + - + - - - +

Oligol + - + + + -

Oligo2 - + - - + - - - - - +

Ca2VMg2* - - + + + + + + + + +

75KDaI | 25KD3, J 15KDa J b 0 3 •¡Oligo2-PCDH15 Oligol-CDH23 TEV Protease >PCDH15 ,CDH23

1 a* ■01

Fig. 8. SDS-PAGE verification of protein-DNA coupling. Successful coupling requires an LPETG-containing protein, a GGG-oligo, the sortase enzyme, and divalent cations (primarily Ca2+). Only when all the requisite constituents are present (green box), do coupled products appear.

functionalized oligos. If the protein is not thermostable an alternate approach can be taken as described below in Section 2.4.

2.4. Protocol for hybridization of DNA-protein hybrid to scaffold (non-thermostable proteins)

In this example the CDH23 and PCDH15 fragments are not very thermostable and hybridization through temperature annealing was not an option. For this system the GGG-oligos were hybridized onto the scaffold and the sortase coupling was done in situ directly on the scaffold. Performing the coupling on the oligos before hybridization allows one to easily control which protein is attached to which oligo. For this system selective coupling was achieved using the Flag-tag as a protecting group. That is, oligo 1 was processed fully, resulting in a GGG-oligo, while oligo 2 did not undergo TEV cleavage of its Flag tag (Fig. 1).

An additional concern is ensuring that each site on the scaffold receives its complimentary oligo. To accomplish this, the oligos are added at 50-fold excess. This, however, results in a large surplus of

free floating oligos. This can be a problem if there is an excess of GGG-oligo floating around which will compete with the in situ reaction. To overcome this issue excess oligos had to be removed from the solution.

2.4.1. Preparation of reagents

a. The GGG- and Flag-TEV-GGG- oligos should be concentrated to ~10 iM. This can be achieved by using a speedvac (Thermo-Savant, SC210A) or a 3 kDa spin column (Vivaspin 500, VS0191). Note: If a speedvac is used the oligos should first be dialyzed into water to remove salts before concentration.

2.4.2. Annealing oligos

a. Mix the following in a low-bind 250 iL PCR tube: 5 iL of 20 nM origami scaffold (linear M13 in this case), 1.19 iL of the unfunctionalized oligo mixture (consisting of equal volumes of the 119 unfunctionalized oligos stocks, each at a starting concentration of 100 iM—see reference [6]), 0.5 iL of 10 nM GGG-oligo 1, and 0.5 iL of 10 nM Flag-TEV-GGG-oligo 2.

b. Subject the mixture to a temperature ramp from 95 °C to 20 °C at 0.5 degree increments to anneal the oligos to the scaffold.

2.4.3. Removal of excess oligos by PEG-precipitation (modified from 1171)

a. Dilute the product of the annealing step in 115 iL of 4%, by weight, 8 K PEG (Amresco, 0159) in 30 mM MgCl2.

b. Mix thoroughly.

c. Centrifuge at 16,000g for 30 min at 25 °C.

d. Remove the top 112 iL leaving the bottom 10 iL which should contain the precipitated scaffold.

e. Dilute the remaining 10 iL with another 115 iL of 4%, by weight, 8 K PEG (Amresco, 0159) in 30 mM MgCl2. Note: Be sure to mix thoroughly.

f. Centrifuge at 16,000g for 30 min at 25 °C.

g. Remove the top 115 iL of supernatant.

Anti-His Beads

Fig. 9. Removal of reactants and catalysts. Upon completion of coupling the product is amidst TEV, sortase, and excess protein. While all of these components initially bear a His-tag, sortase cleaves the His-tag when coupling the protein and oligo. Thus, by running the sample over anti-His beads, everything but the product of interest will bind to the affinity matrix leaving the DNA-protein hybrid in the supernatant.

h. The remaining 10 iL should have the scaffold free of any detectable amount of unhybridized oligo.

2.4.4. In situ coupling of protein 1-LPETG-HHHHHH to the GGG-Oligo

a. Mix the following, 40 iL 1 M Tris HCl pH 7.5, 0.8 iL 1 M CaCl2, 8 iL 3 M NaCl, 10 iL of PEG-precipitated scaffold, 50 iL of 14 mg/ml sortase, and 15 iL of 0.1 mM protein1-LPETG-HHHHHH.

b. Place the mixture into a dialysis membrane (Spectra/Por MicroFloat-a-lyzer, F235053).

c. Place the Floatalyzer in 1 L of sortase reaction buffer.

d. Allow this reaction to run for 0.5-1 h at RT before moving to 4 °C for an additional 2 h (upon transferring to 4 °C it is best to transfer to a pre-chilled liter of sortase reaction buffer).

e. Add 4 iL of 2 mg/ml TEV and allow to sit at room temperature for 1 h.

f. If your protein buffer contains Ca2+ wash 1 ml of magnetic Anti-His beads (GenScript, L00275) 2 times with 1 ml of TBS. If, however, your protein buffer does not contain Ca2+ PBS can be used for these washes and the washes in step g. This is to remove any phosphate from the storage solution to prevent calcium-phosphate crystal formation. Omitting this step will result in large losses in later steps.

g. Wash 3 more times with 1 ml of TBS + 5 mM CaCl2.

h. Apply the product of the sortase reaction and allow 2 h rotating at 4 °C for binding.

i. The supernatant contains the DNA-protein hybrid, free of TEV and excess CDH23.

To this supernatant, add the following and place in a new Floatalyzer: 15 iL of 0.1 mM protein 2-LPETG-HHHHHH, and 50 iL of 14 mg/ml sortase.

j. Repeat steps b, c, d, f, g, and h.

k. The supernatant contains the pure site-directedly bi-functionalized DNA-protein hybrid.

l. Functionality of the nanoswitch can be assayed by gel electrophoresis as previously described in [6].

3. Conclusions

We have presented detailed protocols for a reliable approach for linking proteins to DNA-oligos, while preserving protein function.

Additionally we outline methods for the incorporation of these chimeras into self-assembling nanostructures. These techniques frontload all harsh chemistries to synthetic oligos and peptides, which are more amenable to these non-physiological conditions. The use of click chemistry ensures that linkages are bioorthogonal, site directed, and efficient. The use of an evolved sortase allows for protein coupling to occur under conditions favorable for protein stability. The protocols have been designed to be resilient to changes in the protein of interest. All materials are commercially available, with the exception of the sortase (for which the plasmid is available from the Liu Lab [8]). The built in purification schemes allow for fast and efficient purification, allowing for the immediate use of the chimeric product. When combined with a library of sor-tase compatible oligos and peptides, this flexible and modular approach could enable the creation of a wide range of functional nanostructures on demand. Furthermore, our approach expands the range of functional DNA-protein chimeras that can be constructed, enabling the incorporation of previously inaccessible protein machinery to generate nanostructures with previously unobtainable functionalities.

Acknowledgments

The authors gratefully acknowledge the help of Brent Dorr, for providing the plasmid for the evolved sortase, Rachelle Gaudet, for allowing us to use her facilities for protein production and purification; David P. Corey, Gary Yellen, Rachel Wilson, Jeffrey Holt, Hidde Ploegh, Ahmed Badran, Zhi-Yang Tsun, and members of the Wong and Corey Labs for critical discussions; The labs of Sun Hur and Timothy Springer, for assistance in protein production; Lynne Oland, for scientific training, discussions, and instruction on the proper use of hyphens. Additionally, the authors thank M. Plugh Humde-Ha. Funding for this project was provided by NIH R01 DC02281 to D.P.C and R.G.; NIH K99/R00 DC012534-01 to M.S.; M.K. was supported by NSF GRFP 2012147612; W.W. was supported by BCH startup funds.

References

[1] P.W.K. Rothemund, Nature 440 (2006) 297-302.

[2] S.M. Douglas, I. Bachelet, G.M. Church, Nature 335 (2012) 831-834.

[3] V. Linko, H. Dietz, Curr. Opin. Biotechnol. 24 (2013) 555-561.

[4] C.M. Niemeyer, Trends Biotechnol. 20 (2002) 395-401.

[5] C.M. Niemeyer, B. Sacca, Chem. Soc. Rev. 40 (2011) 5910-5921.

[6] K. Halvorsen, D. Schaak, W. Wong, Nanotechnology 22 (2011) 494005-494012.

[7] C.G. Gordon, J.L. Mackey, J.C. Jewett, E.M. Sletten, K.N. Houk, C.R. Bertozzi, J. Am. Chem. Soc. 134 (2012) 9199-9208.

[8] I. Chen, B.M. Dorr, D.R. Liu, Acad. Sci. USA 108 (2011) 11399-11404.

[9] M. Popp, J. Antos, G. Grotenbreg, E. Spooner, H. Ploegh, Nat. Chem. Biol. 3 (2007) 707-708.

[10] P. Kazmierczak, H. Sakaguchi, J. Tokita, E. Wilson-Kubalek, R. Milligan, U. Müller, B. Kachar, Nature 449 (2007) 87-91.

[11] M. Sotomayor, W.A. Weihofen, R. Gaudet, D.P. Corey, Nature 492 (2012) 128132.

[12] S. Pritz, Y. Wolf, O. Kraetke, J. Klose, M. Bienert, M. Beyermann, J. Org. Chem. 72 (2007) 3909-3912.

[13] C. Theile, M. Witte, A. Blom, L. Kundrat, H. Ploegh, C. Guimaraes, Nature Protocols 8 (2013) 1800-1807.

[14] C. Guimaraes, M. Witte, C. Theile, G. Bozkurt, L. Kundrat, A. Blom, H. Ploegh, Nature Protocols 8 (2013) 1787-1799.

[15] M. Witte, C. Theile, T. Wu, C. Guimaraes, A. Blom, H. Ploegh, Nature Protocols 8 (2013) 1808-1819.

[16] M. Sotomayor, W.A. Weihofen, R. Gaudet, D.P. Corey, Neuron 66 (2010) 85-100.

[17] J.L. Hartley, H. Bowen, Focus 66 (1996) 27-28.