Synthesis of Fluorescently Labeled Antibodies Using Non-Canonical Amino Acids in Eukaryotic Cell-Free Systems

Cell-free protein synthesis (CFPS) enables the development of antibody conjugates, such as fluorophore conjugates and antibody-drug conjugates (ADCs), in a rapid and straightforward manner. In the first part, we describe the cell-free synthesis of antibodies containing fluorescent non-canonical amino acids (ncaa) by using pre-charged tRNA. In the second part, we describe the cell-free synthesis of antibodies containing ncaa by using an orthogonal system, followed by the site-specific conjugation of the fluorescent dye DyLight 650-phosphine. The expression of the antibodies containing ncaa was analyzed by SDS-PAGE, followed by autoradiography and the labeling by in-gel fluorescence. Two different fluorescently labeled antibodies could be generated.


Introduction
ADCs represent one of the most promising strategies in the pharmaceutical industry to treat solid and hematological cancers. Up to date, nine of them are already approved [1]. The concept of ADCs combines the effect of highly specific tumor-targeting immunotherapy with the concept of chemotherapy relying on highly cytotoxic drugs. ADCs are considered as a strategy to provide a better cytotoxicity to immunotherapy and a better specificity to chemotherapy. ADCs are composed of a monoclonal antibody conjugated to a cytotoxic drug by a linker. In the early stages of the development of ADCs, non-toxic fluorescent agents can be used to optimize the conjugation efficiency, to analyze the internalization efficiency, and to locate and quantify the antibody in vitro and ex vivo, prior to the conjugation to the cytotoxic agent.
While full-length Immunoglobulin G (IgG) is mainly produced in the mammalian system [2], such as Chinese Hamster Ovary (CHO) cell lines, the drug is produced either by hemi-synthesis or by chemical synthesis. One strategy to conjugate the latter to the former is the amber suppression technology [3][4][5][6]. This technology is based on the introduction of an amber stop codon into the gene sequence at a desired position. By adding an engineered tRNA and synthetase the ncaa is incorporated in the antibody sequence during its synthesis exactly at the position of the amber stop codon. Following protein synthesis, the drug containing the corresponding reactive group can be conjugated to the antibody by the reactive group of the ncaa. The basis of the amber suppression technology is an orthogonal tRNA/synthetase pair. The orthogonal synthetase is engineered to amino-acylate specifically the ncaa at the 3 0 -end of the orthogonal tRNA. The latter is engineered to recognize the amber stop codon, thus allowing for the incorporation of the ncaa in a site-specific manner. Most importantly, the orthogonal tRNA/ synthetase pair should not show cross-reactivities between endogenous amino acids, tRNAs, and synthetases.
Orthogonal systems and the resulting products can be developed by using a CFPS system [5]. The CFPS system based on CHO lysate contains microsomes derived from the endoplasmic reticulum of CHO cells as previously described [7]. By translocating de novo synthesized proteins into the lumen of these microsomes by using a melittin signal peptide, post-translational modifications such as disulfide bridge formation and glycosylation can be performed [7]. Using this system, full-length post-translationally modified antibodies can be produced within hours, making the screening of orthogonal tRNAs, or synthetases, ncaa, antibody candidates, and ADCs rapid and straightforward [5].
In this chapter, we describe the proof-of-concept for the cellfree synthesis of antibodies containing ncaa. To allow for the sitespecific introduction of ncaa, the chosen model antibody contained an amber stop codon at amino acid position 134 in the CH1 domain of antibody heavy chain (HC). In the first part, we expressed antibodies containing a Bodipy-TMR-lysine, pre-charged on a tRNA. Autoradiography and in-gel fluorescence analysis showed the expression of the fluorescently labeled antibody of interest. In the second part, we expressed antibodies containing p-azido-L-phenylalanine (AzF) by using an orthogonal system, composed of an engineered E. coli tRNA [8] amino-acylated by an engineered E. coli tyrosine synthetase [9]. Autoradiography confirmed the synthesis of suppression and full-length product. Subsequently, the fluorophore DyLight 650-phosphine was conjugated to the antibody by Staudinger ligation. The successful conjugation was shown by in-gel fluorescence.

Materials
Prepare all buffers and solutions using ultrapure water. 4. CHO lysate prepared as described previously [10,11] (see Note 1). Shock-freeze in liquid nitrogen after every usage and store at À80 C.

Cell-free Synthesis and Fluorescence
Labeling of IgG Using Pre-Charged tRNA

Batch-Based CFPS
Batch-based cell-free reactions are carried out as coupled transcription-translation reaction, in which transcription and translation take place simultaneously in the same reaction compartment ("one-pot").
1. Thaw all components of the cell-free reaction on ice. Mix all components thoroughly before usage. Protect pre-charged tRNA Bodipy-TMR-lysine-tRNACUA from light (see Note 4).
4. For fluorescence labeling, supplement the cell-free reaction with 2 μM Bodipy-TMR-lysine-tRNACUA. Protect the translation mixture from light during pipetting and incubation.
5. Add HC and LC encoding plasmid at a final concentration of 60 nM each (see Note 6).
6. Adjust the final volume of the reaction mix with ultrapure water to 50 μL (see Note 7).
7. Mix all components thoroughly and incubate the reaction at 27 C and 500 rpm for 3 h in a thermomixer (see Note 8). After completing the cell-free reaction place reaction vessels on ice and proceed with the procedure described in Subheadings 3.2 and 3.4 (Fig. 2).

Fractionation of Translation Mixtures
1. Centrifuge translation mixtures at 16,000Âg for 10 min at 4 C. Take off the supernatant (SN1) and discard.
2. Carefully resuspend the microsomal pellet in PBS supplemented with 0.2% DDM by pipetting up and down for several times. Incubate the solution for 45 min at room temperature (RT) under intense agitation (see Note 9).
3. Centrifuge the solution at 16,000Âg for 10 min at 4 C to separate soluble antibodies, located in the supernatant (supernatant 2, SN2), from the microsomes. Take off the supernatant (containing soluble antibodies) and place it on ice. Analyze cell-free synthesized antibodies as described in Subheading 3.4.  5. Fill the reaction chamber (red lid) with the complete volume of reaction mix and the feeding chamber (colorless lid) with the complete volume of feeding mix. Insert the chamber into the RTS 500 adapter in a thermomixer. Incubate the reaction at 30 C for 24 h at 1000 rpm. Harvest the reaction mix from the reaction chamber and place the reaction mixture on ice.
6. In order to separate soluble from insoluble protein, centrifuge the translation mix at 16,000 Â g for 10 min at 4 C. Harvest the supernatant containing soluble eAzFRS by pipetting.
8. Equilibrate each column with 2 Â 800 μL washing buffer and add 500 μL of the supernatant from this step to each column.
9. Once the supernatant has completely entered the column, wash each column 5 Â with 200 μL washing buffer (see Note 10).
10. To elute the synthetase, add 6 Â 100 μL elution buffer to each column. Collect each flow-through and analyze separately. Afterwards, pool all fractions containing the target protein.
11. To regenerate the column, add 3 Â 1 mL regeneration buffer to each column, followed by addition of 2 Â 800 μL of 1Â washing buffer. For storage add 2 mL washing buffer and place the columns at 4 C.
12. To exchange the elution buffer to synthetase storage buffer, apply pooled elution fractions to Zeba™ Spin Desalting Columns. First, remove the storage solution of the Zeba™ Spin Desalting Column by centrifugation of the column at 1500 Â g for 1 min. Subsequently, add 300 μL synthetase storage buffer to the resin bed and centrifuge the column at 1500 Â g for 1 min. After repeating this step 2Â place the column into a new collection tube and apply 100 μL of the pooled synthetase solution to each column. Collect target proteins by a final centrifugation step at 2000 Â g for 2 min.
13. Concentrate target protein by using Amicon Centrifugal Filter Devices 0.5 mL. Adjust the volume of the synthetase from step 12 to 500 μL with storage buffer and add this solution to the concentrator. Centrifuge at 14,000 Â g for 10 min and 4 C. Collect the flow-through. The concentration of the flowthough can be determined by photometric measurement using NanoDrop based on the calculated molecular mass of the synthetase (48.6 kDa) and the extinction coefficient (54.3) (see Note 11). Store synthetase at À80 C after shock freezing in liquid nitrogen.

Preparation of Orthogonal tRNA
1. First, suitable DNA templates of the tRNA gene need to be generated. For this, a reverse primer containing a 2'-OMe group has to be used in order to prevent unspecific addition of nucleotides to the 3 0 end by the T7 RNA polymerase.   2. Analyze labeling reaction by autoradiography and in-gel fluorescence as described in Chap. 3, Subheading 4. (Fig. 3).

SDS-PAGE and Autoradiography
1. Take a 6 μL aliquot of the sample (e.g., TM, MF, SN2) and mix with 6 μL 2Â non-reducing LDS sample buffer. Incubate on a shaker for 15 min at RT (see Note 17).
3. After electrophoresis, remove the gel from the plastic cassette and put in 100 mL water for in-gel fluorescence analysis (3.4.2).

In-Gel Fluorescence Analysis and Autoradiography
1. Analyze the gel using Amersham Typhoon RGB Biomolecular Imager. Place the gel onto the scanning surface and scan using the following parameters: 633 nm extinction and 670 nm emission for DyLight-phosphine and 532 nm extinction and 580 nm emission for Bodipy-TMR-lysine.
2. Rinse the gel three times with deionized water to remove SDS and buffer salts.
3. Dry gels for 90 min at 70 C using a vacuum filtration system.

4.
Label the marker bands of the dried gel with radioactive ink.
5. Place dried gel into a phosphorimager cassette and incubate for at least 3 days. 6. Scan the screens using the Typhoon Trio + Variable Mode Imager or Amersham Typhoon RGB Biomolecular Imager.
1. CHO lysates were prepared as described previously [10]. In brief, CHO cells were grown in a Biostat B-DCU II bioreactor (Sartorius Stedium Biotech GmbH) at 37 C using a chemically defined and serum-free cell medium. Cells were grown to a density of 3.5-5Â 10 6 cells/mL and harvested by centrifugation at 200 Â g for 5 min. Cells were washed twice and resuspended in a HEPES-based homogenization buffer (40 mM HEPES-KOH (pH 7.5), 100 mM NaOAc and 4 mM DTT). Resuspended CHO cells were lysed mechanically by applying a 20-gauge needle and a syringe. By using the syringe, cells were manually passed through the needle. After cell disruption, the homogenate was centrifuged at 6500 Â g for 10 min to remove cell nuclei and debris. The resulting supernatant was applied to a Sephadex G-25 column which was equilibrated with homogenization buffer. Elution fractions with the highest RNA/protein ratios were pooled. In order to remove the endogenous mRNA, cell lysates were mixed with S7 micrococcal nuclease (f.c. 10 U/mL) and CaCl 2 (f.c. 1 mM) and incubated for 2 min at RT. Inactivation of micrococcal nuclease was achieved by the addition of EGTA (f.c. 6.7 mM). Afterwards, creatine kinase (f.c. 100 μg/mL) was added to the lysate in order to ensure the regeneration of ATP out of creatine phosphate. Aliquots of the CHO lysate were immediately shock frozen in liquid nitrogen and subsequently stored at -80 C until further usage.
2. Coding sequences of HC and LC should be N-terminally fused to the melittin signal sequence to allow for the translocation of de novo synthesized polypeptide chains into the lumen of the microsomal vesicles [7]. Furthermore, HC and LC sequences should be fused to regulatory sequences necessary for CFPS (Fig. 1). The 5 0 untranslated region (5 0 UTR) of HC/LC templates contained a T7 promotor sequence and an internal ribosomal entry site (IRES from the intergenic region (IGR) of the Cricket paralysis virus (CrPV), (Genbank accession no. AF218039, nucleotides 6025-6216)) as regulatory sequences to allow for efficient transcription based on T7 RNA polymerase and factor-independent translation initiation, respectively. The 3'UTR contained a T7 terminator sequence and a multiple cloning site for subsequent cloning. HC/LC sequences were synthesized de novo by Geneart (Life technologies, Thermo Fisher) and cloned into pMA vector backbone. 8. In general, the optimal temperature of the CHO cell-free system is 30 C [11] but the optimal incubation temperature may be different for different proteins. For antibody synthesis, 27 C was found to result in the highest yields of active antibodies.
9. Cell-free synthesized antibodies which have been translocated and trapped inside the lumen of the microsomes can be released by re-solubilization of the microsomal vesicles using PBS supplemented with 0.2% DDM. It is important to thoroughly resuspend the microsomes within the buffer in order to release translocated antibodies quantitatively. 10. We find that it is beneficial to collect all fractions throughout the purification, buffer exchange, and concentration procedure (e.g., flow-through, washing fractions, elution fractions). Aliquots of these solutions should be diluted in SDS-PAGE sample buffer and analyzed by SDS-PAGE in order to monitor the purity of the aminoacyl-tRNA-synthetase during the preparation procedure. 11. We recommend to concentrate the synthetase up to a concentration of 5 g/L to ensure a minimal final concentration of 100 μM. If necessary, repeat the concentration step.
12. We purify the PCR product using QIAquick PCR Purification Kit and determine the concentration of the PCR product by using a NanoDrop 2000c. For further analysis, prepare a 1% (w/v) agarose gel and load 1 μL of the PCR product. The expected band size is 123 bps.
13. It is important to work in an RNase-free environment. Use RNase-free filter tips and reaction vessels.
14. Prepare a 2% (w/v) agarose gel. For sample preparation, mix 2 μL of the RNA with 6 μL MOPS sample buffer and load the sample to the agarose gel. Use an RNA ladder. The expected band size is around 200 bps.
15. After centrifugation, three phases will be visible: the aqueous phase on top with approximately 50% of the total volume containing the RNA; a middle interphase which is nearly invisible and below the red phenol/chloroform phase. Try to isolate only the aqueous phase.
16. Reactive groups of ncaa are often sensitive to light and might become instable upon light exposure. Thus, protect solutions from light by using colored tubes or wrap the tubes with aluminum foil.
17. The use of non-reducing sample buffer is important to maintain the disulfide bonds which connect the polypeptide chains of the antibodies. Heating of samples before gel electrophoresis is not necessary.