Improving the Developability of an Antigen Binding Fragment by

Subscriber access provided by UNIV OF SOUTHERN INDIANA

Article

Improving developability of an Antigen Binding Fragment by Aspartate Substitutions Laila Ismail Sakhnini, Per Greisen, Charlotte Wiberg, Zoltan Bozoky, Søren Lund, Adriana-Michelle Wolf Pérez, Hanne Sophie Karkov, Kasper Huus, JensJacob Hansen, Leif Bülow, Nikolai Lorenzen, Maria Dainiak, and Anja K Pedersen Biochemistry, Just Accepted Manuscript • Publication Date (Web): 22 May 2019 Downloaded from http://pubs.acs.org on May 24, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Improving developability of an Antigen Binding Fragment by Aspartate Substitutions Laila I. Sakhnini1,2*, Per J. Greisen1, Charlotte Wiberg1, Zoltan Bozoky3, Søren Lund1, AdrianaMichelle Wolf Perez1, Hanne S. Karkov1, Kasper Huus1, Jens-Jacob Hansen1, Leif Bülow2, Nikolai Lorenzen1, Maria B. Dainiak1*, Anja K. Pedersen4* 1Global

2Department

3Current

Research Technologies, Novo Nordisk A/S, Måløv, Denmark of Pure and Applied Biochemistry, Lund University, Lund, Sweden

Affiliation, Canada’s Michael Smith Genome Sciences Centre, BC Cancer, Vancouver Canada

4Chemistry,

Keywords:

Manufacturing and Control, Novo Nordisk A/S, Gentofte, Denmark

Protein

aggregation,

Aspartate

substitutions,

antigenic

binding

affinity,

complementary determining regions, thermodynamic stability, Antigen binding fragment Corresponding Authors *Laila I. Sakhnini ([email protected]); Anja. K. Pedersen ([email protected]); Maria Dainiak ([email protected]).

ACS Paragon Plus Environment

1

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 33

Abstract Aggregation can be a major challenge in the development of antibody-based pharmaceuticals as it can compromise the product quality during bioprocessing, formulation and drug administration. To avoid aggregation, developability assessment is often run in parallel with functional optimization in the early screening phases to flag and de-select problematic molecules. As developability assessment can be time- and resource demanding, there is a high focus on development of molecule design strategies to engineer molecules with a high developability potential. Previously, Dudgeon et al. (2012) demonstrated how Asp substitutions at specific positions in human variable domains and single-chain variable fragments could decrease the aggregation propensity. Here, we have investigated whether these Asp substitutions would improve the developability potential of a murine antigen binding fragment (Fab). A full combinatorial library consisting of 393 Fab variants with single, double and triple Asp substitutions was first screened in silico with Rosetta, thereafter 26 variants with highest predicted thermodynamic stability were selected for production. All variants were subjected to a set of developability studies. Interestingly, most variants had on-par or improved thermodynamic stability relative to wild-type. 25 of the variants were improved with regards to non-specificity. Half of the variants exhibited improved aggregation resistance. Strikingly, while we observed remarkable improvement in the developability potential, the Asp substitutions had no substantial effect on the antigenic binding affinity. Altogether, by combining the insertion of negative charges and the in silico screen based on computational models, we were able to improve the developability of the Fab in a fast manner.


2

Page 3 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Introduction Monoclonal antibody (mAb)-based reagents are widely used as therapeutic drugs1,2, probes in diagnostic applications3,4,5, and affinity ligands in preparative downstream processes6 in the biopharmaceutical industry. In addition, fragments of mAbs, such as single-chain variable fragment (scFv) and antigen binding fragment (Fab), are of high interest as their smaller size can reduce steric hindrance7,8,9, improve column capacity during downstream processing6, and be used for multivalent formatting10. In the development of pharmaceutical antibodies, there are several important attributes, apart from desired antigen binding and functionality, that need to be achieved such as low aggregation propensity, low non-specificity, and high thermodynamic stability11. The measurement of these biophysical properties during early-stage screenings is known as developability assessment11,12,13. Protein aggregation is considered a major challenge as it can compromise product quality, safety, and efficacy14, e.g. enhanced immune responses in patients15. Protein aggregation is a complex process as proteins often form different types of aggregates following different pathways depending on intrinsic factors, such as amino acid sequence and structure, as well as extrinsic factors such as protein concentration, pH, temperature, and excipients16,17,18,19. It has been reported that interfaces for protein-protein interactions, such as the complementarity determining regions (CDRs) of antibodies, can mediate aggregation20,21,22,23,24,25. A common strategy to decrease the aggregation propensity of antibodies is the insertion of charged residues. For instance, Miklos et al.26 generated supercharged scFvs with increased aggregation resistance by substitution of surface-exposed residues to charged amino acids. NonCDR positions were targeted and the mutations were selected based on computational models in Rosetta to avoid the introduction of unfavorable interactions26. Perchiacca et al.27 reported


3

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 33

insertion of negatively charged residues near the edges of the hydrophobic CDR3 loop of a human variable heavy domain (VH) to reduce aggregation propensity. The key finding was that the surface net charge of the scaffold is the determinant of the charge-type of the mutations in the CDRs. Thus, if the scaffold has a positive surface net charge, then positively charged mutations in the CDRs will increase aggregation resistance and vice versa27. In another report, Perchiacca et al.28 found that aggregation resistance was not only dependent on net surface charge; highly specific charge mutations at the positions 29 and 31-33 of VH conferred high aggregation resistance of human VH antibodies28. Dudgeon et al.29 proposed a general strategy for the generation of human variable domains with increased aggregation resistance by aspartate (Asp) substitutions at specific positions clustering in the CDR1 of heavy chain (CDR1-H) and CDR2 of light chain (CDR2-L). These positions were identified by phage display screening of which the library design was based on the most common germline families in the human repertoire (VH3 and Vκ1). High affinity binding to antigen was retained by exclusion of mutations in the CDR3H, where the main antigenic contacts were located. The study showed preference for Asp over other charged residues for the reduction of aggregation propensity, which was measured as the binding activity after heating to 80 °C29. In this study, we investigated whether Asp substitutions at the specific positions reported by Dudgeon and co-workers could improve the developability of a murine Fab. The approach consisted of the following steps: (i) assembly of a full combinatorial in silico library of single, double, and triple Asp substitutions in the FRs and CDRs, (ii) a priori selection of mutations from the assembled library by in silico structural modelling in Rosetta for retention of thermodynamic stability, and (iii) production of selected Fab variants for developability assessment of antigenic affinity, thermodynamic stability, non-specificity and aggregation


4

Page 5 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

propensity. We report on the developability potential of Asp substitutions for decreasing nonspecific binding, and aggregation propensity while retaining or improving antigenic binding affinity. We show that the number of variants for in vitro assessment can be considerably reduced by performing a structural pre-screen based on computational models.

Materials and Methods Materials Expi293™ Expression System Kit was obtained from Life Technologies (USA). Corning™ Disposable Vacuum Filter (1-L scale, 0.22 µm), CaptureSelect™ Biotin Anti-LC-kappa (Murine) Conjugate and materials for SDS-PAGE and IEF gel were obtained from Thermo Fisher Scientific (USA). IEF Marker 3-10 (Liquid Mix) was obtained from Serva (Germany). Vivaspin 6 (MWCO: 10 kDa) was obtained from Sartorius Stedim Biotech. MEP HyperCel and Eshmuno HCX Media were obtained from Pall (USA) and Merck (USA), respectively. SP Sepharose HP, HiLoad 16/600 Superdex 200 pg, PD-10 desalting and PD MiniTrap G-25 columns and Biotin CAPture Kit (Series S) were obtained from GE Healthcare (Sweden). Chemicals used for buffer preparation were purchased from Sigma Aldrich (USA) unless otherwise stated. Structural Modelling To generate a reliable homology model, the top 20 closest sequential homolog antibody structures were searched in the Protein Data Bank (PDB) for both the heavy and the light chain variable domain sequences. Sequence alignments were carried out using Muscle (version 3.8.31; Edgar, R.C. (2004) BMC Bioinformatics, (5) 113.). These sequences served as inputs for homology modelling using MODELLER (version 9.15, BIOVIA). The best antibody model was


5

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 33

selected among the models based on their energy minimization scores calculated by Rosetta version 3.7. The following mutations were fully combined as double- and triple-substitutions; 28/30/31/32/33/35D in heavy chain 1 (CH1) and 24/45/49/50/51/52/53/56D in light chain (LC). The mutations were introduced into the homology model and calculation of the relative difference in Gibbs free energies (ddG) between mutant and wild-type (WT) was carried out using Rosetta version 3.7 with an energy function30 optimized towards thermodynamic properties. Single- and double/triple-substitutions causing more than 5 and 1.5 Rosetta Energy Units (REU), respectively, decrease in ddG were eliminated. Amino acid numbering was based on Kabat numbering31. Protein Expression DNA sequences encoding WT anti-protein X LC and WT anti-protein X CH1 were designed based on the DNA sequences of a full-length anti protein X antibody raised in mice. DNA sequences containing the following mutations were ordered in the pBOK85_S250 vector as plasmid preparations (0.5-10 mg) from Thermo Fisher Scientific GENEART (Germany); [T28D], [A30D], [S31D], [T28D;S31D] in HC and [K45D], [K50D], [S52D], [N53D], [S56D], [K50D;S56D], [S52D;S56D], [N53D;S56D], [K50D;N53D], [K50D;S52D] in LC. The plasmid constructs encoding anti-protein X LC was co-transfected with plasmid constructs encoding antiprotein X HC into the Expi293™ Expression System (Thermo Fischer Scientific, USA) in a 1:1 ratio. This was done for the WT and mutant sequences. The transfections were made following the instructions of the supplier of the expression system in a 0.5-L expression scale. The transfected HEK293 EXPI cells were grown for five days at 8 % CO2 and 36.5 °C in a shaker incubator. At day five, supernatants were harvested by centrifugation, and sterile filtered through a 0.22 µm polyethersulfone filter system (EMD Millipore, USA).


6

Page 7 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Protein Purification Harvested Fab variant supernatants from HEK293 cells were adjusted to pH 4.5-5.0 using 1 M hydrochloric acid, centrifuged (1000 × g, 20 min) and sterile filtered for removal of precipitate. Capture of Fab variants was performed by multimodal chromatography using MEP HyperCel (26 mm × 5.7 cm; column volume 30 mL) with linear flow velocity of 34 cm/h (3 mL/min). Running buffers were equilibration buffer (54 mM sodium dihydrogen citrate, 45 mM disodium phosphate; pH 7.5), wash buffer (54 mM sodium dihydrogen citrate, 45 mM disodium phosphate, 1 M sodium chloride; pH 7.5) and elution buffer (54 mM citric acid, 45 mM sodium dihydrogen phosphate; pH 3.0), respectively. Prior to loading, sodium chloride was added to the harvest materials to a final concentration of 1 M in order to promote hydrophobic interactions. The Fab variants were eluted from the multimodal resin by isocratic elution (90 % elution buffer) and the collected elution fractions were pooled and adjusted to pH 4.5-5.0 using 0.5 M disodium phosphate (pH 9.0). Elution fractions that contained precipitate were centrifuged (1000 × g, 30 min) and sterile filtered prior to pooling. Captured Fabs were purified by cation exchange (CIEX) chromatography using SP Sepharose HP (16 mm × 2.7 cm; column volume 5.4 mL) with linear flow velocity of 90 cm/min (3 mL/min). Running buffers were equilibration buffer (25 mM sodium acetate, 10 mM acetic acid; pH 4.5-5.0) and elution buffer (25 mM sodium acetate, 10 mM acetic acid, 1 M sodium chloride; pH 5.0), respectively. Prior to loading, the capture pools were dialyzed into the equilibration buffer overnight at 4 ºC using Slide-A-Lyzer™ G2 Dialysis Cassettes (10K MWCO, 70 mL, Thermo Fisher Scientific, USA) to achieve a conductivity < 3 mS/cm. The Fab variants were eluted from the CIEX resin by linear elution gradient (0-30 % elution buffer for 50 column volumes) and the collected elution fractions were pooled and buffer-exchanged into a neutral


7

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 33

buffer (20 mM sodium phosphate, 140 mM sodium chloride; pH 7.4) using PD-10 desalting columns (GE Healthcare, Sweden). All centrifugations were performed with the Heraeus Multifuge 3SR Plus from Thermo Fisher Scientific (USA). Qualitative and quantitative analysis of all purified proteins was performed by SDS-PAGE, SE-UPLC and LC-MS analyses. SDS-PAGE SDS-PAGE analyses were performed using precast NuPAGE 4-12 % Bis-Tris Gels (1.0 mm × 10 and 12 wells), NuPAGE MES SDS Running Buffer 20X, Sample Buffer 4X, Sample Reducing Agent 10X and Mark12 Unstained Standard from Thermo Fisher Scientific (USA) according to the instructions of the manufacturer. 2 µg of protein was loaded to each well. SE-UPLC Analyses Monitoring the ratio of high molecular weight proteins (HMWP) to monomers as well as determination of protein concentrations were carried out by SE-UPLC analyses (Waters Acquity H Class UPLC, Waters Corporation, USA) on a ACQUITY UPLC Protein BEH SEC (200 Å, 1.7 µm, 4.6 mm × 150 mm) column from Phenomenex (USA). The flow rate was 0.4 mL/min and the temperature was 25 ℃. The running buffer was 20 mM sodium phosphate, 100 mM sodium chloride and 5% isopropanol (pH 6.8). Sample volume was 10 µL. The HMWP were defined as entities migrating with a retention time, tR, lower than tR for the monomeric protein. The amount of protein was calculated based on the area under the curve monitored at 280 nm relative to a standard solution. LC-MS Analyses LC-MS analysis was performed on an Agilent 6224 TOF LC/MS system (Agilent Technologies, USA). The LC system was an Agilent 1200 Series with a MassPREP Micro


8

Page 9 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Desalting column from Waters Corporation (USA). The flow rate was 0.4 ml/min and the temperature was kept at 20 °C. Buffer A was 0.1 % formic acid and buffer B was 0.1 % formic acid in acetonitrile. The column was equilibrated in 5 % buffer B for 3 min and the protein was eluted with a linear gradient of 5-60 % buffer B over 1 min. Buffer B was increased to 95 % and the column was regenerated with four consecutive runs from 95 % (0.1 min) to 5 % (0.9 min) buffer B. Prior to analysis of purified Fab fragments, deglycosylation was carried out by incubation of the 30 μL sample containing Fab fragments with 2.5 μL of N-glycosidase F (Roche, Switzerland) for 1 h at 37 ºC. Surface Plasmon Resonance Kinetic studies of Fab variants were performed using Surface Plasmon Resonance (SPR) by the Biacore 8K instrument from GE Healthcare (Sweden). CaptureSelect™ Biotin Anti-LCkappa (Murine) Conjugate (10 µg/mL) from Thermo Fisher Scientific (USA) was captured in both flow cells of all 8 channels on Series S Sensor Chip CAP (GE Healthcare, Sweden) to 3 minutes injection at a flow rate of 10 µL/min. Murine Fab variants (3 nM) were subsequently injected for 60 s at a flow rate of 5 µL/min, over flow cell 2 of either channel respectively, resulting in capture levels of 20-30 RU in flow cell 2 and leaving flow cell 1 of each channel as a reference with only CaptureSelect™ Biotin Anti-LC-kappa Conjugate bound. Target protein X was diluted to 0, 20 and 40 nM and injected over the different Fab variants in duplicate at a flow rate of 30 µL/min to both flow cells for 240 s. The dissociation was monitored for 600 s. The WT Fab was included for target protein X binding in all channels first and last of each run as well as in one channel in each cycle throughout the run as a control. Running buffer was 50 mM Tris-Buffered Saline (TBS) at pH 7.4 containing 5 mM calcium chloride, 1 mg/mL bovine serum albumin (BSA) and 0.005 % Tween-20. Conditioning and regeneration of the chip was


9

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 33

performed according to the manufacturer’s instructions. Obtained data from association and dissociation of target protein X were subjected to double referencing and fitted to 1:1 Langmuir binding model using the Biacore 8K Evaluation Software (GE Healthcare, Sweden). Isoelectric Focusing Determination of the isoelectric point (pI) of the Fab variants was performed by isoelectric focusing (IEF) using precast Novex® pH 3-10 IEF Protein Gel (1.0 mm, 10 wells), Novex® IEF Sample Buffer pH 3-10 (2X), Novex® IEF Anode Buffer (50X), Novex® IEF Cathode Buffer pH 3-10 (10X), Colloidal Blue Staining Kit, IEF Marker 3-10, fixation solution (12 % trichloroacetic acid, 3.5 % sulfosalicylic acid) and 96 % ethanol according to the instructions of the manufacturer. Invitrogen PowerEase 500 from Thermo Fisher Scientific (USA) was used as a power source to run the gel. 2 µg of protein was loaded to each well. Differential Scanning Fluorimetry Determination of thermal stability was performed by differential scanning fluorimetry using Prometheus NT.48 (NanoTemper Technologies, Germany). 10 µL of each protein sample (0.5 mg/mL) was loaded into nanoDSF grade standard capillaries and then subjected to a temperature ramp from 20 to 95 °C with a heating rate of 1.5 ºC/min. Unfolding was measured with tryptophan fluorescence using the ratio of florescence at 350 and 330 nm. The midpoint of the thermal unfolding reaction (Tm) was determined from the first derivative spectrum by fitting the maximum using the software PR control (NanoTemper Technologies, Germany). Non-Specificity SE-UPLC* Analyses Investigation of non-specific interaction was performed by SE-UPLC* analyses, defined here as analyses in absence of organic solvents and high concentrations of salts. The column was an


10

Page 11 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

ACQUITY UPLC Protein BEH SEC (200 Å, 1.7 µm, 4.6 mm × 150 mm) column from Phenomenex (USA) and the system was Waters Acquity H Class UPLC (Waters Corporation, USA). The running buffer was 20 mM sodium phosphate (pH 6.8). The flow rate was 0.4 mL/min and the temperature was 25 ℃. Sample volume was 10 µL. The SE-UPLC* retention was monitored with UV detection at 280 nm. Cross-interaction chromatography Investigation of non-specific interaction to the IgG pool purified from human serum (Sigma Aldrich) was performed by cross-interaction chromatography (CIC). The CIC column was prepared as described by Wolf Pérez et al.32 and set up on Waters Alliance HPLC system (Waters Corporation, USA). Running buffer was 20 mM sodium phosphate, 140 mM sodium chloride (pH 7.4). 5 µg of protein was injected to the column. The flow rate was 0.1 mL/min and the run time was 30 min with UV detection at 280 nm for monitoring of column retention. Storage Stability Study 65 µL of Fab sample (1 mg/mL) in triplicates in 96-well PCR semi-skirted plates (Thermo Fischer Scientific, USA) were incubated at 45 ºC using the Applied Biosystems Veriti Thermal Cycler PCR instrument (Thermo Fischer Scientific, USA). Adhesive PCR plate seals (Thermo Fischer Scientific, USA) were used to seal the PCR plates. The lid temperature was set to 55 ºC to prevent evaporation. After 6 days, the samples were analyzed according to the described SEUPLC method. The percentage of HMWP was calculated based on the area under the curve monitored at 280 nm relative to the sample prior to incubation at 45 ºC. Variants were ranked based on number of standard deviations from the HMWP level of Fab WT. The standard


11

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 33

deviation was calculated as an average of the sample standard deviations of the obtained HMWP levels (n = 3) for all the variants. In Silico Protein Surface Property Calculation The spatial aggregation propensity (SAP) was calculated in BIOVIA Discovery Studio 2017 (Accelrys, USA). A radius of 10 Å was selected as the most appropriate for identifying the hydrophobic patches as previously reported33.

Results and discussion In silico Screen of Asp Substitutions A full combinatorial in silico library of single, double, and triple Asp substitutions in positions 28/30/31/32/33/35 in VH and positions 24/45/49/50/51/52/53/56 in VL of a murine model Fab was assembled and screened using Rosetta. Positions were selected based on the Dudgeon et al. study29, adding one additional variant, [K45D]-FR2 of VL, since it was of interest to investigate whether there was a preference for Asp over Lys as previously reported by Dudgeon and coworkers for decrease of aggregation propensity. Based on computational models (see Methods), the relative difference in Gibb’s free energy between mutant and WT (ddG) was calculated. A total of 26 different single- and double/triple-mutations with ddG values of < 5 and < 1.5 REU, respectively, were selected, adding one additional one [S31D;K45D]-VH/VL, for production and biophysical characterization (see Figure 1A-B). The aim was to retain the thermodynamic stability of the Fab by selecting the most favorable mutations predicted by the computational models (see calculated ddG values can be found in Table S1 in Supporting Information).


12

Page 13 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 1. (A) A graphical overview of the obtained ddG values for the in silico Asp substitution library is illustrated in a scatter plot. Light and dark blue dashed lines correspond to the screening criteria for single (< 5 REU, light blue dots) and double/triple (< 1.5 REU, dark blue dots) Asp substitutions, respectively. (B) The molecular structure of the variable domain of the Fab is visualized with the selected Asp substitutions (red spheres). Quality of Produced Variants All Fab variants were successfully produced and assessed qualitatively and quantitatively by SE-UPLC and SDS-PAGE. Obtained chromatograms showed monodispersed Fab variants with purity > 98.5 % (see Table S2 in Supporting Information). In addition, a high level of purity could be observed according to SDS-PAGE, with some variants showing one vague band of molecular weight between 31-21.5 kDa, corresponding to free LC (see Figure S1A in Supporting Information). The identities of the produced variants were confirmed by LC-MS analyses (data not shown). Based on IEF gel, all Asp substitutions confirmed to be surface-exposed as negative


13

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

shifts in pI were observed relative to WT (see Table 1). In addition, all variants showed a similar heterogeneity pattern, i.e. two main charge isoforms with an average difference of 0.1 in pI (see Figure S1B in Supporting Information). Table 1. Isoelectric points (pI), binding properties (kinetic parameters ka, kd and equilibrium constant KD), thermodynamic stability (Tm) and storage stability study (HMWP content after incubation at 45 ºC for six days) for Fab WT and the variants selected from the in silico structural screen. The variants are ranked based on the number of standard deviations from WT. Storage stability study ka kd KD Tm %HMWPd [°C] [%] [ ∙ 106 M-1 s-1] [ ∙ 10 ―2 s-1] [ ∙ 10 ―6 mM] 1 Fab [S31D;N53D]-VH/VL 5.7 1 2 11 63.0 0.6 ± 0.1 1 Fab [S31D]-VH 6.7 1 1 12 62.0 0.7 ± 0.2 1 Fab [T28D;N53D;S56D]-VH/VL 5.2 2 2 13 63.3 0.7 ± 0.3 1 Fab [T28D;S52D;S56D]-VH/VL 5.2 2 2 10 62.9 0.8 ± 0 1 Fab [T28D]-VH 6.7 2 2 12 62.1 0.9 ± 0.9 1 Fab [S31D;S56D]-VH/VL 5.7 1 1 11 62.5 0.9 ± 0 1 Fab [T28D;N53D]-VH/VL 5.7 2 2 13 62.9 0.9 ± 0.1 1 Fab [T28D;S56D]-VH/VL 5.7 2 2 11 62.5 0.9 ± 0.2 1 Fab [T28D;S31D;S56D]-VH/VL 5.3 1 1 7 62.6 1.1 ± 0.2 1 Fab [N53D;S56D]-VL 5.6 2 3 18 62.8 1.1 ± 0.1 1 Fab [N53D]-VL 6.6 1 3 18 62.5 1.2 ± 0.5 1 Fab [S31D;N53D;S56D]-VH/VL 5.2 2 2 11 63.3 1.2 ± 0.1 1 Fab [T28D;S31D]-VH 5.8 1 ± 0.3 1 ± 0.2 7 ± 0.2 61.9 1.2 ± 0.3 2 Fab [S31D;K45D]-VH/VL N/A 2 1 10 62.0 1.6 ± 0.2 2 Fab [S56D]-VL 6.6 1 2 17 62.0 1.6 ± 0.1 2 Fab [S52D]-VL 6.6 2 2 15 61.8 1.9 ± 0.3 2 Fab [A30D]-VH 6.9 1 2 15 60.3 2.0 ± 0.8 2 Fab WT 7.5 1 ±0.3 2 ± 0.3 17 ± 0.2 61.5 2.0 ± 0.5 3 Fab [K45D]-VL 5.6 1 2 17 61.1 3.1 ± 0.7 3 Fab [T28D;S31D;K50D]-VH/VL 5.2 2 ± 0.2 2 ± 0.1 9 ± 0.1 58.3 3.2 ± 0.5 3 Fab [S31D;K50D]-VH/VL 5.7 1 2 17 58.4 3.7 ± 0.2 3 Fab [S31D;K50D;S56D]-VH/VL 5.2 1 2 15 59.0 4.0 ± 0.4 3 Fab [T28D;K50D;S56D]-VH/VL 5.4 2 4 18 59.3 4.1 ± 0.5 3 Fab [T28D;K50D]-VH/VL 5.6 1 ± 0.4 3±1 20 ± 0.3 58.7 4.1 ± 0.4 3 Fab [K50D]-VL 6.4 1 3 32 56.9 4.5 ± 0.5 3 Fab [T28D;K50D;S52D]-VH/VL 5.4 2 4 19 58.7 5.3 ± 0.9 3 Fab [T28D;K50D;N53D]-VH/VL 5.3 2 2 13 59.6 7.8 ± 0.8 3 Fab [K50D;S56D]-VL 5.6 2 ± 0.5 5 ± 1.1 27 ± 1.3 57.2 12 ± 0.6 aRanking of the variants was performed based on the number of standard deviations (n ) from WT in terms of average percentage σ of HMWP; rank (1) better than WT (2 ≤ nσ < 4), rank (2) on-par with WT (0 ≤ nσ < 2), rank (3) worse than WT (nσ < 0). bData is presented as average of two charge isoforms (see Figure S1B in Supporting Information). cData from Biacore SPR is presented as mean values of separate experiments (n = 2). For statistical support of data, Fab wt and four variants, [T28D;K50D]-VH/VL, [T28D;S31D]-VH, [K50D;S56D]-VL and [T28D;S31D;K50D]-VH/VL, were analyzed in three separate experiments. Relative standard errors of parameters ka and kd were < 2 % in the individual experiments and Chi/Rmax < 0.03 RU. Full data set can be found in Figure S3 in Supporting Information. dData is presented as mean ± σ (n = 3) determined from SE-UPLC analyses after incubation at 45 ºC for six days. Not available is indicated by N/A. Ranka

Protein variant

Biacore SPRc

NanoDSF

pIb


14

Page 15 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Antigenic Binding Affinity Affinity and kinetic properties of the antigenic binding were evaluated by SPR. All variants retained the affinity within a factor of two relative to WT (KD = 17 nM), with the exception of the Fabs [T28D;S31D]-VH and [T28D;S31D;S56D]-VH/VL, which resulted in a 2.4-fold improvement in affinity (KD = 7 nM) (see Table 1 and Figure 2A). In summary, the investigated single-, double-, and triple-mutations did not have any substantial impact on the affinity to the target protein. This may be dependent on the epitope and paratope interactions and may differ for other protein interaction pairs.


15

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 33

Figure 2. Antigen binding properties from Biacore SPR experiments. (A) A graphical overview of the association rate constants (ka) versus dissociation rate constants (kd) is shown for the Fab variants. A scale for the dissociation equilibrium constant (KD) is illustrated as black dashed lines, whereas the variants are illustrated as grey dots. Variants with highest and lowest obtained affinity relative to WT (blue dot) are highlighted as green ([T28D;S31D]-VH) and red ([K50D;S56D]-VL) dots, respectively. Their corresponding sensorgrams are visualized in (C)(D) at two target protein concentrations (40 and 20 nM). The experimentally obtained data is shown as black curves, whereas the adapted fits are shown as blue (WT), green ([T28D;S31D]VH) and red ([K50D;S56D]-VL) curves.


16

Page 17 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Thermodynamic Stability To investigate whether the designed variants retained the thermodynamic stability relative to WT, the thermodynamic stability was assessed by the thermal stability. 15 out of the 26 produced Fab variants had a retained or even improved Tm relative to WT (0.5-1.8 ºC, see Table 1), which is in contrast to the study reported by Dudgeon et al.29 as no improvement in thermodynamic stability was observed. The mutations responsible for the observed increase in Tm stability were identified as [T28D]-VH, [S31D]-VH, [N53D]-VL and [S56D]-VL. The 11 remaining Fab variants exhibited decreased thermal stability by up to 4.6 ºC lower Tm relative to WT. The mutations responsible for this decrease in Tm were identified as [A30D]-VH and [K50D]-VL. For all double- and triple- substitutions, additive effects could be observed for Tm. The variants with highest obtained Tm consisted of double- and triple-substitutions of the singlesubstitutions identified as most stable due to their additive effect on Tm (Table 1). Furthermore, the selected FR2 mutation, [K45D]-VL, was tested in combination with [S31D]-VH despite not being selected in the initial in silico screen by Rosetta. The affinity was retained within a factor of 2, whereas the thermodynamic stability in terms of Tm was improved by 0.5 ºC relative to WT (see Table 1). A Fab variant with [F46D]-VL substitution was produced as a control. The mutation was manually selected based on the homology model of the Fab. For this variant, a decrease in Tm by 5 ºC was observed while no change in pI was observed relative to WT, thereby verifying that the substitution was unfavorable most likely due to it being buried in the interior (VH-VL interface) of the Fab (see Table S3 in Supporting Information). Overall, all variants showed an acceptable thermal stability compared to WT, suggesting that it is valuable to perform an in silico pre-screen of a large number of mutations prior to in vitro screening.


17

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 33

Non-Specific Binding Properties An important attribute for the developability of the Fab is low non-specific binding11. This was investigated by assessing the retention of the variants on cross-interaction chromatography (CIC) and SE-UPLC*. All variants exhibited a shift towards lower CIC and SE-UPLC* retention relative to WT, with an exception of [K50D]-VL, clearly demonstrating that the introduced Asp substitutions reduced non-specific binding (Figure 3A). Interestingly, the SE-UPLC* retention decreased with increasing number of Asp substitutions (Figure 3B), however this was not evident based on the CIC retentions (Figure 3C). The discrepancy between CIC and SE-UPLC* is in good agreement with a previous report by Jain et al.11, where the two non-specificity assays were found to give different rankings of a mAb library. Moreover, lowest non-specific binding, i.e. lowest CIC and SE-UPLC* retentions, was obtained for the variants with triple Asp substitutions (Figure 3A). This observed effect of non-specific binding is in line with a recent study reported by Rabia et al.34, showing that Asp residues in the CDRs played a major role in reducing nonspecificity of antibodies.


18

Page 19 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Figure 3. Non-specific binding properties. (A) CIC retention versus SE-UPLC* retention. (B) SE-UPLC* retention versus number of Asp substitutions and (C) CIC retention versus number of Asp substitutions. Variants with single, double and triple Asp substitutions are shown as crosses, asterisks and triangles, respectively, whereas WT is shown as the circle. The direction of reduced non-specificity is indicated by the blue arrows.


19

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 33

Aggregation Propensity In the study by Dudgeon et al.29, the effect of the Asp mutations on aggregation propensity was assessed after heating to 80 °C, where the molecules were primarily unfolded and therefore following non-native state aggregation pathways. Aggregation is often measured at elevated temperatures (60-95 °C) which are above or close to the Tm of the tested molecules26,27,28. To investigate the effects of these mutations at conditions that are more relevant for bioprocessing and formulation, aggregation was evaluated at a temperature below the onset of the thermal unfolding transition to ensure that the protein molecules are primarily in the folded native state. In our study, an incubation temperature of 45 °C was selected as this was well below the Tm of the variants and still high enough to accelerate aggregation such that pronounced amounts of HMWP was formed within short time (Figure S2). After six days of incubation at 45 °C, the HMWP levels were measured with SE-UPLC, and the variants were ranked by the number of standard deviations from WT (Table 1). A total of 13 different variants resulted in improved aggregation resistance by up to 1.4 %-units lower HMWP level relative to WT (see rank 1 in Figure 4 and Table 1). The mutations that exhibited these stabilizing effects were identified as [T28D]-FR1-H, [S31D]-CDR1-H and [N53D]-CDR2-L, both as single mutations and combined in both VH and VL. Moreover, a total of four different variants were on-par with WT in terms of HMWP formation (see rank 2 in Figure 4 and Table 1). These mutations were identified as [A30D]-CDR1-H, [S52D]-CDR2-L, [S56D]-CDR2-L and [S31D;K45D]-CDR1-H/FR2-L. Furthermore, a total of 10 different mutants resulted in a substantially increased aggregation propensity by up to 10 %-units higher HMWP level relative to WT (see rank 3 in Figure 4 and Table 1). The mutations responsible for this elevated aggregation propensity were identified as


20

Page 21 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

[K50D]-CDR2-L and [K45D]-FR2-L, both as single mutations and combined with other mutations. Although the investigated Asp mutations were found to decrease the aggregation propensity of human variable domains29, some of these mutations radically increased the aggregation propensity of our Fab. This discrepancy is likely due to two factors; (1) different WT molecules are used as starting point of mutagenesis in these two studies and it is likely that the insertion of Asp mutations will have varying effects on molecules with different spatial distribution of hydrophobic and charged properties, and (2) different aggregation pathways.


21

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 33

Figure 4. Aggregation propensity. The percentage of HMWP formed after incubation at 45 °C for six days is shown by the bar chart. The different Asp variants are grouped according to their ranking (see Table 1); (1) improved, (2) on-par and (3) worse aggregation propensity relative to WT. The standard deviation (n = 3) is shown by the error bars. The red dashed lines represent the upper and lower limits of the standard deviation for Fab WT.


22

Page 23 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Next, the stability storage data was compared with the selected developability assays using Spearman correlation statistics, which takes the monotonicity of the data into account rather than assuming linearity of a reaction. Interestingly, decreased aggregation propensity correlated well with increased thermodynamic stability (Tm) (Spearman correlation coefficient, SCC = - 0.87, p < 0.001, Figure S4-5 in Supporting Information), meaning that for this set of Fab variants, optimization of Tm is a strategy to improve the aggregation resistance (see Figure 5). The remarkable correlation between aggregation propensity and Tm determination observed for this variant set does not appear to apply generally as no such correlation was observed when a set of 137 different therapeutic antibodies were evaluated

in a nearly similar study design11. In

addition, aggregation correlated with lower CIC retentions (SCC = 0.64, p < 0.001), which has previously been suggested to be related with protein self-association and solubility32,35. Furthermore, the aggregation was compared from the storage stability study with the in silico SAP tool, which has been proposed to predict aggregation prone regions36,37,38. For this model system no significant correlation could be found (SCC = 0, p > 0.001). Interestingly, SAP correlated remarkably well with SE-UPLC* retention (SCC = 0.94, p < 0.001), which is in agreement with previous findings32.


23

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 33

Figure 5. Influence of Tm on aggregation propensity. The percentage of HMWP formed after incubation at 45 °C for six days is shown as a function of Tm. The Fab variants are shown as circles, whereas the linear adaption of shown as the dashes line (R2 = 0.64). Application of the Design Strategy Altogether, we show that Asp substitutions in the CDRs and FRs can successfully improve important developability attributes such as non-specificity and aggregation resistance, while retaining thermodynamic stability and antigenic binding affinity. Although the Asp mutations unambiguously reduced non-specific interactions, only half of the variants exhibited reduced aggregation propensity. This highlights that several design cycles, including variant design and experimental analysis of aggregation, might be needed for successful engineering of aggregation resistance. In addition, we show that the number of variants for in vitro assessment can be considerably reduced by performing an in silico screen based on computational models. One aspect to consider when introducing Asp residues is the potential risk of isomerization. The


24

Page 25 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

effect of isomerization may affect biological function and in vitro biophysical stability39,40. However, isomerization of Asp residues is dependent on the sequence position as hydrogen bonding of the Asp side-chain has shown to prevent this41. A way to avoid potential isomerization is to substitute residues into Glu instead of Asp. Dudgeon et al.29 showed that both Asp and Glu substitutions could improve the biophysical properties of human variable domains, although they observed a detectable preference towards Asp residues. These findings show that negative charges have important implications in the aggregation propensity of antibodies. However, it still remains to be investigated whether such design strategies can be applied generically to improve the developability of a broad range of different antibodies and their fragments.

Conclusions In this study, we show that by structurally screening an in silico library of 393 Asp substitutions (single, double and triple) in the FRs and CDRs of a murine Fab, 26 different mutations could be pre-selected for in vitro screening with minimal effect on overall thermodynamic stability (Tm). All variants resulted in fully retained or improved affinity relative to Fab WT. Based on the six-day storage stability study at 45 °C, 13 of the Fab variants exhibited improved aggregation resistance relative to WT. These improved variants were characterized with 1.1 °C higher Tm in average as well as reduced CIC and SE-UPLC* retentions relative to WT. Moreover, four of the other variants were aggregating on-par with WT, whereas the remaining 10 variants exhibited elevated aggregation propensity due to [K50D]-VL and [K45D]VL. All Asp substitutions resulted in substantially reduced non-specific binding, which is important for the developability potential of biopharmaceuticals. Altogether, the results


25

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 33

presented here demonstrate how the introduction of negative charges improved the developability potential of the Fab fragment.


26

Page 27 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

ASSOCIATED CONTENT The following file is available free of charge. Supporting Information (PDF) AUTHOR INFORMATION Corresponding Authors *Laila I. Sakhnini ([email protected]); Anja. K. Pedersen ([email protected]); Maria Dainiak ([email protected]). Funding Sources Financial support from the Novo Nordisk STAR office and the Danish Innovation Fund (grant number 5016-00127B) is acknowledged. ACKNOWLEDGMENT Lone Elisabeth Christiansen and Anne Lee Andersen from Novo Nordisk A/S are gratefully acknowledged for their help involving the molecular biology work. ABBREVIATIONS Asp, aspartate; BSA, bovine serum albumin; CDRs, complementarity determining regions; CIC, cross-interaction chromatography; CIEX, cation-exchange chromatography; Fab, antigenbinding fragment; FR, framework; mAb, monoclonal antibody; HMWP, high molecular weight protein; IEF, isoelectric focusing; pI, isoelectric point; SAP, spatial aggregation propensity; SCC, Spearman correlation coefficient; SE-UPLC, size-exclusion ultra-performance liquid chromatography; scFv, single-chain variable fragment; Tm, temperature of midpoint of thermal unfolding; VH, heavy variable chain; VL, light variable chain; WT, wild-type


27

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 33

References

(1) Friedman, M., Ståhl, S. (2009) Engineering affinity proteins for tumour-targeting applications. Biotechnol. Appl. Biochem. 53, 1-29. (2) Golay, J., Introna, M. (2012) Mechanism of action of therapeutic monoclonal antibodies: Promises and pitfalls of in vitro and in vivo assays. Arch. Biochem. Biophys. 526, 146-153. (3) DeNardo, S.J. (2005) Radioimmunodetection and therapy of breast cancer. Semin. Nucl. Med. 35, 143-151. (4) Wu, A.M. and Olafsen, T. (2008) Antibodies for molecular imaging of cancer. Cancer J. 14, 191-197. (5) Saerens, D., Huang, L., Bonroy K. and Muyldermans, S. (2008) Antibody fragments as probe in biosensor development. Sensors 8, 4669-4686. (6) Sakhnini, L.I., Pedersen, A.K., Ahmadian, H., Hansen, J-J., Bülow, L. and Dainiak, M. (2016) Designing monoclonal antibody fragment-based affinity resins with high binding capacity by thiol-directed immobilisation and optimisation of pore/ligand size ratio. J. Chromatogr. A 1468, 143-153. (7) Chames, P.,Van Regenmortel, M., Weiss, E., Baty, D. (2009) Therapeutic antibodies: successes, limitations and hopes for the future. Br. J. Pharmacol. 157, 220-233. (8) Wu, A.M., Senter, P.D. (2005) Arming antibodies: prospects and challenges for immunoconjugates. Nat. Biotechnol. 23, 1137-1146.


28

Page 29 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

(9) Holliger, P., Hudson, P.J. (2005) Engineering antibody fragments and the rise of single domains. Nat. Biotechnol. 23, 1126-1136. (10) Cuesta, Á., M., Sainz-Pastor, N., Bonet, J., Oliva, B., Álvarez-Vallina, L. (2010) Multivalent antibodies: when design surpasses evolution. Trends Biotechnol.28, 355-362. (11) Jain, T., Sun, T., Durand, S., Hall, A., Rewa Houston, N., Nett, J.H., Sharkey, B., Bobrowicz, B., Caffry, I., Yu, Y., Cao, Y., Lynaugh, H., Brown, M., Baruah, H., Gray, L.T., Krauland, E.M., Xu, Y., Vásquez, M., Wittrup, K.D. (2017) Biophysical properties of the clinical-stage antibody landscape. Proc. Natl. Acad. Sci. USA 114, 944-949. (12) Jarasch, A., Koll, H., Regula, J.T., Bader, M., Papadimitriou, A., Kettenberger, H. (2015) Developability Assessment During the Selection of Novel Therapeutic Antibodies. J. Pharm. Sci. 104, 1885-1898. (13)

Ponsel, D., Neugebauer, J., Ladetzki-Baehs, K., Tissot, K. (2011) High Affinity,

Developability and Functional Size: The Holy Grail of Combinatorial Antibody Library Generation. Molecules 16, 3675-3700. (14)

Vázquez-Rey, M., Lang, D.A. (2011) Aggregates in Monoclonal Antibody

Manufacturing Processes. Biotechnol. Bioeng. 108, 1494-1508. (15) Hermeling, S., Crommelin, D.J.A., Schellekens, H., Jiskoot, W. (2004) StructureImmunogenicity Relationships of Therapeutic Proteins. Pharm. Res. 21, 897-903. (16) Demarest, S.J. and Glaser, S.M. (2012) Antibody therapeutics, antibody engineering, and the merits of protein stability. Curr. Opin. Drug Discov. Devel. 11, 675-687.


29

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 33

(17) Gil, D., Schrum, A.G. (2013) Strategies to stabilize compact folding and minimize aggregation of antibody-based fragments. Adv. Biosci. Biotechnol. 4, 73-84. (18) Wang, W., Nema, S. and Teagarden, D. (2010) Protein aggregation ─ Pathways and influencing factors. Int. J. Pharm. 390, 89-99. (19) Chi, E.Y., Krishnan, S., Randolph, T.W. and Carpenter, J.F. (2003) Physical stability of proteins in aqueous solution: mechanism and driving forces in nonnative protein aggregation. Pharmaceut. Res. 20, 1325-1336. (20) Pechmann, S., Levy, E.D., Gaetano Tartaglia, G., Vendruscolo, M. (2009) Physicochemical principles that regulate the competition between functional and dysfunctional association of proteins. Proc. Natl. Acad. Sci. USA 106, 10159-10164. (21) Rouet, R., Lowe, D. and Christ, D. (2014) Stability engineering of the human antibody repertoire. FEBS Lett. 588, 269-277. (22) Wang, X., Das, T.K., Singh, S.K. and Kumar, S. (2009) Potential aggregation prone regions in biotherapeutics. MAbs 1, 254-267. (23) Lowe, D., Dudgeon, K., Rouet, R., Schofield, P., Jermutus, L. and Christ, D. (2011) Aggregation, stability, and formulation of human antibody therapeutics. Adv. Protein Chem. Struct. Biol. 84, 41-61. (24) Geng, S.B., Cheung, J.K., Narasimhan, C., Shameem, M. and Tessier, P.M. (2014) Improving Monoclonal Antibody Selection and Engineering using Measurements of Colloidal Protein Interactions. J. Pharm. Sci. 103, 3356-3363.


30

Page 31 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

(25) Jiang, L., Cao, S., Cheung, P.P.H., Zheng, X., Leung, C.W.T., Peng, Q., Shuai, Z., Tang, B.Z., Yao, S., Huang, X. (2017) Real-time monitoring of hydrophobic aggregation reveals a critical role of cooperativity in hydrophobic effect. Nat. Commun. 8, 15639. (26) Miklos, A.E., Kluwe, C., Der, B.S., Pai, S., Sircar, A., Hughes, R.A., Berrondo, M., Xu, J., Codrea, V., Buckley, P.E., Calm, A.M., Welsh, H.S., Warner, C.R., Zacharko, M.A., Carney, J.P., Gray, J.J., Georgiou, G., Kuhlman, B. and Ellington, A.D. (2012) Structure-Based Design of Supercharged, Highly Thermoresistant Antibodies. Chem. Biol. 19, 449-455. (27) Perchiacca, J.M., Lee, C.C. and Tessier, P.M. (2014) Optimal charged mutations in the complementarity determining regions that prevent domain antibody aggregation are dependent on the antibody scaffold. Protein Eng. Des. Sel. 27, 29-39. (28) Perchiacca, J.M., Bhattacharya, M. and Tessier, P.M. (2011) Mutational analysis of domain antibodies reveals aggregation hotspots within and near the complementarity determining regions. Proteins 79, 2637-2647. (29) Dudgeon, K., Rouet, R., Kokmeijer, I., Schofield, P., Stolp, J., Langley, D., Stock, D. and Christ, D. (2012) General strategy for the generation of human antibody variable domains with increased aggregation resistance. Proc. Natl. Acad. Sci. USA 109, 10879-10884. (30) Park, P. Bradley, P., Greisen P.J., Liu, Y., Khipple Mulligan, V., Kim, D.E., Baker, D., DiMaio, F. (2016) Simultaneous optimization of biomolecular energy function on features from small molecules and macromolecules. J. Chem. Theory Comput. 12, 6201-6212.


31

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 33

(31) Kabat, E., Wu, T.T., Perry, H.M., Kay, S., Gottesman, C.F. (1992) Sequences of Proteins of Immunological Interest, 5th ed., pp 2719, Diane Publishing, Pennsylvania. (32) Wolf Pérez, A.M., Sormanni P., Sonne Andersen, J., Sakhnini, L.I., Rodriguez-Leon, I., Rose Bjelke, J., Juhl Gajhede, A., De Maria, L., Otzen, D.E., Vendruscolo, M., Lorenzen, N. (2019) In vitro and in silico assessment of the developability of a designed monoclonal antibody library. mAbs 11, 388-400. (33) Chennamsetty, N., Voynov, V., Kayser, V., Helk, B., Trout, B.L. (2010) Prediction of aggregation prone regions of therapeutic proteins. J. Phys. Chem. B 114, 6614-6624. (34) Rabia, L.A., Zhang, Y., Ludwig, S.D., Julian, M.C., Tessier, P.M. (2019) Net charge of antibody complementarity-determining regions is a key predictor of specificity. Protein Eng. Des. Sel., DOI:10.1093/protein/gzz002. (35) Jacobs, S.A, Wu, S.J., Feng, Y., Bethea, D., O’Neil, K.T. (2010) Cross-interaction chromatography: a rapid method to identify highly soluble monoclonal antibody candidates. Pharm. Res. 27, 68-71. (36) Chennamsetty, N., Helk, B., Voynov, V., Kayser, V. and Trout, B.L. (2009) Aggregationprone motifs in human immunoglobulin G. J. Mol. Biol. 391, 404-414. (37) Chennamsetty, N., Voynov, V., Kayser, V., Helk, B. and Trout, B.L. (2009) Design of therapeutic proteins with enhanced stability. Proc. Natl. Acad. Sci. 106, 11937-11942. (38) Chennamsetty, N., Voynov, V., Kayser, V., Helk, B. and Trout, B.L. (2010) Prediction of aggregation prone regions of therapeutic proteins. J. Phys. Chem. B 114, 6614-6624.


32

Page 33 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

(39) Paborji, M., Pochopin, N.L., Coppola, W.P., Bogardus, J.B. (1994) Chemical and Physical Stability of Chimeric L6, a Mouse−Human Monoclonal Antibody. Pharm. Res. 11, 764-771. (40) Kroon, D.J., Baldwin-Ferro, A., Lalan, P. (1992) Identification of Sites of Degradation in a Therapeutic Monoclonal Antibody by Peptide Mapping. Pharm. Res. 9, 1386–1393. (41) Cacia, J., Keck, R., Presta, L.G., Frenz, J. (1996) Isomerization of an Aspartic Acid Residue in the Complementarity-Determining Regions of a Recombinant Antibody to Human IgE: Identification and Effect on Binding Affinity. Biochemistry 35, 1897-1903.

For Table of Contents Use Only


33

Improving the Developability of an Antigen Binding Fragment by

Recommend Documents