WO2015144719A1

WO2015144719A1 - A method of preparing a library of biosensors

Info

Publication number: WO2015144719A1
Application number: PCT/EP2015/056282
Authority: WO
Inventors: Laurens H. LINDENBURG; Maarten Merkx
Original assignee: Technische Universiteit Eindhoven
Priority date: 2014-03-24
Filing date: 2015-03-24
Publication date: 2015-10-01

Abstract

The present invention relates to methods for developing a library of fluorescent antibody sensors for efficient epitope mapping. Specifically, the present invention relates to a method of preparing a library of biosensors, each biosensor comprising at least two epitopes and a linker separating the at least two epitopes. Also disclosed are a library of biosensors, and uses thereof, including methods of identifying an antibody.

Description

A method of preparing a library of biosensors

Field of the invention This invention relates to methods for developing a library of fluorescent antibody sensors for efficient epitope mapping. Specifically, the present invention relates to a library of multivalent antibody binding proteins with antibody binding epitopes that vary between molecules yet are identical within the same molecule. Background to the Invention

Antibodies bind antigens at relatively short (typically 6 residue) recognition sites that are either "continuous", i.e. having a short antigen primary structure or non-continuous, i.e. a particular fold in the antigen's tertiary structure. These recognition sites are known as epitopes. There is much interest in methods that can establish the identity of epitopes ("epitope mapping") for applications such as antibody detection, vaccine development and the use of antibodies as drugs. Furthermore, identification of the epitope allows further development, such as establishing the shortest antigenic version of the epitope and development of synthetic higher affinity versions of the epitope, or mimotopes, in a process known as "analoging". Mimotopes could also be used as convenient, genetically encodable alternatives to carbohydrate-based antigens. The availability of high affinity epitope sequences is desirable for the construction of sensitive probes that can detect the presence of antibodies for diagnostic purposes. For example, in the diagnosis of blood group types, mimotopes might one day replace the expensive, purified red blood cells in current use. For vaccine development, it is very important that the precise epitope sequences are known. For example, the recent discovery that a so-called broadly neutralizing antibody recognized a short stretch of highly conserved hydrophobic amino acids on an HIV envelope protein (gp41 ) allowed animal immunization studies to focus on this particular antigen as a promising means to develop an HIV vaccine.

Mimotopes based on tumor-derived epitopes recognized by T cell receptors (TCRs) may also be exploited therapeutically, to stimulate anti-tumor T cells, generating tumor protective activities. Antibodies are increasingly used as cancer treatment but binding of the antibody at sites other than the tumor may cause severe side effects. To improve the therapeutic index of such antibody treatments, a masking strategy was recently devised. This masking is achieved through use of mimotopes that bind the antibody's paratopes with high affinity until such time as some other, site- specific event (e.g. protease activity), causes the mimotopes to be released and the antibody to become available for binding. A sufficiently high affinity of the synthetic mimotopes is critical for the success of this latter strategy.

Current epitope discovery strategies can be broadly grouped into one of two categories. In the first, large, completely randomized libraries of potential epitope sequences are screened for binding to a surface-immobilized antibody, using techniques such as phage display and ribosome display. The second strategy involves screening partially overlapping, 6 to 20-mer synthetic peptides covering a known antigenic protein. In this case, the peptides are each immobilized in a separate well of a well- plate or on beads on a chip and tested for antibody binding in an ELISA-like format. Although often yielding useful results, the need to individually synthesize and purify each peptide is both labor- intensive and expensive. Avidity effects likely aid in the recovery of useful binders in both phage display and peptide fragment mapping. In the latter, this increase in affinity is caused by the increased effective concentration of the second paratope for a mimotope once the first paratope has bound, an effect that can be modeled and is well-understood. In phage display, the avidity effect results from the fact that each M13 phage particle displays 5 copies of variable mimotope, allowing each phage to bind multiple paratopes, as long as target antibody is sufficiently densely coated on the well surface. However, the tendency for a subset of parasitic phages to outgrow useful binders complicates phage display. Furthermore, complicated strategies are typically required to allow correct surface immobilization of antibodies. Also, ELISA-like screening assays tend to be highly laborious due to multiple wash steps and limited throughput. A solution-based, bivalent screening assay for novel genetically encoded epitopes would be highly desirable.

A group at the Technical University of Eindhoven previously developed CerAbsCit (previously referred to as Absense3), a single-chain fluorescent protein-based sensor that detects antibodies. This sensor has two epitopes connected by a semi-rigid peptide linker and flanked by fluorescent proteins Citrine and Cerulean, both bearing hydrophobic surface mutations Q204F and V224L. In the absence of antibody, the fluorescent domains form an intramolecular dimeric complex and consequently undergo a high degree of Forster Resonance Energy Transfer (FRET). Binding of antibody to both epitopes disrupts that complex, resulting in a readily detectable decrease in FRET efficiency (FIG. 1A).

Introduction of epitopes with sequence WEKIRLR resulted in a sensor that could detect an antibody against HIV's P17 protein. The modularity of the sensor's design allowed straightforward exchange of both the reporter domains and epitopes, as recently demonstrated when the fluorescent proteins were replaced with an enzyme/inhibitor complex and P17 epitopes were exchanged for

hemagglutinin and Dengue fever related epitopes.

As a genetically encoded sensor, CerAbsCit is amenable to directed evolution and screening sensor variants for a ratiometric response would offer a direct method to develop antibody sensors for any antibody for which high affinity, specific epitopes are not yet available (FIG. 1 B).

There are well-established techniques to introduce variation in DNA sequence, including approaches such as error prone PCR or use of NNK codons, making the construction of DNA libraries straightforward in most cases. However, the bivalent nature of the CerAbsCit-antibody interaction represents a unique challenge in DNA library construction: the variation is to be identical at both epitopes within the single-chain sensor. Summary of the invention

The present invention provides a cloning technique that allows the introduction of identical variation at the two separate epitope sites of the DNA encoding the antibody sensor. The resulting library was validated through sequencing and by screening library members for antibody binding in a 96-well plate based assay.

The availability of short linear mimotope (synthetic epitope) peptide sequences that bind antibodies with high affinity is important in diagnostic tests, for the development of vaccines and in therapeutic applications. The invention pertains to the development of a library of genetically encoded FRET- based antibody sensors to discover novel epitope sequences in a solution-based assay. First, a technique was developed that allowed creation of a DNA library encoding a randomly varied, yet perfectly repeated epitope sequence at two separate sites on the FRET sensor. This was achieved by using rolling circle amplification to create a tandem repeat of a circularized template. As an exemplary embodiment of this method, a library was created in which each 7-residue epitope was mutated at a single position. This library was validated through sequencing and expression of single clones in a 96-well plate scale. One way to screen large libraries of this FRET sensor would be to express clones in microdroplets and use microfluidic manipulations. Alternatively, large libraries of this FRET sensor could be screened by displaying the FRET sensor on a cell surface and screen sensors with improved epitopes using Fluorescence-activated cell sorting (FACS). Embodiments of this invention pave the way for the high throughput screening of sensors with improved epitopes.

Accordingly, the present invention provides a method of preparing a library of biosensors, each biosensor comprising at least two epitopes and a linker separating the at least two epitopes; the method comprising the steps of:

(a) providing a linear polymeric oligonucleotide; each oligonucleotide comprising: a nucleic acid sequence encoding at least one epitope and having at least one mutation; and at least one restriction site; amplifying the linear polymeric oligonucleotide by polymerase chain reaction (PCR) to provide multiple copies of the linear polymeric oligonucleotide; introducing the linear polymeric oligonucleotide into a vector comprising at least one marker; and

(d) further introducing a nucleic acid sequence encoding the linker separating the at least two epitopes into the linear polymeric oligonucleotide to provide a library of vectors, each vector encoding a biosensor comprising at least two epitopes and a linker separating the at least two epitopes. By "polymeric oligonucleotide" is meant a nucleic acid molecule comprising at least two monomeric oligonucleotides. For example, a monomeric oligonucleotide is a nucleic acid molecule comprising a given nucleic acid sequence. A polymeric oligonucleotide is a nucleic acid molecule comprising at least two monomeric oligonucleotides, wherein the monomeric oligonucleotides may be the same or different monomeric oligonucleotides. A dimeric oligonucleotide is a nucleic acid molecule comprising two monomeric oligonucleotides, wherein the monomeric oligonucleotides may be the same or different monomeric oligonucleotides. Optionally, the dimeric oligonucleotide comprises two monomeric oligonucleotides, wherein the monomeric oligonucleotides are the same monomeric oligonucleotide, also referred to as a "tandem repeat", "single tandem repeat", and/or "tandem epitope repeat".

Optionally, the providing step comprises providing a linear dimeric oligonucleotide comprising:

(i) a nucleic acid sequence encoding two epitopes, each epitope having at least one mutation; and

(ii) at least one restriction site.

Optionally, each epitope has at least one mutation, wherein each mutation can be the same or different mutations. Optionally, each mutation is the same mutation.

Optionally, the at least one restriction site is located between the two epitopes having at least one mutation.

Optionally, the introducing step comprises introducing the linear polymeric oligonucleotide into a vector comprising two markers. Further optionally, the introducing step comprises introducing the linear polymeric oligonucleotide into a location between the two markers within the vector.

Optionally, the further introducing step comprises introducing the nucleic acid sequence encoding the linker into a location between the two epitopes of the linear polymeric oligonucleotide within the vector.

Optionally, the nucleic acid sequence encoding the linker comprises the nucleic acid sequence 5'- GGTGGAAGTGGGGGCAGCGGAGGGGGCTCCGGCGGTAGCGGTGCGGAGGCTGCAGCGAAA GAAGCTGCAGCGAAGGAAGCTGCAGCGAAAGAGGCTGCAGCGAAGGAGGCTGCAGCGAAAGA AGCTGCAGCGAAGGCGGGATCCGGCGGTAGCGGTGGATCCGGTGGCCCGCAGGGTAT TCTGGGCCAGGGATCCGGTGGTTCTGGCGGTAGTGGTGGTTCTGGTGCAGAAGCCGCGGC GAAGGAGGCCGCGGCGAAAGAGGCCGCGGCGAAAGAAGCCGCGGCGAAGGAGGCCGCGGC GAAGGAAGCCGCGGCGAAAGCAGGGTCGGGCGGTAGCGGTGGCTCGGGCGGGAGCGGCGG CTCCGGCGGTTCCGGAGGTGAG-3' . Optionally, the nucleic acid sequence encoding the linker comprises the nucleic acid sequence defined in SEQ ID NO: 1.

According to a first aspect of the present invention, there is therefore provided a method of preparing a library of biosensors, each biosensor comprising at least two epitopes and a linker separating the at least two epitopes; the method comprising the steps of:

(a) providing a linear monomeric template oligonucleotide comprising: a nucleic acid sequence encoding at least one epitope and having at least one mutation; and at least one restriction site;

(b) ligating the linear monomeric template oligonucleotide to form a circular template

oligonucleotide; (c) subjecting the circular template oligonucleotide to rolling circle amplification (RCA) in the presence of a first primer and a second primer to provide a linear polymeric oligonucleotide; wherein the nucleic acid sequence of each of the first and second primers comprises at least one restriction site; (d) isolating the linear polymeric oligonucleotide;

(e) amplifying the linear polymeric oligonucleotide by polymerase chain reaction (PCR) in the presence of a third primer and a fourth primer to provide multiple copies of the linear polymeric oligonucleotide;

(f) introducing the linear polymeric oligonucleotide into a vector comprising at least one marker; and

(g) further introducing a nucleic acid sequence encoding linker separating the at least two

epitopes into the linear polymeric oligonucleotide to provide a library of vectors, each vector encoding a biosensor comprising at least two epitopes and a linker separating the at least two epitopes.

Optionally, the method further comprises the step of (h) expressing each of the vectors to provide a library of biosensors, each biosensor comprising at least two epitopes and a linker separating the at least two epitopes. Optionally, the expressing step comprises expressing each of the vectors in a bacterium. Optionally, the bacterium is E. coli. Optionally, the bacterium is selected from E. coli BL21 (DE3); E. coli NovaBlue; and E. coli DH5G. Optionally, the method comprises the additional step of purifying the linear polymeric oligonucleotide prior to the amplifying step. Optionally, the additional purifying step comprises contacting the linear polymeric oligonucleotide from the subjecting step with at least one restriction enzyme capable of recognizing the at least one restriction site of the nucleic acid sequence of each of the first and second primers. Further optionally, the additional purifying step comprises contacting the linear polymeric oligonucleotide with the restriction enzyme capable of recognizing the at least one restriction site of the nucleic acid sequence of the first primer; and with the restriction enzyme capable of recognizing the at least one restriction site of the nucleic acid sequence of the second primer. Optionally, the additional purifying step comprises contacting the linear polymeric oligonucleotide with at least one restriction enzyme selected from Spel (Sphaerotilus natans) and Sacl (Streptomyces achromogenes). Further optionally, the additional purifying step comprises contacting the linear polymeric oligonucleotide with Spel (Sphaerotilus natans) and Sacl

(Streptomyces achromogenes).

Optionally, the method comprises the additional step of heating the linear polymeric oligonucleotide prior to the isolating step. Optionally, the additional heating step comprises heating the linear polymeric oligonucleotide to 95°C. Further optionally, the additional heating step comprises heating the linear polymeric oligonucleotide to 95°C for 5 minutes. Optionally or additionally, the additional heating step comprises cooling the heated linear polymeric oligonucleotide. Further optionally or additionally, the additional heating step comprises cooling the heated linear polymeric

oligonucleotide to room temperature. Still further optionally or additionally, the additional heating step comprises cooling the heated linear polymeric oligonucleotide to 16°C to 26°C. Still further optionally or additionally, the additional heating step comprises cooling the heated linear polymeric oligonucleotide to 20°C. Optionally or additionally, the additional heating step comprises cooling the heated linear polymeric oligonucleotide at a rate of 0.5°C/20 seconds. Optionally, the linear monomeric template oligonucleotide is a single-stranded oligonucleotide. Optionally, the linear template oligonucleotide is at least 51 nucleotides in length. Optionally, the linear polymeric oligonucleotide is a linear dimeric oligonucleotide. Further optionally, the linear polymeric oligonucleotide is a linear dimeric oligonucleotide at least 160 nucleotides in length.

Optionally, the linear template oligonucleotide further comprises a phosphate group. Further optionally the linear template oligonucleotide further comprises a phosphate group attached to a terminal nucleic acid. Still further optionally the linear template oligonucleotide further comprises a phosphate group covalently attached to a terminal nucleic acid. Optionally the linear template oligonucleotide further comprises a phosphate group attached to the 5' terminal nucleic acid. Further optionally the linear template oligonucleotide further comprises a phosphate group covalently attached to the 5' terminal nucleic acid. Optionally or additionally, the linear template oligonucleotide further comprises a hydroxyl group. Further optionally or additionally, the linear template oligonucleotide further comprises a hydroxyl group attached to a terminal nucleic acid. Still further optionally or additionally, the linear template oligonucleotide further comprises a hydroxyl group covalently attached to a terminal nucleic acid. Optionally, the linear template oligonucleotide further comprises a hydroxyl group attached to the 3' terminal nucleic acid. Further optionally the linear template oligonucleotide further comprises a hydroxyl group covalently attached to the 3' terminal nucleic acid.

Optionally, the nucleic acid sequence encoding at least one epitope encodes the amino acid sequence ELDRWEKIRLRP with at least one mutation. Optionally, the nucleic acid sequence encoding at least one epitope encodes the amino acid sequence defined in SEQ ID NO:4 with at least one mutation. For the avoidance of doubt, the amino acid sequence defined in SEQ ID NO:4 is a wild-type sequence having no mutations and the nucleic acid sequence encoding at least one epitope encodes the amino acid sequence defined in SEQ ID NO:4 with at least one mutation. Optionally, the nucleic acid sequence of the at least one epitope is 5'-

GAACTGGATCGCTGGGAAAAAATCCGTCTGCGTCCT-3' and has at least one mutation.

Optionally, the nucleic acid sequence of the at least one epitope is the nucleic acid sequence defined in SEQ ID NO:3 and has at least one mutation. For the avoidance of doubt, the nucleic acid sequence defined in SEQ ID NO:3 is a wild-type sequence having no mutations and the nucleic acid sequence of the at least one epitope is the nucleic acid sequence defined in SEQ ID NO:3 with at least one mutation. Optionally, the at least one mutation comprises substitution of at least one nucleotide of the linear template oligonucleotide. Further optionally, the at least one mutation comprises substitution of at least one nucleotide of a codon of the linear template oligonucleotide. Optionally, the at least one mutation comprises substitution of at least one codon of the linear template oligonucleotide. Further optionally, the at least one mutation comprises substitution of at least one codon of the linear template oligonucleotide with a degenerate codon having the nucleic acid sequence NNK, wherein N is selected from adenine, thymine, guanine, and cytosine; and K is selected from thymine and guanine. Alternatively, the at least one mutation comprises substitution of at least one codon of the linear template oligonucleotide with a degenerate codon having the nucleic acid sequence NNN, wherein N is selected from adenine, thymine, guanine, and cytosine. Further alternatively, the at least one mutation comprises substitution of at least one codon of the linear template oligonucleotide with a degenerate codon having the nucleic acid sequence NNS, wherein N is selected from adenine, thymine, guanine, and cytosine; and S is selected from cytosine and guanine.

Optionally, the nucleic acid sequence encoding at least one epitope is selected from at least one of: 5'-GAGACTGGAGCAACGNNKGAAAAGATAAGGTTGAGGGGCACCGCAACTCAC-3'; 5'- GAGACTGGAGCAACGTGGNNKAAGATAAGGTTGAGGGGCACCGCAACTCAC-3'; 5'- GAGACTGGAGCAACGTGGGAANNKATAAGGTTGAGGGGCACCGCAACTCAC-3'; 5'- GAGACTGGAGCAACGTGGGAAAAGNNKAGGTTGAGGGGCACCGCAACTCAC-3'; 5'- GAGACTGGAGCAACGTGGGAAAAGATANNKTTGAGGGGCACCGCAACTCAC-3'; 5'- GAGACTGGAGCAACGTGGGAAAAGATAAGGNNKAGGGGCACCGCAACTCAC-3'; and 5'- GAGACTGGAGCAACGTGGGAAAAGATAAGGTTGNNKGGCACCGCAACTCAC-3'. Optionally, the nucleic acid sequence encoding at least one epitope is selected from the nucleic acid sequence defined in at least one of: SEQ ID NO: 17; SEQ ID NO: 18; SEQ ID NO: 19; SEQ ID NO:20; SEQ ID NO:21 ; SEQ ID NO:22; and SEQ ID NO:23. Optionally, the ligating step comprises ligating the respective terminal nucleic acids of the linear template oligonucleotide. Further optionally, the ligating step comprises ligating the 5' and 3' respective terminal nucleic acids of the linear template oligonucleotide. Still further optionally, the ligating step comprises ligating the 5' and 3' respective terminal nucleic acids of the linear template oligonucleotide to each other.

Optionally, the ligating step comprises providing a ligase enzyme. Optionally, the ligating step comprises providing a deoxyribonucleic acid (DNA) ligase enzyme. Optionally, the DNA ligase enzyme is a single strand (ss) DNA ligase enzyme. Optionally or additionally, the DNA ligase enzyme is a thermostable DNA ligase enzyme. Optionally or additionally, the DNA ligase enzyme is an ATP-dependent DNA ligase enzyme. Optionally or additionally, the DNA ligase enzyme is selected from CircLigase™ and CircLigasell™ (EPICENTRE® Biotechnologies). Alternatively, the ligating step comprises providing a ribonucleic acid (RNA) ligase enzyme. Optionally, the RNA ligase enzyme is RNA ligase (ATP). Optionally, the at least one restriction site comprises a non-palindromic restriction sequence.

Optionally, the at least one restriction site comprises a first restriction sequence and a second restriction sequence. Optionally, the first and second restriction sequences together form a restriction site. By "restriction site" is meant a nucleic acid sequence recognizable by a restriction enzyme. Optionally, the first and second restriction sequences each flank the nucleic acid sequence encoding at least one epitope and having at least one mutation. Further optionally, the first and second restriction sequences are located adjacent the respective ends of the nucleic acid sequence encoding at least one epitope and having at least one mutation. Optionally, the at least one restriction site is a Baul (Bacillus aquaemaris RFL1 ) restriction site. Further optionally, the at least one restriction site comprises the nucleic acid sequence 5'-CACGAG-3'. Still further optionally, the at least one restriction site is a Baul restriction site and comprises a first restriction sequence and a second restriction sequence. Still further optionally, the at least one restriction site comprises the nucleic acid sequence 5'-CACGAG-3' and comprises a first restriction sequence and a second restriction sequence. Optionally, the at least one restriction site comprises the nucleic acid sequence 5'-CACGAG-3' and comprises a first restriction sequence comprising the nucleic acid sequence 5'- CAC-3' and a second restriction sequence comprising the nucleic acid sequence 5'-GAG-3'.

Optionally, the subjecting step comprises providing a polymerase enzyme. Further optionally, the subjecting step comprises providing a DNA polymerase enzyme. Still further optionally, the subjecting step comprises providing a thermostable DNA polymerase enzyme. Still further optionally, the subjecting step comprises providing a replicative DNA polymerase enzyme. Still further optionally, the subjecting step comprises providing DNA polymerase III. Optionally, the nucleic acid sequence of the first primer is reverse complimentary to the nucleic acid sequence encoding at least one epitope and having at least one mutation. Optionally or additionally, the nucleic acid sequence of the first primer comprises a restriction sequence. Optionally, the restriction sequence is located at the 5' terminal nucleic acid of the first primer. Optionally, the nucleic acid sequence of the first primer comprises a non-palindromic restriction sequence.

Optionally, the nucleic acid sequence of the first primer comprises is a Spel (Sphaerotilus natans) restriction site. Further optionally, the nucleic acid sequence of the first primer comprises the nucleic acid sequence 5'-ACTAGT-3'.

Optionally, the nucleic acid sequence of the first primer comprises the nucleic acid sequence 5'- GTTCCTCGCCTTTGGACACCATGAGCTCACCGTGAGTTGCGGTGCC-3'. Further optionally, the nucleic acid sequence of the first primer comprises the nucleic acid sequence defined in SEQ ID NO:24.

Optionally or additionally, the nucleic acid sequence of the second primer comprises at least part of the nucleic acid sequence encoding at least one epitope and having at least one mutation. Optionally or additionally, the nucleic acid sequence of the second primer comprises a restriction sequence. Optionally, the restriction sequence is located at the 5' terminal nucleic acid of the second primer. Optionally, the nucleic acid sequence of the second primer comprises a non-palindromic restriction sequence. Optionally, the nucleic acid sequence of the second primer comprises is a Sacl

(Streptomyces achromogenes) restriction site. Further optionally, the nucleic acid sequence of the second primer comprises the nucleic acid sequence 5'-GAGCTC-3'. Optionally, the nucleic acid sequence of the second primer comprises the nucleic acid sequence 5'- GGCATGGACGAGCTGTACAAGACTAGTGAGACTGGAGCAACG-3'. Further optionally, the nucleic acid sequence of the second primer comprises the nucleic acid sequence defined in SEQ ID NO:25.

Optionally, the isolating step comprises isolating a linear dimeric oligonucleotide. Optionally, the isolating step comprises subjecting the linear polymeric oligonucleotide to nucleic acid

electrophoresis in order to separate the linear dimeric oligonucleotide. Further optionally, the isolating step comprises subjecting the linear polymeric oligonucleotide to polyacrylamide gel electrophoresis (PAGE) in order to separate the linear dimeric oligonucleotide. Still further optionally, the isolating step comprises subjecting the linear polymeric oligonucleotide to polyacrylamide gel electrophoresis (PAGE) in order to separate the linear dimeric oligonucleotide at least 160 nucleotides in length.

Optionally, the nucleic acid sequence of the third primer is complementary to the nucleic acid sequence of the first primer. Optionally, the nucleic acid sequence of the third primer comprises the nucleic acid sequence 5'- GGCATGGACGAGCTGTACAAGACTAGT-3'. Further optionally, the nucleic acid sequence of the third primer comprises the nucleic acid sequence defined in SEQ ID NO:26. Optionally or additionally, the nucleic acid sequence of the fourth primer is complementary to the nucleic acid sequence of the second primer.

Optionally, the nucleic acid sequence of the fourth primer comprises the nucleic acid sequence 5'- GTTCCTCGCCTTTGG ACACCAT-3' . Further optionally, the nucleic acid sequence of the fourth primer comprises the nucleic acid sequence defined in SEQ ID NO:27.

Optionally, the isolating step comprises purifying the linear dimeric oligonucleotide. Further optionally, the isolating step comprises purifying the linear dimeric oligonucleotide by solid phase extraction. Still further optionally, the isolating step comprises purifying the linear dimeric oligonucleotide by spin column-based nucleic acid purification. Still further optionally, the isolating step comprises purifying the linear dimeric oligonucleotide by silica-membrane-based spin column- based nucleic acid purification. Alternatively, the isolating step comprises purifying the linear dimeric oligonucleotide by affinity chromatography. Further alternatively, the isolating step comprises purifying the linear dimeric oligonucleotide using a commercial gel purification kit.

Optionally, the vector comprises two markers. Further optionally, the vector comprises first and second markers. Optionally, the first and second markers are not the same marker. Optionally, the first and second marker are each a detectable marker. Further optionally, the first and second marker are each a detectable chromophore. Further optionally, the first marker is a donor chromophore and the second marker is an acceptor chromophore. Optionally, the first and second marker are each a detectable fluorophore. Further optionally, the first marker is a donor fluorophore and the second marker is an acceptor fluorophore. Optionally, the first and second maker pair is selected from a cyan fluorescent protein (CFP) - yellow fluorescent protein (YFP) pair; a bioluminescent luciferase -YFP pair; a Cerulean - Citrine pair; and a Cerulean - Venus pair.

Optionally, the first and second markers are separated by a vector backbone.

Optionally or additionally, the vector comprises at least one restriction site. Further optionally or additionally, the vector comprises at least one non-palindromic restriction sequence. Optionally, the vector comprises the same restriction site as the first primer. Further optionally, the vector comprises a Spel (Sphaerotilus natans) restriction site. Still further optionally, the vector comprises the nucleic acid sequence 5'-ACTAGT-3'. Optionally or additionally, the vector comprises the same restriction site as the second primer. Further optionally or additionally, the vector comprises a Sacl

(Streptomyces achromogenes) restriction site. Still further optionally or additionally, the vector comprises the nucleic acid sequence 5'-GAGCTC-3'. Optionally or additionally, the first non-palindromic restriction sequence is located adjacent the first marker. Further optionally or additionally, the second non-palindromic restriction sequence is located adjacent the second marker. Optionally or additionally, the first and second markers are separated by a vector backbone and the first non-palindromic restriction sequence is located adjacent the opposing end of first marker; and the the second non-palindromic restriction sequence is located adjacent the opposing end of the second marker. Further optionally or additionally, the Spel (Sphaerotilus natans) restriction site is located adjacent the opposing end of first marker; and the Sacl (Streptomyces achromogenes) restriction site is located adjacent the opposing end of the second marker.

Optionally, the further introducing step comprises providing a nucleic acid sequence encoding the linker. Optionally, the nucleic acid sequence encoding the linker comprises the nucleic acid sequence defined in SEQ ID NO:1. The nucleic acid sequence encoding the linker can be flanked by one or more nucleic acid sequences that allow, upon treatment of the nucleic acid sequence encoding the linker with a restriction enzyme, generation of single stranded 5'-overhangs necessary for the introduction of the nucleic acid sequence encoding the linker to a location within the vector, optionally between the two epitopes of the linear polymeric oligonucleotide within the vector. Optionally, the nucleic acid sequence encoding the linker further comprises a non-palindromic restriction sequence. Optionally, the nucleic acid sequence encoding the linker comprises a Bsal (Bacillus stearothermophilus 20241 ) restriction site. Still further optionally, the nucleic acid sequence encoding the linker further comprises the nucleic acid sequence 5'-GGTCTC-3'. Optionally, the nucleic acid sequence encoding the linker comprises the nucleic acid sequence 5'- ATATAATAGGTCTCTACGATGGTGGAAGTGGGGGCAGCGGAGGGGGCTCCGGCGGTAGCGGT GCGGAGGCTGCAGCGAAAGAAGCTGCAGCGAAGGAAGCTGCAGCGAAAGAGGCTGCAGCGAA GGAGGCTGCAGCGAAAGAAGCTGCAGCGAAGGCGGGATCCGGCGGTAGCGGTGGATCCGGT GGCCCGCAGGGTATTCTGGGCCAGGGATCCGGTGGTTCTGGCGGTAGTGGTGGTTCTGGTGC AGAAGCCGCGGCGAAGGAGGCCGCGGCGAAAGAGGCCGCGGCGAAAGAAGCCGCGGCGAA GGAGGCCGCGGCGAAGGAAGCCGCGGCGAAAGCAGGGTCGGGCGGTAGCGGTGGCTCGGG CGGGAGCGGCGGCTCCGGCGGTTCCGGAGGTGAGTTATACGATGAGACCTTAAAT-3'.

Optionally, the nucleic acid sequence encoding the linker comprises the nucleic acid sequence defined in SEQ ID NO:32. Optionally, the further introducing step comprises contacting the nucleic acid sequence encoding the linker with Bsal (Bacillus stearothermophilus 20241 ) restriction enzyme. Optionally or additionally, the further introducing step comprises contacting the vector comprising at least one marker with Baul (Bacillus aquaemaris RFL1 ) restriction enzyme. According to a second aspect of the present invention, there is provided a library of biosensors, each biosensor comprising at least two epitopes and a linker separating the at least two epitopes; wherein the library is prepared according to a method according to the first aspect of the present invention. According to a third aspect of the present invention, there is provided a method of identifying an antibody, the method comprising the steps of contacting the antibody with a library of biosensors according to a second aspect of the present invention.

Brief description of the Drawings

Non-limiting embodiments of the present invention will now be described by way of non-limiting examples and the accompanying drawings, in which:

Figures 1A-B represent a schematic representation of the CerAbsCit mechanism of action and proposed further development. (FIG. 1A) In the absence of antibody, the Cerulean and Citrine fluorescent domains form an intramolecular complex and undergo a high degree of FRET. Binding of an anti-HIV-P17 antibody to the P17-based epitopes forces the two fluorescent domains to separate, resulting in a decrease in FRET. (FIG. 1 B) A library of CerAbsCit variants may be screened for binding to an antibody of interest. Dashed lines indicate "clonal units" comprised of sensor protein and encoding DNA. Phenotype-genotype linkage may be achieved by using individual bacterial cells or may be artificially maintained through use of aqueous droplets in oil;

Figures 2A-C represent a schematic representation of rolling circle amplification of circularized DNA, which allows the generation of tandem repeats of the encoded randomized epitope sequence. (FIG. 2A) Synthetic 51 nucleotide oligos bearing a random stretch of sequence are rendered circular by circligase. (FIG. 2B) Primer 1 is reverse complementary to Part 2 of the oligo and primes DNA polymerase-catalyzed rolling circle amplification (RCA), while oligo 2 is of identical sequence to part 1 and primes the synthesis of a strand complementary to the RCA product, resulting in different products bearing n repeats of the template oligo. (FIG. 2C) Discrete products such as (1 ) and (2) are formed, flanked by restriction sites Spel and Sacl and with a Baul site between each repeat.

Following purification of desired product (2), primers 3 and 4 allow amplification by binding to non- repeated, flanking sequences;

Figures 3A-E represent a schematic representation of the insertion of repeated epitope-encoding sequence into acceptor vector, which is followed by insertion of a linker between epitope sequences. (FIG. 3A) The acceptor vector encodes Citrine and Cerulean, between which Spel and Sacl sites are located. Digestion with these restriction enzymes removes a stuffer sequence. (FIG. 3B) Spel & Sacl restricted product (2) (see FIG. 2C) is ligated into the acceptor vector. (FIG. 3C) Linker PCR product is flanked by non-palindromic Bsal restriction sites (5'-GGTCTC, boxed). Bsal digestion results in single-stranded overhangs compatible with Baul-generated overhangs. (FIG. 3D) Bsal- restricted linker and the Baul-digested construct from (FIG. 3B) are incubated in the presence of T4 DNA ligase and Baul restriction enzyme, resulting in the irreversible ligation of linker at the Baul site. (FIG. 3E) Annotated sequence of construct depicted in (FIG. 3B) with DNA in lowercase and single- letter amino acid code in uppercase. The Baul site (CACGAG) formed between Part 2 and Part 1 is shown in bold. Sequence is shown from the last three residues of Citrine to the first three residues of Cerulean;

Figures 4A-F represent gel electrophoretic analysis of steps involved in preparation of an identical random repeat DNA library. (FIG. 4A) Rolling circle amplification results in discrete products with sizes corresponding to multiples of 51 nucleotides, together with a constant 58 nucleotides

(originating from the RCA primer pair's non-annealing 5'-tails). The 160 bp product (^*) represents the desired single repeat fragment. (FIG. 4B) Product (^*) from gel A was gel extracted (lane 1 ), amplified with primers that bind to the unique ends of the RCA products (lane 2) and restricted with Spel and Sacl, generating a 1 15 bp product (lane 3). (FIG. 4C) Colonies (1-10) resulting from transformation of the ligation of Spel & Sacl-restricted single repeat fragment with similarly treated acceptor vector were probed by colony PCR for an expected 251 bp product, as seen for a sequenced control (lane PC). (FIG. 4D) Colonies (1-10) resulting from transformation of the ligation of linker at the Baul site of the single repeat-bearing acceptor vector library were analyzed by colony PCR for an expected 641 bp product. (FIG. 4E) Library DNA probed in (FIG. 4D) was treated with Spel and Sacl, allowing separation on gel of desired 506 bp restriction fragment from undesired 1 15 bp and 64 bp fragments. (FIG. 4F) Part of 96-well-scale colony PCR on colonies resulting from ligation of 506 bp restriction fragment in (FIG. 4E) to a Spel/Sacl-cut acceptor vector, for an expected fragment size of 641 bp. Lanes labeled M1 and M2 contained DNA ladders;

Figure 5 represents a schematic summary of mutations found in the Cit^*Abs^*Cer library. The left- hand column represents each of the 7 residues of the WEKIRLR epitope while the top row represents each of the twenty possible amino acids, in their single letter code. The Amber stop codon is represented by ^*. Mutations are indicated by a black box, those found more than once are indicated by an additional number; Figures 6A-B are graphical representation of Cit^*Abs^*Cer variants' response to antibody addition. (FIG. 6A) Variants' ratiometric response to addition of 0.2 μΜ anti-P17 is plotted as a function of their concentration measured through direct excitation of the Citrine acceptor. The ratiometric response of bacterially expressed and purified CitAbsCer was also monitored at several concentrations. The upper (250 nM) and lower (40 nM) limits of detection are indicated by dashed lines. (FIG. 6B) Overview of the ratiometric response resulting from different mutations. Only those values measured for proteins that were between 40 and 250 nM in concentration are displayed, together with the corresponding mutation; and

Figure 7 represents the amino acid sequence of CitAbsCer. The N-terminal poly-His tag is followed by Citrine (S208F/V224L) (in bold), the linker, including the p17-derived epitopes (underlined) and Cerulean (S208F/V224L) (in bold), followed by a C-terminal Strep tag II. Examples

Example 1 - Cloning of individual constructs

To create a version of CerAbsCit in which the order of the donor and acceptor fluorescent domain was reversed (CitAbsCer), a CPEC-based strategy was used. Four different PCR fragments, citrine, linker, cerulean and opened pET28a vector were produced with primers designed to introduce mutual overlap in sequence, each with a melting temperature of -70 °C. The sequences of the primers used to generate these PCR fragments are summarized in TABLE 1. The linker fragment, which included two copies of the HIV-1 anti-P17 epitopes (ELDRWEKIRLRP), was obtained by using a plasmid template ("Split-beta lactamase" used in an unrelated project in our group), while the fluorescent domain sequences were produced using another plasmid from an unrelated project. Note that in this latter template, the DNA sequences for both fluorescent domains were designed to be as divergent as possible, allowing design of primers that bind with maximal specificity and that both fluorescent domains contained both the S208F and V224L hydrophobic surface mutations. PCR fragments were purified by silica spin column (Qiaquick PCR Purification Kit, Qiagen) and then combined in a standard Phusion polymerase reaction (without primers), with 200 ng of the vector fragment and equimolar amounts of the other fragments. The reaction was subjected to

thermocycling, with 30 seconds at 98 °C followed by 30 cycles of [98 °C for 10 seconds, 72 °C for 3 minutes], 72 °C for 10 minutes. Finally, any remaining circular, methylated plasmid templates were removed by incubating the reaction mixture with Dpnl (FastDigest, Fermentas) for 10 minutes at 37 °C. The final sequence for the ORF is shown is FIG. 7. For the identical random repeat cloning strategy, a CitAbsCer-based acceptor vector was required in which to insert the Spel/Sacl-flanked single repeat. Spel was already present between Citrine and the linker fragment. A Sacl recognition site was introduced by site directed mutagenesis behind the second epitope and just before

Cerulean, using primers Sac_ins_F and Sac_ins_R using a known technique. An undesired Baul (a.k.a. BssSI) site in the pET28a backbone was then destroyed using the same technique and primers Baul_silence_F and Baul_silence_R.

TABLE 1 - Sequences for primers used in both site-directed mutagenesis (SDM) and CPEC-based cloning strategies in this invention.

SEQ

Primer Sequence (5'>3')

ID

7CitCer_Vec_F GACGAACTGTACAAGGGTACCTGGAGCCATCCACAGTT NO:5

2CitCer_Vec_R TCACCATATGGCTGCCGCGCG NO:6

1 CitCer_Cit_F GCGGCAGCCATATGGTGAGCAAGGGTGAAGAATTATTC NO:7

10CitAbsCer_CitR CGCCACCACTAGTTTTATATAACTCATCCATGCCTAACGTGATACC NO:8

9CitAbsCer_AbsF CATGGATGAGTTATATAAAACTAGTGGTGGCGAACTGGATCG NO:9

12CitAbsCer AbsR GAAACCATACCCATGGAACCGCCCG NO: 10 1 1 CitAbsCer_ _CerF GCGGTTCCATGGGTATGGTTTCTAAAGGCGAGGAACTGTTTAC NO: 1 1

8CitCer_Cer_ R CTCCAGGTACCCTTGTACAGTTCGTCCATACCCAGGG NO: 12

Baul_silence _F GAGAGCGTACGATGGAGCTTCCAGGGGGAAACG NO: 13

Baul_silence _R GCTCCATCGTACGCTCTCCTGTTCCGACCC NO: 14

Sacl_ins_F GAGAAGGAGCTCATGGTTTCTAAAGGCGAGGAACTGTTTA NO: 15

SacI ins R AGAAACCATGAGCTCCTTCTCCCAACGATCTAACTCACCTCC NO: 16

Example 2 - Cit*Abs*Cer library cloning

To circularize oligos (listed in TABLE 2), a Circligase II (Epibio, USA) reaction was carried out following the manufacturer's standard reaction conditions (71 nM of each of the 7 different input oligos, 1x Circligase II reaction buffer, 2.5 mM MnCI₂, 100 units Circligase II, in a 20 \iL reaction volume). The reaction was incubated for 1 hour at 60 °C, followed by inactivation of the enzyme by heating to 80 °C for 10 minutes. All rolling circle and PCR reactions were carried out using Phusion DNA polymerase. Rolling circle amplification reactions includes 10 pmol of both primers, 1 μί of the unpurified Circligase reaction, 2 units Phusion DNA polymerase, 200 μΜ of each dNTP, in a total volume of 50 \iL. Cycling conditions were 98 °C for 30 seconds, 30 cycles of [98 °C for 10 seconds, 60 °C for 10 seconds, 72 °C for 15 seconds], 72 °C for 5 minutes. Prior to loading DNA on agarose gel, the rolling circle amplification products were heated to 95 °C for 5 minutes, then slowly cooled to room temperature at a rate of 0.5 °C every 20 seconds. This latter step was found to be desirable in ensuring correct hybridization of strands of identical length. To purify DNA from agarose gels, the Qiaquick Gel Extraction Kit (Qiagen) was used. To increase the quantity of available single repeat fragment, it was amplified with primers 3 and 4 using a thermocycle programme with an incubation at 98 °C for 30 seconds, followed by 30 cycles of 98 °C for 10 seconds and 72 °C for 10 seconds, followed by incubation at 72 °C for 5 minutes. The resulting PCR fragment was purified by silica spin column (Qiaquick, Qiagen). Acceptor vector and purified single repeat insert were restricted with 20 units Spel-HF and SacI at 37 °C in NEB buffer 4 with 0.1 mg/mL BSA. Cleaved acceptor vector DNA was purified by agarose gel electrophoresis followed by gel extraction. Cleaved single insert DNA was purified by silica spin column. DNA ligation was carried out with 400 units of T4 ligase (NEB), 50 ng vector DNA, at 1 :5 vector:insert molar ratio, in 1 x concentrated T4 ligase buffer, in 20 \iL volume. The ligation was transformed to E. coli NovaBlue and directly grown up as 5 mL liquid LB culture. To prepare single repeat-bearing plasmids for ligation with linker, 4.3 \ig single repeat-bearing plasmid library was first treated with 40 units of Baul (Thermo) for 16 hours at 37 °C in a 100 \iL volume. Linearized plasmid was separated from remaining circular plasmid by gel electrophoresis.

Meanwhile, Bsal-flanked linker DNA was prepared by PCR using primers linker_Bsal_F and linker_Bsal_R with pET28a-CitAbsCer as template. The resulting PCR product (2.4 g) was restricted using 40 units of Bsal at 37 °C for 8 hours in a 30 \iL volume. Next a 5 \iL ligation was carried out with 100 units of T4 ligase, 55 ng linear vector and 12 ng insert (i.e. a 1 :4 vector:insert molar ratio), in Baul restriction buffer with 1 mM of added ATP. Baul (2.5 units) was also present in this reaction to restrict self-ligated vector. The reaction was cycled between 30 minutes at 37 °C (for maximal Baul activity, restricting self-ligated vector) and 30 minutes at 16 °C (for maximal ligase activity) for approx. 16 hours and then used to transform E. coli DH5G by electroporation. All colony PCR reactions were carried out using Kapa2G DNA polymerase (Kapa Biosystems, USA), using primers Cit_F208S_F and Oppa_Vec_R, with cycling conditions 95 °C for 30 seconds, followed by 30 cycles of [95 °C for 10 seconds, 60 °C for 10 seconds, 72 °C for 30 seconds].

TABLE 2 - Oligonucleotides used in the creation of the identical random repeat library.

SEQ

Oligo Sequence (5'>3')

ID

PHO- NO : 17

NNK_W1_Bau1 GAGACTGGAGCAACGNNKGAAAAGATAAGGTTGAGGGGCACCGCA

ACTCAC

PHO- NO : 18

NNK_E2_Baul GAGACTGGAGCAACGTGGNNKAAGATAAGGTTGAGGGGCACCGCA

ACTCAC

PHO- NO : 19

NNK_K3_Baul GAGACTGGAGCAACGTGGGAANNKATAAGGTTGAGGGGCACCGCA

ACTCAC

PHO- NO :20

NNK_l4_Baul GAGACTGGAGCAACGTGGGAAAAGNNKAGGTTGAGGGGCACCGCA

ACTCAC

PHO- NO :21

NNK_R5_Baul GAGACTGGAGCAACGTGGGAAAAGATANNKTTGAGGGGCACCGCA

ACTCAC

PHO- NO :22

NNK_L6_Baul GAGACTGGAGCAACGTGGGAAAAGATAAGGNNKAGGGGCACCGCA

ACTCAC

PHO- NO :23

NNK_R7_Baul GAGACTGGAGCAACGTGGGAAAAGATAAGGTTGNNKGGCACCGCA

ACTCAC

Primer 1 GTTCCTCGCC I I I GGACACCATGAGCTCACCGTGAGTTGCGGTGCC NO :24

Primer 2 GGCATGGACGAGCTGTACAAGACTAGTGAGACTGGAGCAACG NO :25

Primer 3 GGCATGGACGAGCTGTACAAGACTAGT NO :26

Primer 4 GTTCCTCGCC I I I GGACACCAT NO :27 linker_Bsal_F ATATAATAGGTCTCTACGATGGTGGAAGTGGGGGCAG NO :28 linker_Bsal_R A I I I AAGGTCTCATCGTATAACTCACCTCCGGAACCG NO :29

Cit_F208S_F CAGAGTGCGTTAAGTAAGGATCCTAATGAAAAGCGTGACC NO :30

Oppa_Vec_R GTAAACAGTTCCTCGCC I I I AGAAAC NO :31

Example 3 - 96-well scale Cit*Abs*Cer protein expression and characterization E. coli BL21 (DE3) was transformed with the DNA library and plated on LB agar medium containing 30 Mg/mL kanamycin. Individual colonies were picked and transferred to 15 μί water in a 96-well plate. Colony PCR reactions were performed in a 96-well PCR plate using 1 μΙ_ of the colony suspension . The colony PCR reactions were subsequently treated with a cocktail of Exonuclease I II and alkaline phosphatase (ExosapIT) to remove primers and unincorporated dNTPs. These enzymes were heat inactivated (10 minutes at 80 °C) and the 96-well plate was submitted for sequencing (Baseclear, Netherlands). The same colony suspensions used for colony PCR were used to inoculate cultures in a 96-deep well plate containing 2 mL LB with 30 Mg/mL kanamycin and 0.1 mM IPTG. The plate was sealed with foil and incubated with shaking at 37 °C overnight. Cells were harvested by centrifugation of the plate at 3000 RPM for 10 minutes and the supernatant was discarded . Next 100 \iL Bugbuster (Novagen) and 1 μί Benzonase (Novagen) were added to each well and the plate was incubated at room temperature for 20 minutes, with shaking. The unclarified cell lysate was mixed with 100 μί binding buffer (0.5 M NaCI , 40 mM Tris-HCI (pH 8), 5 mM imidazole) and transferred to a 96-well plate containing Ni-NTA resin (HisPur, Thermo Scientific). After centrifugation , 250 \iL of 1 x binding buffer was loaded on the wells and the plate was spun again. Subsequently, 250 μί of washing buffer (0.5 M NaCI , 20 mM Tris-HCI (pH 8), 40 mM imidazole) was applied to the wells, a step that was repeated twice. Finally, protein was eluted with 250 ML elution buffer (0.5 M NaCI, 20 mM Tris-HCI (pH 8), 0.4 M imidazole). 50 μί of the Ni-NTA eluate was transferred to a 384-well plate for fluorescence measurements.

Example 4 - Fluorescence spectroscopy

Measurements were performed using a Tecan fluorescence plate reader, with 384-well plates. Unless mentioned otherwise, measurement buffer was 50 mM Tris-HCI (pH 8), 100 mM NaCI and 1 mg/mL BSA. Fluorescence emission ratios were then determined by exciting at 420 nm and measuring either full emission spectra or the emission at 475 nm (Cerulean) and 527 nm (Citrine) only. To detect Citrine directly, it was excited at 490 nm and its emission was measured at 530 nm. Comparing measured values to values determined for a serial dilution of CitAbsCer of known concentration gave a good indication of protein concentration.

Example 5 - Creation of DNA library

To allow directed evolution of the epitope binding sites in the CerAbsCit FRET sensor of anti-P17 antibody, it was desirable to find a method capable of introducing identical random mutations at both epitopes within one molecule. The invention teaches such a method and is based on rolling circle amplification (RCA) that allows the creation of tandem repeats of short, circular, partly randomized DNA templates. First, 51 -nt "template" oligos were designed such that an epitope- encoding, partially NNK-randomized sequence was flanked by two invariant stretches of sequence called "part 1 " and "part 2" (FIG. 2A). Seven different template oligos were designed, with NNK (where N represents A, T, G or C and K represents G or T) replacing either the W, E, K, I , R, L or R-encoding codons of the epitope sequence, with all template oligos carrying a 5'-phosphorylation. The template oligos were intramolecularly circularized using Circ Ligase II (FIG. 2A). At the point of circularization, a restriction site was formed, Baul (CACGAG), that was not present in the linear template oligo. This site served as the point at which a linker was later to be introduced within a tandem epitope repeat (see below). To create a tandem repeat of the randomized epitope sequence, rolling circle-mediated amplification was carried out using the circularized DNA molecules as templates. In this reaction, carried out using a high-fidelity thermostable DNA polymerase, a reverse primer 1 , reverse complementary to Part 2, initially created tandem repeats of the circular template (FIG. 2B). A forward primer 2, present in the same reaction and of the same sequence as Part 1 , together with primer 1 , resulted in amplification of products, whose sizes were discrete and variable, depending on the position where primers 2 and 1 bound. Primers 1 and 2 also carried sequences at their 5'-ends that were non-complementary to the template sequence, and served to introduce restriction sites Spel and Sacl at either end of the tandem repeat (FIG. 2C). As expected, gel electrophoretic analysis of this reaction revealed discretely sized products whose sizes correlated to multiples of 51 bp (the length of the template oligos) plus a constant 58 bp (the total length of the non-template- complementary sequence introduced by the 5'-tails of primers 1 and 2) (FIG. 4A). The 160 bp fragment carried the desired single tandem repeat of the epitope sequence and so was excised from the gel and purified (FIG. 4B, lane 1 ). Next, primers 3 and 4, complementary to the 5'-arms of primers 1 and 2 only, were used to amplify this 160 bp product (FIG. 4B, lane 2). An acceptor vector, pET28a-Citrine(Spel)-L9-(Sacl)Cerulean (FIG. 3A), was restricted with Spel and Sacl, allowing ligation of a similarly treated single tandem repeat (FIG. 4B, lane 3) in between the Citrine and Cerulean sequences (FIG. 3B, FIG. 3E). Since in this construct Citrine is at the N-terminus, while Cerulean is at the C-terminus, resulting constructs will be referred to as Cit^*Abs^*Cer (asterisks represent variability in epitope sequence). The ligation mixture was transformed to 20 μΙ_ chemically competent Escherichia coli NovaBlue cells, which were cultured in 5 mL LB medium. To determine the size of the library at this stage, part of the transformation reaction was plated on selective agar. The transformation yielded a total of 4000 colony forming units (cfu), with 10 out of 10 colonies analyzed bearing the correctly sized insert (FIG. 4C). At this stage, the epitopes' sequences were still located in tandem to one another, yet the Cit^*Abs^*Cer sensor mechanism dictates that they be separated by a semi-flexible 127 amino acid residue linker. A miniprep of the single repeat bearing vector library was treated with Baul to digest the vector at this point. Meanwhile, the semi-flexible linker-encoding fragment was amplified with primers that added Bsal sites at either end of the product. The Bsal sites were designed such that following Bsal restriction they were removed (Bsal cleaves at the 3'-side of the non-palindromic recognition site GGTCTC) and a 4-bp sticky overhang compatible with Baul was created. The non-palindromic nature of the Baul restriction site ensured that the linker could be ligated in only one orientation in the vector (FIG. 3C, FIG. 3D). Furthermore, this ligation was irreversible, as ligation of the Bsal-generated cohesive ends to Baul-generated cohesive ends destroyed the original Baul recognition site. Therefore, in the presence of T4 ligase and Baul restriction enzyme, the linker insertion reaction ought to be driven to completion (FIG. 3D). However, contrary to expectations, PCR analysis of E. coli colonies resulting from transformation of the latter ligation revealed that a significant fraction (50%) of colonies did not yield the expected 641 bp product and were thus incorrect (FIG. 4D). Incorrect products formed in this colony PCR analysis included fragments corresponding to acceptor plasmids bearing zero and single epitope repeats without the linker inserted indicating that an additional library purification step was called for. A library miniprep was restricted with Spel and Sacl and subsequent gel electrophoresis allowed separation of shorter undesired fragments (barely visible) from the desired 641 bp fragment (FIG. 4E). The latter was gel extracted and ligated to a Spel/Sacl treated acceptor vector. Analysis by PCR of colonies resulting from the transformation of this ligation revealed at least 85/92 (92%) of colonies to carry plasmids with correctly sized insert (FIG. 4F).

Example 6 - Validation of library by sequencing and expression

To test the functionality of the library, a small-scale analysis of 96 clones was carried out. Bacterial expression strain E. coli BL21 (DE3) was transformed with library DNA and plasmids of individual colonies were sequenced. Out of 95 clones submitted, 85 were successfully sequenced. Of these 85, 19 did not contain the desired exact repeat of epitope sequence but instead contained two different epitopes. Out of the 66 sequences that did contain the desired repeat, 9 contained an undesired, but exactly repeated, single bp deletion in the epitopes. This deletion is most likely a result of truncated side-products in the oligonucleotide starting material, which was only 80% pure. Mutations found in the remaining 57 sequences are summarized in FIG. 5. The coverage provided by the 85 sequences was 39 out of a possible 148 variants or 26% of the library.

The isolated and sequenced clones were expressed at 2 mL scale and purified by Ni-NTA spin- columns. Fluorescence emission ratios were measured before and after addition of 200 nM anti-P17 antibody, allowing each sensor's DR to be calculated. Protein concentrations were estimated based on the fluorescence intensity of directly excited acceptor. These concentrations were found to range from less than 10 nM to around 350 nM (FIG. 6A). When bacterially expressed CitAbsCer was measured under the same conditions, a decrease in dynamic range was observed below 48 nM (FIG. 6A). At very low concentrations of FRET sensor protein, the larger relative contribution of background fluorescence signal likely interferes with the assay. Further analysis was therefore limited to those proteins that happened to be between 40 and 250 nM (to avoid the need to add large amounts of expensive antibody) after purification (FIG. 6B). Promisingly, the assay appears to be reproducible, as seen by the similarity in ratiometric response obtained for the pairs of W1 L, I4T, I4F and I4V mutants. Although library coverage is limited (19%), it is tempting to draw the preliminary conclusion that position I4 in the epitope is more tolerant of mutagenesis than are positions R5, L6 and R7. Position I4's apparent tolerance might be explained by this residue's position in the P17 immunogen, where it is buried in the protein core and not exposed to the surface. Hence, it seems likely the residue plays no role in antibody binding. Interestingly, out of 6 residues tested, proline was the only mutation of I4 that negatively affected anti-P17 binding.

Claims

34 Claims

A method of preparing a library of biosensors, each biosensor comprising at least two epitopes and a linker separating the at least two epitopes; the method comprising the steps of:

(b) ligating the linear monomeric template oligonucleotide to form a circular template oligonucleotide;

(c) subjecting the circular template oligonucleotide to rolling circle amplification (RCA) in the presence of a first primer and a second primer to provide a linear polymeric oligonucleotide; wherein the nucleic acid sequence of each of the first and second primers comprises at least one restriction site;

(d) isolating the linear polymeric oligonucleotide;

(g) further introducing a nucleic acid sequence encoding the linker separating the at least two epitopes into the linear polymeric oligonucleotide to provide a library of vectors, each vector encoding a biosensor comprising at least two epitopes and a linker separating the at least two epitopes.

A method according to Claim 1 , wherein the method further comprises the step of (h) expressing each of the vectors to provide a library of biosensors, each biosensor comprising at least two epitopes and a linker separating the at least two epitopes.

A method according to Claim 1 or 2, wherein the method comprises the additional step of purifying the linear polymeric oligonucleotide prior to the amplifying step by contacting the linear polymeric oligonucleotide from the subjecting step with at least one restriction enzyme capable of recognizing the at least one restriction site of the nucleic acid sequence of each of the first and second primers. 35

A method according to Claim 3, wherein the additional purifying step comprises contacting the linear polymeric oligonucleotide with at least one restriction enzyme selected from Spel (Sphaerotilus natans) and Sacl (Streptomyces achromogenes).

A method according to any one of Claims 1-4, wherein the method comprises the additional step of heating the linear polymeric oligonucleotide to 95°C.

A method according to any one of Claims 1-5, wherein the nucleic acid sequence encoding at least one epitope encodes the amino acid sequence ELDRWEKIRLRP with at least one mutation.

7. A method according to Claim 6, wherein the at least one mutation comprises substitution of at least one nucleotide of a codon of the linear template oligonucleotide.

8. A method according to Claim 7, wherein the at least one mutation comprises substitution of at least one codon of the linear template oligonucleotide with a degenerate codon having the nucleic acid sequence NNK, wherein N is selected from adenine, thymine, guanine, and cytosine; and K is selected from thymine and guanine.

9. A method according to any one of Claims 1-8, wherein the at least one restriction site

comprises a first restriction sequence and a second restriction sequence.

10. A method according to Claim 9, wherein the at least one restriction site comprises the

nucleic acid sequence 5'-CACGAG-3' and comprises a first restriction sequence comprising the nucleic acid sequence 5'-CAC-3' and a second restriction sequence comprising the nucleic acid sequence 5'-GAG-3'.

1 1. A method according to any one of Claims 1-10, wherein the nucleic acid sequence of the first primer comprises the nucleic acid sequence 5'-ACTAGT-3'.

12. A method according to any one of Claims 1-1 1 , wherein the nucleic acid sequence of the second primer comprises the nucleic acid sequence 5'-GAGCTC-3'.

13. A method according to any one of Claims 1-12, wherein the isolating step comprises

isolating a linear dimeric oligonucleotide.

14. A method according to any one of Claims 1-13, wherein the vector comprises first and

second markers, wherein the first and second marker are each a detectable fluorophore, wherein the first and second maker pair is selected from a cyan fluorescent protein (CFP) - 36

yellow fluorescent protein (YFP) pair; a bioluminescent luciferase -YFP pair; a Cerulean - Citrine pair; and a Cerulean - Venus pair.

15. A method according to Claim 14, wherein the vector comprises the nucleic acid sequence 5'- ACTAGT-3' located adjacent the first marker and the nucleic acid sequence 5'-GAGCTC-3' located adjacent the second marker.

16. A method according to any one of Claims 1-1 1 , wherein the further introducing step

comprises providing a nucleic acid sequence encoding the linker, and the nucleic acid sequence 5'-GGTCTC-3'; and contacting the nucleic acid sequence encoding the linker with Bsal (Bacillus stearothermophilus 20241 ) restriction enzyme.

17. A library of biosensors, each biosensor comprising at least two epitopes and a linker

separating the at least two epitopes; wherein the library is prepared according to a method according to any one of Claims 1-16.

18. A method of identifying an antibody, the method comprising the step of contacting the

antibody with a library of biosensors according to Claim 17.