Preparation of immunoconjugates by Mxe GyrA intein mediated EPL

In vitro cleavage of L19 IgG-GyrA fusion protein was induced by the addition of 50 mM Mesna which has been reported to form a long-living thioester at the C-terminus. This thioester can be employed to link to the IgG a probe that is functionalized with an N-terminal cysteine residue. It is added concomitantly with Mesna but in lower concentration (5 mM). The cysteine replaces the Mesna-thioester, resulting in a stable peptide bond between the IgG and the probe (Figure ​(Figure11).

To study the cleavage efficiency of the L19 IgG-GyrA fusion protein and to test the antibody integrity after EPL, cysteine was used as 'probe'. After 22h, Mesna and cysteine were removed by intensive dialysis to regenerate and stabilize the disulfide bonds of the antibody. L19 IgG cleaved on both heavy chains could be separated from non- or partly-cleaved antibody by His Tag affinity chromatography (Figure ​(Figure3).3). The yield of completely cleaved antibody was about 60%. To test if its functionality had been damaged by the reducing conditions used during EPL it was intensively analyzed. SDS-PAGE under non-reducing conditions resulted in a single band at 150 kDa (calculated: 146.3 kDa), showing that interchain disulfide bonds were intact (Figure ​(Figure4A).4A). Analytical gel filtration confirmed the integrity of the antibody, eluting in a single peak very similar to unmodified L19 IgG (control antibody) (Figure ​(Figure4B).4B). Finally, interaction studies by Surface Plasmon Resonance (SPR) were employed to evaluate antibody functionality after EPL. The antigen binding part was analyzed by investigation of the interaction with the ED-B domain. Binding kinetics and the dissociation constant (~90 nM) were the same as for the L19 IgG control antibody (Figure ​(Figure4C).4C). To test the functionality of the Fc part, the interaction with the high affinity Fcγ RI/CD64 and with the low affinity Fcγ RIIIA/CD16a was analyzed by SPR. Binding kinetics and the dissociation constants for both Fcγ receptors (Fcγ RI/CD64: ~0.2 nM; Fcγ RIIIA/CD16a: ~60nM) did not change significantly demonstrating preservation of full functionality (see Additional file 1). In summary, these results show that antigen binding as well as effector functions of L19 IgG have not been affected by the EPL process.

Open in a separate window

Figure 3

L19 IgG-GyrA after EPL. Analysis by reducing SDS-PAGE (Coomassie stained). Purified fusion protein was treated with 50 mM Mesna and 5 mM cysteine for 22h at RT. After dialysis, cleaved L19 IgG was purified by His tag affinity chromatography (1: L19 IgG-GyrA before cleavage, 2: after cleavage, before purification, 3: after purification, HC: heavy chain, LC: light chain).

Open in a separate window

Figure 4

Analysis of L19 IgG-GyrA integrity after EPL. (A) Non-reducing (ox) and reducing (red) SDS-PAGE (Coomassie stained). (B) Analytical Gel Filtration. Black: L19 IgG-GyrA before cleavage. Dotted: Purified EPL product. Grey: L19 IgG control antibody. (C) SPR analysis. Results were fitted to a one-site-binding model with a K_D of 91.4 nM.

A side reaction of EPL is the hydrolysis of the thioester instead of its derivatization with cysteine. To estimate the yield of L19 IgG-Cys, a pepsin digest assay was employed. Pepsin can be used to prepare Fab dimers from an IgG1 as it cleaves C-terminal to the interchain disulfide bonds connecting the heavy chains [34]. The Fc part of the antibody is cleaved into several fragments. Mass spectrometry analysis of L19 IgG control antibody after pepsin digest (pH 4.5) revealed that the most C-terminal cleavage site is the linker between CH2 and CH3 domain, leaving the CH3 domain intact (Figure ​(Figure5A).5A). This could be used for the observation of cysteine derivatization: If cysteines are attached C-terminal to the heavy chains they are expected to form an interchain disulfide bond, resulting in a 28 kDa fragment under non-reducing conditions. However, if one or both cysteines are missing, 14 kDa fragments are observed. Non-reducing SDS-PAGE after pepsin digest of L19 IgG-Cys revealed the occurrence of a 28 kDa band which was replaced by a 14 kDa band after reduction (Figure ​(Figure5B).5B). Both bands were excised from the gel and analyzed after tryptic digest with HPLC ESI-Q-TOF. Clearly, only peptides from the CH3 domain were found (Figure ​(Figure5C).5C). Also the C-terminal peptide was observed, carrying the cysteine. Interestingly, the 14 kDa band is hardly detectable on the non-reducing SDS-PAGE gel. Mass spectrometry analysis of a non-reduced sample after pepsin digest confirmed this observation: We detected a 28 kDa fragment but no 14 kDa fragment (Figure ​(Figure5C).5C). We conclude that the fraction of hydrolyzed protein is minor and that cysteine derivatized L19 IgG was obtained in high yield. This would result in a homogeneous drug load of two. In addition, a C-terminal disulfide bond is readily formed. This may stabilize the immunoconjugate and hinders undesired side reactions of thiol groups in the oxidizing environment antibodies are held.

Open in a separate window

Figure 5

High yield of cysteine derivatization after L19 IgG-GyrA in vitro cleavage. Purified EPL product was pepsin digested. (A) Scheme indicating observed pepsin cleavage sites. (B) SDS-PAGE analysis (Coomassie stained). Non-reducing (ox) conditions show the Fab dimer and a prominent 28 kDa band. After reduction (red), a 14 kDa band is observed. (C) Mass spectrometry analysis results, illustrated by the amino acid sequence of L19 IgG Fc domain. A sample after pepsin digest was analyzed by ESI-TOF. Two prominent masses were found: The Fab dimer (calculated: 96061.5, observed: 96061.2 Da) and a fragment (calculated: 27857.1 Da, observed: 27856.3 Da) corresponding to a disulfide connected CH3 domain cleaved at the indicated site (asterisk). Sequence analysis was performed by tryptic in-gel digest of the 28 kDa and 14 kDa band in (B) followed by ESI-Q-TOF. Only sequences of the CH3 domain were observed. Bold: Immunoglobuline domains (CH2 and CH3). Underlined: Sequences found in the 28 kDa and 14 kDa band. Dashed underlined: Sequences only found in the 14 kDa band.

The aim of our work was to use GyrA intein for the attachment of probes functionalized with an N-terminal cysteine residue. To show the usability of the protocol for this purpose, a CysLys-Biotin peptide was added instead of cysteine. After SDS-PAGE and Western blotting, the specific attachment of biotin to the heavy chain was verified by Streptavidin-based staining (Figure ​(Figure6).6). In summary, the described EPL protocol is applicable for the site specific modification of full-length IgG by probes attached to an N-terminal cysteine residue.

Open in a separate window

Figure 6

Biotinylation of L19 IgG-GyrA after in vitro cleavage. EPL was performed with 50 mM Mesna and 5 mM CysLys-Biotin, the product was purified by His tag affinity chromatography and analyzed by reducing SDS-PAGE. Western blotting with Streptavidin-Alkaline-Phosphatase revealed specific biotin derivatization of the heavy chain.

Vectors

L19 IgG was derived by transferring the variable domains of L19 scFv [36] into a human IgG1/к format.

Mxe GyrA (from vector pTXB1, New England Biolabs) or Npu DnaE_N (from vector pSKDuet, provided by Hideo Iwai, University of Helsinki) were cloned behind the heavy chain of L19 IgG, separated by a linker coding for GLEGGSGGSEY. GyrA and DnaE_N were extended C-terminally to include a His tag (GSA-H₁₀ and AS-H₁₀, respectively). Constructs encoding L19 IgG (control, without linker) and L19 IgG fused to GyrA or DnaE_N were transferred into pTT5 vector (National Research Council Canada). In the resulting expression vectors, both, heavy and light chain, were under the control of an hCMV promoter.

DNA encoding DnaE_C from Nostoc punctiforme (MIKIATRKYLGKQNVYDIGVERDHNFALKNGFIASN [32]), followed by a linker (CFNPAGSSGVIM) and Glutathione-S-Transferase (GST, Schistosoma japonicum) was purchased from Geneart (Regensburg, Germany), amplified by PCR and ligated into vector pET29a (Novagen) by NdeI/NotI.

Expression and Purification of L19 IgG constructs

0.5 L of HEK293-6E cells were grown in F17 medium with 4 mM GlutaMAX-I and 0.1% Pluronic F-68 (Invitrogen) to a density of 1.5 - 2.0 × 10⁶ cells/mL and transfected with 500 μg plasmid DNA (L19 IgG constructs) as described [37], with minor modifications. After five days cells were harvested by centrifugation (8000 × g, 5'). The supernatant was concentrated to 50 mL by crossflow filtration (Sartorius) and dialyzed against column buffer (500 mM NaCl, 50 mM Hepes-NaOH, pH 7.5). For purification, Ni-loaded IMAC (elution with imidazole gradient) and subsequent gel filtration was applied.

Expression and Purification of DnaE_C-GST

E. coli BL21-Star (DE3) (Invitrogen) was transformed with vector DnaE_C-GST-pET29a. Medium 2xYT (Difco) containing 50 μg/mL kanamycin was inoculated with an overnight culture and grown at 37°C, 220 rpm. At an OD₆₀₀ of 0.4 temperature was decreased to 16°C. Protein expression was induced with 0.3 mM IPTG at an OD₆₀₀ of 0.8. After 48 h cells were harvested (3,000 × g, 20') and suspended in buffer (500 mM NaCl, 50 mM Tris-HCl pH 7.5, 2 mM β-mercaptoethanol). Cell lysate was obtained by high pressure homogenization (Constant cell disruption systems, Constant Systems Ltd, UK) and subsequent centrifugation (20,000 × g, 30'). Purification of DnaE_C-GST protein was achieved by applying Glutathione Sepharose (500 mM NaCl, 50 mM Tris-HCl pH 7.5, 2 mM β-mercaptoethanol, elution with 10 mM reduced glutathione), followed by gel filtration (500 mM NaCl, 50 mM Hepes-NaOH, pH 7.5).

Intein catalyzed protein modification

For GyrA mediated EPL, 5 μM L19 IgG-GyrA was incubated in 500 mM NaCl, 50 mM Hepes-NaOH, pH 8.0, 50 mM Mesna (2-mercaptoethane sulfonate-sodium), 5 mM cysteine or Bio-P1 Peptide (NEB). For DnaE mediated protein trans-splicing, 5 μM L19 IgG-DnaE_N was mixed with 50 μM DnaE_C Peptide (MVKVIGRRSLGVQRIFDIGLPQDHNFLLANGAIAANCFN, kind gift from Prof. Christian Becker, TU Munich, Germany. The sequence corresponds to DnaE_C from Synechocystis sp. strain PCC6803 (Ssp) which is interchangeable with Npu DnaE_C [32]), DnaE_C-Biotin (MVKVIGRRSLGVQRIFDIGLPQDHNFLLANGAIAANCFNK-Biotin, synthesis by JPT Peptide Technologies GmbH, Berlin, Germany) or DnaE_C-GST in 500 mM NaCl, 50 mM Hepes-NaOH, pH 7.5, 5 mM DTT. After incubation for 22-24h at room temperature (RT), intein activity was stopped by removing reducing agents (Mesna, DTT, cysteine) by dialysis over night at RT. For purification, the sample was loaded onto a Ni-IMAC column (500 mM NaCl, 50 mM Hepes-NaOH, pH 7.5). Completely processed L19 IgG was not in the flow through but was eluted at 30 mM imidazole. Partly- or non-processed L19 IgG-GyrA or -DnaE_N was eluted with 300 mM imidazole. For subsequent analysis, protein was dialyzed against 500 mM NaCl, 50 mM Tris-HCl, pH 7.5 and concentrated to 1-2 mg/mL.

SEC

A sample (~50 μg) was loaded onto a gel filtration column connected to an HPLC instrument (Agilent Technologies 1200 Series) equipped with detectors for refractive indices (RI) and multi angle light scattering (MALS) (Optilab rEX and miniDAWN TREOS, Wyatt Technology Corp., Santa Barbara, CA). To calculate the yield of intein processing, a sample after EPL/PTS but before purification was analyzed. The mass of the excised intein was determined by its RI signal and set into relation to the total loaded mass.

Pepsin digestion and mass spectrometry

L19 IgG was dialyzed against 20 mM NaAcetat pH 4.5. Pepsin (Protea Biosciences) was added to give an enzyme-to-antibody ratio of 1:25 (w:w) and incubated for 24h at 37°C. A sample was directly analyzed by ESI-TOF to assess masses of peptides derived by the Pepsin digest. Further samples were separated by reducing and non-reducing SDS-PAGE and gels were Coomassie stained. Protein bands were excised, de-stained, reduced with DTT, alkylated with iodacetamide and digested with trypsine. Protein sequences were determined by ESI-Q-TOF (MS and MS/MS) and search in Sequest (Finnigan) and Mascot Databases.

Western blot

After SDS-PAGE, gels were blotted onto a nitrocellulose membrane (iBlot, Invitrogen) and incubated for 1h in blocking buffer (TBS with 0.05% (v/v) Tween-20 and 5% (w/v) milk powder). Biotin was detected by incubating with 1 μg/ml Streptavidin-Alkaline Phosphatase (Sigma) in blocking buffer over night. GST was detected by over night incubation with anti-GST antibody (GE Healthcare, 1:5000 in blocking buffer) followed by 1h incubation with anti-Goat IgG-Alkaline Phosphatase (Sigma, 1:1000 in blocking buffer) as secondary antibody. BCIP/NBT (SigmaFAST tablet) was used for membrane staining.