METHOD FOR AFFINITY MATURATION OF ANTIBODIES

Abstract

The present invention relates to a novel method of generating libraries of polynucleotides encoding a framework region and at least one adjacent complementarity determining region (CDR) of an antibody of interest. These libraries are suitable for use in affinity maturation procedures in order to obtain maturated antibodies with improved characteristics compared to the parent antibody.

Claims

1. A method of generating a library of polynucleotides each encoding a framework region and at least one adjacent complementarity determining region (CDR) of an antibody of interest comprising a known parent CDR wherein the parent CDR is encoded by a known parent CDR polynucleotide sequence, the method comprising: i) providing a polynucleotide encoding a first framework region of the antibody, ii) providing a first PCR primer for the polynucleotide of (i), iii) providing a mixture of polynucleotides each consisting of elements A-B-C, wherein A) is a polynucleotide capable of hybridizing to a first framework region, each B) is a member of a library of polynucleotides comprising the same number of codons as the parent CDR polynucleotide sequence, wherein the members of said library are designed to comprise at least one randomized codon and C) is a polynucleotide capable of hybridizing to a second framework region, iv) providing a second PCR primer for element C) v) performing a PCR based on the polynucleotides (i) to (iv), thereby obtaining the library of polynucleotides, and wherein such PCR is performed in the absence of the parent CDR polynucleotide sequence.

2. The method of claim 1, wherein said first framework region is either FW1 or FW4, wherein said second framework region is FW2 if the first one is FW1, or is FW3 if the first one is FW4, and wherein said CDR is CDR1 if the first framework region is FW1, or is CDR3 if the first framework region is FW4.

3. The method of claim 1, wherein said first framework region is FW1, wherein said second framework region is FW2, wherein said first primer is a forward primer for FW1 and wherein said second primer is a reverse primer for FW2, and wherein said CDR is CDR1.

4. The method of claim 1, wherein said first framework region is FW4, wherein said second framework region is FW3, wherein said first primer is a reverse primer for FW4 and wherein said second primer is a forward primer for FW3, and wherein said parent CDR is CDR3.

5. A method of generating a library of polynucleotides each encoding a framework region and two adjacent complementarity determining regions (CDRs) of an antibody of interest comprising a first and a second known parent CDR wherein the first and second parent CDRs are encoded by first and second known CDR polynucleotide sequences, the method comprising: i) providing a polynucleotide encoding a first framework region of the antibody, ii) providing a first mixture of polynucleotides each consisting of elements A-B-C, wherein A) is a polynucleotide capable of hybridizing to the first framework region, each B) is a member of a library of first polynucleotides comprising the same number of codons as the first parent CDR, wherein the members of said library are designed to comprise at least one randomized codon and C) is a polynucleotide capable of hybridizing to a second framework region, iii) providing a first PCR primer for element C), iv) providing a second mixture of polynucleotides each consisting of elements A-B-C, wherein A) is a polynucleotide capable of hybridizing to said first framework region, each B) is a member of a library of second polynucleotides comprising the same number of codons as the second parent CDR polynucleotide sequence, wherein the members of said library are designed to comprise at least one randomized codon, and C) is a polynucleotide capable of hybridizing to a third framework region, v) providing a second PCR primer for element C), vi) performing a PCR based on the polynucleotides (i) to (v), thereby obtaining the library of polynucleotides, and wherein such PCR is performed in the absence of any parent CDR polynucleotide sequence.

6. The method of claim 5, wherein said first framework region is FW2, wherein said second framework region is FW1, wherein said third framework region is FW3, wherein the first parent CDR is CDR1, wherein the second parent CDR is CDR2, wherein said first primer for element C) is a forward primer for FW1, wherein said second primer for element C) is a reverse primer for FW3.

7. The method of claim 5, wherein said first framework region is FW3, wherein said second framework region is FW2, wherein said third framework region is FW4, wherein the first parent CDR is CDR2, wherein the second parent CDR is CDR3, wherein said first primer for element C) is a forward primer for FW2, wherein said second primer for element C) is a reverse primer for FW4.

8. The method of claim 1, wherein in element B) one codon or two codons of a parent CDR polynucleotide sequence are randomized.

9. A library of polynucleotides obtainable according to the method of claim 1 encoding one randomized CDR or two adjacent randomized CDRs of a variable chain of an antibody wherein in the library obtained the ratio of parent polynucleotide sequence to other (randomized) polynucleotide sequences is 1:106 or less in case one CDR is randomized and is 1:107 or less in case two CDRs are randomized.

10. Use of a library according to claim 9 for generating a library of polynucleotides encoding the variable chain of an antibody wherein the variable chain is selected from a variable H chain or a variable L chain.

11. A method for generating a library of polynucleotides encoding the variable chain of an antibody by performing an overlapping PCR based on the libraries generated according to claim 3.

12. A library of polynucleotides encoding a variable chain of an antibody obtainable according to claim 11 wherein the variable chain comprises a randomized CDR1, a randomized CDR2 and a randomized CDR3 and wherein in the library obtained the ratio of parent polynucleotide sequence to other polynucleotide sequences in the library is 1:510.sup.7 or less.

13. A method for generating an antibody library wherein the antibody comprises a first variable chain and a second variable chain, wherein a library of polynucleotides encoding the first variable chain of said antibody according to claim 12 is expressed and combined with the second variable chain of said antibody.

14. A method of selecting an antibody comprising a first variable chain and a second variable chain from a library generated according to claim 13 wherein the selected antibody has improved binding characteristics compared to a parent antibody with known parent CDRs.

15. The method of claim 14 wherein the selected antibody exhibits the selected increase of the dissociation complex half-life t/2 of at least 20% compared to the parent antibody. having a first label by comparing the first signal of step c) to the second signal of step d).

16. The method of claim 5, wherein in element B) one codon or two codons of a parent CDR polynucleotide sequence are randomized.

17. A method for generating a library of polynucleotides encoding the variable chain of an antibody by performing an overlapping PCR based on the libraries generated according to claim 4.

18. A method for generating a library of polynucleotides encoding the variable chain of an antibody by performing an overlapping PCR based on the libraries generated according to claim 6.

19. A method for generating a library of polynucleotides encoding the variable chain of an antibody by performing an overlapping PCR based on the libraries generated according to claim 7.

Description

DESCRIPTION OF THE FIGURES

[0279] FIGS. 1A-1C: Construction of a library comprising random amino acid substitutions within one or more of the heavy chain CDRs

[0280] FIG. 1A: Production of the heavy chain fragments required in construction of the mutant library (step 1)

[0281] In the first round (PCR 1) three different heavy chain fragments corresponding to fragments 1, 3 and 4, respectively were generated by aid of corresponding primer sets. The light grey stretches indicate the CDRs. The backbone sequence is given in black. Horizontal arrows indicate the primers used. Vertical arrows point to the results of the PCR. The short 42 bp oligonucleotide (fragment 2) which is crossed out in the Figure was not obtained by PCR but was separately chemically synthesized.

[0282] FIG. 1B: HC library synthesis by CDR single amino acid randomization.

[0283] In the second step PCR 2, the four fragments obtained as described in FIG. 1A served as templates (black lines). Horizontal arrows with a cross indicate the polynucleotide libraries each comprising a degenerated NNK codon for each CDR codon position. These polynucleotide libraries in addition comprise sequence stretches capable of hybridizing to one or two of the fragments of step 1 as required and indicated. Forward and reverse primers, respectively, (small arrows) were used to perform the respective PCRs.

[0284] FIG. 1C: Final step of library synthesis

[0285] The additional sequence stretches capable of hybridizing to one or two of the fragments of step 1 are needed to perform the final step in production of the HC library, i.e. an overlapping PCR using all four products of PCR 2. Terminal primers (F1A; R1A) are used and the fragments themselves act as mega primers in this overlapping PCR.

[0286] FIG. 2: Vector map for periplasmatic Fab expression

[0287] The description in the Figure given is considered self-explaining.

[0288] FIG. 3: ELISA setup for the screening of cTnT binding Fab fragments

[0289] A microtiter plate coated with streptavidin (SA plate) is used to bind biotinylated cardiac troponin T (bi-cTnT) to the solid phase. Fab fragments comprising recombinant anti-cTnT heavy chains (<cTnT>-Fab) bind to TnT and are detected via peroxidase (POD)-labeled anti-human Fab antibodies (Anti huFab-POD).

[0290] FIG. 4: Elecsys sandwich assay

[0291] A scheme showing the assay setup is depicted. The biotinylated (bi) capture antibody is attached to streptavidin (SA) coated beads. Various affinity maturated anti-cTnT antibodies were ruthenylated (Ru) and the effect of the affinity maturations was investigated by ECL analyses.

[0292] FIG. 5: ECL signal counts for the genuine specifier and specifier derivative.

[0293] Counts for the genuine anti-cTnT antibody and a mutant antibody (combination 12, respectively, refer to the Fab fragment identifier used in Table 2) are given. Light grey bars show the assay blank values (noise) in the Diluent Multi Assay reagent, dark grey bars show the counts obtained with Calibrator 1 of the commercial cTnT Elecsys assay (signal). Antibody combination 12 shows an improved signal to noise ratio.

[0294] The following Examples illustrate the invention:

Example 1: Materials & General Methods

Recombinant DNA Techniques

[0295] Standard methods were used to manipulate DNA as described in Sambrook, J. et al., Molecular Cloning: A laboratory manual; Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 1989. The molecular biological reagents were used according to the manufacturer's instructions.

DNA Sequence Determination

[0296] DNA sequences were determined by double strand sequencing performed at Microsynth AG (Balgach, Switzerland).

DNA and Protein Sequence Analysis and Sequence Data Management

[0297] Vector NT1 Advance suite version 11.5.0 was used for sequence creation, mapping, analysis, annotation and illustration.

Protein Chemistry and Labeling Techniques

[0298] Standard protein chemistry and labeling techniques are provided e.g. in Hermanson,

[0299] G. Bioconjugate Techniques 3rd Edition (2013) Academic Press.

Bioinformatics

[0300] Bioinformatics methods are provided in e.g. Keith J. M. (ed.) Bioinformatics Vol. I and Vol. II, Methods in Molecular Biology Vol. 1525 and Vol. 1526 (2017) Springer, and in Martin, A. C. R. & Allen, J. Bioinformatics Tools for Analysis of Antibodies in: Dbel S. & Reichert J. M. (eds.) Handbook of Therapeutic Antibodies Wiley-VCH (2014).

Electrochemiluminescent Immunoassays

[0301] Immunoassays and related methods are provided in e.g. Wild D. (ed.) The Immunoassay Handbook 4th Edition (2013) Elsevier. Ruthenium complexes as electrochemiluminescent labels are provided in e.g. Staffilani M. et al. Inorg. Chem. 42 (2003) 7789-7798. Typically, for the performance of electrochemiluminescence (ECL) based immunoassays an Elecsys 2010 analyzer or a successor system was used, e.g. a Roche analyzer (Roche Diagnostics GmbH, Mannheim Germany) such as E170, cobas e 601 module, cobas e 602 module, cobas e 801 module, and cobas e 411, and Roche Elecsys assays designed for these analyzers, each used under standard conditions, if not indicated otherwise.

Example 2: Library Construction

[0302] The parent antibody variable heavy chain is of murine origin (SEQ ID NO:34). A library comprising mutated HCCDRs was constructed with the goal of a single amino acid randomization in HCCDR1, HCCDR2 and/or HCCDR3, respectively. In a first step four DNA fragments were generated each encoding one of the four different parental antibody framework regions. Framework regions 1, 3 and 4 were obtained by polymerase chain reaction in house, the short fragment 2 (42 bp), representing framework region 2, was ordered at Metabion international AG (cf. FIG. 1A). The fragments were gel purified and quantified. 100 ng of one of these DNA fragments was used as a polynucleotide template in each of the four add-on PCR reaction mixtures. The CDR regions were added by use of a polynucleotide library comprising the same number of codons as the parent CDR, wherein the members of said library were designed to comprise library members with one NNK codon for each of the respective codon position in the respective HCCDR. The polynucleotides in the CDR library in addition comprised sequences capable of hybridizing to the framework region neighboring to the respective CDR. Terminal primers were used for nested PCR amplification. Thereby (cf. FIG. 1B) four DNA fragments with partially overlapping sequences were generated. Overlapping PCR, with terminal primers hybridizing to the 3 end of the FW1 sequence and to the 5 end of the FW4 sequence, was performed to connect the four fragments to a linear DNA library construct (cf. FIG. 1C). A typical PCR reaction was filled with PCR grade water to a 100 l reaction mix containing 10 l 10PCR buffer with MgSO4, 200 M dNTP mix, 0.5 M forward primer and reverse primer, 250 ng DNA template, 5 units Pwo DNA polymerase. A typical PCR started with initial template denaturation at 94 C. for 5 min, employed 30 cycles (94 C. 2 min, 60 C. 45 sec, 72 C. 1 min) and contained a final elongation step at 72 C. for 5 min Primers, templates and fragment sequences are listed in Table 1. The library fragments contained all necessary regulatory sequences for a successful transcription and translation in a cell-free system. The skilled artisan is able to generate such library by following state of the art methods, see e.g. Hanes, J. & Pluckthun, A. (1997), In vitro selection and evolution of functional proteins by using ribosome display, Proc Natl Acad Sci U.S.A. 94, 4937-42. 250 ng of the DNA library thus generated, covering the three HC CDRs and corresponding to about 5.Math.10.sup.11 library members were used for the in vitro display approach.

TABLE-US-00001 TABLE1 Sequencesusedinthegenerationoftheanti-cTnTFabfragmentlibrary SEQIDNO: PCR1 F1A CGAAATTAATACGACTCACTATAGGGAGACCACAACGGTTTCCC 35 1-fr.1rv GGTAAAGGTATAGCCGCTCG 36 1-fr.3fw CCAGAAATTTAAGGATAAAGCGACCC 37 1-fr.3rv GGTCGCGCAATAATACACCG 38 1-fr.4fw CGGTGTATTATTGCGCGACC 39 R1A AACCCCCGCATAGGCTGGGGGTTGGAAAGCCTCTGAGGACCAGC 40 ACG PCR2 Fragment1 Frt GGGAGACCACAACGGTTTCCCTCTAGAAATAATTTTGTTTAACT 41 TTAAGAAGGAGATATACAT 2-H1rvmix 2-H1rv1 GCTCTGTTTCACCCATTTCATATAATAMNNGGTAAAGGTATAGC 42 2-H1rv2 GCTCTGTTTCACCCATTTCATATAMNNATCGGTAAAGGTATAGC 43 2-H1rv3 GCTCTGTTTCACCCATTTCAMNNAATAATCGGTAAAGGTATAGC 44 2-H1rv4 GCTCTGTTTCACCCATTTMNNATAATAATCGGTAAAGGTATAGC 45 2-H1rv5 GCTCTGTTTCACCCAMNNCATATAATAATCGGTAAAGGTATAGC 46 2-fr.1rv CCATGGCTCTGTTTCACCC 47 Fragment2 2-fr.2fw CGAGCGGCTATACCTTTACC 48 2-H1fwmix 2-H1fw1 GCTATACCTTTACCNNKTATTATATGAAATGGGTGAAACAGAGC 49 2-H1fw2 GCTATACCTTTACCGATNNKTATATGAAATGGGTGAAACAGAGC 50 2-H1fw3 GCTATACCTTTACCGATTATNNKATGAAATGGGTGAAACAGAGC 51 2-H1fw4 GCTATACCTTTACCGATTATTATNNKAAATGGGTGAAACAGAGC 52 2-H1fw5 GCTATACCTTTACCGATTATTATATGNNKTGGGTGAAACAGAGC 53 2-H2rvmix 2-H2rv1 CCTTAAATTTCTGGTTATAAAAGGTTTCGCCGTTGTTCGGATTA 54 ATMNNGCCAATCCATTCCAGG 2-H2rv2 CCTTAAATTTCTGGTTATAAAAGGTTTCGCCGTTGTTCGGATTM 55 NNATCGCCAATCCATTCCAGG 2-H2rv3 CCTTAAATTTCTGGTTATAAAAGGTTTCGCCGTTGTTCGGMNNA 56 ATATCGCCAATCCATTCCAGG 2-H2rv4 CCTTAAATTTCTGGTTATAAAAGGTTTCGCCGTTGTTMNNATTA 57 ATATCGCCAATCCATTCCAGG 2-H2rv5 CCTTAAATTTCTGGTTATAAAAGGTTTCGCCGTTMNNCGGATTA 58 ATATCGCCAATCCATTCCAGG 2-H2rv6 CCTTAAATTTCTGGTTATAAAAGGTTTCGCCMNNGTTCGGATTA 59 ATATCGCCAATCCATTCCAGG 2-H2rv7 CCTTAAATTTCTGGTTATAAAAGGTTTCMNNGTTGTTCGGATTA 60 ATATCGCCAATCCATTCCAGG 2-H2rv8 CCTTAAATTTCTGGTTATAAAAGGTMNNGCCGTTGTTCGGATTA 61 ATATCGCCAATCCATTCCAGG 2-H2rv9 CCTTAAATTTCTGGTTATAAAAMNNTTCGCCGTTGTTCGGATTA 62 ATATCGCCAATCCATTCCAGG 2-H2rv10 CCTTAAATTTCTGGTTATAMNNGGTTTCGCCGTTGTTCGGATTA 63 ATATCGCCAATCCATTCCAGG 2-fr.2rv GGGTCGCTTTATCCTTAAATTTCTGG 64 Fragment3 2-fr.3fw GCAAAAGCCTGGAATGGATTGGC 65 2-H2fwmix 2-H2fw1 CCTGGAATGGATTGGCNNKATTAATCCGAACAACGGCGAAACCT 66 TTTATAACCAGAAATTTAAGG 2-H2fw2 CCTGGAATGGATTGGCGATNNKAATCCGAACAACGGCGAAACCT 67 TTTATAACCAGAAATTTAAGG 2-H2fw3 CCTGGAATGGATTGGCGATATTNNKCCGAACAACGGCGAAACCT 68 TTTATAACCAGAAATTTAAGG 2-H2fw4 CCTGGAATGGATTGGCGATATTAATNNKAACAACGGCGAAACCT 69 TTTATAACCAGAAATTTAAGG 2-H2fw5 CCTGGAATGGATTGGCGATATTAATCCGNNKAACGGCGAAACCT 70 TTTATAACCAGAAATTTAAGG 2-H2fw6 CCTGGAATGGATTGGCGATATTAATCCGAACNNKGGCGAAACCT 71 TTTATAACCAGAAATTTAAGG 2-H2fw7 CCTGGAATGGATTGGCGATATTAATCCGAACAACNNKGAAACCT 72 TTTATAACCAGAAATTTAAGG 2-H2fw8 CCTGGAATGGATTGGCGATATTAATCCGAACAACGGCNNKACCT 73 TTTATAACCAGAAATTTAAGG 2-H2fw9 CCTGGAATGGATTGGCGATATTAATCCGAACAACGGCGAANNKT 74 TTTATAACCAGAAATTTAAGG 2-H2fw10 CCTGGAATGGATTGGCGATATTAATCCGAACAACGGCGAAACCN 75 NKTATAACCAGAAATTTAAGG 2-H3rvmix 2-H3rv1 GGTACCCTGGCCCCAATAATCAAACACMNNGGTCGCGCAATAAT 76 ACACC 2-H3rv2 GGTACCCTGGCCCCAATAATCAAAMNNGCGGGTCGCGCAATAAT 77 ACACC 2-H3rv3 GGTACCCTGGCCCCAATAATCMNNCACGCGGGTCGCGCAATAAT 78 ACACC 2-H3rv4 GGTACCCTGGCCCCAATAMNNAAACACGCGGGTCGCGCAATAAT 79 ACACC 2-H3rv5 GGTACCCTGGCCCCAMNNATCAAACACGCGGGTCGCGCAATAAT 80 ACACC 2-fr.3rv CGGTCAGGGTGGTACCCTGGC 81 Fragment4 2-fr.4fw CGGTGTATTATTGCGCGACC 82 2-H3fwmix 2-H3fw1 GGTGTATTATTGCGCGACCNNKGTGTTTGATTATTGGGGCCAGG 83 GTACC 2-H3fw2 GGTGTATTATTGCGCGACCCGCNNKTTTGATTATTGGGGCCAGG 84 GTACC 2-H3fw3 GGTGTATTATTGCGCGACCCGCGTGNNKGATTATTGGGGCCAGG 85 GTACC 2-H3fw4 GGTGTATTATTGCGCGACCCGCGTGTTTNNKTATTGGGGCCAGG 86 GTACC 2-H3fw5 GGTGTATTATTGCGCGACCCGCGTGTTTGATNNKTGGGGCCAGG 87 GTACC Rrt GGAAAGCCTCTGAGGACCAGCACGGATGCCCTGTGC 88 OverlappingPCR F1A CGAAATTAATACGACTCACTATAGGGAGACCACAACGGTTTCCC 89 R1A AACCCCCGCATAGGCTGGGGGTTGGAAAGCCTCTGAGGACCAGC 90 ACG PCR gggagaccacaacggtttccctctagaaataattttgtttaact 91 fragment1 ttaagaaggagatatacatatggaagtgcagctgcagcagagcg gcccggaactggtgaaaccgggcgcgagcgtgaaaatgagctgc aaagcgagcggctatacctttaccGATTATTATATGAAAtgggt gaaacagagccatgg PCR cgagcggctatacctttaccGATTATTATATGAAAtgggtgaaa 92 fragment2 cagagccatggcaaaagcctggaatggattggcGATATTAATCC GAACAACGGCGAAACCTTTtataaccagaaatttaaggataaag cgaccc PCR GcaaaagcctggaatggattggcGATATTAATCCGAACAACGGC 93 fragment3 GAAACCTTTtataaccagaaatttaaggataaagcgaccctgac cgtggataaaagcagcagcaccgcgtatatgcagctgaacagcc tgaccagcgaagatagcgcggtgtattattgcgcgaccCGCGTG TTTGATTATtggggccagggtaccaccctgaccg PCR cggtgtattattgcgcgaccCGCGTGTTTGATTATtggggccag 94 fragment4 ggtaccaccctgaccgtgagcagcgcgaaaaccaccccgccgag cgtgtatccgctg gcgccgggcagcgcggcgcagaccaacagcatggtgaccctggg ctgcctggtgaaaggctattttccggaaccggtgaccgtgacct ggaacagcggcagcctgagcagcggcgtgcatacctttccggcg gtgctgcagagcgatctgtataccctgagcagcagcgtgaccgt gccgagcagcacctggccgagcgaaaccgtgacctgcaacgtgg cgcatccggcgagcagcaccaaagtggataaaaaaattggagct ggtgcaggctctggtgctggcgcaggttctccagcagcggtgcc ggcagcagttcctgctgcggtgggcgaaggcgagggagagttca gtacgccagtttggatctcgcaggcacagggcatccgtgctggt cctcagaggctttcc forward GCTACAAACGCGTACGCTATGGAAGTGCAGCTGCAGCAGAGCG 95 primer for cloning

Example 3: In Vitro Display

[0303] The buffers for Fab display were prepared and incubated overnight at 4 C. with end-over-end rotation. Washing buffer, WB, (60 mM Tris; pH 7.5 adjusted with AcOH, 180 mM NaCl, 60 mM magnesium acetate, 5% Blocker BSA, 33 mM KCl, 200 g t-RNA, 0.05% Tween 20); Bead wash buffer BWB (100 mM PBS, 0.1% Tween 20); Stop buffer SB (50 mM Tris pH 7.5 adjusted with AcOH, 150 mM NaCl, 50 mM magnesium acetate, 5% Blocker BSA (Pierce), 33 mMKCl, 0.5% Tween 20, 8.2 mM ox. glutathione); Elution buffer (55 mM Tris pH 7.5 adjusted with AcOH, 165 mM NaCl, 22 mM EDTA, 1 mg BSA, 5000 U rRNA (5000 U), 50 g tRNA).

[0304] The required volume of magnetic beads (streptavidin coated beads) was blocked with 100 L washing buffer (WB) per 10 L initial suspension with end-over-end rotation at 4 C. overnight. 25 L of the beads were used for the prepanning step and 20 L for panning per target/background sample. To remove the sodium azide of the bead storage buffer, the beads were washed four times with bead washing buffer (BWB) and three times with WB. These steps were performed by applying a magnetic field for collecting the beads for two minutes and subsequently discarding the supernatant. After the final washing step the beads were resuspended in WB to their initial volume.

[0305] PUREfrex DS 2.0 was used according to the manufacturer's instructions, to perform in vitro transcription and translation. A 1.5 mL reaction tubes for the target (T) and one for the background (BG) were prepared.

[0306] The DNA input of expression template (LC) and display template (HC) were applied in a 2:1 molecular ratio. The amount of the DNA, coding for display and expression template were kept constant in all Fab display cycles. The in vitro transcription/translation reaction mix was incubated at 37 C. for 1 h. After incubation, the reaction was stopped by adding 100 L stopping buffer, followed by a centrifugation step at 14 000 rpm for 15 minutes at 1 C. Unless otherwise stated, subsequent steps were performed at 4 C. The stopped supernatant of the translation mix was added to the prepared bead suspension and incubated for 30 minutes on a rocking platform. Afterwards, the suspension was centrifuged at 13 000 rpm and 1 C. for 10 minutes to separate the beads with the unspecific binding molecules from the supernatant with the remaining ternary complexes. The prepanned supernatant (300 L) was transferred into a new 2 mL reaction tube, previously blocked with WB, and kept on ice until further use. The target (recombinant biotinylated cTnT) was added to the 300 L prepanned supernatant in a final concentration ranging from 10 nM to 50 nM. The biotinylated cTnT concentration was decreased in every cycle in order to raise the selection pressure. The suspension was incubated for 30 minutes on a rocking platform. The solution panning step allowed the specific binding between the biotinylated cTnT and the ternary complex. Ternary complexes that bound to the target cTnT were captured with streptavidin beads in a 20 minutes incubation step. A further increase of the selection pressure was achieved in cycle III in two ways: Either by decreasing the antigen concentration to 2 nM or by using a non-biotinylated competitor. In the latter, the panning step was implemented with a low biotinylated cTnT concentration and an excess of the competitor cTnT overnight.

[0307] Washing steps comprise the capturing of the beads with the bound target-ternary complexes in a magnetic field, followed by removal of the supernatant. The beads were washed with 500 L ice-cold WB. The selection pressure was increased in subsequent display cycles by extending the duration of the washing steps from 5 minutes to 1 hour. The final washing step was used to transfer the beads to a new blocked 2 mL reaction tube. Subsequently the beads were captured with a magnetic field and the supernatant was removed. The following elution step was performed by adding 100 L of 1EB containing EDTA and incubating for 10 minutes with shaking. The mRNA was released from the ternary complexes. Afterwards the elution mix was centrifuged at 14 000 rpm for 10 minutes at 1 C. The RNeasy MinElute cleanup kit (Qiagen) was used according to the manufacturer's instructions, to isolate and purify the enriched RNA. The RNA was eluted with 16 L RNase-free water. In order to digest any remaining DNA from the selection step, the Ambion DNA-free kit was used according to the manufacturer's instructions. Remaining DNA cannot be amplified in subsequent PCR reactions. After DNase deactivation the suspension was centrifuged for two minutes at 13 000 rpm and at room temperature. The supernatant (50 L) was transferred to a fresh 1.5 mL reaction tube on ice. The purified RNA was immediately used for the reverse transcription (RT). Any remaining supernatant was stored at 20 C.

[0308] The eluted mRNA was reverse transcribed to cDNA. Two reactions were set up for sample T, containing the target in the panning step. Two further reactions were prepared for sample BG and a negative control contained water. According to the number of samples a master mix was prepared and the premix was distributed to 0.2 mL reaction tubes on ice. Each reaction was inoculated with 12 L of the eluted RNA and 0.5 L of the reverse transcriptase. The negative control was implemented with 12 L of RNase free water instead of RNA. The reverse transcription was performed for 45 minutes at 65 C. in a PCR thermo cycler. Subsequently the cDNA samples were incubated for 5 minutes on ice and amplified in the following steps. Remaining sample was stored at 20 C. Two PCR reactions were implemented: The first PCR PCR on RT was performed with the primers Frt and Rrt to amplify the cDNA of the selection pool. The second PCR PCR on RT-PCR using the primers F1A and R1A was applied in order to reattach the regulatory elements for the in vitro transcription/translation. Both reactions were performed with Pwo DNA polymerase.

[0309] In order to provide a sufficient DNA concentration of the selection pool, four reactions were set up for each of the samples T and BG.

[0310] Additionally, four control samples were set up. The first two samples were derived from the DNA digest after the mRNA isolation of samples T and BG and were verified by PCR to amplify potentially remaining DNA. The third and the fourth were the negative control of RT and a negative control on PCR on RT using PCR grade water.

[0311] The PCR product of T was purified from a preparative 1 agarose gel with the QIAquick gel extraction kit, subsequently quantified and used as a template for PCR on RT-PCR. Three reactions of the selection pool and one negative control with PCR grade water instead of the DNA template were prepared. For each reaction, 250 ng of the previous purified PCR on RT were used. The PCR products were purified from a 1% preparative agarose gel with the QIAquick gel extraction Kit and were further modified for following subcloning into an appropriate expression system.

Example 4: Periplasmatic Expression of Enriched Binders

[0312] In order to isolate enriched Fab binders, the murine variable HCs were cloned into the phoATIR3-9bi Fab TN-T M7chim expression vector (see FIG. 2), containing the human CH1 domain, murine VL domain and human CL domain of the Fab. Each selection pool was provided with a BsiWI restriction site located in the leader sequence Tir9 to enable the cloning into the expression vector.

[0313] The second restriction site KpnI occurs at the end of the variable region of the HC and thus does not have to be attached. Therefore, a PCR was performed using forward primer 5 GCTACAAACGCGTACGCTATGGAAGTGCAGCTGCAGCAGAGCG-3 (SED ID NO: 95), containing the BsiWI restriction site and the reverse primer Rrt 5-GGAAAGCCTCTGAGGACCAGCACGGATGCCCTGTGC-3 (SEQ ID NO:88). Periplasmatic Expression was performed in 96-well deepwell blocks (DWBs). The preculture (master) DWBs were filled with 1 mL LB (100 g/mL ampicillin) per well by using the Integra VIAFlo96 and were inoculated with the isolated clones of the previously implemented subcloning and transformation. About 300 colonies per selection pool were picked. One well was left without inoculation as a negative control; another well was inoculated with an XL1 blue transformed TnT M-7 (wildtype) Fab expression vector as a positive control. The DWBs were sealed with air permeable membranes and incubated in an orbital shaker incubator (750 rpm) overnight at 30 C. Subsequently, 50 L from each well of the master DWBs were transferred to new expression DWBs, prepared with 1150 mL C.R.A.P medium (100 g/mL ampicillin) per well as described by Simmons, L. C., Reilly, D., Klimowski, L., Raju, T. S., Meng, G., Sims, P., Hong, K., Shields, R. L., Damico, L. A., Rancatore, P. & Yansura, D. G. (2002) Expression of full-length immunoglobulins in Escherichia coli: rapid and efficient production of aglycosylated antibodies, J Immunol Methods 263, 133-47. The DWBs were sealed with air permeable membrane and were incubated in an orbital shaker incubator at 30 C. The induction of the Fab expression is based on the phoA promotor with the phosphate-limiting C.R.A.P medium. After 24 hours the cells with the expressed Fabs were harvested by centrifugation at 4000 rpm for 10 minutes and stored at 20 C. until further use.

[0314] The preculture master DWBs were used for glycerol stocks by adding 950 L of 40% glycerol and storing at 80 C. Cell pellets were re-suspended in 50 L B-PERII Bacterial Protein Extraction Reagent (Thermo Fisher Scientific) by vigorous vortexing of the sealed DWBs for 5 minutes and shaking for additional 10 minutes at room temperature. The cell lysates were diluted in 950 L Tris buffer (20 mM Tris pH 7.5, 150 mM NaCl) and incubated for 10 minutes before centrifugation (10 minutes, 4000 rpm). The expression Blocks containing the crude cell extract were kept at 4 C. until further use in SPR kinetic investigations.

Example 5: ELISA Screen

[0315] To uncover the best mutant Fab binders for detailed Biacore analyses a previous Enzyme-linked Immunosorbent Assay (ELISA) was implemented. The ELISA setup is depicted in FIG. 3. Biotinylated recombinant cardiac troponin T (100 nM) was captured at a streptavidinMTP 96 well plate for 1 h at RT by shaking on an orbital shaker. The antigen troponin T was diluted in 100 L IP buffer (PBS pH 7.3, 1% BSA). Subsequently, the wells were washed three times with 300 L lx washing buffer (150 mM NaCl, 0.05% Tween20, 0.2% Bronidox) using the microplate washer BioTek ELx405 Select. After washing, the crude cell extracts containing the mutated anti-cTnT Fab binders were diluted 1:2 in IP buffer and transferred to the troponin T captured wells. Again, the wells were washed three times with 300 L 1 washing buffer. The anti-human IgG (Fab specific)peroxidase-labeled antibody (detection antibody) produced in goat was used at 1:40 000 dilution (in IP buffer) to detect the troponin T bound mutated Fab fragments. The wells were again washed three times with 300 L 1 washing buffer to remove unbound detection antibody. The microplates were incubated with 100 L ABTS per well for 30 minutes at RT. The optical density was measured with the microplate reader BioTek Power wave XS set to 405 nm. The wildtype Fab of the parent anti-cTnT antibody was used as a positive control. First hits were identified and the crude cell extract thereof were submitted to kinetic analysis.

Example 6: SPR Based Functional Analyses

[0316] Detailed kinetic investigations were performed at 37 C. on a GE Healthcare T200 instrument. A Biacore CM-5 series S sensor was mounted into the instrument and was preconditioned according to the manufacturer's instructions. The system buffer was HBS-ET (10 mM HEPES (pH 7.4), 150 mM NaCl, 1 mM EDTA, 0.05% (w/v) Tween 20). The sample buffer was the system buffer supplemented with 1 mg/ml CMD (Carboxymethyldextran, Fluka). In one embodiment an anti-human antibody capture system was established on the CMS biosensor. GAHF(ab)2, (goat anti human F(ab)2) (Code Nr.: 109-005-097, lot #13.12.2005, Jackson Immuno Research) was immobilized according to the manufacturer's instructions using NHS/EDC chemistry. 30 g/ml GAHF(ab)2 in 10 mM sodium acetate buffer (pH 5.0) were preconcentrated to the flow cells 1, 2, 3 and 4 and were immobilized with 10.000 RU GAHF(ab)2. The sensor was subsequently saturated with 1 M ethanolamine pH 8.5.

[0317] Chimeric anti-TnT antibody fragments were periplasmatically expressed in E. coli cells as described and were lyzed by methods known (for technical details see: Andersen, D. C. & Reilly, D. E. (2004); Production technologies for monoclonal antibodies and their fragments. Curr Opin Biotechnol 15, 456-62). The lysates were diluted 1:20 in sample buffer. Fab fragments were captured via their humanized framework regions from the expression lysates on the biosensor at a flow rate of 10 l/min for 1 min followed by a 2 min washing step with 10-fold concentrated HBS-EP buffer at 30 l/min. The Fab fragment capture level (CL) in response units (RU) was monitored. Recombinant human TnT (Roche, 37 kDa) was diluted in sample buffer at 90 nM and a concentration series was produced with 0 nM, 30 nM, 11 nM, 3.3 nM, 1.1 nM, 0 nM, 3.3 nM TnT concentration. The analyte concentration series were 80 l/min for 3 min association phase and the dissociation phase was monitored for 3 min.

[0318] At the end of the analyte association phase a report point, binding late (BL) in response units (RU) was monitored. After each cycle of kinetic rates determination the capture system was regenerated by a 15 seconds injection of 10 mM glycine pH 1.5 followed by two 1 min injections of 10 mM glycine pH 1.7 at 20 l/min.

[0319] The kinetic parameters ka [1/Ms], kd [1/s], t1/2 diss [min], KD [M] and the binding stoichiometry (Molar Ratio) (for details see: Schraeml, M. & Biehl, M. (2012); Kinetic screening in the antibody development process. Methods Mol Biol 901, 171-81.) of the cTnT analyte were determined for each Fab fragment mutant with the Biaevaluation Software (GE healthcare) according to the manufacturer's instructions. Kinetic parameters were correlated to the CDR mutation sites and are listed in Table 3 according to their antigen complex stability (t1/2 diss).

[0320] Kinetic parameters were correlated to the mutations identified in the corresponding CDRs. The mutants obtained in this screening all contained more than one amino acid substitution. Mutant Fab-fragments comprising single substitutions as well as various combinations/variations of all substitution identified in the screening were then made and tested. All mutations/combinations tested are listed in Table 2.

TABLE-US-00002 TABLE 2 Overview over all mutants (with corresponding amino acid substitution(s)) that have been construed and analyzed Numbering of Mutants CDR1 CDR2 CDR3 1 Y34F 2 Y34F F60W 3 Y34F F60W V101Y 4 Y34F F60W Y104F 5 Y34F F60W V101Y + Y104F 6 Y34F V101Y 7 Y34F Y104F 8 Y34F V101Y + Y104F 9 Y34I 10 Y34I F60W 11 Y34I F60W V101Y 12 Y34I F60W Y104F 13 Y34I F60W V101Y + Y104F 14 Y34I V101Y 15 Y34I Y104F 16 Y34I V101Y + Y104F 17 F60W 18 V101Y 19 Y104F 20 V101Y + Y104F 21 F60W V101Y 22 F60W Y104F 23 F60W V101Y + Y104F

[0321] All the above mutants have been analyzed by SPR and ranked according to their antigen complex stability (t1/2 diss), (see Table 3).

TABLE-US-00003 TABLE 3 Kinetic data of affinity maturated cTnT antibody Fab-fragments Fab CDR Capture k k t -diss K R.sub.max Fab CDR positions combinations RU 1/Ms 1/s min M RU MR CDR1 CDR2 CDR3 12 77 1.18E+06 .7E04 31 3.2E10 39 0.5 Y34I F60W Y104F 5 76 2.4E+06 5.7E04 20 2.3E10 40 0.7 Y34F F60W V101Y + Y104F 8 75 2.7E+06 5.2E04 22 2.0E10 41 0.7 Y34F V101Y + Y104F 4 71 2.7E+06 7.2E04 16 2.7E10 3 0.7 Y34F F60W Y104F 3 88 2.1E+06 7.9E04 15 3.7E10 49 0.7 Y34F F60W V101Y 7 84 2.4E+06 1.1E03 11 4.5E10 35 0.7 Y34F Y104F 2 101 2.2E+06 1.0E03 11 4.6E10 53 0.7 Y34F F60W 22 47 3.0E+06 1.1E03 10 3.8E10 28 0.7 F60W Y104F 19 49 2.9E+06 1.4E03 8 4.8E10 29 0.8 Y104F 1 105 2.3E+06 1.4E03 8 6.1E10 57 0.7 Y34F parental 119 2.5E+06 1.4E03 8 5.8E10 59 0.6 17 47 2.9E+06 1.6E03 7 5.4E10 27 0.7 F60W 13 42 3.0E+06 1.9E03 6 6.2E10 22 0.7 Y34I F60W V101Y + Y104F 9 74 2.0E+06 2.3E03 5 1.2E09 35 0.6 Y34I 11 59 2.2E+06 2.2E03 5 1.0E09 32 0.7 Y34I F60W V101Y 15 45 2.2E+06 2.3E03 5 1.1E09 23 0.7 Y34I Y104F 23 42 2.8E+06 3.4E03 3 1.2E09 20 0.6 F60W V101Y + Y104F 20 33 3.1E+06 3.4E03 3 1.1E09 15 0.6 V101Y + Y104F 21 42 2.7E+06 4.3E03 3 1.0E09 17 0.5 F60W V101Y 18 45 2.9E+06 4.7E03 2 1.6E09 19 0.5 V101Y 16 35 1.7E+06 5.9E03 2 3.4E09 17 0.6 Y34I V101Y + Y104F indicates data missing or illegible when filed

[0322] Abbreviations in Table 3: ka: association rate constant [M1s1], kd: dissociation rate constant [s1], KD: dissociation equilibrium constant KD [M], t/2-diss: complex half-life, ln(2)/kd*60 [min], Rmax: Response maximum of analyte [RU], MR: Molar Ratio=Ratio of Response maximum (Rmax) of analyte.

[0323] When separately analyzing the individual substitutions comprised in antibody combination 12, i.e. the mutations comprised in numbers, 9, 17 and 19 (see Table 3) it becomes clear, that there is a synergistic effect of the three mutation sites that improves the affinity, complex stability and ECL assay performance of this mutated antibody. This also demonstrates the synergistic effect of the mutations comprised therein.

Example 7: Expression of Chimeric Antibodies in HEK Cells

[0324] Chimeric human/mouse antibodies were obtained according to standard procedures.

[0325] The corresponding vector and the cloning processes are described in Norderhaug et al. J Immunol Methods. 1997 May 12; 204(1):77-87.

[0326] From several Fab fragments selected by SPR full length murine/human chimeric antibodies, i.e. antibodies with a human IgG CHL CH2 & CH3 domains, have been constructed and produced. The cDNAs coding for the heavy and light chains were obtained from hybridoma clone 7.1 A 12.2-22 (ECACC 89060901) by RT-PCR and were cloned into separate vectors downstream of a human cytomegalovirus (CMV) immediate-early enhancer/promoter region and followed by a BGH polyadenylation signal.

[0327] The suspension-adapted human embryonic kidney FreeStyle 293-F cell line (Thermo Fisher Scientific) was used for the transient gene expression (TGE) of the antibody: The cells were transfected at approx. 210E6 viable cells/ml with equal amounts of the both expression plasmids (in total 0.7 mg/L cell culture) complexed by the PEIpro (Polyplus-transfection SA, Strasbourg) transfection reagent according to the manufacturer's guidelines. Three hours post-transfection, valproic acid, a HDAC inhibitor, was added (final concentration: 4 mM) in order to boost the expression. Each day, the culture was supplemented with 6% (v/v) of a soybean peptone hydrolysate-based feed. Seven days after the transfection the culture supernatant was collected by centrifugation and antibodies were purified therefrom according to standard procedures.

Example 8: ECL Measurements

[0328] The antibodies produced according to Example 7 were tested in a sandwich immuno assay (see FIG. 4). IgG Ruthenium conjugates were generated and used in place of and in comparison to the original standard ruthenylated conjugate comprised in the genuine Roche Elecsys assay, catalogue number 05092744190 (Roche Diagnostics GmbH, Mannheim, Germany) in order to compare the performance of the parental anti-cTnT antibody with the mutated anti-cTnT antibodies. The mutated mAbs were conjugated to ruthenium at different labeling stoichiometries. In one embodiment the ruthenium labeling molar ratio was 1:10 antibody IgG:label. The ruthenium conjugates from anti-cTnT antibody variants were diluted in the Elecsys R2 reagent and measurements performed on a Cobas E170 Module using the Troponin T hs assay protocol with a blank control (Diluent Universal, Id. 11732277122, Diluent Multi Assay, Id. 03609987170, Roche Diagnostics GmbH, Mannheim, Germany), Call and Cal2 from Troponin T hs CalSet (Id. 05092752190, Roche Diagnostics GmbH, Mannheim, Germany) using the Troponin T hs assay specifications. Results are given in FIG. 5. Antibodies comprising the mutations present in combinations number 11 and 12, respectively, show an improved signal to noise ratio as compared to the parent (non-mutated) antibody.